Oxoanion binding by flexible guanidiniocarbonyl pyrrole-ammonium bis-cations in water

Article abstract of DOI:10.1021/jo070981z

(Chemical Equation Presented) The syntheses of several bis-cations 1-5 with a simple primary ammonium cation attached via flexible linkers of varying length to a guanidiniocarbonyl pyrrole oxo anion binding site are reported. In UV-binding studies in aqueous buffer solution these bis-cations showed efficient binding of various N-acetyl amino acid carboxylates. However, complex affinity is significantly depending on both the anion and the length of the linker in the bis-cation. With increasing linker length, complex stability first increases until an optimum is reached for bis-cation 3 with a C4-linker. Then the complex stability decreases again. The best binding substrate in this series is N-acetyl phenyl alanine, most likely due to additional cation-π-interactions between the aromatic ring and the guanidiniocarbonyl pyrrole cation. The formation of the complex between bis-cation 3 and N-acetyl phenyl alanine carboxylate was investigated further by fluorescence titrations and NOE studies, as well as molecular mechanics calculations.

Full text of DOI:10.1021/jo070981z

Oxoanion Binding by Flexible Guanidiniocarbonyl
Pyrrole-Ammonium Bis-Cations in Water
Carsten Schmuck* and Volker Bickert
UniVersita¨t Wu¨rzburg, Institut fu¨r Organische Chemie, Am Hubland, 97074 Wu¨rzburg, Germany
schmuck@chemie.uni-wuerzburg.de
ReceiVed May 15, 2007
The syntheses of several bis-cations 1-5 with a simple primary ammonium cation attached via flexible
linkers of varying length to a guanidiniocarbonyl pyrrole oxo anion binding site are reported. In UV-
binding studies in aqueous buffer solution these bis-cations showed efficient binding of various N-acetyl
amino acid carboxylates. However, complex affinity is significantly depending on both the anion and the
length of the linker in the bis-cation. With increasing linker length, complex stability first increases until
an optimum is reached for bis-cation 3 with a C4-linker. Then the complex stability decreases again. The
best binding substrate in this series is N-acetyl phenyl alanine, most likely due to additional cation-π-
interactions between the aromatic ring and the guanidiniocarbonyl pyrrole cation. The formation of the
complex between bis-cation 3 and N-acetyl phenyl alanine carboxylate was investigated further by
fluorescence titrations and NOE studies, as well as molecular mechanics calculations.
Introduction
weakened in more polar solvents.4 To achieve strong complex-
ation also in aqueous solvents, metal ligand interactions are often
used as their strength can approach that of covalent bonds.5
However, the use of metals can have other disadvantages such
as unfavorable complexation kinetics or bioincompatibility if
toxic metals are used. Another possibility is to use hydrophobic
contacts6 or aromatic interactions7 which are especially strong
The vast majority of supramolecular anion complexation1 has
been and still is based on directed electrostatic interactions,
namely H-bonds2 and ion pair formation,3 respectively. This,
however, often limits such host-guest systems to organic
solvents of low polarity as due to the competitive solvation of
donor and acceptor sites H-bonds and ion pairs are significantly
(4) For two recent review articles on supramolecular recognition in water
see: (a) Oshovsky, G. V.; Reinhoudt, D. N.; Verboom, W. Angew. Chem.,
Int. Ed. 2007, 46, 2393. (b) Kubik, S.; Reyheller, C.; Stuewe, S. J. Incl.
Phenom. Macro. Chem. 2005, 52, 137.
(5) Some recent examples for the use of metal-ligand interactions in
supramolecular chemistry can be found in: (a) Schubert, U. S.; Hofmeier,
H. Chem. Commun. 2005, 2423. (b) Buryak, A.; Severin, K. Angew. Chem.,
Int. Ed. 2004, 43, 4771. (c) Imai, H.; Munakata, H.; Uemori, Y.; Sakura,
N. Inorg. Chem. 2004, 43, 1211. (d) Wright, A. T.; Anslyn, E. V. Org.
Lett. 2004, 6, 1341. (e) Sirish, M.; Chertkov, V.; Schneider, H.-J. Chem.
Eur. J. 2002, 8, 1181. (f) Ogoshi, H.; Mizutani, T. Acc. Chem. Res. 1998,
31, 81.
* Address correspondence to this author. Fax: +49 931 8884625.
(1) (a) Fabbrizzi, L.; Amendola, V.; Bonizzoni, M.; Esteban-Go´mez, D.;
Licchelli, M.; Sanceno´n, F.; Taglietti, A. Coord. Chem. ReV. 2006, 250,
1451. (b) Bowman-James, K. Acc. Chem. Res. 2005, 38, 671-678.
(2) (a) Prins, L. J.; Reinhoudt, D. N.; Timmermann, P. Angew. Chem.,
Int. Ed. 2001, 40, 2382. (b) Zimmermann, S. C.; Corbin P. S. Struct. Bond.
2000, 63. (c) Rebek, J., Jr. Acc. Chem. Res. 1999, 32, 278. (d) Meot-Ner,
M. Chem. ReV. 2005, 105, 213. (e) Cooke, G.; Rotello, V. M. Chem. Soc.
ReV. 2002, 31, 275.
(3) Mainly by using organic cations such as guanidinium or amidinium
cations for anion binding: (a) Schug, K. A.; Lindler, W. Chem. ReV. 2005,
105, 67. (b) Houk, R. J. T.; Tobey, S. L.; Anslyn, E. V. Top. Curr. Chem.
2005, 255, 199. (c) Blondeau, P.; Segura, M.; Perez-Fernandez, R.; de
Mendoza, J. Chem. Soc. ReV. 2007, 36, 198. (d) Schmidtchen, F. P.; Berger,
M. Chem. ReV. 1997, 97, 1609.
(6) For reviews on hydrophobic interations see: (a) Widom, B.;
Bhimalapuram, P.; Koga, K. Phys. Chem. Chem. Phys. 2003, 5, 3085. (b)
Pratt, L. R.; Pohorille, A. Chem. ReV. 2002, 102, 2671. (c) Southall, N. T.;
Dill, K. A.; Haymet, A. D. J. J. Phys. Chem. B 2002, 106, 521.
10.1021/jo070981z CCC: $37.00 © 2007 American Chemical Society
Published on Web 08/02/2007
6832
J. Org. Chem. 2007, 72, 6832-6839

Syntheses of SeVeral Bis-Cations
in aqueous solvents. They are, however, difficult to use
deliberately as they are in general less directional and less
specific than, e.g., H-bonds, respectively. Therefore the selective
complexation of a given substrate solely based on solvophobic
contacts in aqueous solvents is rather difficult. Furthermore,
substrates which allow for extensive hydrophobic interactions
often have only limited solubility in aqueous solvents. With the
development of more efficient binding motifs, it was demon-
strated in recent years that also electrostatic interactions can be
used for substrate binding in water.8 In general, several such
weak electrostatic interactions have to be combined to achieve
a strong and selective substrate binding under these competitive
conditions (“Gulliver-Effect”).2a
amino acid side chain with a pronounced optimum for the
binding of phenylalanine by the longer bis-cation 3 with a C4
spacer.
In this context, we are exploring how a clustering of ion pairs
can be used for oxoanion binding in aqueous solvents.9 For
example, we recently showed that tetrapeptides with a free
carboxylate terminus can be efficiently recognized in water by
multi-cationic receptors solely based on electrostatic interac-
tions.10 In one of our works we have quantitatively screened a
combinatorial receptor library derived from a tripeptide with
an appended tailor-made guanidiniocarbonyl pyrrole carboxylate
binding site11,12 for the binding of the polar tetrapeptide N-Ac-
D-Glu-Lys-D-Ala-D-Ala.13 From these screenings, it appears that
the guanidiniocarbonyl pyrrole anion binding site could be
further improved by an additional second positive charge, which
can simultaneously interact with the bound carboxylate thereby
strengthing the complex. Whenever a lysine was directly next
to the guanidiniocarbonyl pyrrole cation, complex stability
significantly increased compared to that of other, non-cationic
amino acids in this position.14 We have therefore now system-
atically investigated the influence of a second charge on
oxoanion binding by guanidiniocarbonyl pyrrole cations. For
this purpose, we have synthesized a series of flexible guani-
diniocarbonyl pyrrole-ammonium bis-cations 1-5 in which the
distance between the ammonium cation and the guanidiniocar-
bonyl pyrrole cation varies from two to five carbons. These
bis-cations were then tested for their binding affinity toward
amino acid carboxylates and other oxoanions by using UV
titrations in buffered water. We find that complex stability
significantly depends on both the length of the linker and the
Results and Discussion
Synthesis of Receptors 1-4. The general structure of bis-
cations 1-4 is shown in Figure 1: a guanidiniocarbonyl pyrrole
cation is attached to a diamino alkane with a chain length
varying from two to five carbons. The synthesis of these cations
is described in Scheme 1: Pyrrole dicarboxylic acid monobenzyl
ester 6 was reacted with tBoc-guanidine 7 with PyBOP as the
coupling reagent (93% yield). The benzyl ester in 8 was
quantitatively cleaved off by hydrogenation with a catalytic
amount of palladium on activated charcoal. The resulting
carboxylic acid 9 was then reacted with the mono-tBoc protected
diamines 10 with PyBOB as the coupling reagent. The yields
varied from 52% to 74% yield for the different diamines. Finally,
the remaining protecting groups were cleaved off by acidic
treatment with concentrated hydrochloric acid in methanol. After
lyophilization the desired bis-cationic receptors 1-4 were
obtained as their hydrochloride salts.
The lysine derivative 5 was synthesized accordingly (Scheme
2). The pyrrole carboxylic acid 9 was reacted with H-Lys(Cbz)-
OMe 12 with PyBOP as the coupling reagent in a mixture of
DCM and DMF (yield 86%). Removal of the Cbz-protecting
group with H2/Pd/C provided the desired receptor 5 as the
chloride salt after lyophilization from aqueous HCl in methanol.
The moderate yield (60%) of the cleavage reaction is due to
the fact that under these conditions also the methyl ester is
cleaved to some extent, giving the free carboxylic acid as the
byproduct.
Binding Studies. The binding properties of the five bis-
cations 1-5 were investigated by using UV-titration studies15
in buffered water with 10% DMSO added to increase the
solubility of the N-acetyl amino acid carboxylate used as
substrate. The bis-cations (c ) 7 × 10-5) were dissolved in
bis-tris buffer (c ) 5 × 10-3) and the pH was adjusted to 6.0
to make sure that the receptor is fully protonated and the
substrates are deprotonated. Control studies have shown that
the pKa of the guanidiniocarbonyl pyrrole cation is around 6.5
whereas the ammonium group has a pKa of 9.2-9.7. These
values were determined with use of the Henderson-Hasselbalch
equation by measuring the pH at the half equivalence point of
each host in water. As substrates a variety of oxo-anions (c )
1.5 × 10-3) such as several N-acetyl amino acid carboxylates
as well as sulfate and acetate were used. Aliquots of a stock
solution of the corresponding anion were addedd to the solution
of the bis-cation and the changes in the UV spectra were
(7) For reviews on aromatic interactions in supramolecular model systems
see: (a) Waters, M. L. Curr. Opin. Chem. Biol. 2002, 6, 736. (b) Hunter,
C. A.; Lawson, K. R.; Perkins, J.; Urch, C. J. J. Chem. Soc., Perkin Trans.
2001, 2, 651.
(8) Schmuck, C.; Heil, M.; Scheiber, J.; Baumann, K. Angew. Chem.,
Int. Ed. 2005, 44, 7208.
(9) (a) Schmuck, C. Coord. Chem. ReV. 2006, 250, 3053. (b) Schmuck,
C.; Geiger, L. Curr. Org. Chem. 2003, 7, 1485.
(10) For recent examples of peptide receptors by other groups that
function in water see: (a) Krattiger, P.; Wennemers, H. Synlett 2005, 706.
(b) Schmuck, C.; Wich, P. Angew. Chem., Int. Ed. 2006, 45, 4277. (c)
Shepherd, J.; Gale, T.; Jensen, K. B.; Kilburn, J. D. Chem. Eur. J. 2006,
12, 713. (d) Rodriguez-Docampo, Z.; Pascu, S. I.; Kubik, S.; Otto, S. J.
Am. Chem. Soc. 2006, 128, 11206. (e) Darbre, T.; Reymond, J.-L. Acc.
Chem. Res. 2006, 39, 925.
(11) (a) Schmuck, C. Chem. Commun. 1999, 843. (b) Schmuck, C. Chem.
Eur. J. 2000, 6, 709.
(12) For recent examples on guanidinium-based receptors see: (a)
Schmuck, C.; Schwegmann, M. J. Am. Chem. Soc. 2005, 127, 3373. (b)
Schmuck, C.; Machon, U. Chem. Eur. J. 2005, 11, 1109. (c) Schmuck, C.;
Dudaczek, J. Tetrahedron Lett. 2005, 46, 7101. (d) Suhs, T.; Ko¨nig, B.
Chem. Eur. J. 2006, 12, 8150. (e) Schmuck, C.; Schwegmann, M. Org.
Biomol. Chem. 2006, 4, 836. (f) Mazik, M.; Cavga, H. J. Org. Chem. 2007,
72, 831.
(13) (a) Schmuck, C.; Heil, M. Org. Biomol. Chem. 2003, 1, 633. (b)
Schmuck, C.; Heil, M. ChemBioChem. 2003, 4, 1232.
(14) Schmuck, C.; Heil, M. Chem. Eur. J. 2006, 12, 1339.
(15) (a) Hirose, K. J. Inclusion Phenom. Macrocycl. Chem. 2001, 39,
193. (b) Connors, K. A. Binding Constants; Wiley: New York, 1987.
J. Org. Chem, Vol. 72, No. 18, 2007 6833

Schmuck and Bickert
SCHEME 1. Synthesis of the Bis-Cationic Receptors 1-4
SCHEME 2. Synthesis of the Lysine Derivative 5
FIGURE 1. Top: Change of the UV absorption in the course of the
titration of receptor 3 with N-acetyl phenyl alanine carboxylate.
Bottom: Resulting binding isotherm at λ ) 300 nm used to calculate
the complex stability based on a nonlinear curve-fitting procedure. The
dashed line represents the expected change in absorbance if only a
dilution of the sample had taken place.
recorded after each addition. Interaction of the anion with the
guanidiniocarbonyl pyrrole cation causes a distinct decrease in
the molar absorbance at 300 nm, which can be used to follow
the complexation quantitatively. From this decrease in UV
absorbance the binding constants were calculated by using a
nonlinear curve-fitting procedure for a 1:1 complexation taking
into account the dilution of the sample during titration. All
measurements were repeated twice, and the results were identical
within the experimental error of the method. A representative
binding isotherm is shown in Figure 1. The complex stoichi-
ometry was independently confirmed by a Job plot16 (Figure 2)
as well as ESI-MS experiments. The resulting binding constants
are summarized in Table 1 and in Figure 3.
Data Interpretation. As the data show, the binding of
oxoanions by these bis-cations depends quite significantly on
the length of the linker. One might expect a steady decrease of
complex stability with increasing length and flexibility of the
linker as has been observed in other cases. It is generally
assumed that freezing of an internal torsion upon guest
complexation reduces complex stability. For example, Schneider
has studied the influence of alkyl spacers on the binding between
flexible bis-ammonium cations and bis-anions in water. He
indeed found a steady decrease of complex stability with
increasing linker length.17 However, in this case for all anions
complex stability first increases until a pronounced maximum
in complex stability is reached for bis-cation 3 with a linker
length of four methylene groups (Figure 3). Only afterward does
the complex stability drop again when the linker becomes even
longer. In general, the smaller bis-cations 1 and 2 do not bind
anions very efficiently under these experimental conditions (K
(16) McCarthy, P. Anal. Chem. 1978, 50, 2165.
6834 J. Org. Chem., Vol. 72, No. 18, 2007
(17) Hossain, M. A.; Schneider, H.-J. Chem. Eur. J. 1999, 5, 1284.

Syntheses of SeVeral Bis-Cations
becomes even longer (bis-cation 4 and the lysine derivative 5,
both with five atoms in the linker) complex stability decreases
again. Perhaps now the entropic costs required to orientate the
longer and more flexible linker compensate to some extent the
additional attractive charge interaction. These entropic costs
should be larger in 4 and 5 than in 3.18
Among the different anions, N-acetyl phenyl alanine car-
boxylate is bound much more efficiently than the other amino
acid anions by all bis-cations (except the lysine derivative 5).
Again, the maximum complex stability is seen with bis-cation
3 (Figure 3) with an association constant of K ) 10.700 M-1
,
which is surprisingly large for amino acid binding in aqueous
buffer. This binding constant is nearly twice as large as that
with alanine (K ) 5.800 M-1), for example, and more than five
times larger as that for the binding of valine (K ) 2.200 M-1),
respectively. Valine in general is bound only rather weakly by
these bis-cations. The higher affinity for phenylalanine most
likely reflects additional cation-π-interactions of the aromatic
phenyl ring with the guanidinium cation or aromatic stacking
interactions with the pyrrole (vide infra).
FIGURE 2. Job plot for the complex formation between receptor 3
and N-acetyl phenyl alanine carboxylate confirming a 1:1 complex
stoichiometry.
For the two bis-carboxylates, N-acetyl aspartate and N-acetyl
glutamate, also 1:1 complex stoichiometries were found ac-
cording to Job plots calculated from the titration data. These
two polar amino acids are also bound most efficiently by bis-
cation 3. For example, the complex stability for glutamate by
the shorter bis-cation 2 is about five times smaller than that
with 3 (K ) 1300 vs 6100 M-1, respectively). For the longer
bis-cations 4 and 5 complex stability drops again as seen with
the mono-carboxylates. An interesting aspect is that the shorter
bis-cations 1 and 2 interact more strongly with aspartate than
glutamate whereas 3-5 bind glutamate more strongly than
aspartate (e.g., for 3 K ) 6100 vs 3900 M-1, respectively). This
change in the relative binding affinities is not unreasonable. The
shorter bis-cations (1 and 2) prefer the shorter bis-anion (Asp)
whereas the longer bis-cations (3-5) prefer the longer bis-anion
(Glu). Molecular modeling studies (Macromodel 8.0, Amber*
force field, GB/SA water solvation)19 suggest that the distance
between the two charges in 1 or 2 is not sufficient to allow a
simultaneous strain free interaction with both carboxylates in
glutamate whereas the longer linker in 3 just has the right length
(Figure 4). This could explain the significant increase in complex
stability for glutamate going from bis-cation 2 to 3. For aspartate
with one methylene group less between the two carboxylates
even in bis-cation 1 both positively charged groups can interact
with both carboxylates of the substrate. However, at least
according to the modeling studies the ion pair between the
guanidiniocarbonyl pyrrole cation and the R-carboxylate is
somewhat distorted. The geometry is not the usual planar
bidentate ion pair indicating some strain introduced by the
additional interaction with the second positive charge in the side
chain.
FIGURE 3. Binding constants for bis-cationic receptors 1-5 and
various N-acetyl amino acid carboxylates calculated from UV-titration
experiments in buffered water (Kass in M-1).
TABLE 1. Binding Constants (Kass in M-1) for Complex
Formation between the Bis-Cationic Hosts 1-5 and Several
Oxo-Anions As Determined from UV-Titration Experiments in
Buffered Water
host
ALA
PHE
VAL
GLU
ASP
NaOAc
Na2SO4
1
2
3
4
5
1400
1400
5800
2100
3700
1400
2600
10700
4700
1600
1400
2200
2700
1100
1100
1300
6100
2100
3100
2200
3200
3900
1700
3000
1500
<1000
3600
3300
3300
1300
<1000
1800
2600
2400
The two inorganic anions, acetate and sulfate, were also
studied. They are not efficiently bound by the two shorter bis-
cations 1 and 2 (K < 1500 M-1). The binding affinity
significantly increases for bis-cation 3. However, complex
stability does not vary much for the longer bis-cations 4 and 5.
1850
< 1500 M-1). Complex stability is similar to what is expected
for anion binding by a simple guanidiniocarbonyl pyrrole mono-
cation.9 Most likely, the linker is too short to allow a significant
additional interaction of the ammonium group with the bound
carboxylate. With four carbon atoms in the linker as in 3, the
linker is now long enough to allow a strain-free charge
interaction of the NH3 cation with the anions. If the linker
(18) However, it might also be that the increased flexibility in the host
allows for more favorable intramolecular interactions within the host, which
also weakens intermoelcular substrate binding. See discussion and references
in: Mardis, K. L. J. Phys. Chem. B 2006, 110, 971.
(19) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R. M.
J.; Lipton, M.; Caufiled, C.; Chang, G.; Hendrickson, T.; Still, W. C. J.
Comput. Chem. 1990, 11, 440.
J. Org. Chem, Vol. 72, No. 18, 2007 6835

Schmuck and Bickert
FIGURE 4. Calculated energy minimized structure of the complexes
between receptor 1 and N-acetyl aspartate (top) and between receptor
3 and N-acetyl glutamate (bottom) as obtained from a Monte Carlo
conformational search.
For example, acetate is bound by 3 with K ) 3600 M-1 and by
4 and 5 with K ) 3300 M-1, respectively. The binding of acetate
is stronger than that of sulfate, which most likely reflects the
increased basicity of the former or differences in the solvation
properties of the anions.
Further Studies of Complex 3@Phe. As the complex
between bis-cation 3 and N-acetyl phenyl alanine carboxylate
is significantly more stable than any other complex in this series,
this complex was investigated further. Unfortunately, we were
not able to determine any binding constants using isothermal
microcalorimetry (ITC)20,21 as only very small heat changes were
observed (see the Supporting Information). No reliable data
interpretation was possible based on these small effects. At best
the data hinted to a slightly exothermic and hence mildly
enthalpic and entropic driven association, which would not be
surprising for the combination of both ion pair formation and
cation-π-interactions in aqueous solvents.
FIGURE 5. Top: Change of the fluorescence in the course of the
titration of receptor 3 with N-acetyl phenyl alanine carboxylate by a
synchronous excitation of ∆λ ) 20 nm. Bottom: Resulting binding
isotherm at λ ) 285 nm.
the fluorescence band of the pyrrole moiety of host 3 at λ )
335 nm decreases. After subtracting the changes due to simple
dilution of the sample, one sees that this emission band actually
slightly increases upon complex formation (Figure 5). On the
basis of this binding isotherm the association constant Kass was
calculated by using a nonlinear curve-fitting procedure. With
Kass ) 4600 M-1 the fluorescence titration also indicates the
formation of a rather strong complex between host 3 and
N-acetyl phenyl alanine carboxylate. The somewhat smaller
value compared to the UV titrations might reflect the differences
in the experimental conditions (different concentration of host
and guest as well as the buffer).
We then turned to NMR studies. Upon addition of the
carboxylate to a solution of 3 in water (with 10% DMSO-d6),
shift changes are observed for various protons of 3. For example,
the pyrrole CH protons show distinct high-field shifts as
observed in other cases of oxoanion binding by guanidiniocar-
bonyl pyrrole cations. However, the shift changes are too small
too allow a reliable calculation of a binding constant. Signals
which would be expected to show more significant shift changes
(such as the amide or guanidinium NHs) are not detectable under
these protic conditions due to fast exchange with the solvent.
However, the shift changes again indicate complex formation.
In line with this, a NOESY spectrum (d3-MeOH, 213 K, 300 ms
mixing time) revealed distinct cross-peaks between the phenyl-
We also examined complex formation between 3 and N-acetyl
phenyl alanine using a fluorescence titration experiment measur-
ing light emission from 250 to 400 nm with a synchronous
excitation of ∆λ ) 20 nm (aqueous bis-tris buffer, 10 mM, pH
6.0 with 10% DMSO added; [3]0 ) 7 µM; [guest]0 ) 0.15 mM).
Upon the addition of the guest an intense band at λ ) 285 nm
due to the fluorescence of the phenylalanine occurs, whereas
(20) Schmidtchen, F. P. Isothermal Titration Calorimetry in Supramo-
lecular Chemistry. In Analytical Nethods in Supramolecular Chemistry;
Schalley, C. A., Ed.; Wiley-VCH: New York, 2007.
(21) Some recent examples for the use of ITC in supramolecular
chemistry can be found in: (a) Sessler, J. L.; Gross, D. E.; Cho, W.-S.;
Lynch, V. M.; Schmidtchen, F. P.; Bates, G. W.; Light, M. E.; Gale, P. A.
J. Am. Chem. Soc. 2006, 128, 12281. (b) Schmidtchen, F. P. Org. Lett.
2002, 3, 431.
6836 J. Org. Chem., Vol. 72, No. 18, 2007

Syntheses of SeVeral Bis-Cations
FIGURE 6. Parts of the NOE spectrum for a 1:1 mixture of receptor 3 and N-acetyl phenyl alanine carboxylate in d3-MeOH with selected
intermolecular NOEs (red circles) indicating complex formation.
alanine and bis-cation 3 confirming the formation of a defined
complex (Figure 6). For example, cross-peaks between the
N-acetyl group of the anion and the methylene groups of the
linker in 3 are observed. These methylene groups of the linker
also show cross-peaks with the phenyl alanine amide NH as
well as the phenyl CH protons. The aromatic ring gives cross-
peaks with the linker amide NH of 3 as well as one of the pyrrole
CH protons. Hence, these intermolecular cross-peaks, which are
schematically summarized in Figure 6, are a clear indication
for the formation of a complex with a defined structure.
A possible complex structure was elucidated by using
molecular mechanics calculations (Macromodel 8.0, Amber*
force field, GB/SA water solvation). The energy minimum
structure obtained from a MC conformational search with
100.000 steps is shown in Figure 7. At least according to these
calculations, the aromatic ring of the phenyl alanine indeed
stacks with the guanidiniocarbonyl pyrrole moiety. A similar
effect has already been observed for the binding of N-acetyl
amino acid carboxylates by a simple guanidiniocarbonyl pyrrole
mono-cation.11
by the corresponding NOE signals and the stacking of the phenyl
ring below the guanidiniocarbonyl pyrrole moiety brings it into
close proximity of the pyrrole CH as well as the amide NH
and the methylene groups of the linker. The benzylic methylene
protons, however, point away from the bis-cation and accord-
ingly no intramolecular NOE signals were observed.
Conclusion
It is generally assumed that flexible single bonds in anion
hosts should be avoided as the increasing flexibility should cause
a decrease in complex stability. However, we show here that
within the series of bis-cations 1-5 complex stability first
significantly increases with increasing linker length and flex-
ibility. Whereas the two shortest bis-cations 1 and 2 show only
very weak binding of N-acetyl amino acid carboxylates in
aqueous buffer solution, for the longer bis-cation 3 a significant
increase in complex stability is observed. The complex stability
then drops again for the longer bis-cations 4 and 5. Hence, a
minimum linker length of four carbons is necessary to allow
efficient anion binding by both positive charges. This underlines
that not just unspecific Coulomb interactions are responsible
for anion binding in this case but rather the formation of directed
ion pairs which require a certain relative geometry and arrange-
All the observed intermolecular NOEs (Figure 6) are in good
agreement with the calculated complex structure shown above
in Figure 7. For example, in this structure, the N-acetyl group
of the amino acid is orientated next to the linker, as indicated
J. Org. Chem, Vol. 72, No. 18, 2007 6837

Schmuck and Bickert
atmosphere. The resulting solution was filtered through a celite pad,
which was washed several times with methanol with 2% triethyl-
amine. The solvent was evaporated under reduced pressure. After
adding water (10 mL) to the resulting oil, the solution was
lyophilized giving the product 9 as a white solid (1.89 g, 95%).
1
Mp >300 °C; H NMR (400 MHz, DMSO-d6) δ 1.08 (t, 9 H,
3J(H,H) ) 7.20 Hz, CH3), 1.45 (s, 9 H, CH3), 2.79 (q, 6 H, 3J(H,H)
3
) 7.20 Hz, CH2), 6.47 (d, J(H,H) ) 3.64 Hz, 1 H, CH),6.77 (d,
3J(H,H) ) 3.68 Hz, 1 H, CH), 8.58 (br s, 1 H, NH), 9.31 (br s, 1
H, NH), 10.84 (br s, 1 H, NH); 13C NMR (100 MHz, DMSO-d6)
δ 9.7 (CH3), 27.8 (CH3), 45.2 (CH2), 80.2 (quat C), 112.1, 114.1
(both CH), 128.8, 133.0, 158.5, 160.6, 163.9, 167.2 (all quat C);
FT-IR (KBr disk) ν˜ [cm-1] 3393 (m), 2958 (w), 1650 (s), 1542
(s), 1319 (s); MS (negative ESI) m/e calcd for C12H15N4O5- (M -
H+) 295.11, found 295.1.
General Procedure of the Synthesis of the tBoc-Protected
Receptors 11 and 13. 5-(tBoc-Guanidiniocarbonyl)-1H-pyrrole-
2-carboxylate 9 (397 mg, 1 mmol, 1 equiv) was added to a solution
of PyBOP (780 mg, 1.5 mmol, 1.5 equiv) and triethylamine (278
µL, 2 mmol, 2 equiv) in dry dichloromethane (30 mL) with dry
DMF (2 mL) and the solution was stirred for 15 min. This slightly
yellow suspension was then added dropwise to a solution of an
amine 10/12 (2 mmol, 2 equiv) and triethylamine (278 µL, 2 mmol,
2 equiv) in dichloromethane (30 mL) with dry DMF (2 mL) and
the resulting mixture was stirred at room temperature overnight.
The solvent was evaporated and the residue taken up in water (60
mL) and ethyl ether (90 mL). For better solubility ultrasound was
used. The water layer was extracted with ethyl ether (five times
with 90 mL) and then the combined organic layers were dried with
magnesium sulfate and afterward the solvent was evaporated.
Further purification by filtration column (silica gel, 3:2 dichlo-
romethane/acetone with 1% triethylamine) yielded white to slightly
yellow solids (52% to 86%).
FIGURE 7. Calculated energy minimized structure of the complex
between receptor 3 and N-acetyl phenyl alanine carboxylate as obtained
from a Monte Carlo conformational search. Whereas the carboxylate
is interacting with both cationic groups, the aromatic phenyl ring
π-stacks with the planar guanidiniocarbonyl pyrrole moiety.
General Procedure of the Synthesis of the Receptors 1-4.
The cleavage of the N-protecting group was done twice with the
tBoc-protected receptors 11 (1 mmol, 1 equiv) in 2 mL of concd
hydrochloric acid in 10 mL of methanol. These solutions were each
stirred at room temperature for 2 h. After repeated lyophilization
with 20 mL of water the receptors were obtained as white chloride
salts (95% to quant.).
ment between the charged groups in the host and guest. Hence,
increasing flexibility in the design of anion hosts is not always
detrimental.
Experimental Section
Bis-Cation 1. White solid (52%). Mp 210 °C dec; 1H NMR (400
Synthesis of 5-((tert-Butyloxycarbonyl)guanidiniocarbonyl)-
1H-pyrrole-2-carboxylic Acid Benzylester 8. A mixture of the
1H-pyrrole-2,5-dicarboxylic acid monobenzyl ester 6 (2.45 g, 10.0
mmol, 1 equiv), PyBOP (5.72 g, 11 mmol, 1.1 equiv), and N-methyl
morpholine (2.45 mL, 22 mmol, 2.2 equiv) was stirred in DMF
(50 mL) at room temperature for 30 min then tert-butyloxycarbo-
nylguanidine 7 (3.19 g, 20.0 mmol, 2 equiv) was added and the
resulting solution was stirred over night. The yellow solution was
slowly poured into vigorously stirred water (150 mL) causing the
formation of a white solid that was dissolved in diethyl ether (150
mL). After phase separation and twice extracting the water/DMF
phase with diethyl ether (150 mL) the solvent was evaporated and
the crude product was further purified by flash chromatography
yielding the colorless powder 8 (3.59 g, 93%).
3
MHz, DMSO-d6) δ 2.98 (m, 2 H, CH2), 3.50 (q, J(H,H) ) 5.68
3
Hz, 2 H, CH2), 6.92 (d, J(H,H) ) 2.28 Hz, 1 H, CH), 7.06 (s, 1
H, NH), 7.56 (s, 1 H, CH), 8.06 (br s, 3 H, NH3), 8.48 (br s, 2 H,
3
NH2), 8.67 (br s, 2 H, NH2), 8.83 (t, J(H,H) ) 5.04 Hz), 1 H,
NH), 12.10 (s, 1 H, NH), 12.35 (s, 1 H, NH); 13C NMR (100 MHz,
DMSO-d6) δ 36.6, 38.5 (all CH2), 112.7, 115.9 (both CH), 125.5,
132.5, 155.6, 159.3, 159.7 (all quat C); FT-IR (KBr disk) ν˜ [cm-1
]
3336 (s), 3191 (s), 3005 (m), 1697 (s), 1656 (s), 1558 (s), 1478
(m), 1295 (s), 1257 (w), 1201 (w), 1067 (w), 812 (w), 759 (m),
+
665 (w); HR-MS (positive ESI) m/e calcd for C9H15N6O2 (M +
H+) 239.125, found 239.122; HR-MS (positive ESI) m/e calcd for
+
C9H14N6NaO2 (M + Na+) 261.107, found 261.107.
1
Bis-Cation 2. White solid (63%). Mp >250 °C dec; H NMR
Mp 88 °C; 1H NMR (400 MHz, DMSO-d6) δ 1.46 (s, 9 H, CH3),
5.31 (s, 2 H, CH2), 6.85 (m, 2 H, CH), 7.30-7.46 (m, 5 H, CH),
8.58 (br s, 1 H, NH), 9.31 (br s, 1 H, NH), 10.74 (br s, 1 H, NH),
11.62 (br s, 1 H, NH); 13C NMR (100 MHz, DMSO-d6) δ 27.7
(CH3), 65.5 (CH2), 81.2 (quat C), 113.8, 115.8 (both CH), 124.5
(quat C), 127.8, 128.0, 128.4 (all CH), 134.0, 136.2, 155.6, 158.3,
159.8, 168.0 (all quat C); FT-IR (KBr disk) ν˜ [cm-1] 3393 (m),
3256 (m), 2360 (s), 2340 (m), 1719 (s), 1717 (s), 1635 (s), 1540
(s), 1286 (s), 1149 (s), 842 (w); HR-MS (positive ESI) m/e calcd
3
3
(400 MHz, DMSO-d6) δ 1.81 (quin, J(H,H) ) 6.84 Hz, J(H,H)
) 7.38 Hz, 2 H, CH2), 2.84 (quin, 3J(H,H) ) 5.56 Hz, 3J(H,H) )
3
6.68 Hz, 2 H, CH2), 3.31 (t, J(H,H) ) 6.60 Hz, 2 H, CH2), 6.88
3
4
(dd, J(H,H) ) 4.04 Hz, J(H,H) ) 2.28 Hz, 1 H, CH), 7.54 (dd,
3J(H,H) ) 4.04 Hz, 4J(H,H) ) 2.40 Hz, 1 H, CH), 7.94 (br s, 3 H,
NH3), 8.48 (br s, 2 H, NH2), 8.66 (br s, 2 H, NH2), 8.73 (t, 3J(H,H)
) 5.62 Hz), 1 H, NH), 12.08 (s, 1 H, NH), 12.36 (s, 1 H, NH); 13
C
NMR (100 MHz, DMSO-d6) δ 27.3, 35.9, 36.8 (all CH2), 112.5,
116.0 (both CH), 125.5, 132.7, 155.5, 159.3, 159.6 (all quat C);
FT-IR (KBr disk) ν˜ [cm-1] 3343 (s), 3164 (s), 2995 (m), 1698 (s),
1646 (s), 1560 (s), 1476 (m), 1287 (s), 1262 (w), 1199 (w), 1059
(w), 808 (w), 761 (m); HR-MS (positive ESI) m/e calcd for
C10H17N6O2+ (M + H+) 253.141, found 253.140; HR-MS (positive
+
for C19H22N4NaO5 (M + Na+) 409.146, found 409.149.
Synthesis of 5-((tert-Butyloxycarbonyl)guanidiniocarbonyl)-
1H-pyrrole-2-carboxylate 9. A mixture of the benzyl ester 8 (1.94
g, 5.00 mmol, 1 equiv), a catalytic amount of Pd/C (∼200 mg),
and triethylamine (1.05 mL, 7.50 mmol, 1.5 equiv) in methanol
(30 mL) was vigorously stirred at 40 °C for 5 h under hydrogen
+
ESI) m/e calcd for C10H16N6NaO2 (M + Na+) 275.123, found
275.123.
6838 J. Org. Chem., Vol. 72, No. 18, 2007

Syntheses of SeVeral Bis-Cations
1
Bis-Cation 3. White solid (71%). Mp >250 °C dec; H NMR
(400 MHz, DMSO-d6) δ 1.58 (m, 2 H, CH2), 2.79 (m, 2 H, CH2),
0.2 mmol) was solved in methanol (20 mL) and Pd/C (∼10 mg)
was added. This mixture was vigorously stirred at 40 °C for 5 h
under hydrogen atmosphere. The resulting solution was filtered
through a suction filter (Celite 545), which was washed several
times with methanol. After that further tBoc-cleavage was done in
this solution as described above (49 mg, 60%). White solid (52%).
3
3
3.31 (q, J(H,H) ) 5.92 Hz, 2 H, CH2), 6.86 (dd, J(H,H) ) 4.04
4
3
Hz, J(H,H) ) 2.40 Hz, 1 H, CH), 7.56 (dd, J(H,H) ) 4.04 Hz,
4J(H,H) ) 2.40 Hz, 1 H, CH), 7.92 (br s, 3 H, NH3), 8.50 (br s, 2
3
H, NH2), 8.62 (t, J(H,H) ) 5.62 Hz), 1 H, NH), 8.69 (br s, 2 H,
NH2), 12.12 (s, 1 H, NH), 12.34 (s, 1 H, NH); 13C NMR (100
MHz, DMSO-d6) δ 24.5, 26.0, 38.0, 38.5 (all CH2), 112.4, 116.0
(both CH), 125.3, 132.9, 155.6, 159.1, 159.6 (all quat C); FT-IR
(KBr disk) ν˜ [cm-1] 3327 (s), 3098 (s), 2931 (m), 1698 (s), 1655
(s), 1558 (s), 1476 (m), 1286 (s), 1256 (w), 1194 (w), 1063 (w),
853 (w), 751 (m), 605 (w); HR-MS (positive ESI) m/e calcd for
C11H19N6O2+ (M + H+) 267.156, found 267.157; HR-MS (positive
Mp 190 °C dec; H NMR (400 MHz, DMSO-d6) δ 1.43 (m, 2 H,
1
CH2), 1.59 (m, 2 H, CH2), 1.79 (m, 2 H, CH2), 2,77 (q, 3J(H,H) )
5.96 Hz), 2 H, CH2), 3.66 (s, 3 H, CH3), 4.41 (m, 1 H, CH), 6.93
3
4
(dd, J(H,H) ) 3.80 Hz, J(H,H) ) 2.24 Hz, 1 H, CH), 7.58 (dd,
3J(H,H) ) 3.68 Hz, 4J(H,H) ) 2.16 Hz, 1 H, CH), 7.93 (br s, 3 H,
NH3), 8.53 (br s, 2 H, NH2), 8.71 (br s, 2 H, NH2), 8.86 (d, 3J(H,H)
) 7.32 Hz), 1 H, NH), 12.15 (s, 1 H, NH), 12.49 (s, 1 H, NH); 13
C
+
ESI) m/e calcd for C11H18N6NaO2 (M + Na+) 289.138, found
NMR (100 MHz, DMSO-d6) δ 22.5, 26.4, 30.1, 38.4 (all CH2),
52.0 (CH3), 52.1 (CH), 113.5, 115.8 (both CH), 125.8, 131.9, 155.5,
159.2, 159.7, 172.4 (all quat C); FT-IR (KBr disk) ν˜ [cm-1] 3327
(s), 3147 (s), 2957 (m), 1699 (s), 1654 (s), 1556 (s), 1475 (m),
1292 (s), 1248 (s), 1225 (s), 1155 (w), 814 (w), 754 (m), 602 (m);
HR-MS (positive ESI) m/e calcd for C14H23N6O4+ 339.178, found
339.177.
289.137.
Bis-Cation 4. White solid (74%). Mp 205 °C dec; 1H NMR (400
MHz, DMSO-d6) δ 1.36 (m, 2 H, CH2), 1.51-1.60 (m, 4 H, 2
CH2), 2.76 (m, 2 H, CH2), 3.24 (q, 3J(H,H) ) 6.68 Hz, 2 H, CH2),
3
4
6.85 (dd, J(H,H) ) 4.04 Hz, J(H,H) ) 2.40 Hz, 1 H, CH), 7.51
(dd, 3J(H,H) ) 4.04 Hz, 4J(H,H) ) 2.40 Hz, 1 H, CH), 7.83 (br s,
3
3 H, NH3), 8.44 (br s, 2 H, NH2), 8.50 (t, J(H,H) ) 5.50 Hz), 1
H, NH), 8.63 (br s, 2 H, NH2), 12.01 (s, 1 H, NH), 12.31 (s, 1 H,
NH); 13C NMR (100 MHz, DMSO-d6) δ 23.2, 26.6, 28.4, 38.4,
38.6 (all CH2), 112.4, 115.9 (both CH), 125.3, 132.9, 155.5, 159.0,
159.6 (all quat C); FT-IR (KBr disk) ν˜ [cm-1] 3363 (s), 3047 (s),
2931 (m), 1697 (s), 1626 (s), 1561 (s), 1475 (m), 1286 (s), 1286
(s), 1199 (m), 1070 (w), 853 (w), 758 (m), 598 (m); HR-MS
(positive ESI) m/e calcd for C12H21N6O2+ (M + H+) 281.172, found
281.173.
Acknowledgment. Financial support by the Deutsche For-
schungsgemeinschaft and the Fonds der Chemischen Industrie
is gratefully acknowledged.
Supporting Information Available: 1H and 13C NMR spectra
of compounds 1-5, 8, and 9 and ITC data for complex formation
between 3 and N-acetyl phenyl alanine carboxylate. This material
is available free of charge via the Internet at http://pubs.acs.org.
Synthesis of the Receptor 5. For the cleavage of the CBz-
protecting group by hydrogenation the benzyl ester 13 (115 mg,
JO070981Z
J. Org. Chem, Vol. 72, No. 18, 2007 6839