Ion pairing between the chain ends induces folding of a flexible zwitterion in methanol

Article abstract of DOI:10.1002/ejoc.200700164

A well defined folded loop structure can be induced in a flexible zwitterion 10 in the polar and protic solvent methanol by charge interactions between the two termini of the zwitterion. In 10 a (guanidiniocarbonyl)pyrrole moiety, a highly efficient oxoan

Full text of DOI:10.1002/ejoc.200700164

FULL PAPER
DOI: 10.1002/ejoc.200700164
Ion Pairing Between the Chain Ends Induces Folding of a Flexible Zwitterion
in Methanol
Carsten Schmuck*[a] and Jürgen Dudaczek[a]
Dedicated to Prof. Dr. Peter Welzel on the occasion of his 70th birthday
Keywords: Ion pairing / Guanidinium cations / Supramolecular chemistry / Foldamers / Amino acids
A well defined folded loop structure can be induced in a flex-
ible zwitterion 10 in the polar and protic solvent methanol by
charge interactions between the two termini of the zwitter-
ate and the (guanidiniocarbonyl)pyrrole cation leads to the
formation of a well defined loop as could be shown by NMR
analysis (NOESY and H/D-exchange experiments) as well as
ion. In 10 a (guanidiniocarbonyl)pyrrole moiety, a highly ef- molecular modelling calculations. Without this intramolecu-
ficient oxoanion binding site, and a pyrrole-2-carboxylate
unit serve as complementary binding sites at the ends of a
flexible strand in which two amino acids, alanine and valine,
lar charge interaction, as in the protected and hence un-
charged precursor 9, no loop is formed but rather weak inter-
molecular dimerization is observed.
are linked via a butylene spacer in a head-to-head orienta- (© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
tion. Intramolecular ion pair formation between the carboxyl-
Germany, 2007)
without the need for a structure-inducing linker.[6] This,
however, requires a very efficient ion pair formation in po-
lar solution.[7] For this purpose we chose a (guanidiniocar-
bonyl)pyrrole– carboxylate ion pair,[8] as this modified gua-
nidinium cation is one of the most efficient binding motifs
for carboxylate anions.[9] Hence, in zwitterion 10 two amino
acids were linked in a head-to-head orientation via a fully
flexible 1,4-diaminobutane linker and were further modified
at the chain ends with a (guanidiniocarbonyl)pyrrole and a
pyrrole-2-carboxylate moiety, respectively.
Introduction
The design of intrinsically flexible molecules which adopt
a predictable and defined conformation in solution based
on specific intramolecular non-covalent interactions is still
a challenge despite the recent progress in this area over the
last years,[1,2] especially in cases where structure formation
in polar and protic solvents is considered. Most systems
reported so far adopt specific structures only in organic sol-
vents (e.g. chloroform), because they rely mainly on hydro-
gen bonds, which are easily disrupted when the polarity of
the solvent increases. One way to overcome this problem
is to use the much stronger metal–ligand interaction as a
structure-dictating principle.[3] Also rigid scaffolds[4] or
structurally biased turn-elements[5] have been used to in-
duce a certain conformation within otherwise flexible mole-
cules. We want to present here a zwitterion 10 with a flexi-
ble linker in between the two opposite charges which due
to strong and directed charge interactions between the two
termini of the molecule adopts a specific and well defined
folded loop-structure even in the polar protic solvent meth-
anol.
Figure 1. Inducing a folded structure by ion pairing between the
chain ends of an otherwise flexible zwitterion.
Results and Discussion
We reasoned that in a molecule of the general structure
shown in Figure 1 strong interactions between the two
chain ends should lead to an ordered, folded structure even
The synthesis of zwitterion 10 is described in Scheme 1.
Cbz–Val–OH was coupled with mono-tBoc-protected 1,4-
diaminobutane 1[10] using HCTU in DMF to obtain 2 in
excellent yields of 98%. The tBoc-protecting group was
cleaved off (with TFA) and the resulting amine was treated
with tBoc-Ala–OH to give 3 (yield 50%). Again, the tBoc
group was removed with TFA and the free amine 4 was
[a] Universität Würzburg, Institut für Organische Chemie,
Am Hubland, 97074 Würzburg, Germany
Fax: +49-931-8884625
E-mail: schmuck@chemie-uni-wuerzburg.de
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
3326
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3326–3330

Folding of a Flexible Zwitterion in Methanol
FULL PAPER
coupled with the tBoc-protected (guanidinocarbonyl)pyr- ing a weak intermolecular self-association probably by for-
role 5[11] using PyBOP in DMF to give 6 in good yields of mation of a H-bonded dimer. However, even though zwit-
72%. The Cbz-goup on the valine residue in 6 was removed terion 10 is present as a monomeric species, the NMR shifts
(H2/Pd) and the resulting amine 7 was treated with pyrrole show that 10 adopts a conformation in which the guanidi-
dicarboxylic acid mono benzyl ester 8 again using PyBOP nium cation forms a strong ion pair with the carboxylate.
as the coupling reagent to provide the protected model sys- For example, the guanidinium amide NH occurs at δ =
tem 9. Finally, the remaining protecting groups were re- 12.95 which is a clear indication of an ion pair interaction
moved with first H2 and then TFA to give zwitterion 10.
which therefore must occur intramolecularly in this case.
The reference value for a non-ion paired guanidiniocar-
bonyl pyrrole cation is around δ = 10.7, respectively.[8]
The structure of both the zwitterion 10 and the protected
monomer 9 were then probed by NOESY NMR experi-
ments (Figure 2). Significant diagnostic interstrand cross
peaks are found for 10 which are absent in the protected
compound 9. For example, among others NOE signals are
observed between the two alkyl groups of the amino acid
side chains and also between the two pyrrole NHs. Further-
more, the NOEs observed for the two amino acid amide
NHs are significantly different from each other. Whereas
the valine amide–NH shows cross peaks to the methyl
group of the alanine as well as the two different types of
methylene groups of the butylene linker, the alanine amide-
NH does only show a cross peak with the alanine methyl
group but neither with the isopropyl group of the valine
nor the methylene groups of the linker. These selected diag-
nostic NOEs, summarized also in Figure 2, are indicative of
Scheme 1. Synthesis of the flexible zwitterion 10.
The conformation of zwitterion 10 was probed by NMR
in CD3OH and compared to the fully protected precursor
9, in which now charge interactions are possible. A dilution
1
study showed, that for zwitterion 10 the shifts in the H
NMR are concentration independent at least in the range
0.1–25 m. Accordingly, no intermolecular self-association
occurs and 10 exists as a monomeric species at least in this
concentration range. This is in contrast to the protected
molecule 9. 1H NMR dilution studies showed small concen-
Figure 2. Top: Part of the NOESY spectrum of 10 (600 MHz,
CD3OH, room temp.) showing the different NOEs and hence the
different environment of the two amino acid amide NHs. Bottom:
Schematic representation of selected non-sequential diagnostic
tration-dependent shift changes in the same range indicat- NOEs.
Eur. J. Org. Chem. 2007, 3326–3330
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3327

C. Schmuck, J. Dudaczek
FULL PAPER
a folded loop-like conformation in which the chain ends are
close to each other. Otherwise, in a more extended random
structure no NOE signals between the two pyrrole NHs or
the two alkyl groups of the amino acid side chains would
be observable. This situation is found for the protected com-
pound 9, which hence most likely only adopts an unordered
random-coil like structure in solution. None of the diagnos-
tic NOEs shown in Figure 2 for zwitterion 10 are observed
for the protected precursor 9.
Structural differences between the two amid NHs are
also obvious from a comparison of the coupling constants
between the protected compound 9 and zwitterion 10.[12]
Whereas the coupling constant of the valine amide NH with
the α-CH significantly increases from 3J = 7.3 to 8.6 Hz
going from 9 to 10 (in CD3OH), it only slightly changes for
3
the alanine amide NH (from J = 7.9 to 7.6 Hz). This indi-
cates a more ordered, probably more beta-sheet like struc-
ture around the valine amide NH in 10 compared to 9, but
not for the alanine amide NH which seems to exhibit a
similar local environment in 10 as in 9.
Also a H/D-exchange experiment confirmed a strikingly
different chemical environment for the two amino acid
amide-NHs. The kinetics with which an amide-NH ex-
changes with the solvent depends on whether it is H-
bonded or not.[2h] And indeed, whereas the alanine amide
NH rapidly exchanges upon the addition of D2O to a NMR
sample of 10 in [D6]DMSO, the valine amide NH exchanges
much more slowly (Figure 3). This suggests that the valine
Figure 4. (Top): Calculated structure of zwitterion 10 showing in-
tramolecular H-bonds (dotted lines) and the experimentally ob-
served diagnostic non-sequential NOEs (double-headed arrows).
The more rapidly exchanging alanine amide NH pointing back-
wards is also indicated. (Bottom): MD simulation (500 ps at 300 K)
of structure of zwitterion 10, snapshot taken every 20 ps [non-polar
hydrogen atoms omitted for clarity].
amide NH is engaged in intramolecular hydrogen bonding mized structure shown in Figure 4 (top) is fully consistent
as shown in Figure 4 (top), which slows down the exchange, with the observed experimental data discussed above.[15]
whereas the alanine amide NH is only interacting with the The two alkyl side chains of the amino acids are facing each
solvent.[13] Also the two butylene NHs exchange with the other, giving rise to non-sequential NOEs between them.
solvent, much more slowly but both with similar rates. As Whereas the valine amide NH points inwards and is in-
they are less acidic than the amino acid amide NHs, no volved in hydrogen bonding to the pyrrole carbonyl group
direct comparison can be made with the absolute exchange at the other chain end, the alanine amide NH points out-
rate of the two amino acid amides. But the fact that both wards and is interacting with the solvent. Hence, the alanine
linker amide NHs exchange with similar rates indicates a amide NH should exchange more easily with the solvent in
similar hydrogen bonding situation for both of them.
contrast to the valine amide NH. Also, the formation of
this intramolecular H-bond is fully consistent with the ob-
served NOEs: only the valine amide NH shows a cross sig-
nal with the alanine methyl group as it points towards the
alanine but not vice versa. The alanine amide NH points
to the backside and hence no NOE with the isopropyl
group of the valine is seen. The different environment of
the two amino acid amide NHs is also in agreement with
3
the larger J coupling constant for the valine amide NH
relative to the alanine amide NH and the non-structured
protected precursor 9.
The loop structure that zwitterion 10 adopts even in a
polar and protic solvent such as methanol is also conforma-
tionally rather rigid at least according to modelling stud-
ies.[16] A MD simulation (300 K, 500 ps, water solvation)
was performed to probe the flexibility of the loop. As
shown in Figure 4 (bottom) the general structure of the
Figure 3. H/D exchange experiment of 10 (D2O added to a 10 m
solution of 10 in [D6]DMSO at room temperature, 400 MHz).
The most likely structure of zwitterion 10 was calculated loop stays intact, even over a rather long time period of 500
using molecular mechanics [Macromodel V8.0, MMFFs ps. There are no significant changes in the principal binding
force field, GB/SA water solvation, MC conformational interactions (ion pairs, H-bonds, van der Waals interac-
search with 100.000 steps].[14] The resulting energy mini- tions) that hold the loop together.
3328
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3326–3330

Folding of a Flexible Zwitterion in Methanol
FULL PAPER
give the desired product: 7.62 g (92%); m.p. 88 °C. 1H NMR ([D6]-
DMSO, 400 MHz): δ = 6.75 (s, 1 H), 3.30 (d, 2 H), 2.88 (m, 2 H),
1.37 (m, 13 H) ppm. 13C NMR ([D6]DMSO, 100 MHz): δ = 155.54
(Cq), 77.21 (Cq), 41.43 (CH2), 30.70 (CH2), 28.25 (CH3), 27.02
(CH2) ppm. MS (EI) = m/z 189.3 [M + H+].
Conclusions
In conclusion, we have demonstrated here that the con-
formation of a flexible linear molecule can be controlled by
using directed charge interactions between the termini of
the molecule. This however requires very efficient charge
Compound 2: Procedure C. Cbz-Val–OH: 251 mg, HCTU: 414 mg,
interactions that are strong enough even under competitive NMM: 330 µL, 1: 188 mg; yield 413 mg (98%); m.p. 193 °C. 1H
NMR ([D6]DMSO, 400 MHz): δ = 7.87 (t, 1 H), 7.33 (m, 5 H),
7.18 (d, J = 8.96 Hz, 1 H), 6.75 (t, 1 H), 5.02 (s, 2 H), 3.77 (dd, 1
H), 3.07 (m, 2 H), 2.98 (m, 2 H), 2.89 (m, 4 H), 1.91 (m, 1 H), 1.37
(s, 9 H), 0.83 (d, J = 6.84 Hz, 6 H) ppm. 13C NMR ([D6]DMSO,
100 MHz): δ = 170.60 (Cq), 136.84 (Cq), 128.03 (CH), 127.45 (CH),
127.34 (CH), 77.03 (Cq), 65.05 (CH2), 60.04 (CH), 37.96 (CH2),
29.97 (CH), 27.98 (CH3), 27.08 (CH2), 26.86 (CH2), 18.93 (CH3),
17.97 (CH3) ppm. MS (EI): m/z = 206.2 (M+ – C10H19N2O3).
solvation conditions. Using a guanidiniocarbonyl pyrrole/
carboxylate ion pair, zwitterion 10 which does not possess
any rigid or structurally biased linker was shown to adopt
a predictable and specific loop-conformation even in meth-
anol. We are currently investigating how this approach can
be used to build even larger self-assembled structures.
Compound 3: Procedure A and C. Boc-Ala–OH: 189 mg, HCTU:
414 mg, NMM: 330 µL, 2: 493 mg; yield 288 mg (50%); m.p.
177 °C. H NMR ([D6]DMSO, 400 MHz): δ = 7.87 (t, 1 H), 7.70
Experimental Section
1
General Procedure A for the Deprotection of the tBOC Group: A
solution of the corresponding protected compound (100 mg) in tri-
fluoroacetic acid (5 mL) and dichloromethane (5 mL) was stirred
for 1 h at room temperature under TLC control. After completion
of the reaction, the trifluoroacetic acid was evaporated in vacuo
yielding a slightly yellow oil. 10 mL of water were added, the solu-
tion was lyophilised and the white solid was dried in vacuo (yield:
quantitative).
(t, 1 H), 7.33 (m, 5 H), 7.17 (d, J = 8.96 Hz, 1 H), 6.76 (d, J =
7.08 Hz, 1 H), 5.02 (s, 2 H), 3.90 (m, 1 H), 3.78 (m, 1 H), 3.08 (m,
2 H), 3.00 (m, 2 H), 1.91 (m, 1 H), 1.37 (s, 13 H), 1.14 (d, J =
7.04 Hz, 3 H), 0.83 (m, 6 H) ppm. 13C NMR ([D6]DMSO,
100 MHz): δ = 170.62 (Cq), 136.84 (Cq), 128.03 (CH), 127.46 (CH),
127.34 (CH), 65.05 (CH2), 60.04 (CH), 37.88 (CH2), 29.98 (CH),
27.89 (CH3), 26.23 (CH2), 26.06 (CH2), 18.94 (CH3), 18.15 (CH3),
+
17.96 (CH3) ppm. HRMS (ESI): calculated for C25H40N4NaO6
[M + Na+]: m/z = 515.284, found 515.285.
General Procedure B for the Deprotection of the Cbz Group: A solu-
tion of the corresponding protected compound (100 mg) and 10%
Pd/C in methanol (10 mL) was hydrogenated at room temp. for
0.5 h (TLC control). The mixture was filtered through a pad of
Celite to remove the catalyst (Pd/C) and the solvent was removed
under reduced pressure. The crude product was used in the follow-
ing steps without further purification.
Compound 6: Procedure A and D. 5: 201 mg, PyBOP: 354 mg,
NMM: 224 µL, 3: 267 mg; yield 283 mg (62%); m.p. 184 °C. 1H
NMR ([D6]DMSO, 400 MHz): δ = 11.61 (br., 1 H), 10.83 (br., 1
H), 9.37 (br., 1 H), 8.63 (br., 1 H), 8.46 (d, J = 7.44 Hz, 1 H), 7.94
(t, 1 H), 7.88 (t, 1 H), 7.34 (m, 5 H), 7.17 (d, J = 8.84 Hz, 1 H),
6.83 (m, 2 H), 5.02 (s, 2 H), 4.41 (m, 1 H), 3.78 (m, 1 H), 3.04 (m,
4 H), 1.92 (m, 1 H), 1.46 (s, 9 H), 1.39 (s, 4 H), 1.28 (d, J = 7.04 Hz,
3 H), 0.83 (m, 6 H) ppm. 13C NMR ([D6]DMSO, 100 MHz): δ =
172.07 (Cq), 170.94 (Cq), 158.20 (Cq), 156.09 (Cq), 137.15 (Cq),
128.34 (CH), 127.77 (CH), 127.65 (CH), 65.37 (CH2), 60.36 (CH),
48.45 (CH), 38.22 (CH2), 30.29 (CH), 27.80 (CH3), 26.58 (CH2),
26.47 (CH2), 19.25 (CH3), 18.35 (CH3), 18.27 (CH3) ppm.
General Procedure C for Coupling of Amino Acids: A mixture of
the corresponding N-protected amino acid (1.0 mmol), HCTU
(1.0 mmol) and NMM (3.0 mmol) in DMF (5 mL) was stirred at
room temp. for 15 min. Then the corresponding amine component
(1.0 mmol) was added to the solution and the reaction mixture
stirred overnight (at room temp.). Then 30 mL of water were added
and the solution was stirred in an ice-bath for further 30 min. The
white precipitate was filtered, washed several times with water and
lyophilised to give the desired product.
HRMS (ESI): calculated for C32H46N8NaO8 [M + Na+]: m/z =
+
693.333, found 693.335.
Bis-Protected Zwitterion 9: Procedure B and D. 8: 167 mg, PyBOP:
354 mg, NMM: 224 µL, 6: 365 mg; yield 306 mg (59%); m.p.
186 °C. 1H NMR ([D3]MeOH, 600 MHz): δ = 12.74 (br., 1 H),
12.03 (br., 1 H), 9.09 (br., 1 H), 8.95 (s, 1 H), 8.90 (s, 1 H), 8.76 (s,
2 H), 8.61 (s, 1 H), 7.43 (m, 5 H), 6.90 (m, 4 H), 5.19 (s, 2 H), 4.37
(m, 1 H), 4.21 (m, 1 H), 3.19 (m, 4 H), 2.10 (m, 1 H), 1.52 (s, 9
H), 1.41 (s, 4 H), 1.30 (s, 3 H), 0.97 (d, J = 24 Hz, 6 H) ppm. 13C
NMR ([D6]DMSO, 63 MHz): δ = 172.10 (Cq), 170.72 (Cq), 160.11
(Cq), 159.14 (Cq), 158.90 (Cq), 158.47 (Cq), 136.29 (Cq), 131.03
(Cq), 128.52 (Cq), 128.10 (CH), 127.93 (CH), 124.17 (CH), 115.29
(CH), 113.89 (CH), 112.82 (CH), 65.49 (CH2), 58.08 (CH), 48.42
(CH), 38.23 (CH2), 30.54 (CH), 27.78 (CH3), 26.60 (CH2), 26.45
(CH2), 19.29 (CH3), 18.57 (CH3), 18.34 (CH3) ppm. HRMS (ESI):
General Procedure D for Coupling of the Pyrrolecarboxylic Acids 5
and 8: A solution of the pyrrole derivative 5 or 8 (0.68 mmol),
PyBOP (0.68 mmol) and NMM (2.04 mmol) in DMF (10 mL) was
stirred at room temp. for 15 min. Then the corresponding amine
component (0.68 mmol) was added and the solution was stirred at
room temp. overnight. After adding 60 mL of water the solution
was stirred in an ice-bath for another 30 min. The white precipitate
was filtered, washed several times with water and lyophilised to give
the desired product.
tert-Butyl (4-Aminobutyl)carbamate (1): A solution of 1,4-diami-
nobutane (19.33 g, 220 mmol) in chloroform (150 mL) was cooled
to 0 °C and then treated under continuous stirring with a solution
of Boc2O (9.60 g, 44 mmol) in chloroform (50 mL). The reaction
was stirred overnight at room temperature. The suspension was fil-
tered and the solution was concentrated under reduced pressure.
The colourless oil was dissolved in ethyl acetate (500 mL) and
washed twice with brine (60 mL). The aqueous phases were com-
bined and extracted once with ethyl acetate. The combined organic
layers were dried with sodium sulfate and evaporated under re-
duced pressure. The resulting oily residue was dried in vacuo to
+
calculated for C37H49N9NaO9 [M + Na+]: m/z = 786.355, found
768.356.
Zwitterion 10: Procedure B and A. Yield 75 mg (quant.); m.p.
242 °C. 1H NMR ([D6]DMSO, 400 MHz): δ = 12.74 (br., 1 H),
12.25 (br., 1 H), 11.78 (br., 1 H), 8.50 (d, J = 7.68 Hz, 1 H), 8.30
(d, J = 8.6 Hz, 1 H), 8.07 (t, 1 H), 7.93 (t, 1 H), 6.81 (br., 2 H),
6.77 (dd, 1 H), 6.72 (dd, 1 H), 4.41 (m, 1 H), 4.26 (m, 1 H), 3.06
(m, 4 H), 2.03 (m, 1 H), 1.40 (s, 4 H), 1.29 (d, J = 7.08 Hz, 3 H),
Eur. J. Org. Chem. 2007, 3326–3330
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3329

C. Schmuck, J. Dudaczek
FULL PAPER
0.88 (dd, 6 H) ppm. 13C NMR ([D6]DMSO, 63 MHz): δ = 172.04
(Cq), 170.80 (Cq), 162.11 (Cq), 159.11 (Cq), 158.93 (Cq), 130.03
(Cq), 125.99 (Cq), 114.42 (CH), 113.74 (CH), 113.10 (CH), 58.14
(CH), 48.47 (CH), 38.23 (CH2), 30.42 (CH), 26.54 (CH2), 26.45
(CH2), 19.30 (CH3), 18.58 (CH3), 18.34 (CH3) ppm. HRMS (ESI):
4236–4237; d) T. S. Haque, S. H. Gellman, J. Am. Chem. Soc.
1997, 119, 2303–2304; e) E. De Alba, M. A. Jimenez, M. Rico,
J. Am. Chem. Soc. 1997, 119, 175–183.
Charge interactions between side chain residues – but not the
chain ends – have also been used to stabilize secondary struc-
tures in larger peptides: a) B. Ciani, M. Jourdan, M. S. Searle,
J. Am. Chem. Soc. 2003, 125, 9038–9047; b) S. E. Kiehna,
M. L. Waters, Protein Sci. 2003, 12, 2657–2667; c) C. A. Blasie,
J. M. Berg, Biochemistry 1997, 36, 5245–5250.
[6]
calculated for C25H36N9O7 [M + H+]: m/z = 574.273, found
+
574.274.
Supporting Information (see also the footnote on the first page of
this article): NMR spectra (1H, 13C) of compounds 1–3, 6, 9 and
10.
[7]
[8]
C. Schmuck, J. Org. Chem. 2000, 65, 2432–2437.
a) C. Schmuck, L. Geiger, J. Am. Chem. Soc. 2004, 126, 8898–
8899; b) C. Schmuck, Chem. Eur. J. 2000, 6, 709–718.
a) K. A. Schug, W. Lindner, Chem. Rev. 2005, 105, 67–113; b)
R. J. Fitzmaurice, G. M. Kyne, D. Douheret, J. D. Kilburn, J.
Chem. Soc., Perkin Trans. 1 2002, 841–864.
[9]
Acknowledgments
[10]
[11]
J.-F. Pons, J.-L. Fauchére, F. Lamaty, A. Molla, R. Lazaro, Eur.
J. Org. Chem. 1998, 853–859.
Pyrrolecarboxylic acid 5 was obtained from the corresponding
benzyl ester after hydrogenolysis: C. Schmuck, T. Rehm, F.
Gröhn, F. Klein, F. Reinhold, J. Am. Chem. Soc. 2006, 128,
1431–1431.
Financial support by the Fonds der Chemischen Industrie and by
the Deutsche Forschungsgemeinschaft (DFG) is gratefully ac-
knowledged.
[1] M. Searle, B. Ciani, Curr. Opin. Struct. Biol. 2004, 14, 458–
464.
[12] Unfortunately, due to signal overlap of the α-CHs with the OH
signal of the solvent only the coupling constants of the amino
acid amide NHs could be analysed reliably.
[13] As the exchange rate of both amides in the protected monomer
9 is very similar, the observed difference in 10 can not be due
to a different intrinsic reativity caused by the different steric
environment around the two amide NHs but must result form
different H-bond features.
[2] For recent reviews on this topic see: a) T. Moriuchi, T. Hirao,
Chem. Soc. Rev. 2004, 33, 294–301; b) W. A. Loughlin, J. D. A.
Tyndall, M. P. Glenn, D. P. Fairlie, Chem. Rev. 2004, 104, 6085–
6118; c) P. Chitnumsub, W. R. Fiori, H. A. Lashuel, H. Diaz,
J. W. Kelly, Bioorg. Med. Chem. 1999, 7, 39–59; d) E. Lacroix,
T. Kortemme, M. Lopez de la Paz, L. Serrano, Curr. Opin.
Struct. Biol. 1999, 9, 487–493; e) K. D. Stigers, M. J. Soth, J. S.
Nowick, Curr. Opin. Chem. Biol. 1999, 3, 714–723; f) S. H.
Gellman, Curr. Opin. Chem. Biol. 1998, 2, 717–725; g) C. L.
Nesloney, J. W. Kelly, Bioorg. Med. Chem. 1996, 4, 739–766; h)
C. L. Nesloney, J. W. Kelly, Bioorg. Med. Chem. 1996, 4, 739–
766; i) J. P. Schneider, J. W. Kelly, Chem. Rev. 1995, 95, 2169–
2187; j) D. S. Kemp, Trends Biotechnol. 1990, 8, 249–255.
[3] Some recent examples:a) M. Kruppa, C. Bonauer, V. Michlova,
B. Koenig, J. Org. Chem. 2005, 70, 5305–5308; b) T. Moriuchi,
V. Nomoto, K. Yoshida, A. Ogawa, T. Hirao, J. Am. Chem.
Soc. 2001, 123, 68–75; c) A. Nomoto, T. Moriuchi, S. Yama-
zaki, A. Ogawa, T. Hirao, Chem. Commun. 1998, 1963–1964.
[4] For example:a) J. M. Langenhan, S. H. Gellman, Org. Lett.
2004, 6, 937–940; b) J. S. Nowick, J. O. Brower, J. Am. Chem.
Soc. 2003, 125, 876–877; c) J. S. Nowick, S. K. Lam, T. V.
Khasanova, W. E. Kemnitzer, S. Maitra, H. T. Mee, R. Liu, J.
Am. Chem. Soc. 2002, 124, 4972–4973; d) B. R. Huck, J. D.
Fisk, S. H. Gellman, Org. Lett. 2000, 2, 2607–2610.
[14] F. Mohamadi, N. G. J. Richards, W. C. Guida, R. Liskamp, M.
Lipton, C. Caufield, G. Chang, T. Hendrickson, W. C. Still, J.
Comput. Chem. 1990, 11, 440–467.
[15] The energy minimized structure in Figure 4 as obtained from
the force field calculations contains one cis-amide linkage. In-
tramolecular ion pairing is not possible in the same way if that
amide adopts a trans conformation. At least there is no energy
minimum within 20 kJ/mol of the one shown here, which con-
tains a trans linkage. Hence, the energetical cost for cis-amide
formation is obviously more than overcome by the strong intra-
molecular ion pairing.
[16] The rather high stability of the loop structure is also supported
by NMR experiments. The 1H NMR spectra of 10 do not
change upon the addition of up to 50% water to the methanol
solution. With even higher water contents precipitation occurs.
Even though the exchanging acidic NH protons, which are
most diagnostic for ion pairing, cannot be observed in the pres-
ence of water, there are no noticeable shift changes for any
proton, suggesting that the loop most likely also exists in water/
methanol mixtures.
[5] For example, the rigid NG or -Pro-Gly turn elements have
been used in this context:a) J. D. Fisk, D. R. Powell, S. H. Gell-
man, J. Am. Chem. Soc. 2000, 122, 5443–5447; b) E. De Alba,
M. Rico, M. A. Jimenez, Protein Sci. 1999, 8, 2234–2244; c)
H. E. Stanger, S. H. Gellman, J. Am. Chem. Soc. 1998, 120,
Received: February 21, 2007
Published Online: May 15, 2007
3330
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 3326–3330