Selection of synthetic receptors capable of sensing the difference in binding of d-Ala-d-Ala or d-Ala-d-Lac ligands by screening of a combinatorial CTV-based library

Article abstract of DOI:10.1016/j.tet.2004.08.074

Screening of a combinatorial CTV-based artificial, synthetic receptor library 1 {1-13, 1-13, 1-13} for binding of a variety d-Ala-d-Ala and d-Ala-d-Lac containing ligands (6-11) was carried out in phosphate buffer (0.1 N, pH=7.0). After screening and Edma

Full text of DOI:10.1016/j.tet.2004.08.074

Tetrahedron 60 (2004) 11145–11157
Selection of synthetic receptors capable of sensing the difference
in binding of D-Ala-D-Ala or D-Ala-D-Lac ligands by screening
of a combinatorial CTV-based library
*
Cristina Chamorro and Rob M. J. Liskamp
Department of Medicinal Chemistry, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, P.O. Box 80082,
3508 TB Utrecht, The Netherlands
Received 23 January 2004; revised 24 June 2004; accepted 20 August 2004
Available online 7 October 2004
Abstract—Screening of a combinatorial CTV-based artificial, synthetic receptor library 1 {1-13, 1-13, 1-13} for binding of a variety D-Ala-
D-Ala and D-Ala-D-Lac containing ligands (6–11) was carried out in phosphate buffer (0.1 N, pHZ7.0). After screening and Edman
sequencing, synthetic receptors were found containing amino acid sequences, which are either characteristic for binding dye labeled D-Ala-D-
Ala or D-Ala-D-Lac containing ligands. For example, receptors capable of binding D-Ala-D-Ala containing ligands 6, 7, 9 and 11 contained—
almost in all cases—at least one basic amino acid residue—predominantly Lys—in their arms. This was really a striking difference with the
arms of the receptors capable of binding D-Ala-D-Lac containing ligands 8 and 10, which usually contained a significant number of polar
amino acids (Gln and Ser), especially in ligand 8, but hardly any basic amino acids. Use of different (fluorescent) dye labels showed that
the label has a profound, albeit not decisive, influence on the binding by the receptor. A hit from the screening of the CTV-library with
FITC-peptidoglycan (6) was selected for resynthesis and validation.
q 2004 Elsevier Ltd. All rights reserved.
1. Introduction
Recently, we have prepared a 2197-member split-mix3
combinatorial library of artificial tripodal receptors 1{1-13,
1-13, 1-13} on the solid phase,4 using cyclotriveratrylene
(CTV)5 as a scaffold (Scheme 1). Thus, every bead contains
one unique CTV-based receptor with three peptide arms,
one of them attached to the solid phase and different from
the other two identical peptide arms. Selectivity and affinity
for ligands might be enhanced, since the CTV-scaffold is
clearly capable of aligning peptide arms.6
The identification of artificial synthetic receptor molecules
capable of binding specific peptide sequences has attracted
considerable interest.1 Many of these receptors have been
designed, prepared and screened for their ability to bind
resin attached peptides from combinatorial libraries.
Combinatorial solid phase synthesis offers an attractive
strategy for the generation of libraries of synthetic receptor
molecules capable of binding specific peptide sequences.2
The CTV-based receptors were attached to Argogelw-NH2
resin, therefore an ‘on-bead screening assay’3c,7 for binding
of different peptide sequences by the CTV-based combi-
natorial library 1{1-13, 1-13, 1-13} can be performed.
Beads containing receptors capable of binding particular
ligands are visually identified and removed for structural
elucidation by Edman degradation.8 In order to visualize the
best CTV-receptor binders, fluorescence or staining of the
library beads resulting from binding a specific peptide
sequence or ligand in solution is an attractive approach.
For this purpose labeling with a (fluorescent) dye of these
peptides or ligands is required.
Keywords: Synthetic receptors; CTV-based library; Screening; Fluorescent
ligand; D-Ala-D-Ala-OH; D-Ala-D-Lac-OH.
Abbreviations: Boc, tert-butoxycarbonyl; BOP, benzotriazol-tris-(dimethyl-
t
amino)phosphonium hexafluoro-phosphate; Bu, tert-butyl; Cbz, benzyl-
oxycarbonyl; CTV, 2,7,12-trihydroxy-3,8,13-trimethoxy-10,15-dihydro-
5H-tribenzo[a,d,g]cyclononene; DCM, dichloromethane; DiPEA, N,N-
diisopropyl-N-ethylamine; DMF, dimethylformamide; Ds, Dansyl; ES-
MS, elctrospray ionization mass spectrometry; EtOAc, ethyl acetate; Et2O,
diethyl ether; FITC, fluorescein isothiocyanate; Fmoc, N-fluoren-9-
ylmethoxy-carbonyl; NBD, 7-nitrobenz-2-oxa-1,3-diazol; NMP, N-methyl-
pirrolidone; Rt, retention time; rt, room temperature; TEA, triethylamine;
TFA, trifluoroacetic acid.
As a still very challenging example of screening for binding
of specific peptide sequences by the CTV-based library
1{1-13, 1-13, 1-13}, peptide sequences D-Ala-D-Ala and
* Corresponding author. Tel.: C31-30-2537396; fax: C31-30-2536655;
e-mail: r.m.j.liskamp@pharm.uu.nl
0040–4020/$ - see front matter q 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2004.08.074

11146
C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
Scheme 1. Combinatorial synthesis of a 2197-member library of CTV-based tripodal receptors 1{1-13, 1-13, 1-13}.4
D-Ala-D-Lac were chosen. Vancomycin and other closely
related glycopeptide antibiotics9 bind to the D-Ala-D-Ala
end of the Gram-positive bacterial peptidoglycan cell wall
precursor thereby interfering with its cross-linking, resulting
in an incomplete bacterial cell wall, leading to osmotic lysis
and bacteria death.10 Unfortunately, the terminal D-alanine
is substituted by D-lactate in vancomycin-resistant bacterial
strains, resulting in a 1000-fold decrease in affinity and
essentially useless antibiotic activity of vancomycin.11
Thus, identification of synthetic receptors capable of
binding with high affinity to the above mentioned sequences
could have a potential application for overcoming vanco-
mycin resistance problems. This is nevertheless an extre-
mely difficult problem to address, since the number of
interactions of the small D-Ala-D-Ala and D-Ala-D-Lac
ligands with either ‘synthetic’ or ‘natural’ receptors (viz.
vancomycin) is very limited. Despite this, Nature managed
to come up with the natural ‘receptor’ vancomycin having a
high affinity for D-Ala-D-Ala (Kaw106 MK1).12
systems and finding hits for these. We found that it is no
longer difficult to find hits of synthetic receptors with good
to excellent binding properties in organic solvents.1c,f,h,2b–c
The challenge, however, is to discover molecules with good
binding properties in an aqueous environment, since
intermolecular interactions (hydrogen bond, electrostatic)
between ligand and receptor are much weaker here than in
apolar solvents.
Since D-Ala-D-Ala or D-Ala-D-Lac ligands are relatively
small, the influence of the (fluorescent) dye was evaluated
by varying it from fluorescein (FITC), dansyl (Ds), or
7-nitrobenz-2-oxa-1,3-diazol (NBD), to disperse red (DR).
The best synthetic receptors were selected and subjected to
Edman degradation8 and the binding sequences were
analyzed. Validation of ‘hits’ by resynthesis on the solid
phase of the selected library member has been carried out.
The resynthesized receptors have been tested in vitro against
Staphylococcus aureus Newman.
In this report a 2197-member library of CTV-based
synthetic tripodal receptor 1{1-13, 1-13, 1-13} attached to
the solid phase was screened for binding of D-Ala-D-Ala and
D-Ala-D-Lac containing ligands, in phosphate buffer (0.1 N,
pHZ7.0). So far there are not many examples of screening
for binding properties by synthetic receptors in aqueous
systems.2a–b,13 However, this was deemed absolutely
essential for moving ultimately to biologically relevant
One can philosophize how Nature did find such a good
candidate for binding as vancomycin. However, it seems
plausible that the ‘natural’ equivalent of combinatorial
approaches, i.e. evolution led to the emergence of such a
good binder for binding D-Ala-D-Ala. Similarly, we propose
that ultimately combinatorial approaches will lead to new
binders and possibly antibiotics. Here, we describe the
screening of a CTV library as a step in this direction.

C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
11147
2. Results and discussion
(2) (92%), followed by BOP coupling of Cbz-Gly-OH to 2,
and again hydrogenolysis.
In order to investigate the binding properties for D-Ala-D-
Ala and D-Ala-D-Lac containing ligands by the fully
deprotected recently described CTV-based receptor library
1{1-13, 1-13, 1-13}4, fluorescent FITC-peptidoglycan,14
(6), NBD-(CH2)5-C(O)-Gly-D-Ala-D-Ala-OH (7), and
Ds-Gly-D-Ala-D-Ala-OH (9), as well as red-colored ligand
DR-C(O)–(CH2)2–C(O)-Gly-D-Ala-D-Ala-OH (11) were
prepared corresponding to cell wall precursors of vanco-
mycin sensitive bacteria. In addition, fluorescent NBD-
(CH2)5-C(O)-Gly-D-Ala-D-Lac-OH (8), and Ds-Gly-D-Ala-
D-Lac-OH (10) were synthesized as a small parts of cell wall
precursors of vancomycin resistant bacteria. The (fluor-
escent) dye labels were chosen to study their possible
influence on binding.
NBD-(CH2)5-C(O)-Gly-D-Ala-D-Ala-OH (7) was syn-
thesized by BOP coupling of fluorescent 6-(7-NitroBenzo-
2-oxa-1,3-Diazole) hexanoic acid to 4 and cleavage of the t-
Bu-ester under acidic conditions. Treatment of 4 with a
slight excess of dansyl chloride in the presence of NaHCO3
and cleavage of the t-Bu-ester led to the dansylated D-Ala-D-
Ala containing ligand (Ds-Gly-D-Ala-D-Ala-OH) 9. Similar
approaches were used, for the preparation of the NBD and
Ds-labeled D-Ala-D-Lac containing ligands 8 and 10. For
these ligands H-Gly-D-Ala-D-Lac-OtBu (5) was used as a
precursor. This precursor was synthesized analogously to 4,
but usingH-D-Lac-OtBu insteadofH-D-Ala-OtBu (Scheme 2).
Disperse red D-Ala-D-Ala containing ligand (11) was
synthesized by treatment of 4 with succinic anhydride
under basic conditions followed by BOP coupling of
2.1. Synthesis of the (fluorescent) dye labeled D-Ala-D-
Ala and D-Ala-D-Lac containing ligands (7–11)
16
disperse red-NH2 and final cleavage of t-Bu-ester.
The D-Ala-D-Ala containing ligands (7, 9 and 11) were
prepared by attachment of the different fluorescent or dye
labels to H-Gly-D-Ala-D-Ala-OtBu (4) (Scheme 2). The
common intermediate 4 was obtained by BOP coupling15 of
commercially available Cbz-D-Ala-OH and H-D-Ala-OtBu
and hydrogenolysis to the dipeptide H-D-Ala-D-Ala-OtBu
2.2. Screening of library 1{1-13, 1-13, 1-13} for binding
of D-Ala-D-Ala or D-Ala-D-Lac containing ligands
The fully deprotected CTV-based receptor library 1{1-13,
1-13, 1-13} was screened for binding of the dye containing
Scheme 2. Synthesis of the (fluorescent) dye labeled D-Ala-D-Ala and D-Ala-D-Lac containing ligands (7–11).

11148
C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
Figure 1. Screening for binding for D-Ala-D-Ala and D-Ala-D-Lac containing ligands (6–11) by the CTV-based receptor library 1{1-13, 1-13, 1-13}.
D-Ala-D-Ala or D-Ala-D-Lac ligands (6–11) in phosphate
buffer (0.1 N, pHZ7.0). Although this aqueous system is a
biologically more meaningful environment, peptide recog-
nition by the artificial receptors is more difficult than in the
frequently used apolar organic systems due to the ability of
water molecules to compete for hydrogen bonding by
solvation and thus disruption of these non-covalent
interactions between peptides and synthetic receptors. For
the fluorescent dyes the overexposure mode of the Leica
DC-100 digital camera system and image analysis was used
for estimation of the relative fluorescence intensities in a
semi-quantitative manner. Different ligand concentrations
were used to determine the optimal selectivity, as judged
by the number of highly fluorescent beads against either
no or slightly fluorescent beads. For NBD- or Ds-labeled
ligands (7–10), no selectivity of binding at higher
concentrations than 50 mM was observed. At concen-
trations lower than 10 mM, fluorescent beads were barely
visible. Good discrimination between fluorescent and
non-fluorescent beads was possible at 50 mM, and there-
fore this concentration was used for screening experiments
with fluorescent ligands (7–10). FITC-labeled peptido-
glycan fragment ligand (6) was tested at concentrations of
10, 1 and 0.1 mg mLK1, leading to the best discrimination
when a concentration of 1 mg mLK1 was used. The best
discrimination between red and colorless beads was
obtained when 5 mM of the disperse-red labeled ligand 11
was used. In general, the observed fluorescence intensity
was slightly higher for D-Ala-D-Ala containing ligands
than for D-Ala-D-Lac containing ligands. This might be
Table 1. Identity of the peptide arms as obtained by Edman degradation of the selected solid phase attached CTV-based receptors
NF, non-fluorescent bead.

C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
11149
a qualitative indication of stronger binding of the
D-Ala-D-Ala sequence. A higher fluorescence intensity
was generally observed when NBD-label ligands were
used instead of the Ds-label. The best selectivity was
found when FITC-peptidoglycan (6) was used as ligand
(Fig. 1).
colored beads, contained amino acids especially Glu and
Asp in their receptors, which were not found in the
fluorescent or colored beads. Second, the receptors capable
of binding of D-Ala-D-Ala containing ligands 6, 7, 9 and 11
contained—almost in all cases—at least one basic amino
acid residue (indicated in blue in Table 1)—predominantly
Lys—in their arms. This was really a striking difference
with the arms of the receptors capable of binding D-Ala-D-
Lac containing ligands 8 and 10, which usually contained a
significant number of polar amino acids (Gln and Ser),
especially in binding ligand 8, but hardly any basic amino
acids. Although there was a significant influence of the label
(vide infra) on binding of the ligand to the synthetic
receptors, fortunately, the labels did not conceal the
differences in binding of D-Ala-D-Ala or D-Ala-D-Lac
ligands, i.e. they did not obscure the selectivity of binding.
The differences in the sequences of the receptors in binding
of D-Ala-D-Ala or D-Ala-D-Lac remained manifest irrespec-
tive of the (fluorescent) dye label. Nevertheless, there were
appreciable differences in sequences of receptors capable of
binding the D-Ala-D-Ala with the various labels. For
example, the synthetic receptors capable of binding
For every independent screening nine or ten of the most
intensively fluorescent beads were selected and sequenced
by Edman degradation.8 Edman sequencing was carried out
in all the cases for a fourth cycle, which always showed no
amino acid for all the beads, thus confirming the tripeptide-
identity of the library members. In addition, a non-
fluorescent bead was selected and sequenced as a negative
control.
Table 1 shows the identity of the peptide arms of the
selected solid phase attached CTV-based receptors. In
Figure 2, the relative abundance of amino acids at each
residue position of the receptor is shown. The general trends
are very clear and can be summarized as follows. First, the
negative controls, i.e. the selected non-fluorescent or non-
Figure 2. Relative abundance of amino acids at each residue position of the CTV-based receptor.

11150
C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
‘NBD-D-Ala-D-Ala’ (7) contained—in addition to Lys—
virtually only hydrophobic amino acids (Leu, Phe, Val),
whereas the synthetic receptors capable of binding ‘Ds-D-
Ala-D-Ala’ (9) contained—in addition to Lys—a number of
more polar Ser residues as well as the smaller hydrophobic
amino acids such as Ala. The behavior of synthetic receptors,
which bound ‘DR-D-Ala-D-Ala’ (11) was somewhat
intermediate: Lys residues, large hydrophobic amino acids
(Leu, Phe) but also the smaller hydrophobic amino acid Ala
was found, but no polar amino acids like Ser. Thus, we found
synthetic receptors selective for binding either the D-Ala-D-
Ala or the D-Ala-D-Lac sequence. As such these compounds
can be considered as ‘sensors’ for these particular peptide
sequences and synthetic receptors acting as sensors for other
peptide ligands might be accessible using this combinatorial
approach. Possibly, by simply screening this library with
another peptide ligand one might even uncover synthetic
receptors acting as sensors for other peptides. The binding
behavior of the synthetic receptors—as reflected in the amino
acid sequences—was influenced by the character of the
(fluorescent) dye label, but the selectivity remained.
hydrophobic amino acids was very striking in the receptors
capable of binding D-Ala-D-Lac, i.e. 8 and 10. In the latter
case even 50% of the sequenced receptors contained
hydrophobic amino acids at each of the three amino acid
positions and a good 70% had at least two hydrophic amino
acids, i.e. at the third (AA3) and second (AA2) position of
the synthetic receptors. Finally, the presence of Ser in the
third position in 50% of the synthetic receptors binding
‘Ds-D-Ala-D-Ala’ (9) was also surprising and in 30% of the
receptors the consensus sequence Ser-Lys-AA1 was found.
This cannot be solely ascribed to an effect of the Ds-label, as
the corresponding Ds-containing D-Ala-D-Lac ligand did not
bind to receptors containing Ser residues.
2.3. Validation by resynthesis
Validation of ‘hits’ by resynthesis of the selected synthetic
receptors and reproducing the properties for which the
library (1{1-13, 1-13, 1-13}) was screened is essential in
any combinatorial approach towards finding compounds
with desired—in this case binding—properties. A hit from
the screening of the CTV-library with FITC-peptidoglycan
(6) was selected, which was found twice (Table 1, beads 1
and 2, screening with 6), corresponding to the consensus
sequence Lys-AA2-Lys found in 44% of the sequenced
beads. In addition the ‘negative control’, i.e. a sequenced
non-fluorescent bead giving the CTV-based synthetic
receptor sequence Ser-Ala-Glu, (AA3-AA2-AA1) was
synthesized on Argogelw-NH2 resin (a) as well as on the
Argogelw-Rink resin (b) (Scheme 3). Resynthesis of the
CTV-based receptors 19a and 20a, covalently attached to
Argogelw-NH2 resin, and 17b and 18b on the Argogelw-
Rink resin, was performed analogously to the preparation of
the CTV-based receptor library 1{1-13, 1-13, 1-13}4 except
for the use of the compatible (Fmoc)2-CTV-OH scaffold 14
on the acid labile Argogelw-Rink resin, instead of the
(Boc)2-CTV-OH scaffold 13. The different peptide-binding
arm was built up by three subsequent BOP couplings/Fmoc-
deprotection of Fmoc-L-Ala-OH on both, Argogelw-NH2
and Argogelw-Rink-NH2 resin to obtain 12a and 12b,
respectively (Scheme 3). Then, either the Boc-protected
CTV-scaffold (13) or the Fmoc-protected CTV-scaffold
(14) was coupled to the first peptide-binding arm to give 15a
or 15b. Cleavage of the Boc groups under acidic conditions
from 15a led to 16a, while Fmoc-cleavage under standard
conditions from 15b led to 16b. The construction of the two
identical peptide arms was performed as depicted in Scheme
3. After three coupling/Fmoc-deprotection cycles involving
subsequently Fmoc-L-Lys(Boc)-OH, Fmoc-L-Leu-OH and
Fmoc-L-Lys(Boc)-OH, resin bound synthetic receptors 17a
and 17b were obtained. The same procedure involving the
use of subsequently Fmoc-L-Glu(OtBu)-OH, Fmoc-L-Ala-
OH and Fmoc-L-Ser(tBu)-OH led to resin bound synthetic
receptors 18a and 18b. Treatment of 17a and 18a with TFA
led to the—still resin bound—deprotected synthetic recep-
tors, 19a and 20a, respectively (Scheme 3). In contrast,
treatment with TFA of 17b or 18b led to deprotection and
cleavage of the receptors from the Argogelw-Rink-resin to
afford the trifluoroacetate salts of [Lys-Leu-Lys]-containing
or [Ser-Ala-Glu]-containing synthetic receptors 21 and 22,
respectively (Scheme 3). ES-MS showed the identity of the
trifluoroacetate salts receptors 21 and 22. The purity of them
was 94% purity in both cases, as shown by HPLC. The
In order to reduce the influence of the label, which is
required for screening, different labels which have a
minimal influence on binding, are called for. Alternatively,
the relative influence of the label on binding by a particular
synthetic receptor can be diminished by screening synthetic
receptors capable of binding larger ligands as compared
to the size of the label. The found selectivities for binding
D-Ala-D-Ala vs. binding D-Ala-D-Lac are quite remarkable
considering the fact that the only difference between these
ligands is present in the linkage connecting the two resi-
dues: an amide ‘N–H’ in D-Ala-D-Ala and an ester ‘–O’ in
D-Ala-D-Lac.
It is revealing to look at the relative abundance of the
various amino acids at each residue position of the receptor
shown in Figure 2. In 89% of the sequenced synthetic
receptors capable of binding fluorescein labeled D-Ala-D-
Ala (FITC-peptidoglycan, 6), Lys is found at the first
position (AA1), closest to the CTV-scaffold. Lys in this
position is also found in 44% of the receptors capable of
binding ‘NBD-D-Ala-D-Ala’ (7). Although in 44% of the
receptors binding FITC-peptidoglycan (6), Lys is found at
the third position (AA3), Lys can also be present at the
second position (AA2) in 33% of the receptors binding
‘NBD-D-Ala-D-Ala’ (7), in 30% of the receptors binding
‘Ds-D-Ala-D-Ala’ (9) and in 33% of the receptors
binding ‘DR-D-Ala-D-Ala’ (11). As was discussed above,
the absence of Lys residues in synthetic receptors capable of
binding D-Ala-D-Lac in ligands 8 and 10 is remarkable.
Likewise, the combination of the basic Lys residue with
hydrophobic amino acids indicated in green in the receptor
sequences, when a varying hydrophobic amino acid is
found. Especially with the NBD-ligands 7 and 8, Phe was
found in 33% of the receptors binding D-Ala-D-Ala (in
ligand 7) and in 80% (40% AA1 and 40% AA2) of the
receptors binding D-Ala-D-Lac (in ligand 8). The presence of
this particular amino acid may point to a special
contribution of the NBD label on binding, since this
amino acid was not in majority present in the receptors
binding the other ligands. Nevertheless the presence of

C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
11151
Scheme 3. Resynthesis of resin bound (19a and 20a) and cleaved synthetic receptors (21 and 22).
overall yields (76 and 69%, respectively) were very good
considering the eight steps performed for their solid-phase
synthesis (corresponding to 97 and 95% average yield per
step). Finally, desalting using either Amberlyst–A21 or
Dowex-HC ion exchange resins led to receptors 21 and 22,
respectively (Scheme 3).
Binding experiments of the resynthesized CTV-based
Figure 3. Screening for binding for D-Ala-D-Ala and D-Ala-D-Lac containing ligands (6, 7, 9, and 11) by the resynthesized CTV-based receptor 19a and 20a.

11152
C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
receptors 19a and 20a was carried out by incubation with a
1 mg mLK1 suspension of FITC-peptidoglycan (6). Homo-
geneous fluorescent beads were observed for receptor 19a,
clearly showing affinity for FITC-peptidoglycan while no
fluorescence was observed for the control receptor 20a (Fig.
3). These findings independently confirmed the identity of
the receptors as obtained by the split-mix procedure.
Use of different (fluorescent) dye labels showed that the
label has a profound, albeit not decisive, influence on the
binding by the receptor. In view of the facts that D-Ala-D-
Ala as well as D-Ala-D-Lac are only small ligands, with very
limited of interactions and that the only difference between
these ligands is present in the linkage connecting the two
residues: an amide ‘N–H’ in D-Ala-D-Ala and an ester ‘–O’
in D-Ala-D-Lac, underlines the significance of the findings.
Nevertheless, a resynthetisized receptor based on the
sequences found in the library screening was inactive
against Gram-positive bacteria. If degradation of the
synthetic receptor by proteases is largely responsible for
this observation, then synthetic receptors containing
D-amino acids might be more promising, which is under
study now.
In order to investigate whether the affinity of the
resynthesized CTV-based receptor 19a was due to either
their interaction with D-Ala-D-Ala or with a different part
from the FITC-peptidoglycan ligand (6), the binding of
resin bound receptor 19a for NBD, Dansyl- and disperse-
red-D-Ala-D-Ala containing ligands 7, 9, and 11 was tested.
For that purpose the [Lys-Leu-Lys]-containing CTV
receptor 19a and the ‘control’ [Ser-Ala-Glu]-containing
CTV-based receptor 20a were incubated with a 50 mM
solution of ligands 7 or 9 or a 5 mM solution of ligand 11.
Homogeneous fluorescent beads were observed in all cases
involving synthetic receptor 19a, while no fluorescent beads
were observed with control synthetic receptor 20a (Fig. 3).
Thus, binding is independent from the (fluorescent) dye
label and really seems to involve recognition of the D-Ala-D-
Ala sequence.
4. Experimental
4.1. General
All reagents were purchased from commercial sources and
used without further purification. Argogelw-NH2
(0.40 mmol gK1, average bead diameter 178 mm), Argo-
gelw-Rink-NH-Fmoc (0.33 mmol gK1, average bead
diameter 164 mm) resins were purchased from Argonaut
Technologies, Inc., Amberlys-A21 ion exchange resin
(4.8 mmol gK1) from Aldrich, and Dowex 50!8 (HC
form, 20–50 mesh) from Fluka. The Argogelw-NH2 resin
was dried over P2O5. The protected amino acids were
purchased from Alexis Corporations (La¨ufelfingen,
Switzerland) and Advanced Chemtech Europe (UK). All
reactions on the solid phase were performed in standard
glassware or poly(ethylene) glycol (PE) syringes with PE
frits. Peptide grade solvents, dried on molecular sieves were
used for reactions and resin washing. Kaiser and bromo-
phenol blue (BPB) tests were used for detection of primary
amines on the solid phase.17,18 Analytical thin layer
chromatography (TLC) was performed on Merck pre-coated
silica gel 60 F254 (0.25 mm) plates. Spots were visualised
with UV light, ninhydrine, or Cl2-TDM.19 Column chro-
matography was carried out using Merck Kieselgel 60
In view of these results, it was attempted to inhibit bacterial
growth by the resynthesized [Lys-Leu-Lys]-containing
CTV-receptor 21. [Ser-Ala-Glu]-containing CTV-receptor
22 was also included in the experiments. A range of
concentrations from 0 to 100 mM of 21 or 22 were used to
determine a possible minimal inhibitory concentration
(MIC) against Staphylococcus aureus Newman. Vanco-
mycin was used as a positive control. Unfortunately,
no inhibition of bacterial growth was observed for the
CTV-based receptor 21 indicating no cell wall disruption
and therefore no bacterial death. Negative control synthetic
receptor 22 was also inactive. Although a possible
explanation for the absence of antibacterial activity by
synthetic receptor 21 might be too low a binding affinity
for the D-Ala-D-Ala sequence, an alternative explanation is
proteolytic degradation of the peptide arms of synthetic
receptor 21.
1
(40–63 mm). H NMR and 13C NMR were obtained on a
Varian 300 MHz spectrometer. Chemical shifts are given in
1
ppm with respect to internal TMS for H NMR. 13C NMR
spectra were recorded using the attached proton test (APT)
pulse sequence. ES-MS experiments were performed on a
Shimadzu LCMS QP8000 system. HPLC analyses were
performed on a Shimadzu-10Avp (Class VP) using
PL-ELS-1000 detector and UV detector operating at 220
3. Conclusions
Thus by screening a CTV-based library containing tripep-
tide arms, one attached to the solid phase and two identical
varying arms, we were able to find synthetic receptors
containing amino acid sequences, which are either charac-
teristic for binding dye labeled D-Ala-D-Ala or D-Ala-D-Lac
containing ligands. This opens up possibilities for finding by
combinatorial approaches synthetic receptors, which can act
as ‘sensors’ for particular peptides. By attachment of
fluorescent groups to these sensors or by immobilizing
them to suitable surfaces one might be able to devise read-
out systems for monitoring the extent of binding of peptide
or other ligands. Selectivity of binding by synthetic
receptors as sensors might be conveniently determined by
exposing them to for example to a library of peptides,
followed by measuring the relative binding affinity by mass
spectrometry.
˚
and 254 nm. An Adsorbophere XL C18, 300 A (4.6 mm!
250 mm, 5 mm) was used; flow rate 1 mL minK1; two
mobile phases (mobile phases A, 99.9% water, 0.1% TFA;
mobile phase B, 5% water, 95% acetonitrile, 0.085% TFA)
with a standard protocol, 0% B for 5 min to 100% B in
20 min, 100% B for 5 min to 0% B in 5 min and
reequilibrated at 0% B for 5 min.
4.1.1. H-D-Ala-D-Ala-OtBu (2). To a solution of Cbz-D-Ala-
OH (1.16 g, 5 mmol) and HCl$H-D-Ala-OtBu (0.91 g,
5 mmol) in DCM (30 mL), BOP (2.21 g, 5 mmol) and
DiPEA (2.61 mL, 15 mmol) were added. After stirring at rt
for 3 h, the reaction was evaporated to dryness in vacuo and

C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
11153
the residue was dissolved in EtOAc (60 mL). The solution
was washed with Na2CO3 (1 M), KHSO4 (1 M), H2O, and
brine. After drying (Na2SO4) and evaporating the solvent,
Cbz-D-Ala-D-Ala-OtBu (1.61 g, 92%) was obtained as white
solid and used without further purification. RfZ0.56 (DCM/
methanol, 10:1). 1H NMR (CDCl3, 300 MHz) d: 7.33 (s, 5H,
Ph), 6.42 (bs, 1H, NH(CO)), 5.29 (bs, 1H, NH(CO)O), 5.10
(s, 2H, CH2Ph), 4.40, 4.22 (2m, 2H, CHa), 1.44 (s, 9H,
C(CH3)3), 1.37, 1.34 (2d, 6H, CH3b, JZ6.9 Hz). 13C NMR
(CDCl3, 75 MHz) d: 171.8, 171.5 (COO, NH(CO)), 155.8
(NH(CO)O), 136.2 (Ar-C), 128.5, 128.1, 128.0 (Ar-H), 82.1
(C(CH3)3), 66.9 (CH2Ph), 50.4, 48.6 (Ca), 27.9 (C(CH3)3),
18.4 (CH3b).
CH2Ph), 4.53, 4.38 (2m, 2H, CHa), 3.87 (d, 2H, CH2-Gly,
JZ5.2 Hz), 1.43 (s, 9H, C(CH3)3), 1.35, 1.33 (2d, 6H,
CH3b, JZ6.9 Hz). 13C NMR (CDCl3, 75 MHz) d: 171.8,
171.3, 168.6 (COO, 2NH(CO)), 156.6 (NH(CO)O), 136.1
(Ar-C), 128.5, 128.2, 128.1 (Ar-H), 82.1 (C(CH3)3), 67.2
(CH2Ph), 48.8 (Ca), 44.4 (Ca-Gly), 27.9 (C(CH3)3), 18.7,
18.3 (CH3b). ES-MS (m/z)Z430.35 (100%, [MCNa]C),
408.30 (9%, [MCH]C), 352.25 (42%, [MKtBuCH]C).
Hydrogenolysis of Cbz-Gly-D-Ala-D-Ala-OtBu (1.30 g,
3.19 mmol) gave 4 (0.85 g, 98%) as a colorless syrup.
RfZ0.06 (DCM/methanol, 5:1). 1H NMR (CDCl3,
300 MHz) d: 7.71 (d, 1H, NH(CO), JZ7.7 Hz), 7.04 (d,
1H, NH(CO), JZ6.9 Hz), 4.51, 4.33 (2m, 2H, CHa), 3.29
(s, 2H, CH2-Gly), 1.39 (s, 11H, NH2, C(CH3)3), 1.33 (d, 3H,
CH3b, JZ6.9 Hz), 1.28 (d, 3H, CH3b, JZ7.1 Hz). 13C
NMR (CDCl3, 75 MHz) d: 172.7, 171.8, 171.7 (COO,
2NH(CO)), 81.7 (C(CH3)3), 48.6, 48.2 (Ca), 44.6 (Ca-Gly),
27.8 (C(CH3)3), 18.4, 18.1 (CH3b).
To a cooled solution (ice bath) of Cbz-D-Ala-D-Ala-OtBu
(1.40 g, 4 mmol) in methanol (200 mL), Pd on charcoal
(10%) (20% w/w) was added and the suspension was stirred
under H2 pressure from a balloon at rt for 2 h. The reaction
was filtered and the solvent was evaporated to give 2
(0.86 g, 99%) as a colorless syrup. RfZ0.14 (DCM/
1
4.1.4. H-Gly-D-Ala-D-Lac-OtBu (5). Overnight coupling
(see preparation of 2) of Cbz-Gly-OH (0.67 g, 3.2 mmol) to
3 (0.70 g, 3.2 mmol) gave Cbz-Gly-D-Ala-D-Lac-OtBu
(1.28 g, 98%) as a colorless syrup after column chromatog-
raphy using a gradient DCM/methanol from 40:1 to 20:1.
RfZ0.77 (DCM/methanol, 10:1). 1H NMR (CDCl3,
300 MHz) d: 7.33 (s, 5H, Ph), 6.51 (d, 1H, NH(CO), JZ
7.4 Hz), 5.40 (bs, 1H, NH(CO)O), 5.12 (s, 1H, CH2Ph),
4.95, 4.62 (2m, 2H, CHa), 3.88 (s, 2H, CH2-Gly), 1.45, (2d,
6H, CH3b, JZ7.1 Hz), 1.43 (s, 9H, C(CH3)3). 13C NMR
(CDCl3, 75 MHz) d: 172.2, 169.3, 168.5 (2COO, NH(CO)),
156.6 (NH(CO)O), 136.1 (Ar-C), 128.5, 128.2, 128.1 (Ar-
H), 82.3 (C(CH3)3), 69.8 (Ca-Lac), 67.2 (CH2Ph), 47.9 (Ca-
Ala), 44.4 (Ca-Gly), 27.9 (C(CH3)3), 18.3, 16.8 (CH3b).
Hydrogenolysis of Cbz-Gly-D-Ala-D-Lac-OtBu (1.20 g,
2.94 mmol) gave 5 (0.52 g, 64%) as a colorless syrup.
methanol, 10:1). H NMR (CDCl3, 300 MHz) d: 7.67 (bs,
1H, NH(CO)), 4.40, 3.46 (2m, 2H, CHa), 1.43 (s, 11H, NH2,
C(CH3)3), 1.34, 1.30 (2d, 6H, CH3b, JZ6.9 Hz). 13C NMR
(CDCl3, 75 MHz) d: 175.2, 172.4 (COO, NH(CO)), 81.7
(C(CH3)3), 50.6, 48.1 (Ca), 27.9 (C(CH3)3), 21.7, 18.4
(CH3b).
4.1.2. H-D-Ala-D-Lac-OtBu (3). A solution of Cbz-D-Ala-
OH (1.16 g, 5 mmol), H-D-Lac-OtBu (0.73 g, 5 mmol) and
BOP (2.21 g, 5 mmol) in DCM (30 mL) was cooled to
K20 8C and DiPEA (2.61 mL, 15 mmol) was added. After
stirring for 3 h, the reaction was evaporated to dryness in
vacuo and the residue was dissolved in EtOAc (60 mL). The
solution was washed with Na2CO3 (1 M), KHSO4 (1 M),
H2O, and brine. After drying (Na2SO4) and evaporating the
solvent, the crude 3 was purified by column chromatography
using DCM/methanol, 30:1 to give Cbz-D-Ala-D-Lac-OtBu
(1.55 g, 88%) as a colorless syrup. RfZ0.52 (DCM/
methanol, 10:1). 1H NMR (CDCl3, 300 MHz) d: 7.34–
7.29 (s, 5H, Ph), 5.26 (bs, 1H, NH(CO)O), 5.11 (d, 1H,
CH2aPh, JZ12.2 Hz), 5.06 (d, 1H, CH2bPh), 4.97 (m, 1H,
CHa-Lac), 4.42 (m, 1H, CHa-Ala), 1.51, 1.45 (2d, 6H,
CH3b, JZ6.9 Hz), 1.43 (s, 9H, C(CH3)3). 13C NMR
(CDCl3, 75 MHz) d: 172.5, 169.4 (2COO), 155.6
(NH(CO)O), 136.2 (Ar-C), 128.5, 128.1, 128.0 (Ar-H),
82.2 (C(CH3)3), 69.7 (Ca-Lac), 66.9 (CH2Ph), 49.4 (Ca-
Ala), 27.9 (C(CH3)3), 18.6, 16.8 (CH3b)b). Hydrogenolysis
of Cbz-D-Ala-D-Lac-OtBu (1.40 g, 4 mmol) was performed
following the same procedure as was used for Cbz-D-Ala-D-
Ala-OtBu. 3 (0.76 g, 87%) was obtained as a colorless
1
RfZ0.16 (DCM/methanol, 10:1). H NMR (CDCl3 with a
few drops of CD3OD, 300 MHz) d: 4.88, 4.47 (2m, 2H,
CHa), 3.97 (d, 1H, CH2a-Gly, JgemZ14.0 Hz, Ja,NH
Z
6.9 Hz), 3.92 (d, 1H, CH2b-Gly, Jb,NHZ6.9 Hz), 1.39
(d, 6H, CH3b, JZ7.1 Hz), 1.39 (s, 9H, C(CH3)3).
4.2. Synthesis of D-Ala-D-Ala and D-Ala-D-Lac containing
ligands (7–11)
4.2.1. NBD-(CH2)5-C(O)-Gly-D-Ala-D-Ala-OH (7). To a
suspension of 4 (0.27 g, 1 mmol), 6-(7-nitrobenzo-2-oxa-
1,3-diazole)-hexanoic acid (0.29 g, 1 mmol) and BOP (0.44,
1 mmol) in DCM (6 mL), DiPEA (0.35 mL, 2 mmol) was
added. After stirring the orange solution at rt for 3 h, the
reaction mixture was evaporated to dryness in vacuo and the
residue was dissolved in EtOAc (6 mL). The solution was
washed with KHSO4 (1 M), H2O, and brine. After drying
(Na2SO4) and evaporating the solvent, the crude product
was purified by column chromatography using a gradient of
DCM/methanol from 30:1 to 10:1 and NBD-(CH2)5-C(O)-
Gly-D-Ala-D-Ala-OtBu (0.45 g, 82%) was obtained as an
1
syrup. RfZ0.33 (DCM/methanol, 5:1). H NMR (CDCl3,
300 MHz) d: 4.94 (m, 1H, CHa-Lac), 4.11 (m, 1H, CHa-
Ala), 2.30 (bs, 2H, NH2), 1.45 (d, 3H, CH3b-Lac, JZ
7.7 Hz), 1.43 (s, 9H, C(CH3)3), 1.34 (d, 3H, CH3b-Ala, JZ
6.9 Hz).
4.1.3. H-Gly-D-Ala-D-Ala-OtBu (4). Coupling (see prep-
aration of 2) of Cbz-Gly-OH (0.79 g, 3.8 mmol) to 2 (0.82 g,
3.8 mmol) gave Cbz-Gly-D-Ala-D-Ala-OtBu (1.38 g, 89%)
as a white foam without further purification. RfZ0.76
(DCM/methanol, 5:1). 1H NMR (CDCl3, 300 MHz) d: 7.32
(s, 5H, Ph), 6.82 (d, 1H, NH(CO), JZ7.1 Hz), 6.73 (d, 1H,
NH(CO), JZ7.4 Hz), 5.60 (bs, 1H, NH(CO)O), 5.09 (s, 2H,
1
orange solid. RfZ0.49 (DCM/methanol, 10:1). H NMR
(CDCl3, 300 MHz) d: 8.45 (d, 1H, H-6-NBD, JZ8.8 Hz),
7.22 (bs, 1H, NH), 6.93 (d, 1H, NH(CO)-Ala, JZ7.1 Hz),
6.65 (d, 1H, NH(CO)-Ala, JZ7.1 Hz), 6.43 (t, 1H,
NH(CO)-Gly, JZ5.1 Hz), 6.14 (d, 1H, H-5-NBD), 4.52,
4.40 (2m, 2H, CHa), 4.00 (dd, 1H, CH2a-Gly, Jgem
Z

11154
C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
16.7 Hz, Ja,NHZ5.5 Hz), 3.92 (dd, 1H, CH2b-Gly, Jb,NH
Z
(Na2SO4) and the solvent was evaporated. The crude
product was purified by column chromatography using a
gradient of DCM/methanol from 50:1 to 30:1 to give
Ds-Gly-D-Ala-D-Ala-OtBu (0.41 g, 82%) as a light-green
5.5 Hz), 3.52 (m, 2H, CH2NH-NBD), 2.29 (t, 2H, CH2C(O)-
NBD, JZ6.7 Hz), 1.83–1.48 (m, 6H, CH2-NBD), 1.41 (s, 9H,
C(CH3)3), 1.38, 1.34 (2d, 6H, CH3b, JZ6.9 Hz). 13C NMR
(CDCl3, 75 MHz) d: 173.4, 171.7, 171.5, 168.5 (COO,
NH(CO)), 144.3 (C-NBD), 136.7 (C-6, 5-NBD), 82.3
(C(CH3)3), 53.4 (Ca-Gly), 48.9 (Ca-Ala), 43.0, 35.7 (CH2-
NBD), 27.9 (C(CH3)3), 27.9, 26.2, 24.6 (CH2-NBD), 18.8,
18.3 (CH3b). ES-MS (m/z)Z572.40 (100%, [MCNa]C),
550.25 (45%, [MCH]C), 4.94.35 (96%, [M-tBuCH]C).
1
solid. RfZ0.48 (DCM/methanol, 9:1). H NMR (CDCl3,
300 MHz) d: 8.51, 8.30 (2d, 2H, Ar, JZ8.5 Hz), 8.21 (d,
1H, Ar, JZ7.4 Hz), 7.50, 7.47 (2d, 2H, Ar, JZ16.0 Hz),
7.36, 7.26 (2d, 2H, NH(CO), JZ7.7 Hz), 7.13 (d, 1H, Ar),
6.92 (bs, 1H, NHSO2), 4.52, 4.36 (2m, 2H, CHa), 3.61 (s,
2H, CH2-Gly), 2.85 (s, 6H, N(CH3)2), 1.44 (s, 9H,
C(CH3)3), 1.30, 1.13 (2d, 6H, CH3b, JZ7.0 Hz). ES-MS
(m/z)Z529.35 (16%, [MCNa]C), 451.35 (100%,
[MKtBuCNa]C).
Treatment overnight of NBD-(CH2)5-C(O)-Gly-D-Ala-D-
Ala-OtBu (0.30 g, 0.54 mmol) with DCM/saturated HCl(g)
in ether (1:1) (20 mL) led to an orange precipitate. Filtration
and washing of the precipitate with cold ether gave to 7
(0.19 g, 70%) as an orange solid. RfZ0.04 (DCM/methanol,
5:1). 1H NMR (DMSO, 300 MHz) d: 8.51 (d, 1H, H-6-NBD,
JZ9.1 Hz), 8.33 (d, 1H, NH(CO), JZ6.9 Hz), 8.06 (t, 1H,
NH, JZ6.0 Hz), 7.98 (d, 1H, NH(CO), JZ7.4 Hz), 6.42
(d, 1H, H-5-NBD), 4.34–4.19 (m, 2H, CHa), 3.67 (d, 2H,
CH2-Gly, JZ5.5 Hz), 3.35 (bs, 2H, CH2NH-NBD),
2.14 (m, 2H, CH2C(O)-NBD), 1.66, 1.54, 1.35 (3m, 6H,
CH2-NBD), 1.27, 1.19 (2d, 6H, CH3b, JZ7.1 Hz). 13C
NMR (CDCl3, 75 MHz) d: 174.4, 173.1, 172.4, 169.3
(COOH, NH(CO)), 144.2 (C-NBD), 137.0 (C-6,5-NBD),
58.0 (Ca-Gly), 48.0, 47.9 (Ca-Ala), 42.4, 35.3, 30.3, 26.0,
24.6 (CH2-NBD), 17.7, 16.9 (CH3b). ES-MS (m/z)Z516.30
(40%, [MCNa]C), 494.15 (100%, [MCH]C). HPLC (RtZ
17.75 min, 99% purity).
t
After the cleavage of the Bu ester of Ds-Gly-D-Ala-D-Ala-
OtBu (0.17 g, 0.33 mmol), the reaction mixture was
evaporated in vacuo and KHSO4 (1 M), was added. The
pH of the aqueous layer was adjusted to 3–4 by addition of
NaOH (1 N) and extracted with EtOAc (70 mL) three times.
The combined organic layers were washed with brine, dried
(Na2SO4) and concentrated to afford 9 (0.11 g, 65%) as a
light-green solid. RfZ0.52 (methanol/chloroform/NH4OH,
60:45:20). 1H NMR (CD3OD, 300 MHz) d: 8.57, 8.34
(2d, 2H, Ar, JZ8.5, 8.8 Hz), 8.21 (d, 1H, Ar, JZ7.4 Hz),
7.54–7.63 (m, 2H, Ar), 7.27 (d, 1H, Ar, JZ7.7 Hz), 4.93–
4.22 (m, 2H, CHa), 3.54 (s, 2H, CH2-Gly), 2.86 (s, 6H,
N(CH3)2), 1.37, 1.17 (2d, 6H, CH3b, JZ7.4, 7.1 Hz). 13C
NMR (CD3OD, 75 MHz) d: 177.3 (COOH), 174.1, 170.7
(2NH(CO)), 153.3, 135.98 (Ar-C), 131.6 (Ar-H), 131.2,
130.9 (Ar-C), 130.6, 129.6, 124.3, 120.2, 116.7 (Ar-H), 50.3
(Ca-Ala), 46.4 (Ca-Gly), 45.8 (N(CH3)2), 18.0, 17.8
(CH3b). ES-MS (m/z)Z451.35 (100%, [MCH]C). HPLC
(RtZ15.93 min, 99% purity).
4.2.2. NBD-(CH2)5-C(O)-Gly-D-Ala-D-Lac-OH (8). The
same procedure as was used for the preparation of 7 was
followed. 5 (0.14 g, 0.5 mmol) gave NBD-(CH2)5-C(O)-
Gly-D-Ala-D-Lac-OtBu (0.22 g, 80%) as an orange solid.
RfZ0.59 (DCM/methanol, 10:1). 1H NMR (CDCl3,
300 MHz) d: 8.46 (d, 1H, H-6-NBD, JZ8.5 Hz), 6.87,
6.60, 6.30 (3bs, 3H, NH), 6.14 (d, 1H, H-5-NBD), 4.66–4.53
(m, 2H, CHa), 3.97 (s, 2H, CH2-Gly), 3.50 (m, 2H, CH2NH-
NBD), 2.29 (t, 2H, CH2C(O)-NBD, JZ6.9 Hz), 1.86–1.70
(m, 4H, CH2-NBD), 1.56–1.49 (m, 2H, CH2NBD), 1.49,
1.46 (2d, 6H, CH3b, JZ7.1 Hz), 1.42 (s, 9H, C(CH3)3).
ES-MS (m/z)Z573.35 (100%, [MCNa]C), 517.30 (45%,
[MKtBuCNa]C), 495.13 (30%, [MKtBuCH]C).
4.2.4. Ds-Gly-D-Ala-D-Lac-OH (10). The same procedure
as was used for 9 was followed for the preparation of 10. 5
(0.14 g, 0.5 mmol) gave Ds-Gly-D-Ala-D-Lac-OtBu as a
t
light-green solid. Cleavage of the Bu ester of Ds-Gly-D-
Ala-D-Lac-OtBu led to 10 (0.17 g, 75% over 2 steps) as a
light-green solid. RfZ0.63 (methanol/chloroform/NH4OH,
60:45:20). 1H NMR (CDCl3, 300 MHz) d: 8.52 (2d, 2H, Ar,
JZ8.5 Hz), 8.28 (d, 1H, Ar), 8.20 (1d, 1H, Ar, JZ7.1 Hz),
7.74–7.56 (m, 2H, Ar), 7.19 (1d, 1H, Ar, JZ7.7 Hz), 7.13
(d, 1H, NH(CO), JZ7.4 Hz), 6.49 (bt, 1H, NHSO2), 5.92
(bs, 1H, COOH), 5.01 (1m, 1H, CHa-Lac), 4.45 (1m, 1H,
CHa-Ala), 3.59 (bs, 2H, CH2-Gly), 2.88 (s, 6H, N(CH3)2),
1.48, 1.21 (2d, 6H, CH3b, JZ7.1, 7.4 Hz). 13C NMR
(CDCl3, 75 MHz) d: 173.6 (COOH), 172.2, 169.0
(2NH(CO)), 151.4, 133.6 (Ar-C), 130.7, 130.0 (Ar-H),
129.7, 129.4 (Ar-C), 128.8, 123.4, 118.8, 115.2 (Ar-H), 69.6
(Ca-Lac), 48.0 (Ca-Ala), 45.6 (Ca-Gly), 45.4 (N(CH3)2),
17.0, 16.6 (CH3b). ES-MS (m/z)Z452.3 (100%, [MC
Na]C). HPLC (RtZ15.90 min, 99% purity).
Cleavage of the tBu ester of NBD-(CH2)5-C(O)-Gly-D-Ala-
D-Lac-OtBu (0.18 g, 0.33 mmol) led to 8 (0.11 g, 65%) as an
1
orange solid. RfZ0.03 (DCM/methanol, 10:1). H NMR
(CD3OD, 300 MHz) d: 8.69 (d, 1H, H-6-NBD, JZ8.8 Hz),
6.53 (d, 1H, H-5-NBD), 5.12, 4.66 (2m, 2H, CHa), 4.09 (d,
2H, CH2a-Gly, JgemZ16.7 Hz), 4.02 (d, 2H, CH2b-Gly),
3.72 (bs, 2H, CH2NH-NBD), 2.49 (t, 2H, CH2C(O)-NBD,
JZ7.3 Hz), 1.98, 1.89, 1.66 (3m, 6H, CH2-NBD), 1.63, 1.62
(2d, 6H, CH3b, JZ7.4, 7.1 Hz). ES-MS (m/z)Z517.45
(100%, [MCNa]C), 495.19 (25%, [MCH]C). HPLC (RtZ
18.98 min, 94% purity).
4.2.5. DR-C(O)–(CH2)2–C(O)-Gly-D-Ala-D-Ala-OH (11).
To solution of 4 (0.27 g, 1 mmol) in DCM (2 mL), succinic
anhydride (0.1 g, 1 mmol) and TEA (0.15 mL, 1.1 mmol)
were added. After stirring the mixture at rt for 2 h, the
reaction mixture was evaporated in vacuo and the residue
was dissolved in butanol (3 mL). The solution was washed
with KHSO4 (1 M), H2O, and brine. After drying (Na2SO4)
and evaporation of the solvent, HOOC–(CH2)2–C(O)-Gly-
D-Ala-D-Ala-OtBu (0.32 g, 87%) was obtained as a white
4.2.3. Ds-Gly-D-Ala-D-Ala-OH (9). To a solution of 4
(0.27 g, 1 mmol) in 3% NaHCO3, (10 mL), a solution of
dansyl chloride (0.30 mL, 1.1 mmol) in aceton (10 mL) was
added. After stirring the fluorescent solution in the absence
of light overnight, the reaction mixture was evaporated to
dryness in vacuo and the residue was dissolved in EtOAc
(10 mL). The solution was washed with H2O, dried

C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
11155
1
solid. RfZ0.23 (DCM/methanol, 10:1). H NMR (DMSO,
300 MHz) d: 12.16 (bs, 1H, COOH), 8.16 (d, 2H, 2
NH(CO), JZ6.6 Hz), 7.94 (d, 1H, NH(CO), JZ7.7 Hz),
4.30, 4.07 (2m, 2H, CHa), 3.66 (d, 2H, CH2-Gly, JZ
5.7 Hz,), 2.41–2.35 (m, 4H, CH2), 1.36 (s, 9H, C(CH3)3),
1.23, 1.19 (2d, 6H, CH3b, JZ7.4, 7.1 Hz). 13C NMR
(DMSO, 75 MHz) d: 174.1, 172.1, 171.8, 171.8, 168.6
(COOH, COO, NH(CO)), 80.5 (C(CH3)3), 48.5, 47.78 (Ca-
Ala), 42.3 (Ca-Gly), 30.0, 29.3 (CH2), 27.8 (C(CH3)3), 18.5,
17.0 (CH3b). ES-MS (m/z)Z396.25 (100%, [MCNa]C),
374.25 (7%, [MCH]C), 318.10 (18%, [MKtBuCH]C).
were used and the reaction was shaken at rt overnight.
After 3 h of shaking, the resin was washed with NMP (5!
15 mL/mmol, each 2 min) and DCM (5!15 mL/mmol,
each 2 min). Negative Kaiser and BPB tests indicated
completion of the coupling. The loading of the resin was
determined by Fmoc-spectrophotometric quantification,
after drying the resin under vacuum overnight.
4.4. General procedure for Na-Fmoc deprotection
The Na-Fmoc-protected resin was swollen in NMP (2 min)
and, after draining the solvent, the resin was shaken with
20% piperidine in NMP (3!20 mL per mmol, each
10 min). The resin was washed with NMP (5!10 mL per
mmol, each 2 min) and DCM (5!10 mL per mmol, each
2 min). Positive Kaiser and BPB tests indicated Fmoc-
deprotection.
HOOC–(CH2)2–C(O)-Gly-D-Ala-D-Ala-OtBu (0.31 g,
0.83 mmol), Disperse red-NH2 (0.26 g, 0.83 mmol),
16
BOP (0.37 g, 0.83 mmol) and DiPEA (0.29 mL,
1.66 mmol) were dissolved in DCM (50 mL). After stirring
the dark red solution at rt overnight, the solvent was
evaporated in vacuo and the residue was dissolved in EtOAc
(10 mL). This solution was washed with KHSO4 (1 M),
H2O, and brine. After drying (Na2SO4) and evaporating the
solvent, the crude product was purified by column
chromatography using a gradient of DCM/methanol from
50:1 through 30:1 to 10:1 and DR-C(O)–(CH2)2–C(O)-Gly-
D-Ala-D-Ala-OtBu (0.28 g, 51%) was obtained as a dark red
4.4.1. Loading of Argogelw-NH2 or Argogelw-Rink-NH2
(16a, 16b). Fmoc-L-Ala-OH$H2O (1.05 g, 3.2 mmol) was
coupled to Argogelw-NH2 resin (a) (2 g, 0.8 mmol) or to
Fmoc-deprotected Argogelw-Rink-NH2 (b) (2.42 g,
0.8 mmol). Then the Na-Fmoc group was cleaved, follow-
ing the general procedure and coupling of Fmoc-L-Ala-
OH$H2O was repeated twice. Prior to the last Na-Fmoc
deprotection step, the loading was determined to be 0.31 and
0.25 mmol gK1, respectively. After Na-Fmoc deprotection,
12a was coupled to 13 to give 15a, and 12b was coupled to
14 to give 15b. Treatment of 15a with 50% TFA in DCM,
(7 mL) for 45 min and additional washing of the resin with
NMP (3!7 mL, each 2 min), 25% DiPEA in NMP (4!
7 mL, each 2 min), NMP (5!7 mL, each 2 min), and finally
DCM (5!7 mL, each 2 min) led to resin 16a. Fmoc-
deprotection of 15b using the general procedure gave resin
16b.
1
solid. RfZ0.58 (DCM/methanol, 10:1). H NMR (CDCl3
and a few drops of CD3OD, 300 MHz) d: 8.27, 7.87 (2d, 4H,
Ar, JZ9.1 Hz), 7.84 (d, 2H, Ar, JZ9.9 Hz), 7.39 (bs, 2H, 2
NH(CO)), 7.09 (d, 1H, NH(CO), JZ7.4 Hz), 6.92 (bs, 1H,
NH(CO)), 6.75 (d, 2H, Ar, JZ9.3 Hz), 4.42, 4.32 (2m, 2H,
CHa), 3.95, 3.76 (d, 2H, CH2-Gly, JZ16.8 Hz,), 3.51–3.38
(m, 6H, CH2-N), 2.48 (s, 4H, CH2–C(O)), 1.40 (s, 9H,
C(CH3)3), 1.35, 1.31 (2d, 6H, CH3b, JZ7.1 Hz), 1.21 (t,
3H, CH3, JZ7.1 Hz). ES-MS (m/z)Z691.50 (70%, [MC
Na]C), 669.50 (100%, [MCH]C).
Cleavage of the tBu ester of DR-C(O)–(CH2)2–C(O)-Gly-D-
Ala-D-Ala-OtBu (0.20, 0.3 mmol), as was described above
except for the use of dioxane instead of DCM, gave 11
(0.14 g, 75%) as a dark red solid. RfZ0.05 (DCM/methanol,
10:1). 1H NMR (CDCl3 and a few drops of CD3OD,
300 MHz) d: 8.06, 7.64 (2d, 4H, Ar, JZ9.1 Hz), 7.62 (d,
2H, Ar, JZ9.3 Hz), 6.57 (d, 2H, Ar, JZ9.1 Hz), 4.06, 3.90
(2m, 2H, CHa), 3.66 (s, 2H, CH2-Gly, JZ16.8 Hz,), 3.26–
3.10 (m, 6H, CH2-N), 2.35–2.19 (m, 4H, CH2C(O)), 1.16,
1.09 (2d, 6H, CH3b, JZ7.1 Hz), 0.97 (t, 3H, CH3, JZ
7.1 Hz). 13C NMR (CD3OD, 75 MHz) d: 179.3 (COOH),
175.5, 174.9, 173.7, 171.7 (COO, NH(CO)), 158.1, 153.0,
148.5, 144.7 (Ar-C), 127.2, 125.6, 123.4, 112.5 (Ar-H),
51.8, 50.6 (Ca-Ala), 46.3, 43.6, 38.1, 32.0, 31.9, 30.6
(CH2N, CH2C(O), Ca-Gly), 19.3, 17.9 (CH3b), 12.5 (CH3).
ES-MS (m/z)Z635.40 (15%, [MCNa]C), 613.45 (100%,
[MCH]C). HPLC (RtZ21.70 min, 99% purity).
4.4.2. Resynthesis of receptors 19a and 20a on Argogele-
NH2 resin and 17b and 18b on the Argogelw-Rink resin.
Resins 16a and 16b (each containing 0.051 mmol of amino
groups) were dried overnight under vacuum and were
separately swollen in NMP (for 2 min). After draining the
solvent, three subsequent coupling/Na-Fmoc deprotection
cycles were carried out of Fmoc-L-Lys(Boc)-OH, Fmoc-L-
Leu-OH and Fmoc-L-Lys(Boc)-OH following the general
procedures to give 17a and 17b. The same procedure was
followed for the synthesis of 18a and 18b, except for the use
of Fmoc-L-Glu(OtBu)-OH, Fmoc-L-Ala-OH, and Fmoc-L-
Ser(tBu)-OH as amino acids. The Boc groups from 17a and
18a were cleaved using 95% TFA in H2O, 95:5. The resin
was washed with NMP (3!1.5 mL, each 2 min), 25%
DiPEA in NMP: (4!1.5 mL, each 2 min), NMP (5!
1.5 mL, each 2 min), DCM (5!1.5 mL, each 2 min),
dioxane (4!1.5 mL, each 2 min), dioxane/H2O, 1:1 (4!
1.5 mL, each 2 min), H2O (4!1.5 mL, each 2 min), and
Et2O (5!1.5 mL, each 2 min) to give the receptors 19a and
20a, respectively.
4.3. General procedure for coupling Fmoc amino acids,
(Boc)2-CTV-OH (13) or (Fmoc)2-CTV-OH (14) on the
solid phase
The resin (1 equiv) was swollen in NMP (2 min) and, after
draining the solvent, BOP (4 equiv), the desired Fmoc
amino acid (4 equiv) and NMP (15 mL/mmol) were added.
The mixture was shaken until complete dissolution and then
DiPEA (8 equiv) was added. For coupling of 13 or 14,
2 equiv of 13 or 14, 2 equiv of BOP, and 4 equiv of DiPEA
4.4.3. Deprotection of the acid labile side-chains and
cleavage of the receptors 17b and 18b from the
Argogelw-Rink resin. Receptors 17b and 18b on the
Argogelw-Rink resin were swollen in DCM (2!2 min) and
the solvent was drained. After addition of 95% TFA in H2O
(20 mL/mmol), resins were shaken at rt for 4 h, removed by

11156
C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
filtering and washed with TFA/H2O, 95:5. The filtrates were
combined and Et2O was added to give suspensions of the
cleaved synthetic receptors. After centrifugation
(3500 ppm, 5 min) of the suspensions, the volatiles were
removed and the white solids were dissolved in H2O
followed by lyophilization to give trifluoroacetate salt of 21
(86 mg, 76%); HPLC: RtZ17.80 min, 94% purity, ES-MS
(m/z)Z766.95 (100%, [MC2H]2C), 511.50 (100%, [MC
3H]3C), 383.75 (58%, [MC4H]4C), and trifluoroacetate
salt of 22 (56 mg, 69%); HPLC: RtZ17.68 min, 94%
purity; ES-MS (m/z)Z1390.00 (25%, [MCNa]C), 1368.30
(100%, [MCH]C), 684.60 (60%, [MC2H]2C),
respectively.
Acknowledgements
We thank Arwin Brouwer for the preparation of disperse
red-NH2. Dr. Jos A.G. van Strijp and Erik Heezius of
Eijkman Winkler Institute (Utrecht Medical Center, The
Netherlands) are gratefully acknowledged for providing us
with FITC-peptidoglycan and for testing the synthetic
receptors 21 and 22. This work was supported by the
European Commission (Marie Curie Individual Fellowship,
contract No. HPMFCT-2000-00704).
References and notes
4.5. Procedure for liberation of the amine from the TFA-
salt
1. (a) Arienzo, R.; Kilburn, J. D. Tetrahedron 2002, 58, 711–719.
(b) Henley, P. D.; Waymark, C. P.; Guillies, I.; Kilburn,
J. D. Chem. Soc. Perkin Trans. 1 2000, 1021–1031. (c) Van
Wageningen, A. M. A.; Liskamp, R. M. J. Tetrahedron Lett.
1999, 40, 9347–9351. (d) Iorio, E. J.; Still, W. C. Bioorg. Med.
Chem. Lett. 1999, 9, 2145–2150. (e) Ryan, K.; Still, W. C.
A solution of trifluoroacetate salt of 21 (6.6 mmol) in
acetonitrile/H2O (1:1) (1.2 mL) was added to Amberlyst
A21-ion exchange resin (264 mmol) and the resin was
shaken for 45 min. After filtration, the solution was
lyophilized to give quantitatively 21. In case of the
trifluoroacetate salt of 22, Dowex 50!8 (HC form) ion
exchange resin (1 g) was used. The resin was shaken for 4 h
and the solvent was drained. 25% NH3 in H2O was added to
the resin and shaking was continued for 1 h. After filtration
the solution was lyophilised to give 22.
¨
Bioorg. Med. Chem. Lett. 1999, 9, 2673–2678. (f) Lowik,
D. W. P. M.; Weingarten, M. D.; Broekema, M.; Brouwer,
A. J.; Liskamp, R. M. J. Angew. Chem. Int. Ed. Engl. 1998, 37,
1846–1850. (g) Burger, M. T.; Still, W. C. J. Org. Chem. 1997,
¨
62, 4785–4790. (h) Lowik, D. W. P. M.; Mulders, S. J. E.;
Cheng, Y.; Shao, Y.; Liskamp, R. M. J. Tetrahedron Lett.
4.6. Screening of library 1{1-13, 1-13, 1-13} with D-Ala-D-
Ala and D-Ala-D-Lac containing ligands (6–11)
´
1996, 37, 8253–8256. (i) Pernıa, G. J.; Kilburn, J. D.; Essex,
J. W.; Mortishire-Smith, R. J.; Rowley, M. J. Am. Chem. Soc.
1996, 118, 10220–10227. (j) Shao, Y.; Still, W. C. J. Org.
Chem. 1996, 61, 6086–6087. (k) Yoon, S. S.; Still, W. C.
Tetrahedron 1995, 51, 567–578. (l) Zimmerman, S. C. Top. Curr.
Chem. 1993, 165, 71–102.
A 1 mg mLK1 suspension of 6, a 50 mM solution of 7–10,
and a 5 mM solution of 11 in phosphate buffer (0.1 N, pHZ
7.0) (1.3 mL) were added to different portions of library
1{1-13, 1-13, 1-13} (14 mg, w11.000 beads, w4 copies/
receptor) and incubated at 20 8C for 72 h. The resin was
drained and was washed with phosphate buffer (0.1 N, pHZ
7.0) (5!1.3 mL, each 2 min). Next, the resin was poured
into a petri dish and spread into a monolayer. The beads
were viewed under a fluorescence microscope. By use of a
long needle, most fluorescent or colored beads were isolated
(w100 beads). The fluorescence of the preselected beads
was reevaluated using the overexposure mode of the Leica
DC-100 digital camera system and image analysis to
estimate the relative fluorescence intensities semi-quanti-
tatively. Reevaluation of the red colored beads in the case of
11 was judged by eye. Only nine to ten best fluorescent or
colored beads were selected and analysed by independent
Edman degradation together with a non-fluorescent bead
from each screening.
2. (a) Chamorro, C.;Hofman, J.-W.;Liskamp,R.M. J.Tetrahedron
2004, 60, 8691–8699. (b) Monnee, M. C. F.; Brouwer, A. J.;
Liskamp, R. M. J. QSAR and Combinatorial Science, 2004, 23,
in press. (c) Monnee, M. C. F.; Brouwer, A. J.; Verbeek, L. M.;
Van Wageningen, A. M. A.; Liskamp, R. M. J. Bioorg. Med.
Chem. Lett. 2001, 11, 1521–1525. (d) Braxmeier, T.;
Demarcus, M.; Fessmann, T.; McAteer, S.; Kilburn, J. D.
Chem. Eur. J. 2001, 7, 1889–1898. (e) Botana, E.; Ongeri, S.;
Ariezo, R.; Demarcus, M.; Frey, J. G.; Piarulli, U.; Potenza, D.;
Kilburn, J. D.; Gennari, C. Eur. J. Org. Chem. 2001, 4625–4634.
(f) Fressmann, T.; Kilburn, J. D. Angew. Chem. Int. Ed. 1999, 38,
1993–1996.
3. (a) Houghten, R. A.; Pinilla, C.; Appel, J. R.; Blondelle, S. E.;
Dooley, C. T.; Eichler, J.; Nefzi, A.; Ostresh, J. M. J. Med.
Chem. 1999, 42, 3743–3778. (b) Furka, A.; Sebestyen, F.;
Asgedom, M.; Dibo, G. Int. J. Pept., Protein Res. 1991, 36,
487–493. (c) Lam, K. S.; Salmon, S. E.; Hersh, E. M.; Hruby,
V. J.;Kazmierski, W. M.;Knapp, R. J. Nature1991, 354, 82–84.
(d) Houghten, R. A.; Pinilla, C.; Blondelle, S. E.; Appel, J. R.;
Dooley, C. T.; Cuervo, J. H. Nature 1991, 354, 84–86.
4. Chamorro, C.; Liskamp, R. M. J. J. Comb. Chem. 2003, 5,
794–801.
4.7. Binding affinity for with D-Ala-D-Ala containing
ligands (6, 7, 9 and 11 by 19a and 20a)
A 50 mM solution of 7 or 9 (1.45 mL) or a 5 mM solution of
11, or a 1 mg mLK1 (433 mL) suspension of 6 in phosphate
buffer (0.1 N, pHZ7.0) was added to the resynthesized
receptors 19a or 20a (0.6 mg, w480 beads) and incubated at
20 8C for 72 h. The resins were drained and washed with
phosphate buffer (0.1 N, pHZ7.0) (5!1.3 mL, each
2 min). The resin was poured into a petri dish and spread
into a monolayer. The beads were viewed under a
fluorescence microscope.
5. Colet, A. Tetrahedron Lett. 1987, 43, 5725–5759.
6. Rump, E. T.; Rijkers, D. T. S.; Hilbers, H. W.; de Groot, P. G.;
Liskamp, R. M. J. Chem. Eur. J. 2002, 8, 4613–4621.
7. Rothman, J. H.; Still, W. C. Bioorg. Med. Chem. Lett. 1997, 7,
3159–3164.
8. (a) Edman, P.; Begg, G. Eur. J. Biochem. 1967, 1, 80–91.
(b) Edman, P. Acta Chem. Scand. 1950, 4, 283–293.

C. Chamorro, R. M. J. Liskamp / Tetrahedron 60 (2004) 11145–11157
11157
9. (a) Malabarba, A.; Nicas, T. I.; Thompson, R. S. Med. Res.
Rev. 1997, 17, 69–137. (b) Reynolds, P. E. Eur. J. Clin.
Microbiol. Infect. Dis. 1989, 8, 943–950.
14. Peptidoglycan was isolated from S. aureus according to a
method described by C. Timmermans, ‘Chemilumescence of
human polymorphonuclear leukocytes after stimulation with
whole cells and cell wall components of Staphylococcus
epidermidis’ (from the Thesis entitled: Staphylococcus
epidermidis infections. C.P. Timmermans. 1991, Chapter 7).
Peptidoglycan is labeled afterwards with 1 mg/mL FITC for
1 h and then dialysed against PBS at 4 8C for 24 h. Mattson, E.;
Rollof, J.; Verhoef, J.; van Dijk, H.; Fleer, A. Infect. Immun.
1994, 62, 3837–3843.
10. (a) Reynolds, P. E.; Somner, E. A. Drugs Exp. Clin. Res. 1990,
16, 385–389. (b) Barna, J. C. J.; Williams, D. H. Annu.
Rev. Microbiol. 1984, 38, 339–357.
11. Bugg, T. D. H.; Wright, G. D.; Dutka-Mallen, S.; Arthur, M.;
Courvalin, P.; Walsh, C. T. Biochemistry 1991, 30,
10408–10415.
12. William, D. H.; Bardsley, B. Angew. Chem. Int. Ed. Engl.
1999, 38, 1172–1193.
15. Castro, B.; Dormoy, J. R.; Evin, G.; Selve, C. Tetrahedron
Lett. 1975, 14, 1219–1222.
13. (a) Jensen, K. B.; Braxmeier, T. M.; Demarcus, M.; Frey,
J. G.; Kilburn, J. K. Chem. Eur. J. 2002, 8, 1300–1309.
(b) Anslyn, E. V. J. Am. Chem. Soc. 2000, 122, 542–543.
(c) Dong, D. L.; Ruiping, L.; Sherlock, R.; Wingler, M. H.;
Nestler, H. P. Chem. Biol. 1999, 6, 133–141. (d) Xu, R.;
Greiveldinger, G.; Marenus, L. E.; Cooper, A.; Ellman, J. A.
J. Am. Chem. Soc. 1999, 121, 4898–4899. (e) Davies, M.;
Bonnat, M.; Guillier, F.; Kilburn, J. D.; Bradley, M. J. Org.
Chem. 1998, 63, 8696–8703. (f) Torneiro, M.; Still, W. C.
Tetrahedron 1997, 53, 8739–8750.
16. The Disperse red-NH2 was obtained by Staudinger reaction of
the Disperse red-N3 derivative.
17. Kaiser, E.; Colescott, R. L.; Bossinger, C. D.; Cook, P. I. Anal.
Biochem. 1970, 34, 595–598.
18. Krchnak, V.; Vagner, J.; Safar, P.; Lebl, M. Collect. Czech.
Chem. Commun. 1988, 53, 2542–2548.
19. Arx, E.; von Faupel, M.; Bruggen, M. J. Chromatogr. 1976,
120, 224.