Metal-ion induced amplification of three receptors from dynamic combinatorial libraries of peptide-hydrazones

Article abstract of DOI:10.1039/b300956d

Three building blocks of general structure (MeO)₂CH-aromatic linker-Pro-amino acid-NHNH₂ have been prepared and tested in acid-catalysed dynamic combinatorial libraries. Exposure of these libraries to LiI and NaI led to the amplification of three macrocyclic pseudopeptide receptors. The receptors were isolated and their interactions with LiI and NaI were analysed using NMR, IR and ITC. Binding of the metal ions to the receptors is invariably entropy-driven. Nevertheless, all receptors were found to be flexible with substantial conformational rearrangements accompanying guest binding. This type of receptor is extremely difficult to access through rational design and the fact that dynamic combinatorial chemistry allows facile access to these challenging molecules underlines the power of the dynamic approach.

Full text of DOI:10.1039/b300956d

View Article Online / Journal Homepage / Table of Contents for this issue
Metal-ion induced amplification of three receptors from dynamic
combinatorial libraries of peptide-hydrazones†
Sarah L. Roberts,a Ricardo L. E. Furlan,b Sijbren Ottoa and Jeremy K. M. Sanders*a
a University Chemical Laboratory, Lensfield Road, Cambridge, UK CB2 1EW.
E-mail: jkms@cam.ac.uk; Fax: ϩϩ44 (0)1223-336017; Tel: ϩϩ44 (0)1223-336343
b Universidad Nacional de Rosario, Facultad de Ciencias Bioquímicas y Farmacéuticas,
Suipacha 531, 2000 Rosario, Argentina. E-mail: rfurlan@fbioyf.unr.edu.ar;
Fax: ϩϩ54 (0)3414-375315; Tel: ϩϩ54 (0)3414-375315
Received 24th January 2003, Accepted 18th March 2003
First published as an Advance Article on the web 9th April 2003
Three building blocks of general structure (MeO)2CH–aromatic linker–Pro–amino acid–NHNH2 have been prepared
and tested in acid-catalysed dynamic combinatorial libraries. Exposure of these libraries to LiI and NaI led to the
amplification of three macrocyclic pseudopeptide receptors. The receptors were isolated and their interactions with
LiI and NaI were analysed using NMR, IR and ITC. Binding of the metal ions to the receptors is invariably entropy-
driven. Nevertheless, all receptors were found to be flexible with substantial conformational rearrangements
accompanying guest binding. This type of receptor is extremely difficult to access through rational design and the
fact that dynamic combinatorial chemistry allows facile access to these challenging molecules underlines the power of
the dynamic approach.
Introduction
Dynamic combinatorial chemistry has attracted interest over
recent years as a new strategy for the preparation and identifi-
cation of novel host–guest systems.1 The approach integrates
molecular evolution into combinatorial chemistry, and in prin-
ciple merges the preparation and screening of libraries into one
single process.
Dynamic combinatorial libraries (DCLs) are generated by
the assembly of building blocks through reversible bonds, so
the library product distribution is thermodynamically con-
trolled and responsive to external influences. Stabilization of
one particular member of the library by selective binding to a
guest molecule will shift the equilibrium, leading to an
increased concentration of the selected compound at the
expense of the other library members. In this way, a guest can
Fig. 1 Generation and templating of a dynamic combinatorial library
of macrocyclic hydrazones.
be used as a template to select and amplify its preferred host
from a library of potential receptors.
Several different covalent and non-covalent reactions have
been used to generate DCLs. These include base- and Pd-cata-
lyzed transesterifications,2 transimination of imines, oximes and
hydrazones,3 alkene metathesis,4 disulfide exchange,5 enzyme-
catalyzed aldol condensation and transamination,6 hydrogen-
bond exchange,7 metal–ligand coordination8 and cis–trans
isomerisation.9 However, despite these many reports, the effi-
cient amplification of covalently-assembled receptors or guests
seems to be more elusive and few examples of significant shifts
in library distribution have been reported so far.3a,10 We have
previously reported the amplification of a receptor for Li cat-
ions from a library of pseudopeptide-hydrazones;10d we now
describe (a) the isolation and binding properties of that recep-
tor in more detail, and (b) the amplification, isolation and
properties of two new cation-binding receptors from related
building blocks.
rapid deprotection of the aldehyde and induces hydrazone
formation and exchange, leading to the assembly and inter-
conversion of macrocycles. Neutralization of the reaction
medium switches off exchange, converting the library into a
‘static’ mixture from which the individual members can be
isolated.
Previously we described some properties of a library of
macrocycles prepared from the building block pPFm11 (1), con-
taining -proline (P), -phenylalanine (F), and a p-substituted
(p) aromatic linker (Scheme 1).10d This building block has the
potential to engage in a range of non-covalent interactions
including hydrogen-bonding, Lewis acid–base interactions,
Our approach to hydrazone DCLs is summarized in Fig. 1.
A range of interconverting macrocycles can be generated
from bifunctional building blocks containing one hydrazide
and one protected aldehyde. The addition of acid results in
† Electronic supplementary information (ESI) available: NMR and IR
data of the Liϩ complexes of (mPF)2, (pPF)3 and (pPC)3. See http://
www.rsc.org/suppdata/ob/b3/b300956d/
Scheme 1 Dipeptide building blocks.
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 6 2 5 – 1 6 3 3
1625

View Article Online
Scheme 2 Synthesis of the dipeptide building blocks 1–3.
π–π and cation–π interactions. We have shown that LiI and
NaI select and amplify the cyclic trimer.10d
In order to explore which features of the pPFm system are
important in the formation of the metal ionؒtrimer complex, we
have now studied the response of libraries generated from two
structurally related building blocks. Firstly, we prepared the
monomer mPFm (2) which has a different linker geometry, and
secondly, the phenyl side chain was replaced by a cyclohexyl
side chain, to give the monomer pPCm (3).
Results and discussion
Building block synthesis
The dipeptide building blocks were prepared using standard
peptide chemistry (Scheme 2) starting with EDC coupling
of CBZ-protected -proline (4) to the appropriate amino
acid methyl ester (5a or 5b), followed by deprotection by hydro-
genation and a second EDC coupling to the appropriate
carboxybenzaldehyde dimethoxyacetal (8 or 9). The final
bifunctional monomers 1–3 were then generated by the
hydrazinolysis of the methyl esters introducing the hydrazide
functionality.
Fig. 2 HPLC traces of the DCL made from 5.0 mM mPFm in
CHCl3:MeOH (98:2 v/v) after stirring for 24 h at rt (a) in the absence of
template, (b) in the presence of 15 mM LiI, (c) in the presence of 15
mM NaI.
Preparation and screening of the DCLs
Acid-catalysed cyclisation of the building blocks in CHCl3:
MeOH (98:2 v/v)12 at room temperature generated the desired
DCLs. The mixtures were analysed by electrospray ionisation
mass spectrometry (ESI-MS) and found to consist primarily of
macrocyclic hydrazone oligomers.13 A combination of ESI-MS
and HPLC allowed the assignment of the major peaks in the
chromatograms as shown in Fig. 2(a), 3(a) and 4(a). The HPLC
traces were recorded once thermodynamic equilibrium had
been reached after 1–5 days. After this point no further changes
in the distributions of cyclic products were observed, although
the reactions were monitored for 7 or 14 days. The equilibrium
composition of the mPFm library is dominated by the cyclic
dimer (Fig. 2a), whereas the pPFm (Fig. 3a) and pPCm (Fig. 4a)
libraries contain more of the higher oligomers.
Addition of LiI and NaI as templates affects the product
distribution in all three libraries (traces b and c in Fig. 2–4). In
the pPFm and pPCm libraries the salts induce the amplification
of cyclic trimer. In the case of the pPCm library NaI gives rise
to a slightly greater response than LiI leading to approximately
95% of the peptide library material being present as trimer.
However, in the pPFm library both NaI and LiI give rise to
very similar templating responses, both leading to the trimer
representing approximately 98% of the peptide material.
Fig. 3 HPLC traces of the DCL made from 0.3 mM pPF3 in
CHCl3:MeOH (98:2 v/v) after stirring for 5 days at rt (a) in the absence
of template, (b) in the presence of 2.7 mM LiI (c) in the presence of 2.7
mM NaI.
The HPLC results shown here for the pPFm system with NaI
differ slightly from those reported previously.10d Those initial
HPLC traces turned out not to represent the true equilibrium
position in the presence of NaI, due to precipitation over time.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 6 2 5 – 1 6 3 3
1626

View Article Online
followed by addition of the template to the pre-equilibrated
mixtures. In all cases essentially identical product distributions
were obtained, except for pPFm in the presence of NaI where
solubility problems were encountered (see above).14
No significant differences in templating effects were observed
using templates with different anions, i.e. LiCl, NaCl, Li tri-
fluoroacetate and Na trifluoroacetate, suggesting that iodide
anion is not involved in the recognition process.15
Binding studies
The selected receptors were isolated using semi-preparative
HPLC and characterised by HPLC, NMR and MS analysis.
The interactions between the amplified cyclic compounds and
the metal ions were investigated using 1H NMR, 7Li NMR, 13
C
NMR, FT-IR and isothermal titration calorimetry (ITC). The
limited solubility of NaI in our solvent system restricted NMR
binding studies to the LiI template.
Fig. 4 HPLC traces of the DCL made from 1.0 mM pPCm in
CHCl3:MeOH (98:2 v/v) after stirring for 5 days at rt (a) in the absence
of template, (b) in the presence of 3.0 mM LiI, (c) in the presence of 3.0
mM NaI.
The 1H NMR spectrum of (mPF)2 in CDCl3:MeOD (98:2 v/v)
shows that the compound adopts on average a C2 symmetric
conformation. The spectrum was fully assigned with the aid of
a COSY spectrum. Addition of LiI led to shifts in all 1H NMR
receptor signals (Fig. 5), the system being in fast exchange on
The HPLC traces shown here are for libraries generated at
lower concentration and starting from pure trimer not mono-
mer, conditions which avoid problems with precipitation.14 In
the presence of the NaI the ‘library’ distribution shown in Fig.
3c remains unchanged for at least 14 days, whereas in the
absence of NaI the distribution reverts to its untemplated equi-
librium position in the presence of acid in the course of 5 days
(Fig. 3a).
Introduction of LiI and NaI into the mPFm library induced
a different response. Here it is the cyclic dimer, already domin-
ant in the absence of template, which is amplified by both the
LiI and NaI. Sodium iodide induces the greatest response, lead-
ing to the dimer representing >98% of the peptide material in
the library.
1
the H NMR chemical shift timescale. The Liؒ(mPF)2 complex
also showed averaged C2 symmetry. A binding constant of
5 × 102 MϪ1 was determined for this system by titrating LiI into
a solution of (mPF)2 and assuming a 1:1 stoichiometry.16
Addition of LiI to a solution of (pPC)3 showed that this
system is in slow exchange and forms a 1:1 complex. Both the
(pPC)3 and Liؒ(pPC)3 complex were shown to have averaged C3
symmetric conformations although the two species have very
different 1H NMR spectra (Fig. 6). Similar behaviour was
observed upon titrating (pPF)3 with LiI. All the NMR spectra
were assigned using COSY, and from the integrals of the bound
and unbound 1H NMR signals of a 1:1 host/guest mixture
binding constants of 1.7 × 105 MϪ1 and 4 × 104 MϪ1 for (pPC)3
and (pPF)3, respectively, were obtained.
To prove that the libraries were all operating under thermo-
dynamic control, the same templated library distributions were
generated via two independent routes: firstly by cyclisation of
the monomers in the presence of the templates and secondly,
by cyclisation of the monomers in the absence of template,
For all three receptors addition of LiI leads to significant
1
shifts in every resonance in the H NMR spectra. This sug-
gests that there is a substantial change in the geometry of the
Fig. 5 1H-NMR in CDCl3:CD3OD (98:2 v/v) of (mPF)2 (a) in the absence of LiI and (b) in the presence of 1 equivalent LiI.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 6 2 5 – 1 6 3 3
1627

View Article Online
Fig. 6 1H-NMR in CDCl3:CD3OD (98:2 v/v) of (pPC)3 (a) in the absence of LiI; (b) in the presence of 0.5 equivalent LiI and (c) in the presence of
1 equivalent LiI.
receptors upon complexation with lithium, and therefore that
they are not preorganised.
Further information about the complexation of the receptors
7
with lithium was obtained from Li NMR studies. Titrating
(pPC)3 into a solution of LiI showed a system in fast exchange
on the 7Li NMR timescale, giving rise to an averaged signal for
bound and free lithium (Fig. 7). However, a similar experiment
with (pPF)3 showed a system in slow exchange on the 7Li NMR
timescale, giving rise to separate signals for the bound and free
lithium (Fig. 8). This difference in behaviour seems counter-
intuitive, given that the binding constant for (pPC)3 is higher
than that for (pPF)3. It suggests that there is a difference in the
kinetic stabilities of the two complexes, which does not corre-
spond to their thermodynamic stabilities.17 This may be due to
dissimilarities in the rate constants for binding (kb) and release
(kr) of the two complexes. Since both the binding constant (K)
and the rate constant of release of Li (kr) are greater for the
(pPC)3 system, then as K = kb/kr the rate constant of binding
(kb) to (pPC)3 must also be much higher than that for (pPF)3,
suggesting that the enhanced affinity of (pPC)3 arises from the
greater rate of binding of lithium, possibly due to a less
demanding structural rearrangement for complex formation in
this system (see below).
Fig. 8 7Li NMR in CDCl3:CD3OD (98:2 v/v) of 2 mM LiI (a) in the
absence of (pPF)3 and in the presence of (b) 0.5 and (c) 1.0 equivalent
of (pPF)3.
NMR and FT-IR spectra of the three receptors upon complex-
ation with LiI suggests similar interactions are occurring in
these systems. The 13C NMR spectra of the three cyclic species
show downfield shifts of the carbonyl carbons in the presence
of LiI. In the (mPF)2 spectrum, all three carbonyls shift, by 1.0,
2.0 and 2.0 ppm. In the cases of (pPF)3 and (pPC)3 only the
carbonyl carbons of the proline and phenylalanine units shift:
by 1.0 ppm and 4.5 ppm for (pPF)3 and by 2.0 ppm and 2.0 ppm
for (pPC)3.
There are many precedents for coordination of lithium to the
carbonyl oxygens of peptides.18 Study of changes in the 13C
In addition, in the presence of LiI and NaI the IR carbonyl
stretching frequencies of the receptors shift to lower wave-
numbers (Table 1) also suggesting that the metal ions are
19
coordinating the carbonyl oxygens. For the (pPF)3 and (pPC)3
systems the symmetry of the C᎐O peaks in the IR spectra is
᎐
reduced, which supports the 13C NMR evidence that in these
two cases not all of the carbonyl groups are involved in metal-
ion binding. No such effect is observed for the (mPF)2 system.
In the (mPF)2 system NaI induces a much larger shift in the IR
absorbances than does LiI, which correlates with the greater
templating efficiency observed for sodium over lithium in this
system.
Fig. 7 7Li NMR in CDCl3:CD3OD (98:2 v/v) of 2 mM LiI (a) in the
absence of (pPC)3 and in the presence of (b) 0.4; (c) 0.6; (d) 0.8; (e) 1.0
equivalents of (pPC)3.
Binding constants for the three receptors with the metal-ion
templates were measured using ITC.20 This technique involves
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 6 2 5 – 1 6 3 3
1628

View Article Online
Table 1 IR carbonyl stretching frequencies (cmϪ1) for the dipeptide
that of the more flexible trimers. This difference is also reflected
receptors in the absence and presence of the metal ion templates
in the IR results for the cyclic compounds in the presence
and absence of the metal ions. Significant selectivity for Naϩ
over Liϩ has been reported for other mainly crown ether
based receptors,21 and is generally thought to arise from size
differentiation.
No salt
LiIa
NaIa
(mPF)2
(pPF)3
(pPC)3
1681
1673
1674
1675 (Ϫ5)
1668 (Ϫ5)
1667 (Ϫ7)
1670 (Ϫ11)
1669 (Ϫ4)
1668 (Ϫ6)
Binding of the metal ions to the receptors is invariably
entropy driven, and accompanied by an unfavourable enthalpy
change. Favourable entropy contributions are usually associ-
ated with the liberation of solvent molecules from around both
host and guest, compensating for the cost of the immobilisation
of the substrates. Enthalpic effects are balanced between sol-
vation of the substrates, interactions within the bulk solvent,
and interactions between the host and guest.22 Comparison of
the individual enthalpy and entropy contributions for the form-
ation of the complexes of the (pPF)3 and (pPC)3 receptors high-
lights a striking difference in their otherwise very similar
behaviour. In the case of the (pPC)3 receptor the enthalpy con-
tributions are much more unfavourable for complexation with
both Liϩ and Naϩ and the entropy gains are much greater com-
pared to the (pPF)3 receptor. In addition, it is the Liϩ and not
the Naϩ which gives rise to the greatest unfavourable enthalpic
effect and the greatest entropy gain. This behaviour suggests
that the (pPC)3 receptor has a greater level of preorganisation
for the Liϩ binding compared to the (pPF)3 receptor. This con-
clusion is further supported by differences in the kinetics of
complex formation as observed by 7Li NMR which were much
faster for Liϩ binding to (pPC)3 than to (pPF)3.
a Values in brackets indicate the change in wavenumber from the values
in the absence of template.
the measurement of heat uptake or liberation upon the titration
of one species into another, which, after heats of dilution are
subtracted, gives the enthalpy change upon binding (∆Ho) from
the peak areas; the free energy of binding (∆Go) from the shape
of the titration curve and the stoichiometry from the inflexion
point. The entropy of binding (∆So) can then be calculated
from the Gibbs–Helmholtz equation.
All the titrations were carried out with the cyclic receptors
being titrated into the metal-ion solution in CHCl3:MeOH 98:2
(v/v). The reverse experiment led to large heat effects probably
due to de-aggregation of the salts upon dilution. A represent-
ative ITC titration is shown in Fig. 9 and the results are summar-
ised in Table 2 together with the binding constants obtained by
1H NMR. The stoichiometries obtained from the curve-fitting
procedure were found to be in agreement with 1:1 complexes
being formed in all cases confirming our observations using
NMR. Also the binding constants measured by NMR and
microcalorimetry are in good agreement. For all receptors the
binding constants for Naϩ are higher than those for Liϩ, a result
that correlates well with the templating experiments.
Conclusions and summary
The two orders of magnitude selectivity of the relatively
The results presented here have shed some light on the inter-
actions leading to efficient templating. Replacing the phenyl
group of pPFm by a cyclohexyl group to give pPCm leads to no
significant change in templating behaviour: the trimer species is
still templated efficiently and indeed has slightly higher affin-
ities for Liϩ and Naϩ. Despite similar affinities for the templates
the thermodynamics and kinetics of binding of Liϩ to (pPC)3
and (pPF)3 are surprisingly different suggesting that (pPC)3 is
the better preorganised receptor for this metal ion.
small and rigid (mPF)2 for Naϩ over Liϩ is much higher than
Changing the size and shape of the monomer building block
does have a significant effect on the size of receptor chosen by
the metal ion templates. In the case of mPFm, a dimer rather
than a trimer is selected by Liϩ and Naϩ. The 13C NMR and IR
data suggest that in the (mPF) receptor all six of the C᎐O
᎐
2
groups are involved in the metal ion binding, whilst in both
trimers only two carbonyls from each monomer unit are
involved. Liϩ and Naϩ prefer six hard ligands in their first
coordination sphere. For the mPFm library the smallest macro-
cycle that can satisfy this arrangement is the dimer. Inspection
of CPK models of (pPF)2 and (pPC)2 show that an octahedral
arrangement of the six carbonyls is not possible for these
macrocycles and only by moving to the trimers can these
two systems bind and fully satisfy the metal ion templates.
In summary we have identified three receptors with high
degrees of flexibility that were found to undergo substantial
Fig. 9 Heat effects (a) and enthalpy changes (b) upon titrating 0.7 mM
(pPC)3 into 0.075 mM LiI in CHCl3:MeOH (98:2 v/v), measured at
298 K.
Table 2 Equilibrium constants and thermodynamic data for the binding of LiI and NaI to the peptide receptors measured in CHCl3:MeOH
(98:2 v/v) at 298 K.
Receptor
Guest
KNMR/MϪ1
KITC/MϪ1
∆Ho/kJ molϪ1
T ∆So/kJ molϪ1
∆Go/kJ molϪ1
a
(mPF)2
(pPF)3
(pPC)3
Li
5.0 × 102
—
—
—
6.1
4.7
16.3
50.6
22.4
—
—
Nab
Li
6.2 × 104
1.0 × 105
1.9 × 105
2.2 × 105
6.0 × 105
33.4
34.0
46.4
81.2
55.2
Ϫ27.3
Ϫ29.3
Ϫ30.1
Ϫ30.5
Ϫ32.8
4.0 × 104
—
Nab
Li
1.7 × 105
—
Nac
a Assuming a 1:1 stoichiometry. b 3.0 mM receptor titrated into approximately 0.3 mM NaI or LiI. c 0.7 mM trimer titrated into approximately 0.075
mM NaI/LiI.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 6 2 5 – 1 6 3 3
1629

View Article Online
conformational rearrangements upon binding of the metal-ion
guests. Receptors with these characteristics are difficult to
access through rational design, and form an underexplored but
exciting area in supramolecular chemistry. Our results demon-
strate that dynamic combinatorial chemistry is a very power-
ful tool for the discovery of these challenging ‘induced-fit’
receptors.
with I2 in CHCl3 or Ninhydrin in EtOH. Column chromato-
graphy was carried out using silica gel 60 F (Fluorochem). All
solvents were distilled prior to use and dry solvents freshly dis-
tilled from CaH2 under argon, with the exception of DMSO
(Lancaster) which was used without further purification. HPLC
grade CHCl3 (Aldrich), and HPLC grade acetonitrile (Fisher)
were filtered with a 0.45 µm Millipore filter and used without
further purification. Ultrapure water was obtained from a
Millipore water purification system.
Experimental
4-Carboxybenzaldehyde dimethoxyacetal (8).23,24 To a suspen-
sion of 4-carboxybenzaldehyde (2.00 g, 1.33 mmol) in dry
MeOH (40 ml) was added ammonium chloride (4.00 g, 74.8
mmol) and the reaction was heated under reflux for 20 h. The
solvent was removed under reduced pressure and the resulting
white solid was recrystallised from boiling hexane (1.55 g, 60%).
IR (CHCl ) ν = 1694 (C᎐O) cmϪ1; 1H NMR (400 MHz, CDCl3)
General
HPLC analyses were carried out on a Hewlett Packard 1050 or
1100 system coupled to a UV analyser and processed using HP
Chemstation software. Separations were achieved using a
reversed phase Supelcosil ABZϩPlus column 15.0 cm ×
4.6 mm, 3 µm particle size (analytical scale); 25 cm × 1 cm, 5 µm
particle size, (semi-preparative scale). All separations were
performed with gradient elution of water and acetonitrile
mixtures, typically 75% to 10% water over 25 min, with a flow
rate of 1.0 ml minϪ1 for analytical scale and 5.0 ml minϪ1 for
semi-preparative scale.
᎐
3
δ = 8.11 (2H, d, J = 7.9 Hz, Ar–H), 7.57 (2H, d, J = 7.9 Hz, Ar–
H), 5.46 (1H, s, CH(OMe)2), 3.34 (6H, s, CH(OMe)2); 13C
NMR (100 MHz, CDCl ) δ = 171.7 (C᎐O), 143.8 (Arquat),
᎐
3
130.2 (Ar–H), 129.3 (Arquat), 126.9 (Ar–H), 102.2
(CH(OMe)2), 52.7 (CH(OMe)2); HRMS (EI) [M]ϩ C10H12O4
requires 196.0736, found 196.0740; m.pt. = 119–120 ЊC.
NMR spectroscopy was performed on Bruker DRX 400,
DPX 500 or DRX 800 instruments and chemical shifts are
quoted in parts per million with respect to TMS.
3-Carboxybenzaldehyde dimethoxyacetal (9).23,24 Synthesised
following the procedure for the synthesis of 8, starting from
3-carboxybenzaldehyde (2.00 g, 1.33 mmol). Yield: (1.81 g,
Electrospray mass spectra were recorded on a Micromass
Quattro-LC triple quadrupole mass spectrometer (QUATTRO)
fitted with a z-spray electrospray source. The source was heated
to 100 ЊC and the sampling cone voltage (Vc) was kept between
30–65 V. Samples were introduced into the mass spectrometer
source with an LC pump (Shimadzu LC-9A pump) at a rate of
4 µl minϪ1 of MeCN:H2O 1:1. Calibration was performed using
protonated horse myoglobin. Scanning was carried out from
m/z = 200 to 3000 in 8 s and several scans were summed to
obtain the final spectrum which was processed using MassLynx
V3.0 software. Electrospray mass spectra were also recorded
on a Micromass Q-TOF instrument (QTOF), incorporating
time-of-flight analysis with electrospray ionisation through a
standard z-spray source. Calibration was performed using
erythromycin as the standard. The MSMS spectra were
collected on a Micromass Quattro-LC triple quadrupole appar-
atus (QUATTRO) by selecting the m/z peak for the desired
parent ion isotopomer at the cone voltage where ion current
for that peak was maximum. The collision cell voltage was
manually increased to induce fragmentation and the daughter
ion spectra measured by the second quadrupole analyser. A scan
time of 8 s per spectrum and a low resolution setting were used
and several scans were summed to obtain the final spectrum.
The isothermal calorimetry measurements (ITC) were con-
ducted by using a MCS-ITC calorimeter from MicroCal, LLC,
Northampton, MA, USA. Aliquots of 10 µl were titrated into
the calorimetric cell every 3 min over a one and a half hour
period. The data were analysed using the customised ITC mod-
ule of the Origin 5.0 software package and a least squares fit-
ting procedure to fit the data to the appropriate binding model.
All measurements were carried out at 25 ЊC. For each system
studied a blank run was carried out in which the titrant was
titrated into the cell containing solvent only, to allow correc-
tions for the heat effects due to dilution to be made.
1
70%). IR (CHCl ) ν = 1697 (C᎐O) cmϪ1; H NMR.(400 MHz,
᎐
3
CDCl3) δ = 8.10 (1H, s, Ar–H), 8.07 (1H, d, J = 7.8 Hz, Ar–H),
7.72 (1H, d, J = 7.8 Hz, Ar–H), 7.50 (1H, t, J = 7.8 Hz, Ar–H),
5.46 (1H, s, CH(OMe)2), 3.35 (6H, s, CH(OMe)2); 13C NMR
(100 MHz, CDCl ) δ = 171.9 (C᎐O), 138.8, 132.1 (Arquat),
᎐
3
130.3, 129.4, 128,7, 128.5 (Ar–H), 102.4 (CH(OMe)2), 52.7
(CH(OMe)2); HRMS (EI) [M]ϩ C10H12O4 requires 196.0736,
found 196.0741; m.pt. = 68–69 ЊC.
General procedure for amide coupling. Amide coupling reac-
tions were performed using EDC and DMAP. The acid and
amine components were dissolved in dry DCM under argon. In
examples where the starting amine was purchased as the hydro-
chloride salt, dry Et3N (1.5 equivalents) was added. The result-
ing solution was cooled to 0 ЊC on an ice bath for 30 min after
which time the EDC (1.2 equivalents) and DMAP (20% by
weight with respect to EDC) were added. The reaction was kept
at 0 ЊC for 1 h and allowed to warm to room temperature and
stirred under argon overnight. The work up involved the addi-
tion of DCM (2-fold dilution) and subsequent washing of
the organic solution with three portions of H2O. The organic
layer was dried over MgSO4, filtered, and the solvent removed
in vacuo to give the crude product which was purified by silica
gel column chromatography.
General procedure for hydrogenation. Deprotection of the
CBZ protected intermediates was carried out using hydrogen-
ation. The CBZ protected compounds were dissolved in a
EtOAc:MeOH (4:1) mixture. To the solution was added 5% Pd/
C and the flask was evacuated and flushed with an atmosphere
of hydrogen three times before allowing the reaction to stir
under a hydrogen atmosphere for 6 h. The suspensions were
then filtered through a pad of celite and the solvent removed
in vacuo to afford the products which were used without further
purification.
Solution and solid state infrared spectra were recorded on
a Perkin Elmer Paragon 1000 FT-IR spectrometer at 4 cmϪ1
resolution or better.
General procedure for hydrazinolysis. The methyl esters were
dissolved in MeOH and to the resulting solutions were added
10 equivalents of hydrazine monohydrate. The reactions were
allowed to stir at room temperature overnight or until the reac-
tions had gone to completion, as ascertained by TLC. The sol-
vent was removed in vacuo to afford the crude products as oils
and were purified by silica gel column chromatography.
Materials
All chemicals were purchased from Aldrich, Fluka, Novabio-
chem or Senn Chemicals as reagent grade and used without
further purification. Thin layer chromatography was carried
out on glass sheets precoated with silica gel 60 F254 (Merck)
which were initially inspected by UV light before developing
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 6 2 5 – 1 6 3 3
1630

View Article Online
(S,S )-N-(Carbobenzyloxy-proline)phenylalanine methyl ester
(6a). Carbobenzyloxy--proline (1.0 g, 4.01 mmol) and -
phenyalanine methyl ester hydrochloride (0.87 g, 4.01 mmol)
were coupled in DCM (40 ml) using the standard pro-
cedure described above. Silica gel column chromatography
[EtOAc:Hex 6:4] yielded 6a as a white solid (1.38 g, 84%).
Rf = 0.43 [silica gel, EtOAc:Hex 6:4] (UV); IR (CHCl3) ν = 1740,
1736 (ester C᎐O), 1624 (amide C᎐O) cmϪ1; 1H NMR (400 MHz,
CDCl3) δ = 7.36–7.21 (8H, br m, Ar–H), 7.06–7.03 (2H, m, Ar–
H), 6.33 (1H, br s, NH), 5.17–5.08 (2H, m, CH2Ph), 4.83 (1H, t,
J = 9.7 Hz, α-H), 4.30 (1H, t, J = 8.5 Hz, Pro–α–H), 3.71 (3H, s,
OMe), 3.35 (2H, br m, Pro–NCH2), 3.15 (1H, br m, CHaHbPh),
2.99 (1H, br m, CHaHbPh), 2.04 (1H, m, Pro–CHaHb), 1.81
(2H, br m, Pro–CH2), 1.61 (1H, br m, Pro–CHaHb); 13C NMR
[DCM:MeOH/95:5] afforded pPFm (1) as a white solid (1.00 g,
88%). Rf = 0.38 [silica gel, DCM:MeOH 96:4] (UV, Ninhydrin);
IR (CHCl ) ν = 1681 (hydrazide C᎐O), 1672, 1624 (amide C᎐O)
᎐
᎐
3
cmϪ1; 1H NMR (400 MHz, CDCl3) δ = 7.79 (1H, br, NHNH2),
7.49 (2H, d, J = 8.2 Hz, Ar–H), 7.32 (2H, d, J = 8.2 Hz, Ar–H),
7.18–7.14 (5H, br m, Ar–H), 6.89 (1H, d, J = 8.2 Hz, NH), 5.42
(1H, s, CH(OMe)2), 4.70 (1H, dd, J1 = 10.1 Hz, J2 = 6.7 Hz,
α-H), 4.63 (1H, dd, J1 = 8.5 Hz, J2 = 6.1 Hz, Pro–α–H), 3.85
(2H, br s, NHNH2), 3.46–3.38 (2H, br m, Pro–NCH2), 3.34
(6H, s, CH(OMe)2), 3.23–3.10 (2H, m, CH2Ph), 2.22–2.11 (2H,
m, Pro–CH2), 1.92–1.76 (2H, m, Pro–CH2); 13C NMR (100
᎐
᎐
MHz, CDCl ) δ = 170.1 (hydrazine C᎐O), 169.5, 169.2 (amide
᎐
3
C᎐O), 139.1, 134.8, 133.8 (Arquat), 127.6, 127.1, 125.6, 125.5,
᎐
125.2 (Ar–H), 100.8 (CH(OMe)2), 61.1 (Pro–Cα–H), 53.2
(CH(OMe)2), 53.1 (Cα–H), 50.9 (Pro–NCH2), 37.5 (CH2Ph),
28.5, 25.8 (CH2); HRMS (QTOF) [M ϩ Na]ϩ C24H30N4O5Na
requires 477.2114, found 477.2105; m.pt. = 58–63 ЊC.
(100 MHz, CDCl ) δ = 171.8, 171.2 (ester, C᎐O), 156.1 (amide
᎐
3
C᎐O), 136.4, 136.0 (Arquat), 129.2, 128.5, 128.4, 128.1, 127.9,
᎐
126.9 (Ar–H), 67.4 (CH2Ph), 60.4 (Pro–Cα–H), 52.8 (Cα–H),
50.8 (OMe), 46.8 (Pro–NCH2), 37.9 (CH2Ph), 27.9, 23.4 (Pro–
CH2); HRMS (QTOF) [M ϩ Na]ϩ C23H26N2O5Na requires
433.1739, found 433.1750; m.pt. = 67–71 ЊC.
(S,S )-N-(3-Dimethoxymethyl-benzoyl)proline-phenylalanine
methyl ester (11). Proline-phenylalanine methyl ester (7a)
(1.25 g, 4.54 mmol) and 3-carboxybenzaldehyde dimethoxy-
acetal (9) (0.89 g, 4.54 mmol) were coupled in DCM (40 ml)
following the amide coupling procedure described above. Silica
gel column chromatography [EtOAc:Hex 8:2] yielded 11 as a
colourless oil (1.42 g, 69%). Rf = 0.41 [silica gel, EtOAc:Hex 8:2]
(S,S )-Proline-phenylalanine methyl ester (7a). N-(Carbo-
benzyloxy-proline)phenylalanine methyl ester (6a) (2.63 g, 6.41
mmol) was dissolved in a EtOAc:MeOH (4:1) mixture (100 ml)
and hydrogenated according the procedure given above. Yield:
(1.43 g, 85%). Rf = 0.32 [silica gel, DCM:MeOH 95:5] (UV,
Ninhydrin); IR (CHCl ) ν = 1731 (ester C᎐O), 1624 (amide C᎐
(UV); IR (CHCl ) ν =1739 (ester C᎐O), 1678, 1624 (amide C᎐O)
᎐
᎐
3
cmϪ1; 1H NMR (500 MHz, CDCl3) δ = 7.57 (1H, s, Ar–H), 7.53
(1H, d, J = 6.7 Hz, Ar–H), 7.43–7.37 (2H, br m, Ar–H), 7.34
(1H, d, J = 7.9 Hz, NH), 7.22–7.19 (5H, br m, Ar–H), 5.41 (1H,
s, CH(OMe)2), 4.85 (1H, dd, J1 = 10.5 Hz, J2 = 7.1 Hz, α-H),
4.75 (1H, t, J = 8.4 Hz, Pro–α–H), 3.72 (3H, s, OMe), 3.43–3.38
(2H, m, Pro–NCH2), 3.23 (6H, s, CH(OMe)2), 3.20 (1H, m,
CHaHbPh), 3.03 (1H, m, CHaHbPh), 2.37 (1H, m, Pro–CHaHb),
᎐
᎐
3
O) cmϪ1; 1H NMR (400 MHz, CDCl3) δ = 7.28–7.20 (3H, br m,
Ar–H), 7.10 (2H, m, Ar–H), 6.34 (1H, d, J = 8.1 Hz, NH), 4.85
(1H, dd, J1 = 10.5 Hz, J2 = 6.8 Hz, α-H), 3.71 (3H, s, OMe), 3.68
(1H, dd, J1 = 8.7 Hz, J2 = 6.7 Hz, Pro–α–H), 3.18 (1H, m,
CHaHbPh), 3.03 (1H, m, CHaHbPh), 2.92 (1H, m, Pro–
NCHaHb), 2.74 (1H, m, Pro–NCHaHb), 2.03 (1H, m, Pro–
CHaHb), 1.76 (1H, m, Pro–CHaHb), 1.63–1.47 (2H, br m,
Pro–CH2), 1.35 (1H, br, NH); 13C NMR (100 MHz, CDCl3)
2.03–1.97 (2H, br m, Pro–CH2), 1.78 (1H, m, Pro–CHaHb); 13
C
NMR (125 MHz, CDCl ) δ = 171.8 (ester C᎐O), 170.9, 169.7
᎐
3
δ = 173.5 (ester C᎐O), 160.7 (amide C᎐O), 134.7 (Arquat),
(amide C᎐O), 138.4, 136.1, 136.0 (Arquat), 129.3, 128.8, 128.6,
᎐
᎐
᎐
127.8, 127.0, 125.5 (Ar–H), 58.8 (Pro–Cα–H), 50.9 (Cα–H),
50.8 (OMe), 45.7 (Pro–NCH2), 36.7 (CH2Ph), 29.2, 24.6 (CH2);
HRMS (QUATTRO) [M ϩ H]ϩ C15H21N2O3 requires 277.1153,
found 277.1140.
128.4, 127.3, 126.9, 125.6 (Ar–H), 102.4 (CH(OMe)2), 59.8
(Pro–Cα–H), 53.5 (Cα–H), 52.7 (CH(OMe)2), 52.3 (OMe), 50.2
(Pro–NCH2), 38.0 (CH2Ph), 27.3, 25.3 (CH2); HRMS (QTOF)
[M ϩ Na]ϩ C25H30N2O6Na requires 477.2002, found 477.2012.
(S,S )-N-(4-Dimethoxymethyl-benzoyl)proline-phenylalanine
methyl ester (10a). Proline-phenylalanine methyl ester (7a)
(1.08 g, 3.91 mmol) and 4-carboxybenzaldehyde dimethoxy-
acetal (8) (0.77 g, 3.91 mmol) were reacted together in DCM
(40 ml) following the amide coupling procedure given above.
Silica gel column chromatography [EtOAc:Hex 8:2] yielded 10a
as a white solid (1.23 g, 69%). Rf = 0.43 [silica gel, EtOAc:Hex
mPFm: (S,S )-N-(3-Dimethoxymethyl-benzoyl)proline-phenyl-
alanine carboxylic acid hydrazide (2). N-(3-Dimethoxymethyl-
benzoyl)proline-phenylalanine methyl ester (11) (1.42 g, 3.13
mmol) in MeOH (30 ml) was reacted under the standard
hydrazinolysis conditions given above. Silica gel column
chromography [DCM:MeOH 95:5] afforded mPFm (2) as a
white solid (1.21 g, 85%). Rf = 0.31 [silica gel, DCM:MeOH
8:2] (UV); IR (CHCl ) ν = 1730 (ester C᎐O), 1674, 1625 (amide
95:5] (UV, Ninhydrin); IR (CHCl ) ν = 1681 (hydrazide C᎐O),
᎐
᎐
3
3
C᎐O) cmϪ1; 1H NMR (400 MHz, CDCl3) δ = 7.48 (2H, d, J = 8.1
1672, 1624 (amide C᎐O) cmϪ1; H NMR (400 MHz, CDCl3)
1
᎐
᎐
Hz, Ar–H), 7.41 (2H, d, J = 8.1 Hz, Ar–H), 7.30 (1H, d, J = 7.4
Hz, NH), 7.18–7.13 (5H, br m, Ar–H), 5.41 (1H, s, CH(OMe)2),
4.84 (1H, t, J = 9.8 Hz, α-H), 4.75 (1H, t, J =8.3 Hz, Pro–α–H),
3.70 (3H, s, OMe), 3.41–3.37 (2H, br m, Pro–NCH2), 3.32 (6H,
s, CH(OMe)2), 3.20 (1H, d, J = 14.0 Hz, CHaHbPh), 3.03 (1H, d,
J = 14.0 Hz, CHaHbPh), 2.35 (1H, m, Pro–CHaHb), 2.02–1.90
(2H, m, Pro–CH2), 1.78 (1H, m, Pro–CHaHb); 13C NMR
(125 MHz, CDCl ) δ = 172.3 (ester C᎐O), 169.9, 168.7 (amide
δ = 7.68 (1H, br, NHNH2), 7.55 (1H, s, Ar–H), 7.53 (1H, d,
J = 7.8 Hz, Ar–H), 7.39 (1H, t, J = 7.8 Hz, Ar–H), 7.30 (1H,
d, J = 7.8 Hz, Ar–H), 7.28–7.14 (5H, m, Ar–H), 6.88 (1H, d,
J = 8.3 Hz, NH), 5.40 (1H, s, CH(OMe)2), 4.68 (1H, m, α-H),
4.62 (1H, m, Pro–α–H), 3.83 (2H, br, NHNH2), 3.40 (2H, m,
Pro–NCH2), 3.32 (6H, s, CH(OMe)2), 3.15 (2H, m, CH2Ph),
2.17 (1H, m, Pro–CHaHb), 2.10 (1H, m, Pro–CHcHd), 1.90–1.75
(2H, m, Pro–CH2); 13C NMR (125 MHz, CDCl3) δ = 171.3
᎐
3
C᎐O), 139.4, 134.5, 133.6 (Arquat), 127.5, 127.3, 125.7, 125.5,
᎐
(hydrazine C᎐O), 171.0, 170.9 (amide C᎐O), 138.5, 136.4, 135.3
᎐ ᎐
125.3 (Ar–H), 101.2 (CH(OMe)2), 60.9 (Pro–Cα–H), 53.4
(CH(OMe)2), 53.3 (Cα–H), 50.7 (Pro–NCH2), 37.5 (CH2Ph),
28.4, 25.7 (CH2); HRMS (QUATTRO) [M ϩ H]ϩ C25H31N2O6
requires 455.2183, found 455.2190; m.pt. = 127–129 ЊC.
(Arquat), 129.3, 129.0, 128.7, 128.4, 127.4, 127.1, 125.6 (Ar–H),
102.4 (CH(OMe)2), 60.8 (Pro–Cα–H), 53.1 (Cα–H), 52.8
(CH(OMe)2), 50.5 (Pro–NCH2), 37.0 (CH2Ph), 28.1, 25.3
(CH2); HRMS (QTOF) [M ϩ Na]ϩ C24H30N4O5Na requires
477.2114, found 477.2114; m.pt. = 61–65 ЊC.
pPFm: (S,S )-N-(4-Dimethoxymethyl-benzoyl)proline-phenyl-
alanine carboxylic acid hydrazide (1). N-(4-Dimethoxymethyl-
benzoyl)proline-phenylalanine methyl ester (10a) (1.13 g,
2.49 mmol) underwent hydrazinolysis in MeOH (25 ml) via
the procedure given above. Silica gel column chromography
(S,S )-N-(Carbobenzyloxy-proline)cyclohexylalanine methyl
ester (6b). Carbobenzyloxy--proline (2.69 g, 10.8 mmol) and
-cyclohexylalanine methyl ester hydrochloride (2.00 g, 9.02
mmol) in DCM (100 ml) were reacted using the standard amide
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 6 2 5 – 1 6 3 3
1631

View Article Online
coupling reaction described above. Silica gel column chromato-
graphy [EtOAc:Hex 6:4] yielded 6b as a white solid (3.41 g,
91%). Rf = 0.43 [silica gel, EtOAc:Hex 6:4] (UV); IR (CHCl3)
ν = 1740, 1693 (ester C᎐O), 1618 (amide C᎐O) cmϪ1; 1H NMR
(500 MHz, CDCl3) δ = 7.39–7.25 (5H, m, Ar–H), 7.07 (1H,
br, NH), 5.14 (2H, m, CH2Ph), 4.55 (1H, dd, J1 = 8.4 Hz,
J2 =6.3 Hz, Pro–α–H), 4.34 (1H, dd, J1 = 10.5 Hz, J2 = 8.3 Hz,
α-H), 3.69 (3H, s, OMe), 3.57–3.38 (2H, m, Pro–NCH2), 2.35
(1H, m, Pro–CHaHb), 2.14 (1H, m, Pro–CHaHb), 1.91 (2H, m,
Pro–CH2), 1.73–1.53 (8H, br m, ring and chain CH2), 1.28 (1H,
m, CH(CH2)5), 1.21–1.06 (2H, m, ring CH2), 0.95–0.80 (2H, m,
ring CH2); 13C NMR (125 MHz, CDCl3) δ = 173.2, 171.2
(CH(OMe)2), 50.6 (Pro–NCH2), 50.2 (Cα–H), 38.6 (CH2-
(C6H11)), 34.2 (CH(CH2)5), 33.5, 32.3 (ring CH2), 27.1 (Pro–
CH2), 26.2 (ring CH2), 25.5 (Pro–CH2); HRMS (QTOF)
[M ϩ Na]ϩ C24H36N4O5Na requires 483.2584, found 483.2583;
m.pt. = 70–73 ЊC.
᎐
᎐
General procedures for the preparation of DCLs
All cyclisation reactions were performed at room temperature
on a 1 ml scale. The monomer(s) were dissolved in CHCl3:
MeOH 98:2 (v/v) to give a solution of 20 mM. The stock solu-
tion was then diluted to give 1 ml of a 5 mM solution. TFA was
added (43 equivalents in 20–100 µl of a stock solution in
CHCl3:MeOH 98:2 (v/v)) to the reaction to initiate deprotec-
tion and cyclisation. HPLC or mass spectrometry analysis was
performed by injection of 10–50 µl of the reaction mixture.
Unless otherwise stated cyclisation reactions were stirred at rt
and followed by HPLC analysis every 24 h for up to 7 days.
(ester C᎐O), 156.2 (amide C᎐O), 136.4 (Arquat), 128.5, 128.0,
᎐
᎐
127.8 (Ar–H), 67.3 (CH2Ph), 60.2 (Pro–Cα–H), 52.1 (OMe),
50.3 (Cα–H), 46.8 (Pro–NCH2), 39.7 (CH2(C6H11)), 34.2
(CH(CH2)5), 33.3, 32.4 (ring CH2), 27.8 (Pro–CH2), 26.2 (ring
CH2), 24.6 (Pro–CH2); HRMS (QTOF) [M ϩ Na]ϩ C23H32N2-
O5Na requires 439.2209, found 439.2209; m.pt. = 83–84 ЊC.
General procedure for the large scale isolation of cyclic oligomers
(S,S )-N-(4-Dimethoxymethyl-benzoyl)proline-cyclohexyl-
alanine methyl ester (10b). N-(Carbobenzyloxy-proline)cyclo-
hexylalanine methyl ester (6b) (3.41 g, 8.19 mmol) was dissolved
in a EtOAc:MeOH (4:1) mixture (100 ml) and hydrogenated
as described above, yielding the free amine as a colourless
oil (2.20 g, 95%) which was used without further purifi-
cation. Rf = 0.28 [silica gel, EtOAc:Hex 6:4] (Ninhydrin). The
prolinecyclohexylalanine methyl ester (2.20 g, 7.78 mmol) and
4-carboxybenzaldehyde dimethoxyacetal (8) (1.78 g, 9.34
mmol) were then coupled in DCM (100 ml) according to the
procedure given above. Silica gel column chromatography
[EtOAc:Hex 8:2] yielded 10b as a colourless oil (2.97 g, 83%).
Rf = 0.36 [silica gel, EtOAc:Hex 8:2] (UV); IR (CHCl3) ν = 1740
Large scale cyclisations were carried out on a 25 ml scale with
5 mM monomer, 43 equivalents of TFA and 3 or 5 equivalents
template with respect to monomer, in CHCl3:MeOH 98:2 (v/v).
The reactions were stirred at rt and monitored by HPLC until
thermodynamic equilibrium was reached (24 h). The exchange
reaction was quenched by the addition of 15 g of solid phase
base Amberlyst-21, which was washed thorougly with water,
MeOH and chloroform before use. After stirring for 15 min the
base was filtered off and washed with CHCl3 (100 ml) and
MeOH (100 ml). The filtrates were combined and evaporated to
dryness before redissolution in approximately 2 ml of DMSO.
The crude product was purified by semi-preparative HPLC
which yielded the products as white solids.
(ester C᎐O), 1678, 1618 (amide C᎐O) cmϪ1 1H NMR (500
;
᎐
᎐
MHz, CDCl3) δ = 7.47 (4H, m, Ar–H), 7.31 (1H, d, J = 8.2 Hz,
NH), 5.39 (1H, s, CH(OMe)2), 4.81 (1H, dd, J1 = 8.4 Hz,
J2 = 6.4 Hz, Pro–α–H), 4.57 (1H, dd, J1 = 10.1 Hz, J2 = 8.5 Hz,
α-H), 3.71 (3H, s, OMe), 3.51 (1H, m, Pro–NCHaHb), 3.43 (1H,
m, Pro–NCHaHb), 3.29 (6H, s, CH(OMe)2), 2.48 (1H, m, Pro–
CHaHb), 2.03 (2H, m, Pro–CH2), 1.81 (1H, m, Pro–CHaHb),
1.72–1.52 (8H, br m, ring and chain CH2), 1.29 (1H, m,
(mPF)2:(S,S )-N-(3-Dimethoxymethyl-benzoyl)proline-phenyl-
alanine carboxylic acid hydrazide dimer. Yield: (22 mg, 71%). IR
(CDCl :MeOD 98:2) ν = 1681 (br, C᎐Os) cmϪ1 1H NMR
;
᎐
3
(500 MHz (cryoprobe), CDCl3:MeOD 98:2) δ = 10.52 (2H, br,
NHNC), 8.44 (2H, s, HCN), 8.11 (2H, br, Ar–H), 7.93 (2H,
s, Ar–H), 7.47–7.35 (4H, m, Ar–H), 7.33–7.15 (10H, br m,
Ar–H), 6.20 (2H, br, NH), 4.76 (2H, br m, α-H), 4.37 (2H, br m,
Pro–α–H), 3.90–3.83 (4H, br m, Pro–NCH2), 3.50 (2H, br m,
CHaHbPh), 3.30 (2H, br m, CHaHbPh), 2.32 (2H, br m, Pro–
CHaHb), 2.12 (2H, br m, Pro–CHcHd), 1.89 (2H, br m, Pro–
CHcHd), 1.70 (2H, br m, Pro–CHaHb); 13C NMR (125 MHz
CH(CH2)5), 1.11 (2H, m, ring CH2), 0.87 (2H, m, ring CH2); 13
NMR (125 MHz, CDCl ) δ = 173.2 (ester C᎐O), 170.8, 170.7
C
᎐
3
(amide C᎐O), 140.3, 136.2 (Arquat), 129.9, 127.0 (Ar–H),
᎐
102.4 (CH(OMe)2), 59.5 (Pro–Cα–H), 52.6 (CH(OMe)2), 52.1
(OMe), 50.4 (Pro–NCH2), 50.3 (Cα–H), 39.7 (CH2(C6H11)),
34.2 (CH(CH2)5), 33.4, 32.3 (ring CH2), 26.8 (Pro–CH2), 26.1
(ring CH2), 25.4 (Pro–CH2); HRMS (QTOF) [M ϩ Na]ϩ
C25H36N2O6Na requires 483.2471, found 483.2471.
(cryoprobe), CDCl :MeOD 98:2) δ = 171.0, 169.4, 167.6 (C᎐O),
᎐
3
149.0 (CN), 137.0, 134.8, 133.5, 131.2, 130.2, 129.5, 129.3,
128.5, 126.9, 123.5 (Ar), 63.6 (Pro–Cα–H), 51.3 (Cα–H), 50.7
(Pro–NCH2), 35.7 (CH2Ph), 29.6, 26.0 (CH2); HRMS (QTOF)
[M ϩ Na]ϩ C44H44N8O6Na requires 803.3281, found 803.3282;
m.pt. 258 ЊC (decomposition).
pPCm: (S,S )-N-(4-Dimethoxymethyl-benzoyl)proline-cyclo-
hexylalanine carboxylic acid hydrazide (3). N-(4-Dimethoxy-
methyl-benzoyl)proline-cyclohexylalanine methyl ester (10b)
(2.97 g, 6.46 mmol) was treated with hydrazine monohydrate in
MeOH (30 ml) according to the standard hydrazinolysis pro-
cedure. Silica gel column chromography [DCM:MeOH 95:5]
afforded pPCm (3) as a white solid (2.68 g, 90%). Rf = 0.38
[silica gel, DCM:MeOH 95:5] (UV, Ninhydrin); IR (CHCl3)
(pPC)3: (S,S )-N-(4-Dimethoxymethyl-benzoyl)proline-cyclo-
hexylalanine carboxylic acid hydrazide trimer. Yield: (34 mg,
1
59%). IR (CDCl :MeOD 98:2) ν = 1674 (br, C᎐Os) cmϪ1; H
᎐
3
NMR (500 MHz (cryoprobe), CDCl3:MeOD 98:2) δ = 10.57
(3H, br, NHNC), 8.40 (3H, s, HCN), 7.75 (12H, m, Ar–H), 4.81
(3H, dd, J1 = 11.0 Hz, J2 = 4.0 Hz, α-H), 4.49 (3H, dd, J1 = 10.4
Hz, J2 = 7.3 Hz, Pro–α–H), 3.91 (3H, m, Pro–NCHaHb), 3.78
(3H, m, Pro–NCHaHb), 2.58 (3H, m, Pro–CHaHb), 2.10 (3H, m,
Pro–CHcHd), 2.07 (3H, m, CHaHb(C6H11)), 1.91 (3H, m, Pro–
CHcHd), 1.84 (3H, m, Pro–CHaHb), 1.60 (3H, m, CHaHb-
(C6H11)), 1.18–0.78 (11H, br m, ring CH2); 13C NMR (125 MHz
1
ν = 1681 (hydrazide C᎐O), 1672, 1618 (amide C᎐O) cmϪ1; H
᎐
᎐
NMR (500 MHz, CDCl3) δ = 7.61 (1H, br, NHNH2), 7.52 (4H,
m, Ar–H), 7.20 (1H, d, J = 8.1 Hz, NH), 5.42 (1H, s,
CH(OMe)2), 4.75 (1H, dd, J1 = 8.5 Hz, J2 = 6.1 Hz, Pro–α–H),
4.43 (1H, dd, J1 = 10.3 Hz, J2 = 8.5 Hz, α-H), 3.89 (2H, br,
NHNH2), 3.45 (2H, m, Pro–NCH2), 3.32 (6H, s, CH(OMe)2),
2.43 (1H, m, Pro–CHaHb), 2.15–2.01 (2H, m, Pro–CH2), 1.89
(1H, m, Pro–CHcHd), 1.73 (1H, m, CHaHb(C6H11)), 168
(1H, m, CHaHb(C6H11)), 1.67–1.48 (6H, br m, ring CH2), 1.36
(1H, m, CH(CH2)5), 1.11 (2H, m, ring CH2), 0.96–0.81 (2H, m,
ring CH2); 13C NMR (125 MHz, CDCl3) δ = 171.3 (hydrazide
C᎐O), 171.2, 170.9 (amide C᎐O), 140.7, 135,7 (Arquat), 127.2,
(cryoprobe), CDCl :MeOD 98:2) δ = 171.5, 171.1, 170.0 (C᎐O),
᎐
3
147.8 (CN), 137.2, 135.3, 128.4, 127.4 (Ar), 64.0 (Pro–Cα–H),
51.2 (Pro–NCH2), 49.9 (Cα–H), 37.5, 34.7, 33.8, 31.9,
31.6, 29.6, 27.5 (ring and chain CH2); MS (QUATTRO)
[M ϩ Na]ϩ C66H84N12O9Na requires 1212, found 1212; MSMS
(QUATTRO) [trimer ϩ H]ϩ 1190 (C66H85N12O9), [dimer ϩ H]ϩ
793 (C44H57N8O6), [monomerϪ cyclohexylalanine unit]ϩ 243
(C13H13N3O2); m.pt. 267 ЊC (decomposition).
᎐
᎐
126.9 (Ar–H), 102.4 (CH(OMe)2), 60.0 (Pro–Cα–H), 52.7
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 6 2 5 – 1 6 3 3
1632

View Article Online
8 (a) D. M. Epstein, S. Choudhary, R. M. Churchill, K. M. Keil,
(pPF)3: (S,S )-N-(4-Dimethoxymethyl-benzoyl)proline-phenyl-
alanine carboxylic acid hydrazide trimer. Isolated as described
previously.10d Rf = 0.45 [silica gel, DCM:MeOH 95:5] (UV); IR
A. V. Eliseev and J. R. Morrow, Inorg. Chem., 2001, 40, 1591;
(b) S. Choudhary and J. R. Morrow, Angew. Chem., Int. Ed., 2002,
41, 4096; (c) I. Huc, M. J. Krische, D. P. Funeriu and J.-M. Lehn,
Eur. J. Inorg. Chem., 1999, 1415; (d ) E. Stulz, Y.-F. Ng, S. M. Scott
and J. K. M. Sanders, Chem. Commun., 2002, 524; (e) M. Ziegler,
J. J. Miranda, U. N. Andersen, D. W. Johnson, J. A. Leary
and K. N. Raymond, Angew. Chem., Int. Ed., 2001, 40, 733;
( f ) M. Albrecht, O. Blau and R. Fröhlich, Chem. Eur. J., 1999, 5, 48;
(g) S. Hiraoke and M. Fujita, J. Am. Chem. Soc., 1999, 121, 10239;
(h) Y. Kubota, S. Sakamoto, K. Yamaguchi and M. Fujita, Proc.
Natl. Acad. Sci. USA, 2002, 99, 4854; (i) S. Sakai, Y. Shigemasa
and T. Sasaki, Tetrahedron Lett., 1997, 38, 8145; (j) B. Klekota,
M. H. Hammond and B. L. Miller, Tetrahedron Lett., 1997, 38,
8639; (k) B. Klekota and B. L. Miller, Tetrahedron, 1999, 55, 11687;
(l ) C. Karan and B. L. Miller, J. Am. Chem. Soc., 2001, 123, 7455;
(m) M. A. Case and G. L. McLendon, J. Am. Chem. Soc., 2000, 122,
8089; (n) E. C. Constable, C. E. Housecroft, T. Kulke, C. Lazzarini,
E. R. Schofield and Y. Zimmermann, J. Chem. Soc., Dalton Trans.,
2001, 2864; (o) V. Goral, M. I. Nelen, A. V. Eliseev and J.-M. Lehn,
Proc. Natl. Acad. Sci. USA, 2001, 98, 1347.
1
(CDCl :MeOD 98:2) ν = 1673 (br, C᎐Os) cmϪ1; H NMR (500
᎐
3
MHz (cryoprobe), CDCl3:MeOD 98:2) δ = 10.29 (3H, br,
NHNC), 8.24 (3H, s, HCN), 7.67 (6H, d, J = 8.0 Hz, Ar–H),
7.43–7.14 (21H, m, Ar–H), 4.99 (3H, t, J = 5.6 Hz, α-H), 4.40
(3H, dd, J1 = 10.3 Hz, J2 = 7.0 Hz, Pro–α–H), 3.67 (3H, dd,
J1 = 14.4 Hz, J2 = 7.0 Hz, CHaHbPh), 3.58 (6H, m, Pro–NCH2),
3.11 (3H, dd, J1 = 14.4 Hz, J2 = 10.3 Hz, CHaHbPh), 2.47 (3H,
m, Pro–CHaHb), 2.05 (3H, m, Pro–CHcHd), 1.87 (3H, m, Pro–
CHaHb), 1.78 (3H, m, Pro–CHcHd); 13C NMR (125 MHz (cryo-
probe), CDCl :MeOD 98:2) δ = 171.4, 170.1, 167.2 (C᎐O),
᎐
3
148.3 (CN), 137.1, 135.7, 135.4, 129.4, 129.2, 127.9, 127.5,
127.2 (Ar), 63.2 (Pro–Cα–H), 51.2 (Pro–NCH2), 50.1 (Cα–H),
36.3 (CH2Ph), 29.6, 26.3 (CH2); MS (QUATTRO) [M ϩ Na]ϩ
C66H84N12O9Na requires 1212, found 1212; MSMS (QUAT-
TRO) [trimer ϩ H]ϩ 1172 (C66H67N12O9), [trimer Ϫ phenylalan-
ine unit]ϩ 1025 (C57H58N11O8), [dimer ϩ H]ϩ 781 (C44H45N8O6),
[monomer Ϫ phenylalanine unit]ϩ 243 (C13H14N3O2); m.pt. 262
ЊC (decomposition).
9 (a) A. V. Eliseev and M. I. Nelen, J. Am. Chem. Soc., 1997, 119,
1147; (b) A. V. Eliseev and M. I. Nelen, Chem. Eur. J., 1998, 4, 825.
10 (a) S. Otto, R. L. E. Furlan and J. K. M. Sanders, Science, 2002, 297,
590; (b) R. L. E. Furlan, Y.-F. Ng, G. R. L. Cousins, J. E. Redman
and J. K. M. Sanders, Tetrahedron, 2002, 58, 771; (c) M. Hochgürtel,
H. Kroth, D. Piecha, M. W. Hofmann, C. Nicolau, S. Krause,
O. Schaaf, G. Sonnenmoser and A. V. Eliseev, Proc. Natl. Acad. Sci.
USA, 2002, 99, 3382; (d ) R. L. E. Furlan, Y.-F. Ng, S. Otto and
J. K. M. Sanders, J. Am. Chem. Soc., 2001, 123, 8876; (e) G. R. L.
Cousins, R. L. E. Furlan, Y.-F. Ng, J. E. Redman and J. K. M.
Sanders, Angew. Chem., Int. Ed., 2001, 40, 423; ( f ) H. Hioki and
W. C. Still, J. Org. Chem., 1998, 63, 904.
11 Nomenclature of the dipeptide building blocks: the first m/p denotes
the meta or para position of the protected aldehyde group, the last
character indicates the size of the macrocycle (m for monomer) and
the middle two characters are the one letter codes for the amino
acids, i.e. P for proline, F for phenylalanine, and C for cyclohexyl-
alanine.
12 HPLC grade chloroform (Aldrich), stabilised by 0.1% (by weight)
amylenes was used throughout.
13 Linear oligomers are also present at concentrations which are
generally too low to be detected easily, but they can be amplified by
binding the terminal protonated hydrazinium ion to 18-crown-6.
See: R. L. E. Furlan, G. R. L. Cousins and J. K. M. Sanders, Chem.
Commun., 2000, 1761.
14 In the presence of NaI libraries generated from pPFm pass through
an intermediate stage with relatively large amounts of cyclic dimer,
which is only sparingly soluble. The precipitates formed upon
starting equilibration from the monomer are slow to redissolve,
acting as a kinetic trap, preventing thermodynamic equilibrium from
being reached.
15 The role of the trifluoroacetate anion cannot be assessed, as TFA is
always present in large excess.
16 L. Fielding, Tetrahedron, 2000, 56, 6151.
17 A. J. Goshe, I. M. Steele, C. Ceccarelli, A. L. Rheingold and
B. Bosnich, Proc. Natl. Acad. Sci. USA, 2002, 99, 4823.
18 (a) D. L. Boger, M. A. Patane and J. Zhou, J. Am. Chem. Soc., 1995,
117, 7357; (b) G. Müller, G.-M. Maier and M. Lutz, Inorg. Chim.
Acta, 1994, 218, 121; (c) D. Seebach and H. G. Bossler, Helv. Chim.
Acta, 1994, 77, 291; (d ) D. Seebach, A. K. Beck, H. G. Bossler,
C. Gerber, S. Y. Ko, C. W. Murtiashaw, R. Naef, S. Shoda
and A. Thaler, Helv. Chim. Acta, 1993, 76, 1565; (e) M. Köck,
H. Kessler, D. Seebach and A. Thaler, J. Am. Chem. Soc., 1992, 114,
2676; ( f ) H. Kessler, W. Hehlein and R. Schuck, J. Am. Chem. Soc.,
1982, 104, 4534.
Acknowledgements
We are grateful for support from GlaxoSmithKline and BBSRC
to SLR, from the Fundación Antorchas and the Consejo
Nacional de Investigaciones Científicas y Tecnicas (Argentina)
to RLEF, from the Royal Society (University Research Fellow-
ship) to SO, and from EPSRC to JKMS.
References
1 For reviews, see: (a) S. Otto, R. L. E. Furlan and J. K. M. Sanders,
Curr. Opin. Chem. Biol., 2002, 6, 321; (b) R. L. E. Furlan, S. Otto
and J. K. M. Sanders, Proc. Natl. Acad. Sci. USA, 2002, 99,
4801; (c) S. J. Rowan, S. J. Cantrill, G. R. L. Cousins, J. K. M.
Sanders and J. F. Stoddart, Angew. Chem., Int. Ed., 2002, 41, 898;
(d ) O. Ramström, T. Bunyapaiboonsri, S. Lohmann and J.-M. Lehn,
Biochim. Biophys. Acta, 2002, 1572, 178; (e) J.-M. Lehn and A. V.
Eliseev, Science, 2001, 291, 2331; ( f ) C. Karan and B. L. Miller,
Drug Discov. Today, 2000, 5, 67; (g) A. Ganesan, Angew. Chem., Int.
Ed., 1998, 37, 2828.
2 (a) P. A. Brady, R. P. Bonar-Law, S. J. Rowan, C. J. Suckling and
J. K. M. Sanders, Chem. Commun., 1996, 319; (b) S. J. Rowan,
P. A. Brady and J. K. M. Sanders, Angew. Chem., Int. Ed. Engl.,
1996, 35, 2143; (c) S. J. Rowan, P. A. Brady and J. K. M. Sanders,
Tetrahedron Lett., 1996, 37, 6013; (d ) S. J. Rowan and J. K. M.
Sanders, J. Org. Chem., 1998, 63, 1536; (e) P. A. Brady and J. K. M.
Sanders, J. Chem. Soc., Perkin Trans. 1, 1997, 3237; ( f ) G. Kaiser
and J. K. M. Sanders, Chem. Commun., 2000, 1763.
3 Imine exchange: (a) I. Huc and J.-M. Lehn, Proc. Natl. Acad. Sci.
USA, 1997, 94, 2106; (b) A. Star, I. Goldberg and B. Fuchs, Angew.
Chem., Int. Ed., 2000, 39, 2685; (c) O. Storm and U. Lüning, Chem.
Eur. J., 2002, 8, 793; Oxime exchange (d ) N. Nazarpack-
Kandlousy, J. Zweigenbaum, J. Henion and A. V. Eliseev, J. Comb.
Chem., 1999, 1, 199 . Hydrazone exchange; (e) G. R. L. Cousins,
S.-A. Poulsen and J. K. M. Sanders, Chem. Commun., 1999, 1575;
( f ) V. Berl, I. Huc, J.-M. Lehn, A. DeCian and J. Fischer, Eur.
J. Inorg. Chem., 1999, 3089; (g) T. Bunyapaiboonsri, O. Ramström,
S. Lohmann, J.-M. Lehn, L. Peng and M. Goeldner, ChemBioChem,
2001, 2, 438.
4 T. Giger, M. Wigger, S. Audetat and S. A. Benner, Synlett, 1998, 688.
5 (a) S. Otto, R. L. E. Furlan and J. K. M. Sanders, J. Am. Chem. Soc.,
2000, 122, 12063; (b) O. Ramström and J.-M. Lehn, ChemBioChem,
2000, 1, 41.
6 (a) P. G. Swann, R. A. Casanova, A. Desai, M. M. Frauenhoff,
M. Urbancic, U. Slomczynska, A. J. Hopfinger, G. C. Le Breton and
D. L. Venton, Biopolymers, 1996, 40, 617; (b) R. J. Lins, S. L. Flitsch,
N. J. Turner, E. Irving and S. A. Brown, Angew. Chem., Int. Ed.,
2002, 41, 3405.
7 (a) M. Crego Calama, R. Hulst, R. Fokkens, N. M. M. Nibbering,
P. Timmerman and D. N. Reinhoudt, Chem. Commun., 1998, 1021;
(b) M. Crego Calama, P. Timmerman and D. N. Reinhoudt, Angew.
Chem., Int. Ed., 2000, 39, 755; (c) F. Hof, C. Nuckolls and J. Rebek,
J. Am. Chem. Soc., 2000, 122, 4251.
19 The shifts in the carbonyl stretching frequencies of (pPC)3 measured
here in CHCl3:MeOH (98:2 v/v) are much smaller than those
reported previously (1684 to 1666 cmϪ1
) measured in pure
chloroform (see reference 10d ).
20 M. J. Blandamer, P. M. Cullis and J. B. F. N. Engberts, J. Chem. Soc.,
Faraday Trans., 1998, 94, 2261.
21 J. S. Bradshaw, R. M. Izatt, A. V. Bordunov, C. Y. Zhu and
J. K. Hathaway, in Comprehensive Supramolecular Chemistry, Vol. 1,
eds. J. L. Atwood, J. E. D. Davies, D. D. Macnicol and F. Vögtle,
Elsevier, Oxford, 1996, pp. 65–67 and references therein.
22 B. R. Linton, M. S. Goodman, E. Fan, S. A. van Arman and
A. D. Hamilton, J. Org. Chem., 2001, 66, 7313.
23 A. Clerici, N. Pastori and O. Porta, Tetrahedron, 1998, 54, 15679.
24 J. I. De Graw, L. Goodman and B. R. Baker, J. Org. Chem., 1961, 26,
1156.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 1 6 2 5 – 1 6 3 3
1633