Synthesis of unsymmetrical tweezer receptor libraries and identification of receptors for Lys-D-Ala-D-Ala in aqueous solution

Article abstract of DOI:10.1002/chem.200500876

Libraries of "unsymmetrical" tweezer receptors, featuring a guanidinium head group as a carboxylate binding site and two independently synthesized peptidic arms, have been prepared and screened to identify receptors for the N-Ac-Lys-D-Ala-D-Ala tripeptide

Full text of DOI:10.1002/chem.200500876

FULL PAPER
DOI: 10.1002/chem.200500876
Synthesis of Unsymmetrical Tweezer Receptor Libraries and Identification of
Receptors for Lys-d-Ala-d-Ala in Aqueous Solution
Jon Shepherd, Tom Gale, Kim B. Jensen, and Jeremy D. Kilburn*[a]
Abstract: Libraries of “unsymmetrical”
tweezer receptors, featuring a guanidi-
nium head group as a carboxylate bind-
ing site and two independently synthe-
sized peptidic arms, have been pre-
pared and screened to identify recep-
tors for the N-Ac-Lys-d-Ala-d-Ala tri-
peptide
sequence.
The
binding
d-Ala, were investigated. These studies
demonstrated that when attached to
the solid-phase, the receptor binds dye-
labeled N-Ac-Lys-d-Ala-d-Ala, in buf-
fered aqueous media, with mm binding
affinity.
properties of one such receptor struc-
ture, with dye-labeled N-Ac-Lys-d-Ala-
Keywords: library
receptors · solid-phase synthesis
·
peptides
·
Introduction
recognition site for the carboxylate functionality into the
general receptor structure has facilitated the identification
of receptors for peptides with a free carboxylate terminus.
In this context, we have used bis(aminoalkyl)guanindiniums
as a scaffold to prepare libraries of two-armed “tweezer” re-
ceptor structures 1 (Scheme 1) and hence identified recep-
The selective binding of peptides with synthetic receptors
remains a major challenge, particularly, if binding is to be
successful in competitive aqueous media,[1] which is essential
if such systems are to be of use in the physiological milieu.
In recent years, a variety of receptors, featuring one or more
peptidic arms attached to a rigid template, have proved to
be effective and sequence selec-
tors for
a side-chain-protected tripeptide in aqueous
media.[2e]
tive for peptides, despite the in-
herent flexibility of many of
these receptor systems.[2-4] In
addition, the use of a combina-
torial, split-and-mixapproach
to randomize the amino acid
content in the peptidic arms has
provided a powerful method for
the construction of libraries of
potential peptide receptors,[5]
and this approach has been
used successfully for the identi-
Scheme 1. Two-armed tweezer receptor structures 1.
fication of sequence selective
receptors for a variety of pep-
tides, in both nonpolar organ-
ic[3] and aqueous solvents.[2,4] The incorporation of a specific
Recently, Schmuck has used linear peptides, capped with
a guanidinopyrrole as a carboxylate binding site, to identify
receptors that bind the amyloid peptide (Val-Val-Ile-Ala) in
aqueous solution.[2c,d,6] Although not incorporating a specific
carboxylate binding site, libraries of three-armed receptors,
based on a cyclotriveratrylene scaffold, have also been
screened by Liskamp who utilized the dye-labeled dipep-
tides d-Ala-d-Ala and d-Ala-d-Lac, with a free carboxylate
[a] J. Shepherd, Dr. T. Gale, Dr. K. B. Jensen, Prof. J. D. Kilburn
School of Chemistry
University of Southampton
Southampton, SO17 1BJ (UK)
Fax: ( +44)2380-596-805
E-mail: jdk1@soton.ac.uk
Chem. Eur. J. 2006, 12, 713 – 720
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
713

terminus, as substrates in phosphate buffer.[2a,b] These devel-
opments demonstrate the potential of such receptor systems
to provide potent and sequence selective peptide receptors
for use as novel therapeutics or biosensors.
In our own work with two-armed “tweezer” receptors,
based around a bis(aminoalkyl)guanindinium scaffold, the li-
braries prepared to date have limited diversity; this is be-
cause both arms of the receptor are synthesized simultane-
ously on the solid support, which leads to “symmetrical”
tweezer receptors containing two arms with an identical
amino acid sequence.[2e] To increase the diversity of struc-
tures that can be prepared by using this combinatorial ap-
proach, we have now developed routes to libraries of analo-
gous “unsymmetrical” tweezer receptors in which the two
arms are synthesized independently. These libraries have
been screened to identify receptor structures that are able
to bind to the bacterial cell wall precursor peptide N-Ac-
Lys-d-Ala-d-Ala.[7] One such receptor structure was resyn-
thesized and used to determine the association constant
with N-Ac-Lys-d-Ala-d-Ala. In free solution, although a
UV binding study provides evidence for an association be-
tween the receptor and the tripeptide, the data does not
allow for the determination of a simple 1:1 (receptor:sub-
strate) binding constant. When attached to the solid support,
however, the receptor binds to the tripeptide in an aqueous
buffered solution with a mm association constant. Herein we
describe these studies in detail.
Results and Discussion
Our approach to tweezer receptor libraries has relied on
Edman degradation to identify the amino acid components
in selected “hit” beads from screening experiments. While
this proved to be relatively straightforward for the “symmet-
rical” receptor structures reported previously,[2e] the con-
struction of “unsymmetrical” structures requires a more
complicated procedure and careful use of orthogonal pro-
tecting groups. We have used two strategies for the construc-
tion of such libraries, with both cases utilizing the orthogo-
nally protected guanidinium derivative 6, prepared in five
steps from the previously described[8] thiourea 2, as the
starting point.
In the first strategy, the guanidinium 6 was attached to
tentagel resin and the Fmoc-protecting group was removed
(Scheme 2). Two rounds of split-and-mixsynthesis using
Fmoc-protected amino acids (Fmoc=9-fluorenyloxycarbon-
yl), with acid-sensitive side chain protection where appropri-
ate, were then employed, followed by a third round of split-
and-mixsynthesis using Boc-protected amino acids to give 8
(Boc=tert-butyloxycarbonyl), thus, completing the synthesis
of the first peptidic arm of the receptor structures. Subse-
quent removal of the Dppe-protecting group (Dppe=1-(4,4-
dimethyl-2,6-dicyclohexylidene)phenylethyl)[9] was followed
by three more rounds of split-and-mixsynthesis using Fmoc-
protected amino acids. Finally, treatment with piperidine re-
moved the Fmoc-protecting group on the last amino acid
Scheme 2. a) i) CH3I, acetone; ii) NH4PF6, CH2Cl2, CH3OH; b) TsNH2,
DBU, toluene, CHCl3, reflux; c) 30% CF3CO2H, CH2Cl2; d) Fmoc-l-Glu-
(OtBu)-OH, EDC, HOBt, DIPEA, DMF; e) 60% CF3CO2H, CH2Cl2;
f) Tentagel resin, PyBOP, HOBt, DIPEA, DMF; g) 20% piperidine in
DMF; h) two-fold split-and-mixFmoc peptide synthesis using Gly, l-Leu,
l-Phe, l-Glu(OtBu), and l-Met; i) split-and-mixBoc peptide synthesis
using Gly, l-Leu, l-Phe, l-Glu(OtBu), and l-Met; j) 35% aqueous
NH2NH2; k) three-fold split-and-mixFmoc peptide synthesis using l-Ala,
l-Val, l-Gln, l-Hist, and l-Lys(Boc); l) liquid HF.
and treatment with HF, simultaneously, removed all acid-
sensitive protecting groups, including the tosyl group on the
guanidinium unit, producing library 9.
Identifying the structure of a receptor on a single bead
from this library involved Edman sequencing[10] of both
tweezer arms simultaneously, so that each round of Edman
sequencing identified two amino acids. Therefore, to avoid
any ambiguity as to which arm the respective amino acids
are derived from, the choice of amino acids used in the
split-and-mixsteps is restricted, that is, the set of amino
714
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 713 – 720

Tweezer Receptor Libraries
FULL PAPER
acids used at the first position of the first arm (AA1) must
be different from those incorporated at the first position of
the second arm (AA4). The same restriction applies to AA2
versus AA5 and AA3 versus AA6.
Following this approach, a library containing 15625 mem-
bers was prepared by using the amino acids Gly, l-Leu, l-
Phe, l-Glu(OtBu), and l-Pro for positions AA1–AA3 and
the amino acids l-Ala, l-Val, l-Gln, l-Ser(OtBu), and l-Lys-
(Boc) for positions AA4–AA6. Edman sequencing on ran-
domly selected beads from the library confirmed that the
structure of the receptor on each bead could be unambigu-
ously determined for each case.
An alternative strategy also started from the guanidinium
6, but required the introduction of an independent coding
strand for the structural identification of individual library
members. Thus, tentagel resin was first coupled with N-Boc
phenylalanine (10 mol%), followed by the coupling of the
remaining free amine sites (90 mol%) with the guanidinium
6 (Scheme 3). After the removal of the Fmoc-protecting
group, the resin was split into the desired number of por-
tions and coupled with different Fmoc-protected amino
acids. Each separate portion was then treated with TFA
(TFA=trifluoroacetic acid) to remove the Boc-protecting
group on the coding strand, and the resulting free amine
was coupled with the same amino acid as that used previ-
ously, but with a Boc-protecting group rather than an Fmoc-
protecting group. Thus, on each individual bead the coding
strand contains the same amino acid sequence as on the
peptide arm. On the third cycle of couplings, the Fmoc-pro-
tected amino acids were treated with piperidine and capped
with acetic anhydride, prior to the Boc deprotection and
coupling of the Boc-protected amino acid onto the coding
strand. Synthesis of the second peptidic arm of the receptors
required the removal of the dppe group and Boc deprotec-
tion of the coding strand. Subsequent split-and-mixsynthesis
with Fmoc-protected amino acids produced the second pep-
tidic arm with simultaneous extension of the coding strand.
The use of this procedure avoids restricting the diversity of
library 9, but does, on the other hand, mean that amino
acids with acid-sensitive side-chain protection cannot used
in the construction of the first peptidic arm.
Identifying the structure of a receptor on a single bead
now involved Edman sequencing of the heptapeptide coding
strand. For every bead, the seventh residue from the se-
quencing procedure should be phenylalanine, as this was in-
troduced on all of the beads, providing a useful check that
the library synthesis and sequencing chemistry has worked
efficiently.
Following this approach, a library containing 15625 mem-
bers was constructed by using the amino acids Gly, l-Ala, l-
Val, l-Phe, and l-Gln for posi-
tions AA1–AA3 and amino
acids Gly, l-Ala, l-Ser, l-Met,
and l-Hist for positions AA4–
AA6. Edman sequencing on
randomly selected beads from
the library confirmed that the
structure of the receptor on
each bead could be unambigu-
ously determined for each case.
To demonstrate the potential
of these libraries for the identi-
fication of new receptors, they
were screened in aqueous
buffer (pH 8.5, borate) with the
dye-labeled tripeptide N-Ac-
Lys-d-Ala-d-Ala (13). At this
pH the guanidinium moiety is
still protonated and the carbox-
ylate of the peptidic guest is de-
protonated.
The screening experiments
are simple to perform. First, a
library sample is equilibrated in
the buffer system, and then a
solution of the peptide guest,
dissolved in water and 10%
DMSO, is added. After
a
Scheme 3. a) 10 mol% Boc-L-Phe-OH, PyBOP, HOBt, DIPEA, DMF; b) 6, PyBOP, HOBt, DIPEA, DMF;
c) 20% piperidine in DMF; d) split-and-mixFmoc peptide synthesis using Gly, l-Ala, l-Val, l-Phe, and l-Gln;
e) 30% CF3CO2H, CH2Cl2; f) split-and-mixBoc peptide synthesis using Gly, l-Ala, l-Val, l-Phe, and l-Gln;
g) repeat c–f twice; h) Ac2O, DMAP; i) 35% aqueous NH2NH2; j) three-fold split-and-mixFmoc peptide syn-
thesis using Gly, l-Ala, l-Ser, l-Met, and l-Hist; k) liquid HF.
second equilibration, the selec-
tivity of the peptide binding can
be judged by the observation of
stained beads, visualized under
Chem. Eur. J. 2006, 12, 713 – 720
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
715

J. D. Kilburn et al.
dues, AA5 has a strong preference for histidine (30%) or
methionine (40%), and AA4 has a strong preference for me-
thionine (50%). AA3 and AA2, however, have a strong pref-
erence for hydrophobic residues valine (40%) or alanine
(40%). AA1 is the least well-defined with glycine (30%)
and alanine (30%). Therefore, receptor structure 16 was
chosen for resynthesis and binding studies, since it incorpo-
rated the observed preferences, and was also the exact struc-
ture identified for bead 1 (Scheme 4).
a microscope. Additional aliquots of the peptide guest can
be added to increase the peptide concentration, and thus,
provide optimal selectivity, as adjudged by the number of
highly stained beads against a background of lightly or un-
stained beads.
By using this strategy, both tweezer receptor libraries 9
and 12 were screened with dye-labeled 13. For library 9, the
selectivity was disappointing and there were a large number
of beads stained, with little discrimination between the
degree of staining. By using library 12, however, the selec-
tivity after equilibration (48 h, at a peptide concentration of
30 mm) was high, showing <2% of the highly red-colored
beads. As a control experiment, the library was also incubat-
ed with the disperse red dye 14 (30 mm), but selective bind-
ing of the dye was not observed. Ten of the most intensively
stained beads from the screening experiment of 13 with li-
brary 12 were selected and sequenced by Edman degrada-
tion[10] (Table 1).
Scheme 4. a) 6, PyBOP, HOBt, DIPEA, DMF; b) 20% piperidine in
DMF; c) Fmoc-Gly, DIC, HOBt, DMF; d) Fmoc-l-Val, DIC, HOBt,
DMF; e) Ac2O, DMAP, DMF; f) 35% aqueous NH2NH2; g) Fmoc-l-
Met, DIC, HOBt, DMF; h) Fmoc-l-His, DIC, HOBt, DMF; i) Fmoc-l-
Ser(OtBu), DIC, HOBt, DMF; j) 95% CF3CO2H, CH2Cl2; k) liquid HF.
Table 1. Sequencing results for ten “hit” beads identified from the
screening experiments of library 9 and dye-labeled peptide 12, in aqueous
buffer (pH 8.5, borate).
Bead
AA0
AA1
AA2
AA3
AA4
AA5
AA6
1
2
3
4
5
6
7
8
Phe
Phe
Phe
Phe
Phe
Phe
Phe
Phe
Phe
Phe
Gly
Gly
Ala
Gln
Val
Gln
Ala
Phe
Gly
Ala
Val
Phe
Val
Ala
Ala
Phe
Ala
Val
Ala
Val
Val
Ala
Gln
Val
Val
Ala
Ala
Val
Ala
Phe
Met
His
Gly
Gly
Ser
Met
Met
Gly
Met
Met
His
His
His
Met
Met
Gly
Ser
Met
Met
–
Ser
Ser
Ser
Ser
Ser
Ser
Ser
His
His
–
Receptor 16 was prepared on acid-labile Rink amide
resin.[11] Orthogonally protected guanidine 6[8] was then at-
tached, and the peptide arms constructed by using standard
Fmoc-protected amino acid coupling chemistry. Cleavage
from the resin (TFA) and treatment with HF, to remove the
tosyl group from the guanidine, produced the desired recep-
tor, which was purified by HPLC.
9
10
A UV titration experiment was used to study the binding
of tweezer 16 with dye-labeled 13. As aliquots of the tweez-
er 16 were added, the intensity of the UV absorption maxi-
mum (at 500 nm) of the red dye moiety of the peptide guest
was monitored. This resulted in a decrease in the absorption
at 500 nm, with a clean isobestic point at 400 nm. However,
the data from this experiment did not fit with the presumed
1:1 binding, and the titration curve (Figure 1) suggests that
The sequencing results were successful for each bead,
with the exception of one, for which AA5 and AA6 could
not be unambiguously determined. As expected, for each
bead, the seventh residue identified by Edman sequencing
was phenylalanine. Inspection of the sequencing data reveals
that AA6 has a very high proportion (70%) of serine resi-
716
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 713 – 720

Tweezer Receptor Libraries
FULL PAPER
Figure 1. UV titration for tweezer 16 with peptide 13.
both 1:1 and 1:2 (host:guest) binding stoichiometries are
present. Notably, however, when titration experiments were
also carried out with the tosylated tweezer 15 and peptide
13, no detectable change in the UV absorption was ob-
served, confirming that an interaction between the free gua-
nidinium of the receptor and the carboxylate terminus of
the guest is essential for strong binding. Complexation was
1
also investigated by H NMR spectroscopy, and the addition
of one equivalent of dye-labeled 13 to a solution of the
tweezer receptor 16 did lead to changes in the spectrum of
the latter; however, the limited solubility and poor resolu-
tion precluded further analysis.
Therefore, the receptor structure 16 was resynthesized for
binding studies on the solid-phase. The synthesis essentially
followed the route described above, but used the trifluoro-
acetyl-protected guanidinium as the scaffold for initial at-
tachment to the tentagel resin, since the trifluoroacetyl pro-
tection of guanidines, recently introduced by us,[8] is much
easier to remove than tosyl protection. Thus, guanidine 19
was attached to tentagel resin (0.2 mmolgꢀ1), and the pep-
tide arms constructed by using the standard Fmoc-protected
amino acid coupling chemistry as before (Scheme 5). Final
treatment with K2CO3 (0.15m) cleaved the trifluoroacetyl
group to give the resin-bound tweezer receptor 23.
The binding affinity of the resin-bound receptor 23 for
dye-labeled N-Ac-Lys-d-Ala-d-Ala (13) was measured by
determining the ability of the resin-bound receptor to
absorb the peptide guest from solution, following a previ-
ously described methodology.[3c,12] Thus, a known mass of
resin 23 was incubated with a solution of guest 13 in an
aqueous buffer and the change in the concentration of the
guest in free solution was monitored by UV spectroscopy.[12]
To account for nonspecific absorption of 13 to the polysty-
rene resin matrix, an equivalent mass of underivatized tenta-
gel resin was also incubated with 13. This allowed the
amount of 13 specifically bound by the tweezer receptor to
be determined, and hence the association constant, based on
a 1:1 receptor–substrate stoichiometry, was estimated as
Kass ~1350mꢀ1. An identical binding experiment was carried
out again by using the resin-bound receptor 23, but with the
Scheme 5. a) i) CH3I,
acetone;
ii) NH4PF6,
CH2Cl2,
CH3OH;
b) CF3CONH2, DBU, toluene, CHCl3, reflux; c) H2, Pd/C, DMF;
d) Fmoc-l-Glu(OBn)-OH, EDC, HOBt, DIPEA, DMAP, DMF; e) H2,
Pd/C, DMF; f) Tentagel resin, DIC, HOBt, DIPEA, DMF; g) 20% piper-
idine in DMF; h) Fmoc-Gly, DIC, HOBt, DIPEA, DMF; i) Fmoc-l-Val,
DIC, HOBt, DIPEA, DMF; j) Ac2O, Et3N, DMF; k) 20% CF3CO2H,
CH2Cl2; l) Fmoc-l-Met, DIC, HOBt, DIPEA, DMF; m) Fmoc-l-His(Trt),
DIC, HOBt, DIPEA, DMF; n) Fmoc-l-Ser(OtBu), DIC, HOBt, DIPEA,
DMF; o) 95% TFA, CH2Cl2, p) K2CO3 (0.15m) in MeOH/DMF/H2O
(2:2:1).
diastereomeric dye-labeled N-Ac-Lys-l-Ala-l-Ala, produc-
ing a value for Kass of ~250mꢀ1.
Conclusion
We have described novel solid-phase synthesis methodology
for the construction of “unsymmetrical” gaunidinyl tweezer
receptor structures. A relatively small library of these recep-
tors was prepared and screened to identify a receptor for
dye-labeled N-Ac-Lys-d-Ala-d-Ala. Binding studies on the
solid-phase confirm that the resin-bound receptor has signif-
icant (mm) affinity for the dye-labeled peptide, and reduced
affinity for the diastereoisomeric peptide, dye-labeled N-Ac-
Lys-l-Ala-l-Ala. Binding in free solution, however, proved
hard to establish, and although both UV and NMR spectro-
scopic studies indicated that complexation had occurred,
Chem. Eur. J. 2006, 12, 713 – 720
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
717

J. D. Kilburn et al.
the addition of toluene, the solvents were removed under reduced pres-
sure to give a yellow oil that was redissolved in CH2Cl2 (5 mL) and
binding affinities could not be determined. The failure of
binding on the solid-phase to translate to equivalent binding
in free solution suggests that the environment created by
the resin may have a significant role in influencing binding
affinity—possibly by excluding or organizing solvent mole-
cules within the resin matrix—or may suppress aggregation
of the tweezer receptor molecules, which occurs readily in
free solution. These difficulties might be addressed by at-
taching a polyethylene glycol chain to the receptor in free
solution to mimic more precisely the environment of the
resin-bound receptors[2d] or by linking a number of such
tweezer structures to a dendrimer. This latter strategy
might, ultimately, have an added benefit in creating multiva-
lent Lys-d-Ala-d-Ala receptors. Such studies are underway
in our laboratory. Furthermore, the library described herein
was relatively small given the diversity available just by
using the natural l-amino acids, let alone the structures that
may be produced by introducing other building blocks.
These limitations notwithstanding the results described
herein further emphasize the potential of these tweezer
structures to act as strong and selective peptide receptors,
even in competitive aqueous solvents.
added to
a solution of N-a-Fmoc-l-glutamic acid-w-tert-butyl ester
(320 mg, 0.75 mmol), EDC (144 mg, 0.75 mmol), and HOBt (203 mg,
0.75 mmol) in CH2Cl2 (15 mL) at room temperature. After the addition
of DIPEA (260 mL, 1.5 mmol), the resulting reaction mixture was stirred
for 36 h. CH2Cl2 (50 mL) was then added and the mixture washed with
water (60 mL). The organic layer was dried over magnesium sulfate and
the solvent was removed under reduced pressure to give a yellow solid.
Purification by column chromatography on silica gel (EtOAc/PE, 8:1!
EtOAc) produced 5 as a white hygroscopic foam (510 mg, 70% yield).
1H NMR (400 MHz, CDCl3): d=13.48 (brs, 1H; NH), 7.67 (d, J=7.5 Hz,
2H; ArH), 7.63 (d, J=8.0 Hz, 2H; ArH), 7.49 (d, J=7.5 Hz, 2H; ArH),
7.35–7.05 (m, 9H; ArH), 6.99 (d, J=7.5 Hz, 2H; ArH), 6.35 (brs, 1H;
NH), 4.39 (m, 2H; PhCH2), 4.28 (m, 2H; CH2), 4.13–4.00 (m, 2H), 3.45
(m, 2H), 3.30–3.10 (m, 6H), 2.40–1.80 (m, 11H), 1.36 (s, 9H; (CH3)3C),
0.91 ppm (s, 6H; (CH3)2C); 13C NMR (100 MHz, (CD3)2SO): d=198.1
(C), 174.2 (C), 174.0 (C), 173.0 (C), 156.6 (C), 155.6 (C), 143.8 (C), 142.1
(C), 141.4 (C), 141.1 (C), 135.8 (C), 129.3 (CH), 128.9 (CH), 128.1 (CH),
127.9 (CH), 127.2 (CH), 126.8 (CH), 126.0 (CH), 125.2 (CH), 120.2
(CH), 108.4 (C), 81.3 (C), 67.2 (CH2), 60.5 (CH2), 54.6 (CH2), 53.0 (CH),
47.3 (CH), 42.4 (CH2), 40.9 (CH2), 39.7 (CH2), 35.0 (CH2), 31.7 (CH2),
30.1 (C), 28.4 (CH3), 28.2 (CH3), 27.5 (CH2), 21.6 ppm (CH3); IR (neat):
n˜ =3322 (br), 2949 (w), 2926 (w), 1720 (m), 1561 (s), 1496 (w), 1450 (m),
1419 (w), 1365 (w), 1342 (m), 1245 (s), 1182 (w), 1131 (s), 1081 cmꢀ1 (s);
LRMS (ES): m/z: 947 [M+H]+, 969 [M+Na]+.
4-{2-{{{2-[1-(4,4-Dimethyl-2,6-dioxocyclohexylidene)-2-phenylethylami-
no]ethylamino}(toluene-4-sulfonylimino)methyl}amino}ethylcarbamoyl}-
4-(9H-fluoren-9-ylmethoxycarbonylamino)butyric acid (6): TFA (60% in
CH2Cl2, 11 mL) was added to a solution of ester 5 (470 mg, 0.48 mmol) in
CH2Cl2 (10 mL), and the mixture was stirred vigorously for 18 h at room
temperature. Toluene (150 mL) was then added and the solvents were re-
moved under reduced pressure to give a yellow oil. Column chromatog-
raphy on silica gel (10% MeOH in CH2Cl2) afforded acid 6 as a foam
(380 mg, 86%). 1H NMR (400 MHz, CDCl3): d=7.77 (d, J=8.0 Hz, 2H;
ArH), 6.50 (brs, 1H; NH), 7.73 (d, J=8.0 Hz, 2H; ArH), 7.60–7.05 (m,
11H; ArH), 6.98 (d, J=7.0 Hz, 2H; ArH), 4.74 (s, 2H; PhCH2), 3.55–
3.30 (m, 8H), 2.45–2.20 (m, 9H), 2.15 (m, 1H), 2.03 (m, 1H), 1.02 ppm
(s, 6H; (CH3)2C); 13C NMR (100 MHz, CDCl3): d=198.3 (C), 176.1 (C),
174.7 (C), 173.5 (C), 156.6 (C), 155.6 (C), 143.8 (C), 142.4 (C), 141.4 (C),
140.8 (C), 135.6 (C), 129.4 (CH), 129.0 (CH), 128.1 (CH), 127.9 (CH),
127.2 (CH), 126.8 (CH), 126.0 (CH), 125.2 (CH), 120.2 (CH), 108.3 (C),
67.4 (CH2), 54.2 (CH), 53.6 (CH2), 52.8 (CH2), 47.2 (CH), 42.7 (CH2),
38.7 (CH2), 40.8 (CH2), 36.2 (CH2), 35.0 (CH2), 30.1 (C), 28.3 (CH3) 27.8
(CH2), 21.6 ppm (CH3); IR (neat): n˜ =3319 (br), 2951 (w), 1705 (m),
1561 (s), 1496 (w), 1449 (m), 1341 (m), 1241 (m), 1181 (m), 1128 (s),
1079 cmꢀ1 (s); LRMS (ES): m/z: 914 [M+Na]+.
Experimental Section
General: Solvents for synthesis were purchased from Fisher or Rathburn
Chemicals. TentagelSNH2 or Rink Amide resin was used as the solid
support in peptide synthesis, and was purchased from Rapp Polymere
(Germany) or Novabiochem. Fmoc- and Boc-protected amino acids were
purchased from NovaBiochem or BAChem. All other chemicals were
purchased from Aldrich, Fluka, Lancaster, or NovaBiochem. Peptide syn-
thesis on the solid-phase was performed in polypropylene filtration tubes
with polyethylene frits on a Visiprep SPE vacuum manifold from Supel-
co. The reaction containers were agitated on a blood tube rotator (Stuart
Scientific Blood Tune Rotator SB1). Coupling was monitored by using
the Kaiser ninhydrin test. TLC was performed on aluminum-backed
plates (Merck silica gel 60 F254). Sorbisil C60, 40–60-mesh silica was
used for column chromatography. Melting points were determined in
open capillary tubes by using a Gallenkamp electrothermal melting point
apparatus. IR spectra were recorded on a Bio-Rad FT-IR spectrometer
as neat compounds. 1H NMR spectra were obtained at 300 MHz on a
Bruker AC 300 and at 400 MHz on a Bruker DPX 400. 13C NMR spectra
were recorded at 75 MHz on a Bruker AC300 and at 100 MHz on a
Bruker DPX 400. Coupling constants are given in Hz. The multiplicities
of the signals were determined by using the distortionless enhancement
phase transfer (DEPT) spectral editing technique. ES-MS were obtained
on a Micromass Platform II with a quadrupole mass analyzer or a Waters
ZMD with a quadrupole mass analyzer. UV titration experiments were
recorded on a Shimadzu UV-1601 UV/visible spectrophotometer.
General procedure for Fmoc deprotection on the solid-phase: The Fmoc-
protected resin was suspended in a solution of 20% piperidine in DMF
(20 mL per g resin) and agitated for 30 to 45 min. The resin was then
drained and washed with CH2Cl2 (3”10 mL), DMF (3”10 mL), and ad-
ditional CH2Cl2 (3”10 mL). The procedure was repeated once and the
progress of the deprotection monitored by a ninhydrin test.
General procedure for Boc deprotection on the solid-phase: The Boc-
protected resin was suspended in a mixture of 45% CH2Cl2, 45% TFA,
4% EDT, 3% DMS, and 3% anisole (20 mL per g resin) and agitated
for 60 to 120 min. The resin was then drained and washed with CH2Cl2
(3”10 mL), DMF (3”10 mL), a 20% solution of DIPEA in CH2Cl2 (3”
10 mL), MeOH (3”10 mL), DMF (3”10 mL), and additional CH2Cl2
(3”10 mL). The DIPEA solution wash was omitted for samples requiring
prolonged storage. The progress of the deprotection was monitored by a
ninhydrin test.
Abbreviations: Ac, acetyl; Boc, tert-butyloxycarbonyl; DIC, N,N-diiso-
propylcarbodiimide; DIPEA, diisopropylethylamine; DMAP, 4-(dimethyl-
amino)pyridine; DBU, 1,8-diazabicyclo[5.4.0]undec-7-ene; dppe, 1-(4,4-
dimethyl-2,6-dicyclohexylidene)phenylethyl; EDC, 1-ethyl-3-(3-dimethyl-
aminopropyl)carbodiimide; Fmoc, 9-fluorenyloxycarbonyl; HOBt, 1-hy-
droxybenzotriazole; PyBOP, benzotriazole-1-yl-oxy-tris-pyrrolino-phos-
phonium hexafluorophosphate; TFA, trifluoroacetic acid; Trt, trityl; Ts,
tosyl.
General procedure for tosyl deprotection on the solid-phase using the
HF procedure: The tosyl-protected resin was dried overnight under high
vacuum and placed in a Teflon apparatus. After the addition of p-thiocre-
sol and cresol, the reaction vessel was cooled with liquid nitrogen, and
liquid HF was condensed into the vessel (about 30 mL per g resin). The
cleavage mixture was stirred for 120 min at 08C, after which time the HF
tert-Butyl ester of 4-{2-{{{2-[1-(4,4-dimethyl-2,6-dioxo-cyclohexylidene)-2-
phenylethylamino]ethylamino}(toluene-4-sulfonylimino)methyl}amino}-
ethylcarbamoyl}-4-(9H-fluoren-9-ylmethoxycarbonylamino)butyric acid
(5): Carbamate 4[8] (500 mg, 0.75 mmol) was stirred in a 20% solution of
trifluoroacetic acid in CH2Cl2 (25 mL) at room temperature for 3 h. After
718
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 713 – 720

Tweezer Receptor Libraries
FULL PAPER
was evaporated under a stream of nitrogen. The resin was then washed
with diethyl ether (4”20 mL), 20% DIPEA in CH2Cl2 (3”20 mL),
MeOH (3”20 mL), and CH2Cl2 (3”20 mL).
and then dried. The success of the coupling step was monitored by a
qualitative ninhydrin test. Each coupling cycle was repeated until the nin-
hydrin test showed that no free amino functions were present. Each por-
tion was then Boc-deprotected to release the free amine of the coding
strand (see general procedure) and then coupled with a Boc-protected
amino acid (Gly, Ala, Phe, Val, and then Leu, corresponding with the
Fmoc-protected amino acids used in the previous coupling; 14 mmol) by
using PyBOP (7 mg, 14 mmol), HOBt (2 mg, 14 mmol), and DIPEA
(2 mg, 14 mmol). Once complete, the portions were washed with CH2Cl2
(3”5 mL), DMF (3”5 mL), additional CH2Cl2 (3”5 mL), and then
dried. After the successful coupling step, the resin was recombined and
Fmoc-deprotected, as described in the general procedure, with monitor-
ing by the qualitative ninhydrin test. This split-and-mixprocedure was
then repeated twice, with the same five amino acids, to build up the first
tripeptide side arm and the coding strand. For the next step in the syn-
thesis, the recombined resin was swollen in DMF, Fmoc-deprotected (see
general procedure), treated with excess acetic anhydride and DMAP, and
then agitated on a tube rotator for 4 h to cap the first tripeptide arm. The
resin was then treated with hydrazine hydrate (15 mL of a 35% aq solu-
tion) and agitated on a tube rotator for 4 h, to remove the dppe-protect-
ing group.[9] Once complete, the resin was divided into five portions, pre-
swollen in DMF (2 mL), and subjected to three rounds of split-and-mix
synthesis to build up the second tripeptide arm (Fmoc-protected amino
acids, Gly, Ala, Ser(OtBu), Hist(Trt), and then Met were used) and the
coding strand (with the corresponding Boc-protected amino acids), fol-
lowing the exact same procedures described above for the synthesis of
the first tripeptide arm and coding strand. Next, a solution of Fmoc-pro-
tected amino acid (Ala, Val, Gln, Ser(OtBu), and then Lys(Boc);
0.14 mmol), PyBOP (73 mg, 0.14 mmol), and HOBt (17 mg, 0.14 mmol)
in DMF (3 mL) was preactivated for a few minutes and then added to
each portion of resin, followed by the addition of DIPEA (18 mg,
0.14 mmol). Each portion was then agitated on a tube rotator for 24 h at
room temperature. After this time, the portions were washed with
CH2Cl2 (3”5 mL), DMF (3”5 mL), additional CH2Cl2 (3”5 mL), and
then dried. The success of the coupling step was monitored by a qualita-
tive ninhydrin test. As before, each coupling cycle was repeated until the
ninhydrin test showed that no free amino functions were present. After
the successful coupling step, the resin was recombined and Fmoc-depro-
tected, as described in the general procedure, with monitoring by a quali-
tative ninhydrin test. This split-and-mixprocedure was then repeated
twice more, with the same five Fmoc-protected amino acids, to build up
the second tripeptide side arm. After the recombination of all of the por-
tions of the resin, the protecting groups were removed by liquid HF
treatment (see general procedure) to give library 12.
Receptor library 9: TentaGelSNH2 resin (900 mg, 0.26 mmolgꢀ1
,
0.23 mmol) was preswollen in CH2Cl2 and then drained. A solution of
acid 6 (314 mg, 0.35 mmol), PyBOP (182 mg, 0.35 mmol), HOBt (47 mg,
0.35 mmol), and DIPEA (60 ml, 0.35 mmol) in DMF was added to the
resin and the mixture agitated on a tube rotator for 18 h, before being
drained and washed with CH2Cl2 (3”5 mL), DMF (3”5 mL), and addi-
tional CH2Cl2 (3”5 mL). A ninhydrin test gave a negative result. Follow-
ing Fmoc deprotection (see general procedure) the resin was divided into
five equal portions, and each portion was preswollen in DMF (2 mL). A
solution of Fmoc-protected amino acid (Gly, Leu, Phe, Glu(OtBu), or
Pro; 0.14 mmol), PyBOP (73 mg, 0.14 mmol), and HOBt (17 mg,
0.14 mmol) in DMF (3 mL) was preactivated for a few minutes, and then
added to each portion of the resin, followed by the addition of DIPEA
(18 mg, 0.14 mmol). Each portion was agitated on a tube rotator for at
least 24 h at room temperature. After this time, the portions were
washed with CH2Cl2 (3”5 mL), DMF (3”5 mL), additional CH2Cl2 (3”
5 mL), and then dried. The success of the coupling step was monitored
by a qualitative ninhydrin test. Each coupling cycle was repeated until
the ninhydrin test showed that no free amino functions were present.
After the successful coupling step, the resin was recombined and Fmoc-
deprotected, as described in the general procedure, with monitoring by a
qualitative ninhydrin test. This split-and-mixprocedure was then repeat-
ed with the same five amino acids (first, Fmoc-protected and then Boc-
protected) to build up the first tripeptide side arm. For the next step in
the synthesis, the recombined resin was swollen in DMF (10 mL), treated
with hydrazine hydrate (15 mL of a 35% aq solution), and agitated on a
tube rotator for 4 h, to remove the dppe-protecting group.[9] The resin
was then divided into five portions and each portion was preswollen in
DMF (2 mL). A solution of Fmoc-protected amino acid (Ala, Val, Gln,
Ser(OtBu), or Lys(Boc); 0.14 mmol), PyBOP (73 mg, 0.14 mmol), and
HOBt (17 mg, 0.14 mmol) in DMF (3 mL) was preactivated for a few mi-
nutes and then added to each portion of resin, followed by the addition
of DIPEA (18 mg, 0.14 mmol). Each portion was agitated on a tube rota-
tor for at least 24 h at room temperature. The portions were washed with
CH2Cl2 (3”5 mL), DMF (3”5 mL), additional CH2Cl2 (3”5 mL), and
then dried. The success of the coupling step was monitored by a qualita-
tive ninhydrin test. Each coupling cycle was repeated until the ninhydrin
test showed that no free amino functions were present. After the success-
ful coupling step, the resin was recombined and Fmoc-deprotected, as de-
scribed in the general procedure, with monitoring by the qualitative nin-
hydrin test. This split-and-mixprocedure was then repeated twice more
with the same five Fmoc-protected amino acids to build up the second
tripeptide side arm. After the recombination of all of the portions of the
resin, the protecting groups were removed by liquid HF treatment (see
general procedure) to give library 9.
Screening experiments:
Screening of the tweezer receptor library 18: A sample of library 12
(15 mg) was equilibrated in boraxbuffer solution (300 mL) for 24 h. A so-
lution of the tripeptide guest 13 (20 mM, 500 mL) in a 15% solution of
DMSO/boraxbuffer was then added to the library sample to give a guest
concentration of 12.5 mM. Equilibration was continued for 24 h. Analysis
of the beads was carried out in flat-bottomed glass pots, under a Leica in-
verted DML microscope (magnification”40), and 10 highly red stained
beads were selected and submitted for Edman sequencing.[10]
Receptor library incorporating coding strand 12: TentaGelSNH2 resin
(900 mg, 0.26 mmolgꢀ1, 0.23 mmol), was preswollen in CH2Cl2 and then
drained.
A solution of Boc-Phe (6 mg, 23 mmol), PyBOP (12 mg,
23 mmol), HOBt (31 mg, 23 mmol), and DIPEA (40 ml, 23 mmol) in DMF
(3 mL) was added to the resin and the mixture agitated on a tube rotator
for 18 h, before being drained and washed with CH2Cl2 (3”5 mL), DMF
Benzyl ester of 4-{2-[N’-(2-tert-butoxycarbonylaminoethyl)-N’’-(2,2,2-tri-
fluoroacetyl)guanidino]ethylcarbamoyl}-4-(9H-fluoren-9-ylmethoxycar-
bonylamino)butyric acid (20): Carbamate 19[8] (224 mg, 0.47 mmol) was
dissolved in DMF (10 mL). Palladium on charcoal (10% by wt,
10 mol%, 50 mg) was then added, and the mixture was stirred under a
hydrogen atmosphere for 16 h. After this time, the mixture was filtered
through Celite and the resulting filtrates were evaporated to give a pale
yellow oil. N-a-Fmoc-l-glutamic acid-w-benzyl ester (238 mg,
0.52 mmol), HOBt (127 mg, 0.94 mmol), DIPEA (410 ml, 2.36 mmol),
DMAP (6 mg, 47 mmol), and EDC (99 mg, 0.52 mmol) were added to a
solution of this oil in DMF (10 mL), and the resulting mixture was stirred
for 18 h. After this time, evaporation of all of the solvent and purification
of the residue by column chromatography (EtOAc/PE 1:1) produced 20
as a white hygroscopic foam (73 mg, 20%). 1H NMR (300 MHz, CDCl3):
d=9.65 (brs, 1H; NH), 7.76 (d, 2H, J=7.5 Hz; ArH), 7.57 (d, 2H, J=
(3”5 mL), and additional CH2Cl2 (3”5 mL).
A solution of acid 6
(314 mg, 0.35 mmol), PyBOP (182 mg, 0.35 mmol), HOBt (47 mg,
0.35 mmol), and DIPEA (60 mL, 0.35 mmol) in DMF was then added to
the resin and the mixture agitated on a tube rotator for a further 18 h,
before being drained and washed with CH2Cl2 (3”5 mL), DMF (3”
5 mL), and additional CH2Cl2 (3”5 mL). A ninhydrin test gave a nega-
tive result. Following Fmoc deprotection (see general procedure), the
resin was divided into five equal portions, and each portion was preswol-
len in DMF (2 mL). A solution of Fmoc-protected amino acid (Gly, Ala,
Phe, Val, or Leu; 0.14 mmol), PyBOP (73 mg, 0.14 mmol), and HOBt
(17 mg, 0.14 mmol) in DMF (3 mL) was preactivated for a few minutes
and then added to each portion of the resin, followed by the addition of
DIPEA (18 mg, 0.14 mmol). Each portion was agitated on a tube rotator
for 24 h at room temperature. After this time, the portions were washed
with CH2Cl2 (3”5 mL), DMF (3”5 mL), additional CH2Cl2 (3”5 mL),
Chem. Eur. J. 2006, 12, 713 – 720
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
719

J. D. Kilburn et al.
7.0 Hz; ArH), 7.42–7.16 (m, 9H; ArH), 7.16 (brs, 1H; NH), 7.01 (brs,
1H; NH), 5.76 (brs, 1H; NH), 5.12 (s, 2H; CH2Ph), 4.45–4.30 (m, 2H),
4.19 (m, 2H; CH2), 3.60–3.15 (m, 8H; CH2CH2), 2.49 (m, 2H;
CH2COO), 2.13 (m, 1H), 2.01 (m, 1H), 1.40 ppm (s, 9H; (CH3)3C);
13C NMR (75 MHz, CDCl3): d=173.3 (C), 173.2 (C), 162.7 (C), 157.6
(C), 157.0 (C), 156.6 (C), 143.8 (C), 141.4 (C), 135.7 (C), 128.7 (CH),
128.4 (CH), 128.3 (CH), 127.9 (CH), 127.2 (CH), 125.3 (CH), 120.1
(CH), 116.7 (C), 80.6 (C), 67.3 (CH2), 66.8 (CH2), 54.8 (CH), 47.1 (CH2),
41.9 (CH2), 41.0 (CH2), 39.6 (CH2), 38.9 (CH2), 30.4 (CH2), 28.4 ppm
(CH3); IR (neat): n˜ =1629 (m), 1522 (m), 1449 (m), 1242 (m), 1166 (m),
1140 (m), 848 (w), 739 (m), 514 cmꢀ1 (s); LRMS (ES): m/z: 783 [M+H]+,
805 [M+Na]+.
the resin with K2CO3 (0.15m) in MeOH/DMF/water (2:2:1, 2”3 h). Final-
ly, the resin was washed as above, rinsed with diethyl ether, and dried in
vacuo.
Acknowledgements
We wish to thank EPSRC for supporting this work (GRLꢀ98916/01) and
the BBSRC for a studentship (JS).
4-{2-[N’-(2-tert-Butoxycarbonylaminoethyl)-N’’-(2,2,2-trifluoroacetyl)gua-
nidino]ethylcarbamoyl}-4-(9H-fluoren-9-ylmethoxycarbonylamino)buty-
ric acid (21): Palladium on charcoal (10% by wt, 10 mol%, 8 mg) was
added to a solution of ester 20 (55 mg, 70 mmol) in EtOH (5 mL). The
mixture was stirred under a hydrogen atmosphere for 2 h and then fil-
tered through Celite. Finally, the solvent was removed by evaporation
under reduced pressure to give 21 as an off-white hygroscopic foam
(43 mg, 88%). 1H NMR (300 MHz, CDCl3): d=9.47 (brs, 1H; NH), 7.74
(d, 2H, J=7.5 Hz; ArH), 7.55 (d, 2H, J=6.5 Hz; ArH), 7.44 (brs, 1H;
NH), 7.42–7.22 (m, 4H; ArH), 6.18 (brs, 1H; NH), 5.88 (brs, 1H; NH),
4.45–4.38 (m, 3H), 4.17 (m, 1H; CH), 3.70–2.95 (m, 8H; CH2CH2), 2.44
(m, 2H; CH2COO), 2.18–1.82 (m, 2H), 1.40 ppm (s, 9H; (CH3)3C);
13C NMR (75 MHz, CDCl3): d=180.1 (C), 176.2 (C), 173.7 (C), 171.7
(C), 159.9 (C), 157.8 (C), 156.7 (C), 143.7 (C), 141.4 (C), 127.9 (CH),
127.3 (CH), 125.3 (CH), 120.1 (CH), 80.7 (C), 67.5 (CH2), 55.7 (CH),
54.6 (CH), 47.1 (CH2), 41.9 (CH2), 41.1 (CH2), 39.6 (CH2), 38.7 (CH2),
30.0 (CH2), 28.4 ppm (CH3); IR (neat): n˜ =2977 (br, w), 1628 (s), 1524
(m), 1449 (m), 1243 (m), 1139 (s), 909 (m), 844 (m), 734 cmꢀ1 (s); LRMS
(ES): m/z: 693 [M+H]+, 715 [M+Na]+.
Resin-bound receptor 23: TentaGelSNH2 resin (0.2 mmolgꢀ1, 162.2 mg,
32.4 mmol) was swollen in CH2Cl2 and drained. A solution of acid 21
(34 mg, 48.7 mmol), DIC (15 ml, 97.3 mmol), HOBt (13 mg, 97.3 mmol),
and DIPEA (17 ml, 97.3 mmol) in DMF (2 mL) was then added to the
resin, and the mixture agitated on a tube rotator for 18 h. Any remaining
amine residues were capped by treating the resin with an excess of acetic
anhydride. The resin was then filtered and washed with CH2Cl2 (3”
5 mL), DMF (3”5 mL), and additional CH2Cl2 (3”5 mL). A ninhydrin
test was negative. There then followed three sequential cycles of cou-
pling/Fmoc deprotection to add Fmoc-Gly, Fmoc-Val, and Fmoc-Val to
the resin, and to synthesize the first arm of the receptor. This was a-
chieved by using Fmoc-protected amino acid (97 mmol), DIC (12 mg,
97 mmol), HOBt (13 mg, 97 mmol), and DIPEA (17 ml, 97 mmol) in DMF
(2 mL). Ninhydrin tests were used to monitor progress of each coupling
reaction and the coupling reactions were repeated until complete. After
deprotection of the second Val residue, the chain was capped by the
treatment of the resin with acetic anhydride (9 ml, 97.3 mmol) and
DIPEA (17 ml, 97.3 mmol) in DMF (2 mL). Boc deprotection (see general
procedure) was followed by three sequential coupling/deprotection cycles
to add Fmoc-Met, Fmoc-His(Trt), and Fmoc-Ser(OtBu) to the resin also
by using the Fmoc-protected amino acid (97 mmol), DIC (12 mg,
97 mmol), HOBt (13 mg, 97 mmol), and DIPEA (17 ml, 97 mmol) in DMF
(2 mL); this time to synthesize the second arm of the receptor. Ninhydrin
tests were used to monitor the progress of each coupling reaction. Trt de-
protection was achieved by the treatment of the resin with 95% TFA/
CH2Cl2 (2”1 h), followed by washing with CH2Cl2 (3”5 mL), 50%
DIPEA/DMF (3”5 mL), DMF (3”5 mL), and additional CH2Cl2 (3”
5 mL). Trifluoroacetyl deprotection was carried out by the treatment of
[1] For a recent review on peptide receptors see: a) M. W. Peczuh,
A. D. Hamilton, Chem. Rev. 2000, 100, 2479–2493.
[2] a) C. Chamorro, R. M. J. Liskamp, Tetrahedron 2004, 60, 11145–
11157; b) C. Chamorro, J.-W. Hofman, R. M. J. Liskamp, Tetrahe-
dron 2004, 60, 8691–8699; c) C. Schmuck, M. Heil, Org. Biomol.
Chem. 2003, 1, 633–636; d) C. Schmuck, M. Heil, ChemBioChem
2003, 4, 1232–1238; e) K. B. Jensen, T. M. Braxmeier, M. Demarcus,
J. D. Kilburn, Chem. Eur. J. 2002, 8, 1300–1309.
[3] a) H. Wennemers, M. C. Nold, M. Conza, K. J. Kulicke, M. Neubur-
ger, Chem. Eur. J. 2003, 9, 442–448; b) H. Wennemers, M. Conza, J.
Org. Chem. 2002, 67, 2696–2698; c) R. Arienzo, J. D. Kilburn, Tetra-
hedron 2002, 58, 711–719; d) T. Braxmeier, M. Demarcus, T. Fes-
smann, S. McAteer, J. D. Kilburn, Chem. Eur. J. 2001, 7, 1889–1898;
e) J. D. Kilburn, T. Fessmann, Angew. Chem. 1999, 111, 2170–2174;
Angew. Chem. Int. Ed. 1999, 38, 1993–1996.
[4] Other recent examples of sequence selective peptide receptors:
a) C. Schmuck, L. Geiger, J. Am. Chem. Soc. 2004, 126, 8898–8899;
b) S. Rensing, T. Schrader, Org. Lett. 2002, 4, 2161–2164; c) R. Xu,
G. Greiveldinger, L. E. Marenus, A. Cooper, J. A. Ellman, J. Am.
Chem. Soc. 1999, 121, 4898–4899; d) M. Sirish, H.-J. Schneider, J.
Chem. Soc. Chem. Commun. 1999, 907–908; e) Md. A. Hossain, H.-
J. Schneider, J. Am. Chem. Soc. 1998, 120, 11208–11209; f) R. Bre-
slow, Z. Yang, R. Ching, G. Trojandt, F. Odobel, J. Am. Chem. Soc.
1998, 120, 3536–3537.
[5] a) N. Srinivasan, J. D. Kilburn, Curr. Opin. Chem. Biol. 2004, 8, 305–
310; b) B. Linton, A. D. Hamilton, Curr. Opin. Chem. Biol. 1999, 3,
307–312.
[6] C. Schmuck, P. Frey, M. Heil, ChemBioChem 2005, 6, 628–631.
[7] a) D. H. Williams, Nat. Prod. Rep. 1996, 13, 469–477; b) M. L.
Cohen, Science 1992, 257, 1050–1055; c) H. C. Neu, Science 1992,
257, 1064–1073.
[8] S. Bartoli, K. B. Jensen, J. D. Kilburn, J. Org. Chem. 2003, 68, 9416–
9422.
[9] S. R. Chhabra, B. Hothi, D. J. Evans, P. D. Wite, B. W. Bycroft, W. C.
Chan, Tetrahedron Lett. 1998, 39, 1603–1606.
[10] P. Edman, G. Begg, Eur. J. Biochem. 1967, 1, 80–91; sequencing was
performed by using an applied biosystems 477 A pulsed liquid-phase
sequencer, coupled online to an applied biosystems 120 A phenyl-
thiohydantoin (PTH) derivative analyzer.
[11] H. Rink, Tetrahedron Lett. 1987, 28, 3787–3790.
[12] a) M. Conza, H. Wennemers, Chem. Commun. 2003, 866–867;
b) M. C. F. Monnee, A. J. Brouwer, L. M. Verbeek, A. M. A. van -
Wageningen, R. M. J. Liskamp, Bioorg. Med. Chem. Lett. 2001, 11,
1521–1525; c) Y. Cheng, T. Suenaga, W. C. Still, J. Am. Chem. Soc.
1996, 118, 1813–1814.
Received: July 25, 2005
Published online: October 14, 2005
720
www.chemeurj.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 713 – 720