A cyclopeptide-derived molecular cage for sulfate ions that closes with a click

Article abstract of DOI:10.1002/chem.201000308

The 2:1 sandwich-type complexes formed between a cyclopeptide with alternating L-proline and 6-aminopicolinic acid subunits and inorganic anions can be stabilized by covalently linking a tris-alkyne and a tris-azide derivative of this peptide through coppercatalyzed azide-alkyne cycloaddition. The resulting triply linked bis-cyclopeptide can interact with anions such as sulfate ions in aqueous solution by including them into the cavity between the two cyclopeptide rings, where they can form hydrogen bonds to amide NH groups, distributed along the inner surface. The binding kinetics of this system differ significantly from those of a bis-cyclopeptide that contains only one linker because the rate of guest exchange is considerably slower. Thermodynamically, the stability of the sulfate complex of the triply linked bis-cyclopeptide approaches a log K_a value of 6 in H₂CVCH₃OH 1:1 (v/v) which is, however, only approximately one order of magnitude larger than affinity of the more flexible monolinked analogue. Titration calorimetry revealed that this behavior is mainly due to the change in the binding enthalpy from exothermic to endothermic upon increasing the number of linkers. Results from NMR spectroscopy and X-ray crystallography indicate that the mono- and triply linked bis-cyclopeptides adopt similar conformations in their complexes with sulfate ions, but the complex formation is enthalpically unfavorable for the cage. The substantial entropie contribution to sulfate complexation of this receptor more than compensates for this disadvantage, so that the overall sulfate affinity of both bis-cyclopeptides ends up in the same range. These investigations provide important insight into the structure-property relationships of such receptors, thus leading the way to further structural improvement.

Full text of DOI:10.1002/chem.201000308

FULL PAPER
DOI: 10.1002/chem.201000308
A Cyclopeptide-Derived Molecular Cage for Sulfate Ions That Closes with a
Click
Thomas Fiehn,[a] Richard Goddard,[b] Rꢀdiger W. Seidel,[c] and Stefan Kubik*[a]
Dedicated to Professor Gꢀnter Wulff on the occasion of his 75th birthday
Abstract: The 2:1 sandwich-type com-
plexes formed between a cyclopeptide
with alternating l-proline and 6-amino-
picolinic acid subunits and inorganic
anions can be stabilized by covalently
linking a tris-alkyne and a tris-azide de-
rivative of this peptide through copper-
catalyzed azide–alkyne cycloaddition.
The resulting triply linked bis-cyclo-
peptide can interact with anions such
as sulfate ions in aqueous solution by
including them into the cavity between
the two cyclopeptide rings, where they
can form hydrogen bonds to amide NH
groups, distributed along the inner sur-
face. The binding kinetics of this
system differ significantly from those
of a bis-cyclopeptide that contains only
one linker because the rate of guest ex-
change is considerably slower. Thermo-
dynamically, the stability of the sulfate
complex of the triply linked bis-cyclo-
peptide approaches a logKa value of 6
in H2O/CH3OH 1:1 (v/v) which is, how-
ever, only approximately one order of
magnitude larger than affinity of the
more flexible monolinked analogue. Ti-
tration calorimetry revealed that this
behavior is mainly due to the change in
the binding enthalpy from exothermic
to endothermic upon increasing the
number of linkers. Results from NMR
spectroscopy and X-ray crystallography
indicate that the mono- and triply
linked bis-cyclopeptides adopt similar
conformations in their complexes with
sulfate ions, but the complex formation
is enthalpically unfavorable for the
cage. The substantial entropic contribu-
tion to sulfate complexation of this re-
ceptor more than compensates for this
disadvantage, so that the overall sulfate
affinity of both bis-cyclopeptides ends
up in the same range. These investiga-
tions provide important insight into the
structure–property relationships of
such receptors, thus leading the way to
further structural improvement.
Keywords: anions · calorimetry ·
cycloaddition · molecular contain-
ers · supramolecular chemistry
Introduction
circling the substrate, or by completely encapsulating it
inside a three-dimensional cavity. The latter probably most
closely resembles binding inside the active center of a pro-
tein. Early examples of synthetic receptors that encapsulate
their substrate are the cryptands developed by Lehn[1a] and
subsequently the carcerands developed by Cram[1b] and the
cryptophanes developed by Collet et al.[1c]
A carcerand has a closed surface that completely sur-
rounds the guest molecule, thus preventing every possible
way of escape.[1b] Although carcerand complexes (or carce-
plexes) are fascinating systems that still attract attention, for
example, in the form of endohedral fullerenes,[2] they lack
reversibility in guest exchange, thus precluding an evalua-
tion of thermodynamic stability or binding selectivity. Open-
ing a small portal in the surface of a carcerand furnishes a
hemicarcerand, the complexes of which must be, by defini-
tion, sufficiently kinetically inert at room temperature to be
isolable.[3] Receptors that exhibit reversible binding equili-
bria at ambient temperature can be obtained by increasing
A synthetic receptor can achieve substrate binding by hold-
ing onto the substrate at only few selected positions, by en-
[a] Dipl.-Chem. T. Fiehn, Prof. Dr. S. Kubik
Fachbereich Chemie–Organische Chemie
Technische Universitꢀt Kaiserslautern
Erwin-Schrçdinger-Strasse, 67663 Kaiserslautern (Germany)
Fax : (+49)631-205-3921
E-mail: kubik@chemie.uni-kl.de
[b] Dr. R. Goddard
Max-Planck-Institut fꢁr Kohlenforschung
Kaiser-Wilhelm-Platz 1
45470 Mꢁlheim/Ruhr (Germany)
[c] Dipl.-Chem. R. W. Seidel
Lehrstuhl fꢁr Analytische Chemie
Ruhr-Universitꢀt Bochum
Universitꢀtsstrasse 150, 44780 Bochum (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201000308.
Chem. Eur. J. 2010, 16, 7241 – 7255
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7241

the number and/or the size of the portals or by introducing
a gating mechanism that causes opening and closing of
larger portals on the receptor surface. Terms such as molec-
ular containers, cages, or capsules are used for such recep-
tors for which the complexation/decomplexation rates are
often significantly slower than those of receptors with more
exposed cavities.[4a,b] These unique properties and others
have made molecular containers attractive research objects
in molecular-recognition studies.[4c] Molecular containers
have, for example, been shown to stabilize reactive interme-
diates,[5a,b] act as nanoreactors,[5c] induce guest molecules to
adopt conformations unstable outside the container cavi-
ty[5d–h] or give rise to unusual forms of isomerism if the mo-
bility of the guest is impaired inside the cavity and/or more
than one guest molecule is bound.[5i]
The construction of intricate systems such as molecular
containers requires special synthetic approaches that often
rely on the presence of molecules in the reaction mixture to
act as templates. To assemble containers through covalent ir-
reversible reactions, for example, the template must preor-
ganize the container subunits in a fashion that allows them
to be linked intermolecularly—a strategy used for the prepa-
ration of carcerands, hemicarcerands, and cryptopha-
nes.[1b–c,3] In contrast to this kinetic approach, thermodynam-
ic templation requires reversible interactions between the
container building blocks and is based on the template-in-
duced stabilization of the desired product.[6] Reversible reac-
tions used for the latter approach include boronic ester for-
mation,[7a] imine exchange,[7b–d] disulfide exchange,[7e] metal
coordination,[5b,7f–h] or a combi-
ing sites located inside the cavity as in the active centers of
enzymes are, however, relatively rare. Notable examples of
containers with functionalized interiors are two independ-
ently described coordination cages for sulfate ions,[9a,b] an in-
terlocked sulfate-binding capsule,[9c] self-assembled phenyl-
alanine-substituted resorcarene dimers,[9d–f] and a few other
systems.[9g–i] In addition, bicyclic and tricyclic ammonium- or
amide-based anion receptors possess cavities lined by prop-
erly arranged hydrogen-bond donors.[9j,k]
A promising building block, recently described by us, for
the construction of container molecules with internal bind-
ing sites is cyclic hexapeptide 1.[10a] This peptide has been
shown to bind inorganic anions such as sulfate or halide ions
in the form of 2:1 complexes in competitive aqueous solvent
mixtures in which the anion is sandwiched between two
cyclopeptide rings and is engaged in hydrogen-bonding in-
teractions with six NH groups that point into the cavity. Co-
valent linkage of two cyclopeptide rings through one linker
has furnished bis-cyclopeptides that bind to anions in a 1:1
ratio with considerably higher efficiency than the monotopic
analogue 1.[10b–e] A logical extension of this approach is the
introduction of three linkers between the cyclopeptide rings,
which should give rise to anion-binding molecular contain-
ers. Questions that can be raised in this context are how the
increase in the number of linkages influences the binding
properties and whether it is possible to construct carcerands
with permanently entrapped anions based on such systems.
Connecting two appropriately functionalized analogues of
1 through three linkers can in principle be realized by using
nation of different covalent re-
actions.[7i,j] Noncovalent interac-
tions, such as ion-pairing,[7k–m]
hydrogen-bonding,[5i,7n,o] or hy-
drophobic interactions, have
also been used.[7p]
Although molecular contain-
ers represent a large and struc-
turally diverse family of syn-
thetic receptors, many are syn-
thesized from appropriately
functionalized aromatic build-
ing blocks, thus rendering their
interiors relatively nonpolar.
Substrate binding often has an
electrostatic component, such
as cation–p interactions be-
tween a positively charged sub-
strate and the aromatic cavity
walls,[8a] or electrostatic interac-
tions of the polar or charged
substrate with polar regions
along the container portals[8b]
or with metal cations used for
the assembly of the cage.[8c] Di-
rected attractive interactions
between the substrate and bind-
7242
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7241 – 7255

Cage Receptor for Sulfate Anions
FULL PAPER
irreversible or reversible reactions in conjunction with the
template effects of anions. Both approaches are currently
being pursued in our group. Herein, we present the first bis-
cyclopeptide 2a obtained in this context, which was pre-
pared by using a copper-catalyzed 1,3-dipolar cycloaddition
to connect tris-alkyne 9a and tris-azide 9b.[11a] This reaction,
often referred to as a click reaction,[11b–c] has been shown to
be immensely useful for a variety of purposes. However, to
the best of our knowledge it has only rarely been applied to
the synthesis of molecular containers or cages so far.[9c,12]
pared analogously to a known procedure.[14] This acid was
coupled with allyl 6-aminopicolinate (4), synthesized from 6-
aminopicolinic acid and allyl bromide, by using chlorotripyr-
rolidinophosphonium hexafluorophosphate (PyCloP) as the
coupling agent (Scheme 1). The resulting dipeptide 5 was
used to assemble the cyclic hexamer 6a by following estab-
lished protocols.[15] Hydrogenation of 6a in the presence of
hydrochloric acid gave the trihydrochloride salt of 6b, which
could be converted into the trisubstituted derivatives 9a and
9b by treatment with the pentafluorophenol esters of 4-pen-
tynoic acid (7) or 3-azidopropanoic acid (8), respectively,
the latter being obtained by the addition of hydrogen azide
to acrylic acid.
The monosubstituted analogues 11a and 11b were ob-
tained in a similar manner from the hydrochloride salt of
10b. The Z-protected precursor of 10b, namely, cyclopep-
tide 10a, was assembled from dipeptide 5 and two equiva-
lents of an unsubstituted dipeptide, the synthesis of which
has been described previously.[15]
The copper(I)-catalyzed azido–alkyne cycloaddition to
give 1,4-disubstituted 1,2,3-triazoles is widely used in organic
synthesis, but optimization of the reaction conditions is usu-
ally necessary.[16] Our attempts to identify suitable reaction
conditions initially concentrated on the monofunctionalized
cyclopeptides 11a and 11b, the coupling of which should be
straightforward because it does not strictly require the tem-
plate effect of anions to bring the two cyclopeptide subunits
together. Two aspects important for the subsequent synthe-
sis of the desired cage 2a had to be considered: First, reac-
tions should be carried out in aqueous solvent mixtures,
which favor the anion-promoted self-assembly of two cyclo-
peptide rings, and second they should proceed at or below
room temperature to make sure that most binding partners
are involved in complex formation.
Our results indicate that 2a forms a stable complex with
sulfate ions in H2O/CH3OH mixtures in which the anion is
indeed located inside the cavity between the two cyclopep-
tide rings.[13] Interestingly, the sulfate affinity of 2a is not sig-
nificantly higher than that of analogue 2b, which contains
only one linker despite the improved preorganization that
the three linkers were expected to induce. Characterization
of the thermodynamics of the sulfate complexation of 2a
and 2b with isothermal titration calorimetry (ITC) provided
valuable information about the causes of this unexpected
behavior.
Systematic variation of the conditions showed that by per-
forming the reaction with equimolar amounts of 11a and
11b in H2O/CH3OH (2:3, v/v), two equivalents of copper(II)
sulfate as a source of copper(I) ions and the anionic tem-
plate,[17] sodium ascorbate as a reducing agent, and 2,6-luti-
Results and Discussion
Synthesis and structural charac-
terization: Bis-cyclopeptide 2a
contains two subunits of cyclo-
peptide 6b with alternating 4S-
amino-l-proline and 6-aminopi-
colinic acid subunits. This build-
ing block was assembled prior
to the introduction of the pe-
ripheral functional groups re-
quired for the azide–alkyne cy-
cloaddition. The starting point
of the synthesis of 6b was
(2S,4S)-4-(benzyloxycarbonyl-
ACHTUNGTRENNUNGamino)-1-(tert-butyloxycarbo-
Scheme 1. Synthesis of cyclopeptides 9a and 9b. All=allyl, Boc=tert-butoxycarbonyl, DIEA=diisopropyl-
nyl)proline (3), which was pre-
ACHTUNGTRNEeNUNG thylamine, DMSO=dimethyl sulfoxide, PFP=pentafluorophenol, Z=benzyloxycarbonyl.
Chem. Eur. J. 2010, 16, 7241 – 7255
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7243

S. Kubik et al.
dine as a ligand to stabilize the copper(I) ions and decrease
the risk of coordination of the aminopicolinic acid subunits
in the cyclopeptides to the metal ions leads to complete con-
version after approximately 18 h at room temperature
(Scheme 2). Due to the demanding workup, pure 2b could
only be isolated in 40% yield, which was nevertheless suffi-
cient for the following binding studies.
dence that the structural assignment of 2a is correct comes
from the following results:
1) No azide or alkyne band can be detected in the FTIR
spectrum of the isolated product.
2) A characteristic fragmentation of cyclopeptide azides,
such as 9b, upon collision-induced decay of the molecu-
lar ion in the ESI-MS/MS spectrum corresponds to loss
of nitrogen atoms. Corresponding fragments were absent
in the spectrum of 2a.
3) The relatively simple 1H and 13C NMR spectra of the
product are consistent with a C3 symmetrical structure
containing three intact linkers (the two cyclopeptide
rings are indistinguishable in the NMR spectrum because
directionality of the relatively remote triazole units in
the linkers does not translate into measurable differences
in the chemical shifts of the cyclopeptide protons).
4) Finally, the molecular structure of 2a, determined from
single crystals obtained by slow evaporation of solutions
of 2a in H2O/CH3OH or H2O/CH3CN with and without
added Na2SO4, clearly demonstrated that all three linkers
of the cage are fully formed, and that the three triazole
moieties exhibit the expected 1,4-disubstitution pattern.
The result of the crystallographic analysis is depicted in
Figure 1.
Scheme 2. Syntheses of bis-cyclopeptides 2a and 2b.
In the presence of catalytic amounts of CuSO4 or two
equivalents, as used for the coupling of 11a and 11b, the re-
action between 9a and 9b proved to be very slow and was
still incomplete after approximately one day. Because per-
forming the reaction at higher temperature was undesirable,
the reaction rate was accelerated by increasing the amount
of reagents. Indeed, in the presence of 20 equivalents of
CuSO4 and the corresponding amounts of sodium ascorbate
and 2,6-lutidine, complete disappearance of the starting ma-
terials was observed after 18 h. Moreover, MALDI-TOF
mass spectrometric analysis of the reaction mixture indicat-
ed the formation of a product with the m/z ratio expected
for 2a, but an insoluble material was also formed under
these conditions, which most probably represents a cross-
linked polymer containing both of the cyclopeptide building
blocks. To suppress uncontrolled polymerization, the reac-
tion was carried out under higher dilution conditions and
cyclopeptide 9a was added very slowly (over the course of
8 h) to a solution containing 9b and the other reagents
(Scheme 2). The formation of a precipitate was much less
pronounced under these conditions and HPLC analysis of
the reaction mixture indicated the formation of a soluble
product with the m/z ratio expected for 2a. This compound
was isolated after precipitation of the sulfate ions from the
reaction mixture as BaSO4 and two chromatographic steps,
of which the final step was semipreparative HPLC, to give
the analytically pure form in 28% yield.
Figure 1 shows that 2a adopts an overall C2 symmetrical
conformation in the crystal with the crystallographic symme-
try axis intersecting the triazole ring of one tightly folded
linker. Because the C2 symmetrical conformation is not
compatible with the three triazole linkers oriented similarly
with respect to one of the cyclopeptide rings, we have to
assume that molecules in which the N1 and C4 atoms of the
triazole rings are interchanged are superimposed in the crys-
tal.
Interestingly, the individual cyclopeptide rings in 2a adopt
an almost ideal C3-symmetrical conformation despite the de-
creased symmetry in the crystal. The two cyclopeptide rings
are tilted by 19(1)8 to one another and not by 08 as expected
for an overall C3-symmetrical conformation of the cage. The
receptor crystallizes with 21 water molecules, which appear
to have a very structured arrangement (Figure 1, bottom).
Indeed, apart from one oxygen atom of one water molecule,
which lies outside the cavity and is disordered over two po-
sitions, all the water molecules, including the hydrogen
atoms, could be located on difference Fourier syntheses,
thus suggesting that the water molecules are tightly bound.
À
All six N H groups of the two cyclopeptides and four C=O
groups in the linkers are involved in hydrogen bonding to
water molecules in the cavity. It is worthy of note that there
À
The fact that the m/z ratios of the signals observed in the
MALDI-TOF mass spectrum of the product obtained from
the coupling of 9a and 9b can be assigned to a bis-cyclopep-
tide containing both building blocks is no sufficient proof
that all three linkages between the cyclopeptide rings are
indeed closed because a bis-cyclopeptide unit with residual
alkyne or azide moieties would have the same mass. Evi-
are no direct N H···O=C hydrogen bonds between mole-
cules of 2a in the crystal, thus indicating that water is an im-
portant factor in stabilizing the crystal.
Qualitative evaluation of the binding of sulfate ions: The
first qualitative information on the binding properties of re-
ceptors 2a and 2b came from ESI mass spectrometric and
7244
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7241 – 7255

Cage Receptor for Sulfate Anions
FULL PAPER
Figure 2. ESI mass spectra (negative mode) of a) 2a (0.01 mm) in the
presence of one equivalent of Na2SO4 and b) 2b (0.01 mm) in the pres-
ence of three equivalents of Na2SO4 in H2O/CH3OH (1:1, v/v).
Figure 1. Molecular structure of 2a·21H2O obtained by slow evaporation
of a solution of this bis-cyclopeptide in H2O/CH3OH (1:1, v/v) in the
presence of one equivalent of Na2SO4. Top: the mutual orientation of the
two cyclopeptide rings and the conformations of the linkers in 2a;
bottom: the arrangement of the 21 water molecules including the corre-
sponding hydrogen-bonding pattern (crystallographic C2 axis is vertical).
1
A characteristic effect of anion binding in the H NMR
spectra of our bis-cyclopeptides is usually the pronounced
downfield shift of approximately Dd=1 ppm of the proline
Ha) signals, which is caused by the spatial proximity of the
corresponding protons to the anion in the complex.[10b–c] Be-
cause the rate of complex formation of most systems studied
so far is fast on the NMR timescale, averaging of the signals
of free and complexed bis-cyclopeptides has usually been
observed.
1H NMR spectroscopic measurements. The sulfate ion was
used as model substrate in these investigations because it is
usually bound by our bis-cyclopeptides with the highest affi-
nity.[10b,c] Figure 2 shows the ESI mass spectra of solutions of
2a and 2b in H2O/CH3OH (1:1, v/v) containing sodium sul-
fate (negative mode). In both spectra, the major peak has
an m/z ratio and isotopic pattern that can be assigned to the
1:1 sulfate complexes of both bis-cyclopeptides, namely,
1
Figure 3 shows a series of H NMR spectra of 2a and 2b
in D2O/CD3OD (1:1, v/v) containing increasing amounts of
sodium sulfate. The strong downfield shift of the Ha signals
upon sulfate complexation is clearly visible, thus indicating
that both receptors behave like other structurally related
bis-cyclopeptides and bind the anion in the space between
the two cyclopeptide rings. Additional changes in the ali-
phatic regions of the 1H NMR spectra of 2a and 2b in
which the remaining proline protons and protons in the ali-
phatic parts of the linkers absorb are attributable to a con-
formational reorganization of the receptors upon complex
2À
2À
2a·SO4 and 2b·SO4 (m/z 1009.9 and 802.7, respectively).
Minor signals are attributable to the chloride complexes of
2a and 2b (2a·ClÀ: m/z 1958.7 and 2b·ClÀ: m/z 1544.6, in
Figure 2a and b, respectively). The peak at m/z 1508.6 in
spectrum (b) can be assigned to the deprotonated receptor.
Both spectra, thus, strongly suggest that the major complex
species of both receptors possess a 1:1 stoichiometry.
Chem. Eur. J. 2010, 16, 7241 – 7255
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7245

S. Kubik et al.
Figure 3. 1H NMR spectra of a) 2a and b) 2b in D2O/CD3OD (1:1, v/v) (2a: c=1.6 mm, 2b: c=2.8 mm) containing varying amounts of Na2SO4. The tria-
zole CH and the proline Ha signals are marked as triangles and circles, respectively.
formation (see the Supporting Information). The signal of
the triazole protons experiences an upfield shift, thus sug-
gesting that these protons are not directly involved in the in-
teraction with the anion in contrast to other triazole-con-
taining anion receptors.[18]
conformations were found within an energy window of
5 kJmolÀ1 above the global minimum. In all of these struc-
tures, the cyclopeptide moieties adopt conformations very
similar to that found in the crystal structure of the free mac-
rocycle.[10a] Moreover, the mutual arrangement of both
cyclopeptide rings allows short hydrogen-bonding interac-
tions with the included sulfate anion with N···O distances
that range between 2.6 and 3.1 ꢃ. The hydrogen-bonding
pattern (indicated as dotted lines in Figure 4) comprises
direct hydrogen bonds between two sulfate oxygen atoms
and one NH group in each cyclopeptide ring. The other two
sulfate oxygen atoms are sandwiched between pairs of NH
groups from both rings to which they form bifurcated hydro-
gen bonds. A molecular dynamics simulation indicated that
the linkers retain considerable flexibility even in the com-
plex, which is consistent with the NMR spectroscopic result
that the triazole signal exhibits a continuous shift in the
NMR spectrum upon complex formation.
Differences in the spectra of 2a and 2b (Figure 3) provide
important information about the kinetics of complex forma-
tion. In the case of 2b, there is a continuous shift of the Ha
signals, which is accompanied by pronounced line broaden-
ing in the intermediate region between the free and fully
complexed receptor. The underlying complexation/decom-
plexation equilibrium is thus still fast on the NMR time-
scale, although it clearly approaches a rate that can be re-
solved with the 400 MHz instrument used to record the
spectra. In contrast, clearly separated Ha signals are ob-
served for the free and complexed 2a, with the signals for
the free receptor decreasing in intensity as the substrate
concentration rises and the signal for the complex concomi-
tantly increasing, thus showing that sulfate exchange is
slower than for 2b. Thus, closing the cage affects complexa-
tion kinetics in a similar fashion as observed for other con-
tainer-type molecules.[4a,b] Interestingly, the triazole signal
exhibits a different behavior as it progressively shifts from
the uncomplexed to the fully complexed form. Therefore, it
appears that complex formation and conformational mobili-
ty of the linkers proceed on different timescales, the latter
being faster, which leads to averaging of the linker protons.
To obtain an idea about the structure of the sulfate com-
plex of 2a, molecular modeling studies based on the known
preferred conformation of the cyclopeptide ring and its
mode of interactions with anions were performed.[10a,d]
These calculations yielded the structure depicted in Figure 4
as the global minimum.[19] Seven structurally, closely related
Thermodynamics of sulfate binding: The sulfate affinity of
2a and 2b was evaluated quantitatively by using isothermal
titration calorimetry (ITC). This method has the advantage
of providing the full thermodynamic signature of complex
formation in one measurement.[20a] Changes in complex sta-
bility induced by varying, for example, the structure of the
receptor, the nature of the substrate, or the solvent can thus
be traced to whether they are enthalpic or entropic in
origin. Because ITC measures the overall heat change of a
solution upon complex formation, it does not enable the ef-
fects that result from direct receptor–substrate interactions
and from, for example, the desolvation of the binding part-
ners to be distinguished. Information in this regard can
sometimes be obtained by using systematic trend analyses
7246
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7241 – 7255

Cage Receptor for Sulfate Anions
FULL PAPER
Table 1. Association constants logKa, Gibbs energies DG, enthalpies DH,
and entropies TDS of binding of Na2SO4 to receptors 2a and 2b in H2O/
CH3OH of varying ratios at 298 K.
2a
2b
H2O:CH3OH 35:65
AHCTUNGTERG(NNUN v/v)
50:50
65:35
0.75
35:65
0.88
50:50
65:35
0.85
n[a]
0.89
(Æ0.01)
6.34
(Æ0.02)
À36.2
(Æ0.1)
13.3
(Æ0.1)
49.5
(Æ0.1)
0.92
T
0.87
T
T
A
A
ACHTUNGTRENNUNG
logKa
DG
G
N
N
N
N
ACHTUNGTRENNUNG
[b]
[b]
[b]
A
]
]
]
E
N
E
E
N
ACHTUNGTRENNUNG
DH
[kJmolÀ1
TDS
[kJmolÀ1
N
N
N
T
T
N
ACHTUNGTRENNUNG
E
G
N
A
A
N
ACHTUNGTRENNUNG
[a] Interaction stoichiometry factor obtained from the ITC titrations.
[b] Standard deviations of at least three independent measurements are
specified in brackets.
and has a favorable entropic contribution.[10b,e] Interestingly,
increasing the water content of the solvent mixture from 35
to 65% has almost no effect on the enthalpy of the complex
formation. The decrease in complex stability observed upon
increasing the water content is therefore only due to the en-
tropic term, which becomes smaller in solutions containing
greater concentrations of water.
A comparison of the sulfate affinity of the triply linked
bis-cyclopeptide 2a with that of the monolinked derivative
2b reveals that affinity improves upon introduction of the
additional linkers. The increase of only approximately one
order of magnitude is, however, disappointingly small when
considering the much better preorganization 2a should have
for anion binding.
Insight into the cause of the unexpected low sulfate affini-
ty of 2a can be obtained from Table 1 by comparing the in-
dividual thermodynamic parameters associated with binding
of 2a or 2b to sulfate ions. The most important difference in
the behavior of these bis-cyclopeptides is that sulfate bind-
ing of 2b is exothermic, whereas it is endothermic for 2a,
thus being enthalpically disfavored. Thus, increasing the
number of linkers between the cyclopeptide rings clearly
causes profound changes in the thermodynamics of sulfate
binding. The strongly favorable entropy of complex forma-
tion observed for 2a is cancelled out, to a large extent, by a
highly unfavorable enthalpic term that causes complex sta-
bility to be only slightly larger than that of 2b.
The question arises as to why the enthalpy of binding
changes so strongly upon increasing the number of linkers
between the cyclopeptide rings. Endothermic sulfate binding
has been previously reported for other receptors and was at-
tributed to the fact that the favorable binding enthalpy asso-
ciated with the direct receptor–substrate interactions cannot
compensate for the high enthalpy required for desolvation
of the sulfate anion,[21] but this explanation does not corre-
late structural parameters of a receptor with the thermody-
namics of binding. Because knowledge of the relevant struc-
tural parameters that influence the overall affinity of 2a is
Figure 4. Calculated conformation of the sulfate complex of 2a according
to a Monte Carlo simulation (top) and its conformational mobility ac-
cording to a subsequent molecular dynamics simulation (bottom). Hydro-
gen bonds are indicated by dotted lines. (Macromodel 9.0 with Maestro
7.0 interface, Schrçdinger, Inc.; MMFF94S force-field with GB/SA water
model, 5000 steps, dynamics: 100 ps, 258C; the bottom picture shows an
overlay of 10 snapshots taken after intervals of 10 ps).
that involve changing one parameter, for example, the struc-
ture of the receptor, and keeping all the other parameters
invariable.[20b–e]
The results of the ITC measurements (Table 1) show that
with a logKa value between 4 and 6 the monolinked bis-
cyclopeptide 2b possesses very high sulfate affinity in the
solvent mixtures studied, albeit an approximately one order
of magnitude lower affinity than the previously described
analogue containing
a 2,2’-(1,3-phenylene)diacetic acid
linker, which binds sulfate in H2O/CH3OH (1:1, v/v) with a
log Ka of 5.97.[10e] As for other bis-cyclopeptides investigated
previously, complex formation in H2O/CH3OH is exothermic
Chem. Eur. J. 2010, 16, 7241 – 7255
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7247

S. Kubik et al.
of great importance for the design of such cage-type recep-
tors, we tried to obtain more detailed insight into the under-
lying structure–property relationship. In the concrete case of
2a, the following assumptions could provide plausible ex-
planations for the endothermic sulfate binding:
nificant differences between the spectra of the free and
complexed 2a these spectra provided no information about
a conformational rearrangement of the linkers upon com-
plex formation. Therefore, NMR spectroscopic analysis indi-
cated that weaker receptor–substrate interactions most
probably do not explain the endothermic reaction observed
during sulfate binding of 2a, but it did not allow us to
derive conclusive evidence about the conformation of the
linkers in the complex.
Additional information about the structural factors that
might influence the anion binding of 2a was obtained from
crystal-structure analyses. Seven crystals of 2a, obtained
from both salt-free and solutions containing Na2SO4, were
investigated, and in each case the salt-free form was ob-
tained, thus indicating that the hydrated receptor forms
more stable crystals (Figure 1). It is worth noting that the
larger and better formed crystals were found in the samples
containing Na2SO4, and this outcome may be a result of salt-
ing out.[22] Because there are no direct intermolecular hydro-
gen bonds between the receptor molecules in the crystals of
2a·21H2O, all the hydrogen-bonding interactions of the re-
ceptor are with solute water, and the arrangement indicates
that the water in the cavity is strongly stabilized by coopera-
1) The introduction of three linkages prevents the two
cyclopeptide rings from approaching each other efficient-
ly, thus enthalpically weakening the interactions with an
included anion relative to 2b.
2) Complex formation induces energetically unfavorable
linker conformations, an adverse effect on the enthalpy
of binding that becomes the more pronounced the higher
the number of linkers.
3) The energy required for desolvation of the binding sites
prior to sulfate binding is significantly higher in the case
of 2a.
Unfortunately, calorimetry alone could not reveal how
these factors might play a role. Therefore, additional evi-
dence from independent control experiments was required.
NMR spectroscopic analysis was used to obtain information
about the conformational reorganization of 2a and 2b
during complex formation.
À
tive hydrogen bonding, i.e., N H···O(H)H···O=C, C=
Comparing the 2D NOESY NMR spectrum of mono-
linked bis-cyclopeptide 2b with its sulfate complex revealed
crosspeaks between signals of substituted and unsubstituted
proline units in the spectrum of the complex absent in the
spectrum of uncomplexed 2b (see the Supporting Informa-
tion). Such crosspeaks have been previously observed for
other bis-cyclopeptides and they are a result of a folded con-
formation of the receptor in the complex, which arranges
the two cyclopeptide rings in close proximity and allows the
anion to efficiently bind to all six NH groups simultaneous-
ly.[10b] An additional indication of the proximity of the sub-
strate to protons located inside the receptor cavity, in partic-
ular the Ha protons, is the strong deshielding these proton
experience upon sulfate binding that causes their signals to
O···HOH···O=C, ···O(H)H···O(H)H···, and branched ···O-
AHCTUNGTRENNUNG
(H···O(H)H···)H···O(H)H··· hydrogen bonding).[23] Nine of
the water molecules form a cluster in the center of the re-
ceptor. This cluster is made up of three four-membered
rings and two six-membered rings, but each of these water
molecules forms hydrogen bonds with the receptor, in con-
trast to the endohedral ꢄmolecular iceꢅ recently observed in
a hydrophobic pocket of a self-assembled cage.[24] Removal
of the nine water molecules from one receptor in the crystal
results in a void of 274 ꢃ3,[25] which is almost an order of
magnitude larger than the 33 ꢃ3 calculated by removing the
sulfate anion from the crystal structure of the tetrabutylam-
monium sulfate complex of a monolinked bis-cyclopepti-
de,[10d] thus indicating that the conformation of 2a in the
crystal and the arrangement of the hydrogen-bond donors is
far from optimum for sulfate anion binding.
1
shift in the H NMR spectrum with respect to uncomplexed
bis-cyclopeptide by up to Dd=1.08 ppm to lower field.
The downfield shift of the Ha signals is slightly smaller
when sulfate ions are included in the cavity of 2a (Dd=
0.94 ppm), thus suggesting that the distance between the
anion and corresponding protons is on average larger in this
complex than in 2b. The effect is so small, however, that it
is unlikely to be the sole reason for the differences in the
thermodynamics of sulfate complexation of both receptors.
Unfortunately, 2D NOESY NMR spectroscopic analysis
does not provide information about the mutual orientation
of the cyclopeptide rings in the sulfate complex of 2a be-
cause the two rings are NMR spectroscopically indistin-
guishable. In comparison with the spectra of 2b, there are
stronger crosspeaks in the 2D NOESY NMR spectra of 2a
between the triazole protons and protons in the ethylene
subunits of the linkers, particularly to the protons in the
methylene group at N1. These crosspeaks account for folded
compact linker conformations, but because there are no sig-
It is not unproblematic to use structural information de-
rived from crystal structures to explain behavior in solution,
but because the crystal packing in the case of 2a·21H2O in-
volves strongly “solvated” receptor molecules with a mutual
arrangement that does not rely on directed intermolecular
interactions a cautious analysis should be feasible. Two fac-
tors appear to be relevant: First, 2a prefers to crystallize
from aqueous solution without included sulfate anions in a
conformation that is different from that expected from mo-
lecular modeling of the sulfate complex and derived from
NMR spectroscopic analysis. This behavior may be a result
of energetic differences in the conformations of free and
complexed receptor. The enthalpically more favorable con-
formations of the bis-cyclopeptide in solution appear to be
uncomplexed ones, which are most probably similar to the
conformations found in the crystal. Complex formation in-
duces a pronounced structural reorganization, consistent
7248
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7241 – 7255

Cage Receptor for Sulfate Anions
FULL PAPER
with the NMR spectroscopic results, but leads to conforma-
tions that may be less stable, presumably because the ar-
rangement of the two cyclopeptide rings induces strain in
the three tightly folded linkers. Thus, receptor 2a has to pay
an enthalpic penalty during complex formation, which
causes the interaction with sulfate ions in H2O/CH3OH to
be endothermic overall. In the crystal, in which entropic fac-
tors play a minor role, the sulfate complex of 2a is unstable,
thus explaining why this bis-cyclopeptide prefers to crystal-
lize without included sulfate ions, even if the anion is pres-
ent in solution. Second, the solvate water in the crystal ap-
pears to be tightly bound. Additional contributions to the
endothermicity of complex formation could, therefore, come
from desolvation effects if 2a binds water molecules in solu-
tion as tightly as in the crystal. Whether and to what extent
this effect contributes substantially to binding is difficult to
quantify experimentally. Measurement of solvent isotope ef-
fects (binding in D2O vs. H2O) can provide information
about the contribution of solvation effects to intermolecular
interactions, but because the enthalpic gain of the stronger
hydrogen bonds in deuterated solvents is usually compensat-
ed to a large extent by a more unfavorable entropic term,
we do not expect such measurements to provide clear-cut in-
formation with respect to sulfate binding in the case of
2a.[26] Therefore, we tentatively conclude that enthalpically
unfavorable conformations of this bis-cyclopeptide in its sul-
fate complex possibly in conjunction with the need to
remove tightly bound water molecules from the host upon
anion binding are responsible for the endothermic complex
formation. Evidently, the high stability of the sulfate com-
plex of 2a in H2O/CH3OH is solely a consequence of the
very large entropic factor, which most probably arises in
part from the release of the ordered water molecules from
bis-cyclopeptide and anion upon binding. Because only en-
tropy allows 2a to adopt the conformations necessary for
anion binding, this particular system illustrates a structure-
determining effect of entropy in a synthetic receptor.
It is worth noting that evaluation of solvent dependence
of sulfate affinity revealed the decrease in complex stability
observed for 2a and 2b upon increasing the water content
of the solvent to be entirely due to a decrease in the favora-
ble entropy term. In the case of 2a, binding enthalpy be-
comes even less unfavorable (producing a positive effect on
complex stability) as the water content increases. A similar
effect has been observed previously for a monolinked bis-cy-
clopeptide.[10d] At first sight this trend seems counterintuitive
since one would have most probably predicted the opposite
attributing the weaker binding in solvents containing more
water to weaker interactions between the receptor and the
substrate. This view, however, focuses solely on the direct
receptor–substrate interactions and neglects energetic con-
tributions from desolvation that, as previously pointed out,
cannot be separated from the binding event by calorimetry.
To rationalize how desolvation causes the observed sol-
vent-composition-dependent trends in binding enthalpy and
entropy, we previously proposed that sulfate ions have the
tendency in H2O/CH3OH mixtures to be preferentially sol-
vated by water molecules.[10d] Thus, decreasing the amount
of water should cause the water–sulfate ions interactions to
become enthalpically more favorable (less polar medium)
and entropically more costly (less water available). As a
consequence, desolvation of sulfate ions upon complex for-
mation will be entropically more favorable as the amount of
water decreases, but increasingly endothermic. The same ar-
guments hold when making the reasonable assumption
based on the crystal structure that the receptor cavity of 2a
is preferentially solvated by water molecules.
Unfortunately, only Gibbs energies for the transfer of sul-
fate ions from water to H2O/CH3OH mixtures are avail-
able,[27] but not the individual enthalpic and entropic contri-
butions. Therefore, appropriate data is currently missing to
confirm our interpretation and verification requires further
systematic binding studies. The characteristic effect of the
structures of receptors 2a and 2b on the thermodynamics of
sulfate binding is evident in the different extent to which
the complexation enthalpy for each receptor varies upon in-
creasing the water content of the solvent from 35 to 65%,
with almost no change in the case of 2b, whereas the com-
plexation enthalpy decreases by approximately 6 kJmolÀ1
for 2a.
Kinetics of sulfate binding: The NMR spectroscopic changes
observed upon sulfate complexation of bis-cyclopeptides 2a
and 2b qualitatively indicate that the rate of guest exchange
is slower for the receptor containing three bridges between
the cyclopeptide rings. Temperature-dependent NMR and
quantitative 1H–1H NOESY NMR spectroscopic measure-
ments (exchange spectroscopy, EXSY) were carried out to
gain more quantitative insight into complexation/decom-
plexation kinetics.
The simplest way to spectroscopically estimate the rate of
1
exchange in a dynamic system by using H NMR spectro-
scopic techniques is to determine the distance Dn in sÀ1 be-
tween two peaks of exchanging protons in the slow-ex-
change regime and the temperature at which these peaks co-
alesce. The rate constant at this temperature kc is then pDn/
20.5. Although this correlation strictly only applies to dynam-
ic equilibria in which the interconverting species are equally
populated, are transformed into one another through first-
order reactions, and possess peaks in the NMR spectra that
do not exhibit coupling, this correlation has also been shown
to provide a rough estimate of reaction rates that do not ful-
fill all of these requirements, including binding equilibria.[28]
The problems associated with studying binding equilibria
are: 1) only complex dissociation is usually a first-order re-
action, whereas complex formation follows a second-order
rate law and 2) temperature variation not only affects bind-
ing kinetics but also the ratio of free and complexed binding
partners, namely, the thermodynamics.
Aware of these drawbacks, we studied the effect of tem-
perature on the NMR spectra of solutions of 2a and 2b in
D2O/CD3OD (1:1, v/v) in the presence of 0.5 equivalents of
Na2SO4. In the case of 2b, increasing the temperature
caused the expected sharpening of the Ha signals, whereas
Chem. Eur. J. 2010, 16, 7241 – 7255
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7249

S. Kubik et al.
decreasing the temperature caused the signal to become
broader until two peaks emerged in the spectral region of
the Ha signals at temperatures below 108C, which can be
assigned to free 2b and its sulfate complex (see the Support-
ing Information). Both signals are rather broad even at
À108C, but further lowering of temperature caused the for-
mation of a precipitate in the sample, thus precluding meas-
urements below this temperature. By using the chemical
shifts of the Ha signals of free receptor and those of the re-
ceptor in the presence of two equivalents of Na2SO4 to cal-
culate Dn, a rate constant kc at the coalescence temperature
of 108C between 1400 and 1500 sÀ1 was calculated. The
range derives from the fact that the Ha signals of the substi-
tuted proline units in 2b experience a slightly larger shift in
the NMR spectra upon complex formation than the signals
of the unsubstituted proline units. Analogously, a compara-
ble kc value of 1200 sÀ1 can be calculated from the corre-
sponding spectra of 2a. For this receptor, however, it was
impossible to determine the coalescence temperature be-
Conclusion
These investigations demonstrate that click chemistry can be
successfully used to construct a molecular cage from two cy-
clopeptides, which efficiently interacts with sulfate ions in
highly competitive aqueous solvent mixtures with stability
constants in the range of 105–106 mÀ1. Thus, this compound
enlarges the family of anion cages by another member;
other examples are macrobicyclic polyammonium recap-
AHCTUNGTRENNUNG
tors,[31a] the macrotricyclic receptors described by
Schmidtchen and co-workers,[31b] the coordination cages de-
scribed by the groups of Custelcean,[9a] Wu,[9b] and Sever-
in,[31c] and a number of anion-templated capsules,[9c,31d] in-
cluding the self-assembling receptors described by Rebek
and co-workers.[31e] The characteristic structural feature of
this cage is the polar interior caused by the peptide NH
groups that converge inside the cavity. The binding kinetics
of the cage differ significantly from those of an analogue
containing only one linker between the cyclopeptide rings in
that the rate of guest exchange is considerably slower. This
finding can be explained by the greater difficulty a substrate
experiences when entering or leaving the cavity of a molecu-
lar cage, a notion for which Cram et al. introduced the term
“constrictive binding”.[32] Thermodynamically, the sulfate
complex of the triply linked bis-cyclopeptide is only approx-
imately one order of magnitude more stable than the one of
the more flexible analogue containing only one linker. Titra-
tion calorimetry showed that this behavior is partly due to
the change in the binding enthalpy from exothermic to en-
dothermic upon increasing the number of linkers. Thus, in-
teractions between sulfate ions and the triply linked bis-
cyclopeptide cannot compensate the energy required for de-
solvation of the binding partners. As a result, complex sta-
bility is entirely due to entropy. Structural investigations in-
dicate that this behavior has most probably structural rea-
sons (the receptor conformations in the complex are ener-
getically unfavorable) possibly in combination with solvation
effects (desolvation of the receptor in aqueous solution is
enthalpically costly).
1
cause the H NMR spectrum of 2a in D2O/CD3OD (1:1, v/
v) in the presence of 0.5 equivalents of Na2SO4 still showed
two, albeit very broad, Ha signals even at 608C and a fur-
ther temperature increase was not feasible. Therefore, these
measurements only allow the conclusion that for 2a to ap-
proach a similar rate of guest exchange as observed for 2b
at 108C, the temperature must be more than 508C higher.
To obtain quantitative data for the rate of guest exchange
for 2a, we performed 2D EXSY NMR measurements.[29]
This method has been proven to be very useful to evaluate
the rate of binding equilibria.[4b] However, it can only be ap-
plied to reactions with equilibria that are slow on the NMR
timescale and associated with two separated signals of ex-
changing protons in the spectrum. The optimal mixing time
for the measurement was estimated by recording 1H–1H
NOESY NMR spectra of 2a in D2O/CD3OD (1:1, v/v) in
the presence of 0.5 equivalents of Na2SO4 at different
mixing times (0.03, 0.045, 0.06, 0.08, 0.1, 0.15, 0.25, and 0.5 s)
and determining under which conditions the volumes of the
crosspeaks of the exchanging Ha protons were largest. The
first-order magnetization rate constants k*in and k*out were
calculated from the volumes of the cross and diagonal peaks
of the Ha signals by using the program EXSYCalc.[36]
The spectrum of 2a in D2O/CD3OD (1:1) in the presence
of 0.5 equivalents of Na2SO4 recorded with the optimum
mixing time tm of 0.06 s thus furnished a first-order rate con-
stant for the complexation of sulfate ions by 2a at 228C of
k*in =25.4 sÀ1 and a value for the corresponding decomplex-
ation of k*out =3.8 sÀ1. As dissociation of the complex can be
considered to be a first-order reaction k*out =kout. Thus,
these values confirm that even at 228C, the complexation/
decomplexation equilibrium of the triply linked bis-cyclo-
peptide is considerably slower than that of the more flexible
analogue 2b at 108C. Moreover, they compare favorably
with the rate constants determined for the binding equilibria
of other receptors with a similarly confined cavity, for exam-
ple, the hexameric resorcarene-derived capsule described by
Rebek and co-workers.[4b,30]
Therefore, these investigations show that closing a cage
does not necessarily lead to a significant improvement in
binding if this causes pronounced changes in the energetics
of complex formation. They also demonstrate how impor-
tant knowledge of the thermodynamics of complex forma-
tion is for an understanding of binding behavior. Further in-
vestigations will address strategies to alleviate the enthalpic
disadvantage of the anion binding of 2a, for example, by
structurally optimizing the linkers. In addition, binding stud-
ies will be undertaken to determine whether the mutual ar-
rangement of the two cyclopeptide rings in 2a allows com-
plexation of anions normally not bound by our bis-cyclopep-
tides, such as more sizeable anions including di- or triphos-
phates. Investigations in both directions are currently under-
way.
7250
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7241 – 7255

Cage Receptor for Sulfate Anions
FULL PAPER
C4), 145.8+145.9 (APA C2), 151.9 (APA C6), 152.8+153.3 (tBu CO),
155.6 (Z CO), 164.0 (APA CO), 171.9+172.5 ppm (Apro CO); MS
(MALDI-TOF): m/z (%): 525.2 (100) [M+H]+, 547.2 (69) [M+Na]+,
563.2 (15) [M+K]+; elemental analysis calcd (%): for C27H32N4O7: C
61.82, H 6.15, N 10.68; found: C 61.64, H 6.34, N 10.55.
Experimental Section
General: Analyses were carried out as follows: melting points, Mꢁller
SPM-X 300; NMR, Bruker Avance 600 and Bruker DPX 400; the chemi-
cal shifts were referenced against the shifts of the residual solvent pro-
tons: d([D6]DMSO)=2.54 ppm, d([D4]methanol)=3.34 ppm; MALDI-
TOF MS, Bruker Ultraflex TOF/TOF; ESI quadrupole MS, Bruker Es-
quire 3000; IR, Perkin–Elmer FTIR Spectrometer, Spectrum 1000; ele-
mental analysis, Perkin–Elmer 2400 CHN; ITC, Microcal VP-ITC;
HPLC, Dionex P680 HPLC Pump, ASI-100 Autosampler, TCC-100
Column Oven, UVD 170U UV/Vis detector, Chromeleon V6.70 Soft-
ware, Chromolith SemiPrep 100–10 mm RP-18 endcapped (for 2a), Puro-
Cyclopeptide cyclo[(Z-4S-Apro)-APA]3 ACHTNUTGRNEUNG(6a): Prior to the synthesis of
this cyclopeptide, dipeptide 5 was chain elongated up to the linear hexa-
peptide Boc-[(Z -4S-Apro)-APA]3-OAll according to a reported proce-
dure.[15] For cyclization, this hexapeptide (1.26 g, 1 mmol) was deprotect-
ed at both ends and dissolved in a mixture of degassed DMF (40 mL)
and DIEA (1.04 mL, 6 mmol). The resulting solution was added dropwise
over the course of 4 h to a solution of (O-benzotriazole-1-yl-N,N,N’,N’-
tetramethyluronium tetrafluoroborate; TBTU (1.60 g, 5 mmol) and
DIEA (0.42 mL, 2.4 mmol) in degassed DMF (200 mL) at 808C. If neces-
sary, the pH value of the reaction mixture was adjusted afterward to ap-
proximately 9 by adding more DIEA, and the reaction mixture was
stirred for a further 1 h at 808C. The solvent was evaporated in vacuo,
and the product was isolated from the residue by chromatography. An in-
itial purification step was carried out by column chromatography on
silica gel (acetone). The material recovered was further purified on an
RP-8 column. For this purification, the material was dissolved in a small
amount of DMF and applied to a column conditioned with 1,4-dioxane/
H2O (1:10). The eluent composition was gradually changed until the pure
product was eluted (1,4-dioxane/H2O, 2:1). The material thus obtained
was dissolved in acetone (20 mL), and the resulting solution was poured
slowly in diethyl ether (200 mL) with stirring. Stirring was continued for
15 min, the precipitate was filtered off, and dried in vacuo. In case the
material thus obtained is still impure, it can be recrystallized from H2O/
ethanol (0.37 g, 34%). M.p. 1688C; 1H NMR (600 MHz, [D6]DMSO,
1008C): d=2.06 (m, 3H; Apro Hb), 2.84 (m, 3H; Apro Hb), 3.52 (m,
3H; Apro Hd), 3.96 (m, 3H; Apro Hd), 4.21 (m, 3H; Apro Hg), 5.08 (s,
6H; Z CH2), 5.54 (m, 3H; Apro Ha), 6.88 (s, br, 3H; Z NH), 7.27–7.34
ACHTUNGTRENNUNGspher STAR RP-18 endcapped (5 mm; for 2b). The following abbrevia-
tions are used: Pro=l-proline, Apro=2S,4S-aminoproline, APA=6-ami-
nopicolinic acid, PA=4-pentynoic acid, AzP=3-azidopropanoic acid,
PyCloP=chlorotripyrrolidinophosphonium hexafluorophosphate; EDC=
N-(3-dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride. 1H and
13C NMR spectroscopic assignments of protons and carbon atoms in the
linker:
Allyl 6-aminopicolinate (4): A mixture of 6-aminopicolinic acid[33] (6.90 g,
50 mmol) and NaHCO3 (8.40 g, 100 mmol) was suspended in DMF p.a.
(300 mL). 3-Bromopropene (13.1 mL, 150 mmol) was added to the reac-
tion mixture, which was stirred for seven days at room temperature. The
solvent was evaporated in vacuo and the residue was suspended in ethyl
acetate. The resulting mixture was washed twice with 10% aqueous
Na2CO3, washed with water (3ꢆ), dried, and evaporated to dryness. The
remaining crude product was purified by chromatography (SiO2, pentane/
ethyl acetate, 1:1) and crystallizes as a light yellow solid upon tituration
with hexane (4.90 g, 55%). M.p. 72–738C; 1H NMR (500 MHz, CDCl3,
(m, 18H; APA H3), Z Ph H), 7.48 (d, 3J
ACTHNUTRGNE(UNG H,H)=7.6 Hz, 3H; APA H5),
7.73 (m, 3H; APA H4), 9.13 ppm (s, br, 3H; APA NH); 13C NMR
(151 MHz, [D6]DMSO, 228C): d=37.2 (Apro Cb), 47.9 (Apro Cg), 52.0
(Apro Cd), 60.5 (Apro Ca), 65.5 (Z CH2), 115.9 (APA C3), 119.7 (APA
C5), 127.8+127.9+128.4+136.9 (Z PhC), 139.1 (APA C4), 148.5 (APA
C2), 151.6 (APA C6), 155.8 (Z CO), 166.2 (APA CO), 170.4 ppm (Apro
CO); MS (MALDI-TOF): m/z (%): 1099.6 (28) [M+H]+, 1121.5 (100)
[M+Na]+, 1137.5 (76) [M+K]+; elemental analysis calcd (%): for
C57H54N12O12·3H2O: C 59.37, H 5.24, N 14.58; found: C 59.74, H 5.13, N
14.38.
228C): d=4.78 (s, 2H; NH), 4.89 (d, 3J
5.31 (dd, 2J(H,H)=1.3, 3J(H,H)=10.4 Hz, 1H; All Hcis), 5.43 (dd, 2J-
(H,H)=1.3, 3J
(H,H)=17.2 Hz, 1H; All Htrans), 6.07 (m, 1H; All Hvic),
6.69 (dd, 3J(H,H)=8.2, 4J(H,H)=0.9 Hz, 1H; APA H3), 7.51 (dd, 3J-
(H,H)=7.3, 4J(H,H)=0.9 Hz, 1H; APA H5), 7.56 ppm (t, 3J
(H,H)=
ACHTUNGTREN(UNGN H,H)=5.7 Hz, 2H; All CH2),
G
ACHTUNGTRENNUNG
E
ACHTUNGTRENNUNG
U
ACHTUNGTRENNUNG
E
R
ACHTUNGTRENNUNG
7.8 Hz, 1H; APA H4); MS (EI): m/z (%): 177.8 (21) [M]+; elemental
analysis calcd (%): for C9H10N2O2: C 60.66, H 5.66, N 15.72; found: C
60.60, H 5.72, N 15.47.
Triamine (6b): Cyclopeptide 6a (0.3 g, 0.27 mmol) was dissolved in meth-
anol/dichloromethane (1:1, 80 mL). After the addition of 10% Pd/C
(40 mg), 20% Pd(OH)2/C (40 mg), and 1m HCl (890 mL, 890 mmol,
3.3 equiv), the reaction mixture was hydrogenated at 1 atm for 7 days.
Afterward, the catalysts were removed by filtration over celite, washed
with methanol, and the filtrate was evaporated to dryness. The product
thus obtained was used in the next step without further purification
(0.21 g, 97%). MS (MALDI-TOF): m/z (%): 697.3 (87) [M+H]+, 719.3
(100) [M+Na]+, 735.3 (79) [M+K]+.
Dipeptide Boc-(Z-4S-Apro)-APA-OAll (5): (2S,4S)-4-(Benzyloxycarbo-
nylamino)-1-(tert-butyloxycarbonyl)proline[14] (3; 2.19 g, 6.0 mmol), allyl
6-aminopicolinate (4; 1.17 g, 6.6 mmol), and PyCloP (2.78 g, 6.6 mmol)
were dissolved in CH2Cl2 (100 mL). At room temperature, DIEA
(2.50 mL, 14.4 mmol) was added dropwise, and the reaction mixture was
stirred for 7 days. The solvent was evaporated in vacuo, and the product
was isolated from the residue by chromatography on silica gel (pentane/
ethyl acetate, 1:1) (2.55 g; 81%). M.p. 68–758C; 1H NMR (600 MHz,
[D6]DMSO, 808C): d=1.37 (s, 9H; tBu CH3), 1.93–2.08 (m, 1H; Apro
4-Pentynoic acid pentafluorophenol ester (7): Pentafluorophenol (0.74 g,
4 mmol) and 4-pentynoic acid (0.41 g, 4.2 mmol) were dissolved in dry
ethyl acetate (40 mL), and EDC (0.81 g, 4.2 mmol) was added in small
portions. After stirring overnight, the reaction mixture was washed with
water (3ꢆ), 10% aqueous Na2CO3 (3ꢆ), and water (3ꢆ) then dried and
evaporated to dryness. The product crystallized upon standing at room
temperature (0.99 g, 94%). M.p. 51–558C; 1H NMR (400 MHz, CDCl3,
Hb), 2.54–2.60 (m, 1H; Apro Hb), 3.28 (dd, 2J
6.5 Hz, 1H; Apro Hg), 3.74 (dd, 2J(H,H)=10.6, 3J
Apro Hd), 4.10–4.13 (m, 1H; Apro Hd), 4.53 (dd, 3J
(H,H)=6.5 Hz, 1H; Apro Ha), 4.86 (d, 2J
(H,H)=5.6 Hz, 2H; Z CH2),
5.03–5.08 (m, 2H; All CH2), 5.31 (dd, 2J(H,H)=1.4, 3J
(H,H)=10.5 Hz,
1H; All Hcis), 5.43 (dd, 2J(H,H)=1.4, 3J
(H,H)=17.2 Hz, 1H; All Htrans),
6.02–6.09 (m, 1H; All Hvic), 6.95 (s, 1H; Z NH), 7.27–7.37 (m, 5H; Z Ph
H), 7.78 (d, 3J(H,H)=7.4 Hz, 1H; APA H3), 7.97 (t, 3J
(H,H)=7.9 Hz,
1H; APA H4), 8.27 (d, 3J
(H,H)=8.3 Hz, 1H; APA H5), 10.61 ppm (s,
N
ACHTUNGTRNE(NUNG H,H)=
E
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
228C): d=2.06 (t, 4J
(H,H)=2.6 Hz, 2H; H3), 2.93 ppm (t, 3J
ACHTUNGTRENNUNG
N
(H,H)=7.3, 4J-
ACHTUNGTRENNUNG
T
ACHTUNGTRENNUNG
E
A
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
Ph F3), À157.7 (t, 3J
(F,F)=21.7 Hz, 1F; PhF4), À152.5 ppm (m, 2F,
G
1H; NH); 13C NMR (151 MHz, [D6]DMSO, 228C): d=27.9+28.1 (tBu
CH3), 35.1+35.9 (Apro Cb), 48.8+49.4 (Apro Cg), 50.9+51.6 (Apro
Cd), 58.4+58.7 (Apro Ca), 65.5+65.7 (Z CH2, all C1), 79.0+79.1 (tBu
C), 117.4+117.6 (All C3), 118.6 (APA C3), 120.7 (APA C5), 127.9+
128.4 (Z Ph C2–4), 132.4 (All C2), 136.9 (Z Ph C1), 139.7+139.9 (APA
PhF2); elemental analysis calcd (%): for C11H5F5O2: C 50.02, H 1.91;
found: C 49.94, H 2.00.
3-Azidopropanoic acid pentafluorophenol ester (8): Pentafluorophenol
(0.46 g, 2.5 mmol) and 3-azidopropanoic acid[34] (0.3 g, 2.6 mmol) were
Chem. Eur. J. 2010, 16, 7241 – 7255
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7251

S. Kubik et al.
dissolved in dry ethyl acetate. The solution was cooled in an ice bath,
and EDC (0.5 g, 2.6 mmol) was added in small portions. After the mix-
ture had been stirred for 1 h in the cold, the ice bath was removed and
stirring was continued for another 4 h. The reaction mixture was washed
with water (3ꢆ), 10% aqueous Na2CO3 (3ꢆ), water (2ꢆ), 5% aqueous
KHSO4 (2ꢆ), and water (3ꢆ). After drying, the organic solvent was re-
moved and the product, which was obtained as an oil, was dried in vacuo
preconditioned with aqueous NH3 (2.5%). Aqueous NH3 (2.5%) was
rinsed through the silica pad until the filtrate was almost colorless. The
product was eluted with 1,4-dioxane/water (1:1), the solvent was removed
in vacuo, and the residue was purified by using semipreparative HPLC
(16 mg, 28%). M.p. >3008C; 1H NMR (400 MHz, [D6]DMSO, 228C):
d=1.64 (m, 6H; Apro Hb), 2.38 (m, 6H; H2’), 2.60–2.74 (m, 9H; 6 H7’,
3 H3’), 2.80–2.98 (m, 9H; 6 Apro Hb, 3H3’), 3.30 (6H; Apro Hd), 3.92
(m, 6H; Apro Hd), 4.30–4.57 (m, 12H; 6 Apro Hg, 6 H8’), 5.56 (m, 6H;
(0.68 g, 97%). 1H NMR (400 MHz, CDCl3, 228C): d=2.94 (t, 3J
6.4 Hz, 2H; H2), 3.72 ppm (t, 3J(H,H)=6.4 Hz, 2H; H3); 13C NMR
ACHTUNGTRNE(NUNG H,H)=
Apro Ha) 7.16 (m, 3J
APA H5, H5’), 7.76 (m, 6H; APA H4), 8.14 (d, 3J
AproNH), 8.27 (d, 3J
(H,H)=6.7 Hz, 3H; Apro NH), 9.61 ppm (s, br,
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
(151 MHz, CDCl3, 228C): d=33.3 (C2), 46.4 (C3), 124.8 (Ph C1), 137.2,
138.9, 140.3, 140.6, 142.0 (Ph C2–4), 167.2 (C1); 19F NMR (565 MHz,
AHCTUNGTRENNUNG
6H; APA NH); 13C NMR (151 MHz, [D6]DMSO, 228C): d=20.9 (C3’),
34.5 (C2’), 35.5 (C7’), 37.5 (Apro Cb), 45.5 (C8’), 46.0+46.1 (Apro Cg),
51.4 (Apro Cd), 60.8 (Apro Ca), 116.1 (APA C3), 119.8 (APA C5), 121.9
(C5’), 139.2 (APA C4), 145.6 (C4’), 148.3 (APA C2), 151.4 (APA C6),
166.0 (APA CO), 169.5 (C6’) 170.5 (Apro CO), 171.5 (C1’); MS
(MALDI-TOF): m/z (%): 1926.1 (94) [M+H]+, 1947.9 (100) [M+Na]+,
CDCl3, 228C): d=À162.0 (m, 2F; Ph F3), À157.4 (t, 3J
ACHTUNGTREN(NUNG F,F)=21.7 Hz,
1F; Ph F4), À152.5 ppm (m, 2F; Ph F2); elemental analysis calcd (%):
for C9H4F5N3O2: C 38.45, H 1.43, N 14.95; found: C 38.43, H 1.44, N
14.76.
Tris-alkyne 9a: Cyclopeptide 6b·3HCl (81 mg, 0.1 mmol) and DIEA
(104 mL, 0.6 mmol, 6 equiv) were dissolved in DMSO (5 mL). A solution
of active ester 7 (240 mg, 0.9 mmol, 9 equiv) in dichloromethane (10 mL)
was added and the reaction mixture was stirred at room temperature for
4 h. The solvent was removed in vacuo and the product was isolated from
the residue chromatographically (dichloromethane/methanol, 5:1). Frac-
tions with pure product were evaporated to dryness, the residue was dis-
solved in acetone, and the product was precipitated by addition of diethyl
ether. After filtration, the product was dried in vacuo (61 mg, 65%).
M.p. 188–1948C; 1H NMR (400 MHz, [D6]DMSO, 228C): d=1.90 (m,
3H; Apro Hb), 2.22 (m, 6H; PA H2), 2.28 (m, 6H; PA H3), 2.71 (t, 1H;
1963.9
(37)
[M+K]+;
elemental analysis calcd
(%):
for
C90H93N33O18·13H2O: 50.07, H 5.56, N 21.41; found: C 49.89, H 5.37, N
21.43.
Crystal structure analysis of 2a: [C90H93N33O18]·21
ACHTUGNRTNE[NUGN H2O], Mr =
2303.31 gmolÀ1 colorless prism, crystal size 0.320ꢆ0.322ꢆ0.862 mm3,
,
monoclinic, space group C2, a=30.664(1), 17.7503(6), c=10.5374(4) ꢃ,
b=99.197(1)8, V=5661.7(3) ꢃ3, T=100 K, Z =2, 1calcd =1.351 gcm3, l=
1.54178 ꢃ, m=0.909 mmÀ1
, Gaussian absorption correction (Tmin =
0.50740, Tmax =0.82840), scaling SADABS, Bruker AXS Proteum X8 dif-
fractometer, 2.89<q<66.66, 64165 measured reflections, 9649 independ-
ent reflections, 9570 reflections with I>2s(I). Structure solved by charge
flipping using Superflip,[35a] and the electron-density map was analyzed
with EDMA.[35b] Structure refined by full-matrix least-squares using
SHELXL[35c] against F2 to R1 =0.0256 (I>2s(I)), wR2 =0.0682, 783 pa-
rameters. The molecular symmetry of the cage is not consistent with the
crystal symmetry. Because each molecule sits on a crystallographic two-
fold axis passing through C5 of one of the triazole linkers, the N1 and C4
positions for this triazole ring must be disordered. If one triazole ring is
disordered, all three triazole rings must be disorderd because the capsule
was prepared by the covalent linkage of a tris-alkyne-substituted and a
tris-azide-substituted cyclic hexapeptide with the result that the N1 posi-
tions of the three triazole rings are similarly orientated. The disorder of
the triazole rings was modeled by occupying each N1 and C4 by C and N
atoms, each with half-occupancy, and by giving C and N atoms the same
atomic displacement parameters. The solute water O atom O20 is disor-
dered over two positions (O20A and O20B). The relative occupancies
were refined and converged with an occupancy of 0.820(4) for O20A. All
the hydrogen atoms on the solute water molecules were located on a dif-
ference Fourier synthesis map calculated by using the cage with riding H
atoms and the O atoms of the water molecules. The solute H atoms were
4J
Hd), 3.89 (m, 3H; Apro Hd), 4.32 (m, 3H; Apro Hg), 5.60 (m, 3H; Apro
Ha), 7.22 (d, 3J(H,H)=8.2 Hz, 3H; APA H3), 7.43 (d, 3J
(H,H)=7.3 Hz,
3H; APA H5), 7.74 (m, 3H; APA H4), 8.10 (d, 3J
(H,H)=6.4 Hz, 1H;
ACHTUNGTRENNUNG(H,H)=2.4 Hz, PA H5), 2.85 (m, 3H; Apro Hb), 3.37 (m, 3H; Apro
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
Apro NH), 9.72 ppm (s, br, 3H; APA NH); 13C NMR (100 MHz,
[D6]DMSO, 228C): d=13.9 (PA C3), 34.0 (PA C2), 37.2 (Apro Cb), 46.1
(Apro Cg), 51.8 (Apro Cd), 60.4 (Apro Ca), 70.9 (PA C5), 83.4 (PA C4),
115.7 (APA C3), 119.5 (APA C5), 138.8 (APA C4), 148.4 (APA C2),
151.4 (APA C6), 165.9 (APA CO), 170.4+170. ppm (PA CO, Apro CO);
MS (MALDI-TOF): m/z (%): 937.4 (54) [M+H]+, 959.4 (100) [M+Na]+,
975.4 (80) [M+K]+; elemental analysis calcd (%): for C48H48N12O9·4H2O:
C 57.14, H 5.59, N 16.66; found: C 57.23, H 5.33, N 16.57.
Tris-azide 9b: This peptide was prepared analogously to 9a from
6b·3HCl and active ester 8. For the chromatographic purification, di-
chloromethane/methanol (7:1) was used as the eluent (36 mg, 36%). M.p.
176–1868C; 1H NMR (600 MHz, [D6]DMSO, 228C): d=1.95 (m, 3H;
Apro Hb), 2.31 (m, 6H; AzP H2), 2.86 (m, 3H; Apro Hb), 3.40 (m, 3H;
Apro Hd), 3.43 (t, 6H; 3J
Hd), 4.32 (m 3H; Apro Hg), 5.60 (m, 3H; Apro Ha), 7.25 (d, JAHCTUNGTRENNUNG
ACHTUNGTRENN(UNG H,H)=6.3 Hz, AzP H3), 3.89 (m, 3H; Apro
3
3
8.3 Hz, 3H; APA H3), 7.43 (d, JACHTGNUTRENNUNG
3
3H; APA H4), 8.19 (d, JACHTUNGTRENNUNG
À
first refined isotropically to convergence. They gave an average O H dis-
3H; APA NH); 13C NMR (151 MHz, [D6]DMSO, 228C): d=34.6 (AzP
C2), 37.3 (Apro Cb), 46.3 (Apro Cg), 46.8 (AzP C3), 51.9 (Apro Cd),
60.5 (Apro Ca), 115.7 (APA C3), 119.7 (APA C5), 139.1 (APA C4), 148.6
(APA C2), 151.5 (APA C6), 166.1 (APA CO), 169.8 (AzP CO) 170.6 ppm
(Apro CO); IR (KBr): n˜ =2104 cmÀ1 (azide); MS (MALDI-TOF): m/z
(%): 954.4 (5) [MÀ2N2+Na]+, 982.4 (25) [MÀN2+Na]+, 998.4 (6)
[MÀN2+K]+, 1010.4 (100) [M+Na]+, 1026.4 (21) [M+K]+; elemental
analysis calcd (%): for C42H45N21O9·6H2O: C 46.03, H 5.24, N 26.84;
found: C 46.33, H 5.01, N 26.50.
tance of 0.89(8) ꢃ and an average atomic displacement parameter of
0.06(3) ꢃ2. To decrease the number of parameters without adversely af-
fecting the model, the atomic displacement parameters of the H atoms
on the water molecules were constrained to be 120% of the atomic dis-
placement parameters of the O atoms to which they were attached and
À
the O H distances were restrained to be 0.84 ꢃ with a standard uncer-
tainty of 0.02. There is a void of 35 ꢃ3 in a hydrophobic region of unit
cell but no electron density was visible in the final difference Fourier syn-
thesis. Absolute structure parameter=0.0(1), S=1.016, residual electron
density=+0.20/À0.33 eꢃÀ3. CCDC-763646 (2a) contains the supplemen-
tary crystallographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Smaller crystals grown from water/
acetonitrile in the absence of Na2SO4 gave similar results (CCDC-
763645; see the Supporting Information).
Triply linked bis-cyclopeptide 2a: CuSO4 (150 mg, 0.6 mmol, 20 equiv)
was dissolved in degassed water under argon and solutions of 2,6-lutidine
(193 mg, 1.8 mmol, 60 equiv) in methanol (40 mL), sodium ascorbate
(357 mg, 1.8 mmol, 60 equiv) in water (30 mL), and 9b (30 mg, 30 mmol)
in methanol (15 mL) were added under stirring. A solution of 9a (28 mg,
30 mmol) in methanol (15 mL) was added very slowly to the reaction mix-
ture over 8 h at room temperature by using a syringe pump. After stirring
overnight, an aqueous solution of BaCl2 (156 mg, 0.75 mmol, 25 equiv) in
H2O (2 mL) was added, and the precipitate was filtered off through a
pad of celite. The solvent was removed in vacuo, the residue was dis-
solved in a small amount of aqueous NH3 (2.5%)/1,4-dioxane (20:1), and
the solution was filtered through a pad of RP-8 silica, which had been
Cyclopeptide (10a): This cyclopeptide was prepared from the linear
hexapeptide precursor Boc-(Pro-APA)2-(Z-4S-Apro)-APA-OAll analo-
gously to cyclopeptide 6a. It was eluted from the RP column with 1,4-di-
oxane/H2O (1:1) (0.42 g, 53%). M.p. 184–1908C; 1H NMR (600 MHz,
[D6]DMSO, 228C): d=1.76–1.88 (m, 4H; Pro Hg), 1.88–1.96 (m, 1H; Pro
Hb), 2.00–2.07 (m, 2H; Pro Hb, Apro Hb), 2.52–2.60 (m, 2H; Pro Hb),
7252
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7241 – 7255

Cage Receptor for Sulfate Anions
FULL PAPER
2.83–2.92 (m, 1H; Apro Hb) 3.39–3.41 (m, 1H; Apro Hd), 3.53–3.61 (m,
2H; Pro Hd), 3.65–3.72 (m, 2H; Pro Hd), 3.86–3.93 (m, 1H; Apro Hd),
4.98–5.03 (m, 2H; Z CH2), 5.47–5.54 (m, 2H; Pro Ha), 5.60–5.62 (m,
760.5 (22) [MÀN2+H2+Na]+, 764.6 (41) [M+H]+, 776.5 (4)
[MÀN2+H2+K]+, 786.5 (26) [M+Na]+, 802.5 (17) [M+K]+; elemental
analysis calcd (%): for C36H37N13O7·5H2O: C 50.64, H 5.55, N 21.33;
found: C 50.53, H 5.33, N 21.00.
ACTHNUTRGNEUNG
1H; Apro Ha), 7.16 (d, 3J(H,H)=8.1 Hz, 1H; APA H3), 7.21 (d, 3J-
ACHTUNGTRENNUNG(H,H)=8.1 Hz, 2H; APA H3), 7.27–7.35 (m, 5H; Z PhH), 7.39–7.46 (m,
Monolinked bis-cyclopeptide 2b: CuSO4 (35 mg, 0.14 mmol, 2 equiv) was
dissolved in degassed water under argon. Solutions of 2,6-lutidine (45 mg,
0.42 mmol, 6 equiv) in methanol (7 mL), sodium ascorbate (83 mg,
0.42 mmol, 6 equiv) in water (7 mL), 9a (52 mg, 70 mmol) in methanol
(7 mL), and 9b (53 mg, 70 mmol) in methanol (7 mL) were added, and
the resulting mixture was stirred overnight at room temperature. The sol-
vent was removed in vacuo and the product was isolated from the residue
chromatographically on an RP-8 column. The residue was dissolved in a
small amount of DMF and applied to a column conditioned with 1,4-di-
oxane/H2O (1:10). The eluent composition was gradually changed until
the pure product was eluted (1,4-dioxane/H2O, 1:3). In case the material
thus obtained was still impure, it was additionally purified by semiprepar-
ative HPLC (42 mg, 40%). M.p. 262–2848C; 1H NMR (600 MHz,
[D6]DMSO, 228C): d=1.76–1.93 (m, 9H; 8 Pro Hg, 2 Apro Hb), 2.03 (m,
4H; Pro Hb), 2.35 (m, 2H; H2’), 2.56 (m, 4H; Pro Hb), 2.64 (m, 2H;
H7’), 2.73 (m, 2H; H3’), 2.85 (m, 2H; Apro Hb), 3.34 (2H; Apro Hd),
3.58 (m, 4H; Pro Hd), 3.69 (m, 4H; Pro Hd), 3.89 (m, 2H; Apro Hd),
4.31 (m, 2H; Apro Hg), 4.43 (m, 2H; H8’), 5.52–5.67 (m, 6H; Apro Ha,
Pro Ha), 7.13–7.26 (m, 6H; APA H3), 7.43 (m, 6H; APA H5), 7.66 (s,
3H; APA H5), 7.52–7.57 (m, 1H; Z NH), 7.69–7.77 (m, 3H; APA H4),
9.54+9.57+9.66 ppm (3ꢆs, 3ꢆ1H; NH); 13C NMR (151 MHz,
[D6]DMSO, 228C): d=22.8+22.9 (Pro Cg), 32.9+33.0 (Pro Cb), 37.6
(Apro Cb), 48.3 (Apro Cg), 48.5+48.6 (Pro Cd), 52.5 (Apro Cd), 61.0
(Apro Ca), 61.9 (Pro Ca), 65.9 (Z CH2), 115.9+116.3+116.6 (APA C3),
120.1+120.2 (APA C5), 128.2 (Z C2), 128.3 (Z C4), 128.8 (Z C3), 137.4
(Z C1), 139.4+139.5 (APA C4), 148.9+149.0 (APA C2), 152.0+152.3+
152.5 (APA C6), 156.2 (Z CO), 166.3+166.5 (APA CO), 170.9 (Apro
CO), 171.4+171.5 ppm (Pro CO); MS (MALDI-TOF): m/z (%): 801.4
(98) [M+H]+, 823.4 (100) [M+Na]+, 839.4 (93) [M+K]+; elemental anal-
ysis calcd (%): for C41H40N10O8·2.5H2O: C 58.22, H 5.36, N 16.56; found:
C 58.09, H 5.32, N 16.45.
Monoamine 10b: Removal of the Z-protecting group in cyclopeptide 10a
was achieved analogously to the deprotection of 6a, except that 20%
Pd(OH)2/C was not added to the reaction mixture and reaction time was
decreased to 24 h (0.18 g, 94%). MS (MALDI-TOF): m/z (%): 667.3 (57)
[M+H]+, 689.3 (100) [M+Na]+, 705.3 (22) [M+K]+.
Monoalkyne 11a: This peptide was prepared analogously to 9a from
10b·HCl and active ester 7 by starting from 10b·HCl (0.2 mmol), except
that the cyclopeptide was dissolved in dichloromethane because of better
solubility. For the chromatographic purification dichloromethane/metha-
nol (7:1) was used as eluent (84 mg, 56%). M.p. 196–2088C; 1H NMR
(600 MHz, [D6]DMSO, 228C): d=1.81 (m, 2H; Pro Hg), 1.86 (m, 2H;
Pro Hg), 1.91 (m, 1H; Apro Hb), 2.03 (m, 2H; Pro Hb), 2.21 (m, 2H;
1H; H5’), 7.72 (m, 6H; APA H4), 8.05 (d, 3J
A
3
NH), 8.18 (d, J
(H,H)=6.1 Hz, 1H; Apro NH), 9.63+9.72 ppm (2ꢆs, br,
6H; APA NH); 13C NMR (151 MHz, [D6]DMSO, 228C): d=21.1 (C3’),
22.4 (Pro Cg), 32.5 (Pro Cb), 34.8 (C2’), 35.5 (C7’), 37.1+37.2 (Apro
Cb), 45.5 (C8’), 46.2 (Apro Cg), 48.1 (Pro Cd), 51.9+52.0 (Apro Cd),
60.6+61.5 (Apro Ca), Pro Ca), 115.8+116.0+116.1 (APA C3), 119.7
(APA C5), 122.0 (C5’), 139.0+139.1 (APA C4), 145.7 (C4’), 148.5+148.6
(APA C2), 151.4+151.5+151.9+152.0 (APA C6), 165.9+166.0+166.1
(APA CO), 169.4 (C6’) 170.6 (Apro CO), 171.0+171.1 (Pro CO),
171.5 ppm (C1’); MS (MALDI-TOF): m/z (%): 1510.6 (35) [M+H]+,
1532.6 (100) [M+Na]+, 1548.6 (46) [M+K]+; elemental analysis calcd
(%): for C74H75N23O14·8H2O: C 53.72, H 5.54, N 19.47; found: C 53.80, H
5.52, N 19.20.
PA H2), 2.28 (m, 2H; PA H3), 2.56 (m, 2H; Pro Hb), 2.69 (t, 4J
ACTHNUTRGNE(NUG H,H)=
2.5 Hz, 1H; PA H5), 2.88 (m, 1H; Apro Hb), 3.37 (1H; Apro Hd), 3.58
(m, 2H; Pro Hd), 3.69 (m, 2H; Pro Hd), 3.90 (m, 1H; Apro Hd), 4.32
(m, 1H; Apro Hg), 5.55 (m, 1H; Apro Ha), 5.59 (m, 2H; Pro Ha), 7.22
3
3
(2ꢆd, J
APA H3), 7.41 (2ꢆd, 2ꢆ1H; 3J
(H,H)=7.6 Hz, 1H; APA H5), 7.70–7.76 (m, 3H; APA H4), 8.12 (d, 3J-
(H,H)=6.4 Hz, 1H; Apro NH), 9.69+9.71+9.77 ppm (3ꢆs, 3ꢆ1H;
G
ACHTUNGTREN(NGNU H,H)=8.2 Hz, 1H;
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
APA NH); 13C NMR (151 MHz, [D6]DMSO, 228C): d=14.1 (AP C3),
22.4 (Pro Cg), 32.6 (Pro Cb), 34.2 (AP C2), 37.4 (Apro Cb), 46.3 (Apro
Cg), 48.2 (Pro Cd), 51.9 (Apro Cd), 60.6 (Apro Ca), 61.5 (Pro Ca), 71.3
(AP C5), 83.6 (AP C4), 115.6+115.7+115.9 (APA C3), 119.7 (APA C5),
139+139.1+139.2 (APA C4), 148.6+148.7 (APA C2), 151.5+152.0+
152.1 (APA C6), 166.0+166.1 (APA CO), 170.6+170.7 (Apro CO, AP
CO), 171.0+171.1 ppm (Pro CO); MS (MALDI-TOF): m/z (%): 747.3
(100) [M+H]+, 769.3 (54) [M+Na]+, 785.3 (21) [M+K]+; elemental anal-
ysis calcd (%): for C38H38N10O7·4H2O: C 55.74, H 5.66, N 17.12; found: C
55.59, H 5.79, N 16.67.
Acknowledgements
The generous funding of this study by the Deutsche Forschungsgemein-
schaft is gratefully acknowledged. Thanks are also due to Mr. Fabian
Menges for help with ESI mass spectrometry. We sincerely thank the
ANKA synchrotron facility (Karlsruhe) for the provision of beamtime.
[1] a) J.-M. Lehn, Supramolecular Chemistry, VCH; Weinheim, 1995;
b) D. J. Cram, J. M. Cram, Container Molecules and their Guests,
Monographs in Supramolecular Chemistry, RSC, Cambridge, 1994;
c) A. Collet, J. P. Dutasta, B. Lozach, J. Canceill, Top. Curr. Chem.
1993, 165, 103–129.
Monoazide 11b: This peptide was prepared analogously to 9a from
10b·HCl and active ester 8 by starting from 10b·HCl (0.2 mmol). For the
chromatographic purification, dichloromethane/methanol (5:1) was used
1
as the eluent (133 mg, 87%). M.p. 2108C (decomp); H NMR (600 MHz,
[D6]DMSO, 228C): d=1.81 (m, 2H; Pro Hg), 1.86 (m, 2H; Pro Hg), 1.95
(m, 1H; Apro Hb), 2.02 (m, 2H; Pro Hb), 2.31 (m, 2H; AzP H2), 2.56
(m, 2H; Pro Hb), 2.88 (m, 1H; Apro Hb), 3.39 (m, 1H; Apro Hd), 3.43
[2] a) C. Thilgen, F. Diederich, Chem. Rev. 2006, 106, 5049–5135;
b) G. C. Vougioukalakis, M. M. Roubelakis, M. Orfanopoulos,
Chem. Soc. Rev. 2010, 39, 817–844.
[3] a) A. Jasat, J. C. Sherman, Chem. Rev. 1999, 99, 931–967; b) J. C.
Sherman, Tetrahedron 1995, 51, 3395–3422.
[4] a) M. D. Pluth, K. N. Raymond, Chem. Soc. Rev. 2007, 36, 161–171;
b) L. C. Palmer, J. Rebek, Jr., Org. Biomol. Chem. 2004, 2, 3051–
3059; c) C. Schmuck, Angew. Chem. 2007, 119, 5932–5935; Angew.
Chem. Int. Ed. 2007, 46, 5830–5833.
(t, 3J
ACHTUNGTRENNUNG(H,H)=6.3 Hz, 2H; AzP H3), 3.58 (m, 2H; Pro Hd), 3.69 (m, 2H;
Pro Hd), 3.90 (m, 1H; Apro Hd), 4.33 (m, 1H; Apro Hg), 5.58 (m, 1H;
Apro Ha), 5.63 (m, 2H; Pro Ha), 7.23 (m, 2H; APA H3), 7.27 (d, 3J-
ACHTUNGTRENNUNG ACHTUNGTRENNUNG
(H,H)=8.2 Hz, 1H; APA H3), 7.41 (d, 3J
7.42 (d, 3J
N
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
Apro NH), 9.74 (s, br, 2H; APA NH), 9.82 ppm (s, br, 1H; APA NH);
13C NMR (151 MHz, [D6]DMSO, 228C): d=22.4 (Pro Cg), 32.6 (Pro Cb),
34.6 (AzP C2), 37.4 (Apro Cb), 46.3 (Apro Cg), 46.8 (AzP C3), 48.2 (Pro
Cd), 51.9 (Apro Cd), 60.5 (Apro Ca), 61.5 (Pro Ca), 115.5+115.7+115.9
(APA C3), 119.7 (APA C5), 139.1+139.2 (APA C4), 148.6+148.7 (APA
C2), 151.6+152.0+152.1 (APA C6), 166.0+166.1 (APA CO), 169.9 (AzP
CO), 170.6 (Apro CO), 171.1 ppm (Pro CO); IR (KBr): n˜ =2102 cmÀ1
(m) (azide); MS (MALDI-TOF): m/z (%): 738.6 (100) [MÀN2 +H2 +H]+,
[5] a) R. Warmuth, J. Yoon, Acc. Chem. Res. 2001, 34, 95–105; b) P.
Mal, B. Breiner, K. Rissanen, J. R. Nitschke, Science 2009, 324,
1697–1698; c) M. Yoshizawa, J. K. Klosterman, M. Fujita, Angew.
Chem. 2009, 121, 3470–3490; Angew. Chem. Int. Ed. 2009, 48, 3418–
3438; d) J. Rebek, Jr., Chem. Commun. 2007, 2777–2789; e) S. Ta-
shiro, M. Kobayashi, M. Fujita, J. Am. Chem. Soc. 2006, 128, 9280–
9281; f) S. Tashiro, M. Tominaga, Y. Yamaguchi, K. Kato, M. Fujita,
Angew. Chem. 2006, 118, 247–250; Angew. Chem. Int. Ed. 2006, 45,
Chem. Eur. J. 2010, 16, 7241 – 7255
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7253

S. Kubik et al.
241–244; g) S. Tashiro, M. Tominaga, Y. Yamaguchi, K. Kato, M.
Fujita, Chem. Eur. J. 2006, 12, 3211–3217; h) Y. Hatakeyama, T.
Sawada, M. Kawano, M. Fujita, Angew. Chem. 2009, 121, 8851–
8854; Angew. Chem. Int. Ed. 2009, 48, 8695–8698; i) J. Rebek, Jr.,
Angew. Chem. 2005, 117, 2104–2115; Angew. Chem. Int. Ed. 2005,
44, 2068–2078.
[13] For reviews on anion recognition, see: a) C. Caltagirone, P. A. Gale,
Chem. Soc. Rev. 2009, 38, 520–563; b) S. Kubik, Chem. Soc. Rev.
2009, 38, 585–605; c) Special Issues on Anion Coordination: Coord.
Chem. Rev. 2006, 250, 2918–3244; Coord. Chem. Rev. 2003, 240, 1–
226; d) J. L. Sessler, P. A. Gale, W.-S. Cho, Anion Receptor Chemis-
try, RSC, Cambridge, 2006; e) P. D. Beer, P. A. Gale, Angew. Chem.
2001, 113, 502–532; Angew. Chem. Int. Ed. 2001, 40, 486–516; f) A.
Bianchi, K. Bowman-James, E. Garcꢈa-EspaÇa, Supramolecular
Chemistry of Anions, Wiley, New York, 1997; g) F. P. Schmidtchen,
M. Berger, Chem. Rev. 1997, 97, 1609–1646.
[6] S. J. Rowan, S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders, J. F.
Stoddart, Angew. Chem. 2002, 114, 938–993; Angew. Chem. Int. Ed.
2002, 41, 898–952.
[7] a) N. Nishimura, K. Kobayashi, Angew. Chem. 2008, 120, 6351–
6354; Angew. Chem. Int. Ed. 2008, 47, 6255–6258; b) D. Xu, R. War-
muth, J. Am. Chem. Soc. 2008, 130, 7520–7521; c) X. Liu, Y. Liu, G.
Li, R. Warmuth, Angew. Chem. 2006, 118, 915–918; Angew. Chem.
Int. Ed. 2006, 45, 901–904; d) X. Liu, R. Warmuth, J. Am. Chem.
Soc. 2006, 128, 14120–14127; e) K. R. West, K. D. Bake, S. Otto,
Org. Lett. 2005, 7, 2615–2618; f) M. Fujita, M. Tominaga, A. Hori,
B. Therrien, Acc. Chem. Res. 2005, 38, 369–380; g) S. Leininger, B.
Olenyuk, P. J. Stang, Chem. Rev. 2000, 100, 853–908; h) D. L. Cauld-
er, K. N. Raymond, Acc. Chem. Res. 1999, 32, 975–982; i) B. Icli, N.
Christinat, J. Tçnnemann, C. Schꢁttler, R. Scopelliti, K. Severin, J.
Am. Chem. Soc. 2009, 131, 3154–3155; j) N. Christinat, R. Scopelliti,
K. Severin, Angew. Chem. 2008, 120, 1874–1878; Angew. Chem. Int.
Ed. 2008, 47, 1848–1852; k) R. Zadmard, M. Junkers, T. Schrader,
T. Grawe, A. Kraft, J. Org. Chem. 2003, 68, 6511–6521; l) G. V. Osh-
ovsky, D. N. Reinhoudt, W. Verboom, J. Am. Chem. Soc. 2006, 128,
5270–5278; m) F. Corbellini, R. M. A. Knegtel, P. D. J. Grootenhuis,
M. Crego-Calama, D. N. Reinhoudt, Chem. Eur. J. 2005, 11, 298–
307; n) F. Hof, S. L. Craig, C. Nuckolls, J. Rebek, Jr., Angew. Chem.
2002, 114, 1556–1578; Angew. Chem. Int. Ed. 2002, 41, 1488–1508;
o) J. Rebek, Jr., Chem. Commun. 2000, 637–643; p) S. Liu, B. C.
Gibb, Chem. Commun. 2008, 3709–3716.
[14] A. Fisher, A. Mann, V. Verma, N. Thomas, R. K. Mishra, R. L. John-
son, J. Med. Chem. 2006, 49, 307–317.
[15] S. Kubik, R. Goddard, J. Org. Chem. 1999, 64, 9475–9486.
[16] M. Meldal, C. W. Tornøe, Chem. Rev. 2008, 108, 2952–3015.
[17] For a review about the use of sulfate anions as templates in the syn-
thesis of macrocycles, capsules, interpenetrated, and interlocked
structures, see: K. M. Mullen, P. D. Beer, Chem. Soc. Rev. 2009, 38,
1701–1713.
[18] a) Y. Li, A. H. Flood, Angew. Chem. 2008, 120, 2689–2692; Angew.
Chem. Int. Ed. 2008, 47, 2649–2652; b) Y. Li, A. H. Flood, J. Am.
Chem. Soc. 2008, 130, 12111–12122; c) Y. Li, M. Pink, J. A. Karty,
A. H. Flood, J. Am. Chem. Soc. 2008, 130, 17293–17295; d) H. Ju-
warker, J. M. Lenardt, D. M. Pham, S. L. Craig, Angew. Chem. 2008,
120, 3800–3803; Angew. Chem. Int. Ed. 2008, 47, 3740–3743;
e) R. M. Meudtner, S. Hecht, Angew. Chem. 2008, 120, 5004–5008;
Angew. Chem. Int. Ed. 2008, 47, 4926–4930; f) H. Juwarker, J. M.
Lenhardt, J. C. Castillo, E. Zhao, S. Krishnamurthy, R. M. Jamiol-
kowski, K.-H. Kim, S. L. Craig, J. Org. Chem. 2009, 74, 8924–8934;
g) M. G. Fisher, P. A. Gale, J. R. Hiscock, M. B. Hursthouse, M. E.
Light, F. P. Schmidtchen, C. C. Tong, Chem. Commun. 2009, 3017–
3019.
[19] Macromodel 9.0 with Maestro 7.0 interface, Schrçdinger, Inc.; struc-
tural optimization: Monte-Carlo search by using the MMFF94S
force-field with GB/SA water model, 5000 steps; molecular dynam-
ics: 100 ps, 258C, a snapshot of the structure was taken every 10 ps.
[20] a) F. P. Schmidtchen in Analytical Methods in Supramolecular
Chemistry (Ed.: C. A. Schalley), Wiley-VCH; Weinheim, 2007,
pp. 55–78; b) V. D. Jadhav, E. Herdtweck, F. P. Schmidtchen, Chem.
Eur. J. 2008, 14, 6098–6107; c) V. D. Jadhav, F. P. Schmidtchen, J.
Org. Chem. 2008, 73, 1077–1087; d) V. D. Jadhav, F. P. Schmidtchen,
Org. Lett. 2006, 8, 2329–2332; e) V. D. Jadhav, F. P. Schmidtchen,
Org. Lett. 2005, 7, 3311–3314.
[21] a) M. Berger, F. P. Schmidtchen, Angew. Chem. 1998, 110, 2840–
2842; Angew. Chem. Int. Ed. 1998, 37, 2694–2696; b) M. Berger,
F. P. Schmidtchen, J. Am. Chem. Soc. 1999, 121, 9986–9993.
[22] Y. Zhang, P. S. Cremer, Curr. Opin. Chem. Biol. 2006, 10, 658–663.
[23] G. A. Jeffrey, W. Saenger , Hydrogen Bonding in Biological Struc-
tures, Springer, Berlin, 1991.
[8] a) A. Shivanyuk, J. Rebek, Jr., Proc. Natl. Acad. Sci. USA 2001, 98,
7662–7665; b) F. C. Tucci, D. M. Rudkevich, J. Rebek, Jr., J. Am.
Chem. Soc. 1999, 121, 4928–4929; c) M. Fujita, Chem. Soc. Rev.
1998, 27, 417–425.
[9] a) R. Custelcean, J. Bosano, P. V. Bonnesen, V. Kertesz, B. P. Hay,
Angew. Chem. 2009, 121, 4085–4089; Angew. Chem. Int. Ed. 2009,
48, 4025–4029; b) B. Wu, J. Liang, J. Yang, C. Jia, X.-J. Yang, H.
Zhang, N. Tang, C. Janiak, Chem. Commun. 2008, 1762–1764; c) Y.
Li, K. M. Mullen, T. D. W. Claridge, P. J. Costa, V. Felix, P. D. Beer,
Chem. Commun. 2009, 7134–7136; d) B. Kuberski, A. Szumna,
Chem. Commun. 2009, 1959–1961; e) A. Szumna, Chem. Commun.
2009, 4191–4193; f) A. Szumna, Chem. Eur. J. 2009, 15, 12381–
12388; g) S. M. Butterfield, J. Rebek, Jr., J. Am. Chem. Soc. 2006,
128, 15366–15367; h) J. L. Atwood, L. J. Barbour, A. Jerga, Proc.
Natl. Acad. Sci. USA 2002, 99, 4837–4841; i) P. L. Wash, A. R.
Renslo, J. Rebek, Jr., Angew. Chem. 2001, 113, 1261–1262; Angew.
Chem. Int. Ed. 2001, 40, 1221–1222; j) S. O. Kang, M. A. Hossain,
K. Bowman-James, Coord. Chem. Rev. 2006, 250, 3038–3052; k) P.
Mateus, R. Delgado, P. Brand¼o, S. Carvalho, V. Fꢇlix, Org. Biomol.
Chem. 2009, 7, 4661–4673.
[10] a) S. Kubik, R. Goddard, R. Kirchner, D. Nolting, J. Seidel, Angew.
Chem. 2001, 113, 2722–2725; Angew. Chem. Int. Ed. 2001, 40, 2648–
2651; b) S. Kubik, R. Kirchner, D. Nolting, J. Seidel, J. Am. Chem.
Soc. 2002, 124, 12752–12760; c) S. Otto, S. Kubik, J. Am. Chem.
Soc. 2003, 125, 7804–7805; d) Z. Rodriguez-Docampo, S. I. Pascu, S.
Kubik, S. Otto, J. Am. Chem. Soc. 2006, 128, 11206–11210; e) C.
Reyheller, B. P. Hay, S. Kubik, New J. Chem. 2007, 31, 2095–2102.
[11] a) V. D. Bock, H. Hiemstra, J. H. van Maarseveen, Eur. J. Org.
Chem. 2006, 51–68; b) H. C. Kolb, M. G. Finn, K. B. Sharpless,
Angew. Chem. 2001, 113, 2056–2075; Angew. Chem. Int. Ed. 2001,
40, 2004–2021; c) J. E. Moses, A. D. Moorhouse, Chem. Soc. Rev.
2007, 36, 1249–1262.
[24] M. Yoshizawa, T. Kusukawa, M. Kawano, T. Ohhara, I. Tanaka, K.
Kurihara, N. Niimura, M. Fujita, J. Am. Chem. Soc. 2005, 127, 2798–
2799.
[25] Calculated by using the program PLATON (probe radius 1.2 ꢃ, grid
0.1 ꢃ), A. L. Spek, Acta Crystallogr. Sect. D 2009, 65, 148–155.
[26] a) M. C. Chervenak, E. J. Toone, J. Am. Chem. Soc. 1994, 116,
10533–10539; b) G. I. Makhatadze, G. M. Clore, A. M. Gronenborn,
Nat. Struct. Biol. 1995, 2, 852–855; c) M. M. Lopez, G. I. Makha-
tadze, Biophys. Chem. 1998, 74, 117–125; d) F. P. Schmidtchen,
Chem. Eur. J. 2002, 8, 3522–3529; e) S. Scheiner, M. Cuma, J. Am.
Chem. Soc. 1996, 118, 1511–1521; f) N. Muller, J. Solution Chem.
1991, 20, 669–680.
[27] Y. Marcus, Chem. Rev. 2007, 107, 3880–3897.
[28] For a more recent example, see: M. Lobert, H. Bandmann, U. Bur-
kert, U. P. Bꢁchele, V. Podsadlowski, F.-G. Klꢀrner, Chem. Eur. J.
2006, 12, 1629–1641.
[12] a) J. Morales-Sanfrutos, M. Ortega-MuÇoz, J. Lopez-Jaramillo, F.
Hernandez-Mateo, F. Santoyo-Gonzalez, J. Org. Chem. 2008, 73,
7772–7774; b) K. Banert, J. Wutke, T. Rꢁffer, H. Lang, Synthesis
2008, 2603–2609; c) V. Steinmetz, F. Couty, O. R. P. David, Chem.
Commun. 2009, 343–345.
[29] C. L. Perrin, T. J. Dwyer, Chem. Rev. 1990, 90, 935–967.
[30] a) M. Yamanaka, A. Shivanyuk, J. Rebek, Jr., J. Am. Chem. Soc.
2004, 126, 2939–2943; b) J. Nakazawa, Y. Sakae, M. Aida, Y.
Naruta, J. Org. Chem. 2007, 72, 9448–9455; c) B.-Y. Wang, S. Rieth,
J. D. Badjic, J. Am. Chem. Soc. 2009, 131, 7250–7252.
7254
www.chemeurj.org
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 7241 – 7255

Cage Receptor for Sulfate Anions
FULL PAPER
[31] a) E. Garcꢈa-EspaÇa, P. Dꢈaz, J. M. Llinares, A. Bianchi, Coord.
Chem. Rev. 2006, 250, 2952–2986; b) K. Worm, F. P. Schmidtchen,
A. Schier, A. Schꢀfer, M. Hesse, Angew. Chem. 1994, 106, 360–362;
Angew. Chem. Int. Ed. Engl. 1994, 33, 327–329; c) C. Olivier, Z.
Grote, E. Solari, R. Scopelliti, K. Severin, Chem. Commun. 2007,
4000–4002; d) R. Vilar, Angew. Chem. 2003, 115, 1498–1516;
Angew. Chem. Int. Ed. 2003, 42, 1460–1477; e) O. Hayashida, A.
Shivanyuk, J. Rebek, Jr., Angew. Chem. 2002, 114, 3573–3576;
Angew. Chem. Int. Ed. 2002, 41, 3423–3426.
[33] See the Supporting Information of reference [10a].
[34] J. H. Boyer, J. Am. Chem. Soc. 1951, 73, 5248–5252.
[35] a) L. Palatinus, G. Chapuis, J. Appl. Crystallogr. 2007, 40, 786–790;
b) S. van Smaalen, L. Palatinus, M. Schneider, Acta Crystallogr. Sect.
A 2003, 59, 459–469; c) G. M. Sheldrick, Acta Crystallogr. Sect. A
2008, 64, 112–122.
[36] EXSYCalc, Version 1.0, Mestrelab Research, free download avail-
able from www.mestrelab.com.
[32] D. J. Cram, M. E. Tanner, C. B. Knobler, J. Am. Chem. Soc. 1991,
113, 7717–7727.
Received: February 4, 2010
Published online: May 12, 2010
Chem. Eur. J. 2010, 16, 7241 – 7255
ꢂ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
7255