Molecularly imprinted cavities template the macrocyclization of tetrapeptides

Article abstract of DOI:10.1039/b813439a

Cavities formed using cyclic tetrapeptides (CTPs) or heat-induced conformers act as templates for cyclization; the cavities bind to linear tetrapeptides and enforce turn conformations to enhance cyclization to constrained CTPs. The Royal Society of Chemis

Full text of DOI:10.1039/b813439a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Molecularly imprinted cavities template the macrocyclization
of tetrapeptidesw
Dar-Fu Tai* and Yee-Fung Lin
Received (in Cambridge, UK) 4th August 2008, Accepted 26th August 2008
First published as an Advance Article on the web 29th September 2008
DOI: 10.1039/b813439a
Cavities formed using cyclic tetrapeptides (CTPs) or heat-
induced conformers act as templates for cyclization; the cavities
bind to linear tetrapeptides and enforce turn conformations to
enhance cyclization to constrained CTPs.
These shapes should facilitate the subsequent cyclization
process (Scheme 1). This protocol shows promise as a recogni-
tion-based system involving a cavity of nanometric dimensions
(the distance represented by a b-turn is approximately 10 A).
As previously reported, synthesis of cyclo-(Phe-Phe-Phe-
Phe)15 is a challenging task, as is the synthesis of cyclo-(Gly-
Gly-Gly-Gly)16 and of cyclo-(Pro-Pro-Pro-Pro).16 Therefore,
these three tetrapeptides and the two disputed peptides (Leu-
Pro-Leu-Pro, Pro-Val-Pro-Val)6,7 were selected as examples to
test the new synthesis of CTPs. A three-stage procedure to
facilitate CTPs formation is shown in Scheme 2. The yields of
cyclization can be used as a report on the conformation in a
turn or partial turn of a linear peptide in a nano size cavity.
The first stage is the construction of the desired turn-inducing
cavities for the respective tetrapeptides. Cellulose fibers were
chosen as the platform on which to construct the surface due to
its performance and convenience. The highly hydrophilic char-
acter of cellulose-based composite materials were modified with
3-methacryloxypropyltrimethoxysilane (MPS) to form a hydro-
phobic monolayer on the cellulose surface.17 The grafted
cellulose fiber was hydrophobic enough and suitable for fabri-
cation of molecularly imprinted polymers. This nano-cage was
constructed by polymerization with acrylamide, N-acryltyr-
amine (ATA), N,N0-ethylene bisacrylamide (EBAA) and
2,20-azobisisobutyronitrile (AIBN) in the presence of the target
tetrapeptide at high temperature. To generate hydrophobic
cavities and avoid amino group and carboxylic group inter-
actions on the template the ionic monomer is limited. The linear
Constrained cyclic peptides incorporating b-turn structures have
been designed and studied1 to provide conformational insight2 for
receptor binding.3 In a cyclic tetrapeptide ring the distance
between C alpha (i) and C alpha (i + 3) is less than 7 A4 which
is suitable for metal ion chelation or incorporating small mole-
cules between the respective side chains. Although CTPs have
been isolated and structurally characterized, there has been
difficulty in synthesizing derivatives without functional side
chains.5 Typical examples are the antiproliferative agent cyclo-
(Pro-Leu-;Pro-Leu)6 and tyrosinase inhibitor cyclo-(Pro-Val-
Pro-Val).6,7 Aracil and Francisco6 reported the synthesis of these
two peptides having certain biological activities; other groups6,7
tried to resynthesize these compounds, but obtained compounds
without the reported biological activities and with different
NMR spectra. The cyclo-(Pro-Val-Pro-Val) molecule has a
cis-trans-cis-trans backbone-ring conformation.7 It is possible to
build an all-cis cyclic peptide computationally.8 However, little
effort has been made to construct these diverse and novel
structures.9 Macrocycles, not only those with b-turns, have been
prepared using solid-phase methods.10 Although highly strained
CTPs are not readily available, CTP-like molecules (b-turn
mimetics) are important substitutes11 often used for interaction
with protein targets. The low hit rates commonly encountered in
high-throughput screening of these analogs suggest that improved
macrocyclization approaches to CTP’s will be welcomed.
CTPs are known to undergo conformational interconversions
at high temperature involving a series of cis-trans amide isomeri-
zations.11,12 Recently,
a tripeptide fragment, Ac-Ala-Ala-
Ala-NH2, was folded into a b-turn through encapsulation by a
porphyrin-assembled synthetic host.13 Our strategy to improve
the yield of cyclization has been to maintain such a turn or partial
turn of a linear peptide in a nano sized cavity as a template.
Molecularly imprinted polymers (MIPs)14 are seldom pre-
pared at high temperatures (4 110 1C). At high temperatures,
a linear peptide is prone to adopt a turn conformation owing
to its higher propensity to form cis-amide conformations.
Department of Chemistry, National Dong-Hwa University, Hualien,
Taiwan. E-mail: dftai@mail.ndhu.edu.tw; Fax: 886 38633570;
Tel: 886 38633579
w Electronic supplementary information (ESI) available: Experimental
section. See DOI: 10.1039/b813439a
Scheme 1 Schematic representation of tetrapeptide induced linear/
turn cavities and their potential in cyclization reactions.
ꢀc
This journal is The Royal Society of Chemistry 2008
5598 | Chem. Commun., 2008, 5598–5600

View Article Online
tetrapeptides were used as the template to prepare imprinted
polymers with turn cavities. The amount of effective cavities
generated was around 60% of the template used. The effective
cavities contained both turn cavities and linear cavities.
The second stage is the binding of the linear peptide to their
turn cavities. The orientation and hydrophobicity of the side
chain on the tetrapeptide can have stabilizing interactions with
the MIP to form MIP–peptide complexes. The MIP cage can
accommodate tetrapeptides using solvents such as water,
acetonitrile, benzene, toluene or p-xylene, but it also shows
an unexpected selectivity. Generally, nonpolar solvents per-
formed better at reflux and toluene was found to be the best
solvent for adsorption. Binding of appropriate guests occurs
best in those solvents that cannot fit well inside the cavities,
and refluxing toluene is appropriate to force most tetra-
peptides in but not out. In general, the b-turn conformation
was not favored in water. Here, it was observed that despite
the presence of the linear conformer, the turn conformation
dominated because of efficient host–guest interaction at high
temperature. After refluxing in toluene for 6 h, 60–70% of
linear tetrapeptides were bound to the MIP–cellulose fiber, as
determined by HPLC analysis (Table 2). Rinsing with water
removed the nonspecifically bound tetrapeptides. Further
evidence was obtained using Raman spectroscopy. Captured
turn conformer within the cavity is revealed by the Raman
band1 of amide I for b-turn within B1690 to B1660 cmꢁ1. In
particular, Gly-Gly-Gly-Gly shows two large peaks (1670 and
1679 cmꢁ1) that indicated that the ratio of turn conformers
was increased inside the cavity as compared to their free form
in the solution. In contrast, no increase was observed upon
attaching Pro-Pro-Pro-Pro to its MIPs. Gly-Gly-Pro-Gly, a
b-turn peptide,19 was also examined for comparison (ESIw).
The third stage is the cyclization of the peptide on the surface.
Most of the substrates bound to the MIP-cellulose possess a turn
Scheme 2 Possible participation of linear/turn conformers with
linear/turn cavities for tetrapeptide cyclization.
peptide contains both favorable (turns) and unfavorable (linear)
conformations for cyclization, and the interconversion barriers
from linear to turn conformations are different for each peptide.
Indeed, the polymerization temperature played a crucial role in
the results from the materials synthesized.18 The MIP can
‘‘lock’’ the conformer during polymerization in a manner
similar to memory effects. The higher the curing temperature,
the higher the fraction of turn cavities generated.
The solubility was also important. The tetrapeptide was
mixed in H2O and ethylene glycol (1 : 1) to dissolve all the
components and allow the formation of MIP-cellulose at high
temperature. Some CTPs exhibit one conformation in non-
polar solvents or when a cis-amide is present, but CTPs usually
exist as multiple conformations in polar solvents.11 This
property offers a unique way to control the system. During
polymerization, the equilibrium can be shifted toward the turn
conformer using solvent and temperature effects.18 Generation
of more turn cavities would assure the binding of the turn
conformer and reduce the binding of the linear form. Peptide
bonds, which originally contain less turn character and pre-
ferentially adopt a trans conformation in solution, could be
transformed to a cis conformation at high temperature and
incorporated into the turn cavities. As shown in Table 1, linear
Table 2 Binding of tetrapeptides at high-temperaturea
Gly-Gly- Phe-Phe- Leu-Pro- Pro-Val- Pro-Pro-
Leu-Pro Pro-Val Pro-Pro
Substrate (%) Gly-Gly Phe-Phe
Total
65.9
71.6
16.0
55.6
46.2
21.3
1
70.0
24.1
45.9
24.3
15.9
0
67.5
20.9
46.6
27.8
17.4
0
67.2
19.7
47.5
0
bindingb
Nonspecific 15.9
Table 1 MIPs formed at high-temperaturea
bindingc
Specific
50.0
Template
(%)
Gly-Gly- Phe-Phe- Leu-Pro- Pro-Val- Pro-Pro-
Gly-Gly Phe-Phe Leu-Pro Pro-Val Pro-Pro
bindingd
The yields of 19.6
CTPe
Template in 26.9
solutionb
24.0
65.4
10.6
26.9
62.2
10.9
28.5
58.1
13.4
30.6
56.6
12.8
Total binding 21.3
of NIPf
The yields of
CTPf
13.1
0
Effective
cavitiesc
Template
embeddedd
a
58.1
0
15.0
a
Binding was performed using tetrapeptide (3.2 mmol) in toluene
(1 mL) at reflux for 6 h. The amount of total binding was measured
b
MIP was prepared by heating at 140 1C for 25 min on MPS-cellulose
fiber with 120 mL of H2O and ethylene glycol (1 : 1), containing
template peptide (3.2 mmol)/acrylamide/ATA/EBAA at 8 : 1 : 4 : 15
by HPLC, using N-carbobenzyloxythreoine methyl ester (1 mM) as the
internal standard. Yields were measured based on the amount of
b
c
molar ratio. The amount of template in solution was measured by
HPLC, using N-carbobenzyloxythreoine methyl ester (1 mM) as the
tetrapeptide used (3.2 mmol). The amount of tetrapeptide rinsed
out with water from the MIP. The specific binding was calculated
by deducting the amount of tetrapeptides nonspecifically bound to the
d
c
internal standard. The effective cavities were measured using HPLC,
d
based on the amount of tetrapeptide washed out from the MIP. The
e
cavities from the amount of total binding. The MIP–peptide
complex reacted in DCM–DMF (3 : 1) with HATU–HOAt at room
f
temperature for 6 h. Nonimprinted polymers (NIP) were prepared
under the same conditions without the template.
embedded template was calculated by deducting the amount of
template in solution and the amount of tetrapeptide in the cavities
from the amount of template used (3.2 mmol).
ꢀc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 5598–5600 | 5599

View Article Online
Table 3 Summary of the results of difficult CTPs synthesesa
Gly-Gly-
Gly-Gly
Phe-Phe-
Phe-Phe
Leu-Pro-
Leu-Pro
Pro-Val-
Pro-Val
Pro-Pro-
Pro-Pro
Substrate yield/conv. (%)
By MIPs formed at high-temperature
By MIPs formed and binding at high-temperature
By MIPs formed by UV irradiation
By CTP-templated MIPe
14.1b (25.6)c
19.6b (39.4)c
0b (0)c
40.5b (62.9)c
46.2b (83.1)c
37.1b (70.2)c
49.3b (93.9)c
54.4b (90.2)c
116
—
24.3b (53.1)c
—
—
—
56
—
27.8b (59.7)c
—
—
—
107
0b (0c
0b (0)c
0 (0)c
—
27.7b (87.8)c
42.8b (84.5)c
4.417
By CTP templated MIPe with binding at high-temperature
Lit. yield (%)
—
017
a
The MIP-peptide complex reacted in DCM–DMF (3 : 1) with HATU–HOAt at room temperature for 6 h. Yields were measured using
c
b
HPLC. Yields were measured based on the amount of CTP, divided by the amount of tetrapeptide used for binding (3.2 mmol). Conversions
were measured based on the amount of CTP, divided by the amount of tetrapeptide specifically attached to the MIPs (Table 2). The MIPs were
d
prepared by irradiation of MPS-cellulose fiber (28.8 mg) at 350 nm with 100 mL of the mixed solvent (CH3CN–H2O = 1 : 1) containing template
peptide–acrylamide–ATA–EBAA at 8 : 1 : 4 : 15 molar ratio. The MIPs were prepared by the above method using CTP as template.
e
conformation reducing the number of undesired conformers. The
poor reversibility and high hydrophobicity of the turn cavities
exposes the tetrapeptide amino and carboxylic groups to the
coupling reagents, which can significantly improve the macro-
cyclizations. The linear tetrapeptides that are in contact with the
turn cavities are also shielded from each other and this protection
enhances intramolecular reactions with increased rates and
shorter reaction times. The activated ester could be held in
complexes hidden from hydrolysis and side reactions, so as to
enhance the yield of cyclization. The coupling reagent adopted
for cyclization was O-(7-azabenzotriazol-1-yl)-l,l,3,3-tetramethyl-
uronium hexafluorophosphate (HATU)/1-hydroxy-7-azabenzo-
triazole (HOAt).20 The solvent system was modified by
increasing the ratio of dichloromethane (DCM) to N-dimethyl-
formamide (DMF). As shown in Table 3, DCM–DMF used at a
3 : 1 ratio resulted in higher yields of CTPs. The MIPs effectively
bring the termini of tetrapeptides together for cyclization, over-
coming strain related resistance. In comparison with UV irradia-
tion, the CTP synthesis of Gly-Gly-Gly-Gly was improved from
none to near 20%. However, it was still not possible to achieve
cyclization for Pro-Pro-Pro-Pro to cyclo-(Pro-Pro-Pro-Pro). The
energy difference between the turn conformer and linear con-
former must be too large. This is consistent with Raman spectra.
Fortunately, no heterochiral CTP was generated, indicating no
racemization occurred during the third-stage process. For those
linear peptides with a minimal amount of turn conformation, the
yield for the intramolecular peptide bond formation was
improved using this procedure. With cyclo-(Pro-Leu-Pro-Leu),
the 13C NMR spectrum has nine peaks in the alkyl region; while
In conclusion, this study indicates the possibility of
modulating the MIP systems by raising the temperature
regime and through induced fit molecular recognition for
CTP synthesis.
The present work is partially supported by a Grant from the
National Science Council of Taiwan.
Notes and references
1 G. D. Rose, L. M. Gierasch and J. A. Smith, Adv. Protein Chem.,
1985, 37, 1.
2 E. Vass, M. Hollosi, F. Besson and R. Buchet, Chem. Rev., 2003,
103, 1917.
3 R. M. Freidinger, J. Med. Chem., 2003, 46, 5553.
4 W. Kabsch and C. Sander, Biopolymers, 1983, 22, 2577; C. M.
Wilmot and J. M. Thornton, J. Mol. Biol., 1988, 203, 221.
5 J. S. Davies, J. Pept. Sci., 2003, 9, 471.
6 J. M. Aracil, A. Badre, M. Fadli, G. Jeanty, B. Banaigs, C.
Francisco, F. Lafargue, A. Heitz and A. Aumelas, Tetrahedron
Lett., 1991, 32, 2609; U. Schmidt and J. Langner, J. Pept. Res.,
1997, 49, 67; M. El Haddadi, F. Cavelier, E. Vives, A. Azmani, J.
Verducci and J. Martinez, J. Pept. Sci., 2000, 6, 560.
7 T. Ueda, Acta Crystallogr., 1984, C40, 111; T. Ueda, I. Sada, T.
Kato and N. Izumiya, Int. J. Pept. Protein Res., 1985, 25, 475.
8 R. Poteau and G. Trinquier, J. Am. Chem. Soc., 2005, 127, 13875.
9 K. Ueno and T. Shimizu, Biopolymers, 1983, 22, 633; W. D.
Meutermans, G. T. Bourne, S. W. Golding, D. A. Horton, M.
R. Campitelli, D. Craik, M. Scanlon and M. L. Smythe, Org. Lett.,
2003, 5, 2711.
10 K. Burgess, Acc. Chem. Res., 2001, 34, 826.
11 M. MacDonald and J. Aube, Curr. Org. Chem., 2001, 5, 417; M. P.
´
Glenn, M. J. Kelso, J. D. A. Tyndall and D. P. Fairlie, J. Am.
Chem. Soc., 2003, 125, 640; Y. Che and G. R. Marshall, J. Med.
Chem., 2006, 49, 111.
1
12 U. Link, W. Mastle and M. Rothe, Int. J. Pept. Protein Res., 1993,
42, 475.
13 S. Tashiro, M. Kobayashi and M. Fujita, J. Am. Chem. Soc., 2006,
128, 9280.
14 K. Haupt and K. Mosbach, Chem. Rev., 2000, 100, 2495; M. J.
Whitcombe, C. Alexander and E. N. Vulfson, Synlett, 2000, 5, 911;
G. Wulff, Chem. Rev., 2002, 102, 1.
15 F. Cavelier-Frontin, S. Achmad, J. Verducci, R. Jacquier and G.
Pepe, THEOCHEM, 1993, 286, 125.
´
16 K. Titlestad, Acta Chem. Scand., Ser. B, 1977, 31, 641.
17 M. Abdelmouleh, S. Boufi, A. ben Salah, M. N. Belgacem and A.
Gandini, Langmuir, 2002, 18, 3203.
18 S. A. Piletsky, E. V. Piletska, K. Karim, K. W. Freebairn, C. H.
Legge and A. P. F. Turner, Macromolecules, 2002, 35, 7499.
19 P. Lagant, G. Vergoten, G. Fleury and M. H. Loucheux-Lefebvre,
Eur. J. Biochem., 1984, 139, 137.
the broad H NMR spectrum more closely resembled that of
Haddadi and Cavelier.6 As for cyclo-(Pro-Val-Pro-Val), only one
set of Pro-Val signals was observed and it did not resemble the
spectrum reported by Aracil and Francisco.6
In some cases where the ratio of turn conformers was still
low at 140 1C, the fabrication method was modified by using
CTPs as the templates. As CTPs were available, using them as
the templates during UV irradiation ensured the formation of
the b-turn cavities on MIPs. The yield of cyclization was
improved to 27.7% for the less hydrophobic tetrapeptide:
Gly-Gly-Gly-Gly, and even higher (42.8%) using capture at
high temperature. The yield of cyclo-(Phe-Phe-Phe-Phe) was
also improved to 54.4%. It appears that 85–90% of the
tetrapeptides attached to the MIPs were converted into CTPs.
20 A. Ehrlich, H. U. Heyne, R. Winter, M. Beyermann, H. Haber, L.
A. Carpino and M. Bienert, J. Org. Chem., 1996, 61, 8831.
ꢀc
This journal is The Royal Society of Chemistry 2008
5600 | Chem. Commun., 2008, 5598–5600