Oxazole-based peptide macrocycles: A new class of G-quadruplex binding ligands

Article abstract of DOI:10.1021/ja064713e

Herein we report on the synthesis and DNA binding properties of a new class of water soluble oxazole-based peptide macrocycles that bind selectively to quadruplex DNA, with no detectable binding to duplex DNA. We have recently identified one quadruplex in

Full text of DOI:10.1021/ja064713e

Published on Web 09/30/2006
Oxazole-Based Peptide Macrocycles: A New Class of G-Quadruplex Binding
Ligands
Katja Jantos, Raphae¨l Rodriguez, Sylvain Ladame, Pravin S. Shirude, and
Shankar Balasubramanian*
The UniVersity Chemical Laboratory, UniVersity of Cambridge, Lensfield Road, Cambridge CB2 1EW, U.K.
Received July 12, 2006; E-mail: sb10031@cam.ac.uk
During the past decade, there has been growing interest in the
structure, recognition, and function of DNA G-quadruplexes. The
best-studied example is the human telomeric quadruplex, which is
associated with the inhibition of telomere elongation by telomerase,
an enzyme up-regulated in cancer cells.1 Furthermore, a large
number of other putative quadruplex forming sequences have been
identified in the human genome.2 We have recently identified a
quadruplex forming sequence in the promoter of the c-kit oncogene
Figure 1. Structures of oxazole-based peptide macrocycles 1a-d.
with potential to regulate the expression of the c-kit gene at the
transcription level.3 We have demonstrated that this quadruplex
predominantly exists in a parallel conformation in vitro under near
physiological conditions. In contrast, the human telomeric quadru-
plex has been shown to adopt different conformations in the
presence of either sodium or potassium. Structural studies have
reported that the human telomeric quadruplex can exist as a
parallel,4 antiparallel,5 or even a mixed-type hybrid structure.6
Hitherto, chemists have been addressing the design of small
molecule ligands that selectively recognize quadruplex DNA in
preference to double-stranded DNA. However, owing to the
increasing number of quadruplexes being identified in the human
genome, an emerging goal is to elucidate quadruplex recognition
by small molecules in order to design ligands that can discriminate
between quadruplex sequences and/or quadruplex folds. A number
of other classes of macrocyclic molecules have been reported in
the literature that have shown good potential to bind quadruplexes.7
Herein, we report on the quadruplex binding properties of
oxazole-based peptide macrocycles of general structure 1 (Figure
1). In particular, we have studied their interaction with the c-kit
quadruplex in comparison to the human telomeric quadruplex.
Structural studies on related, biologically active cyclopeptides
using molecular modeling and X-ray crystallography studies
concluded that macrocycles of structure 1 are generally planar.8
This information together with a consideration of the size comple-
mentarity led us to reason that such structures have potential to
recognize quadruplexes via interactions with the top tetrad. The
macrocycles 1a-d are chiral and functionalized with alkylamines
to provide water solubility and potentially also participate in
favorable interactions with the DNA scaffold. As part of the study
we have investigated the influence of stereochemical variations as
well as the importance of the lateral side chains on quadruplex
recognition and stabilization.
double stranded DNA (ds DNA) comprising the oligonucleotide
d(biotin-[G2CATAGTGCGTG3CGT2AGC]) hybridized with its
complementary sequence. The results are summarized in Table 1.
It is noteworthy that while ligands 1a-d all bind quadruplex
DNA, none of the macrocyles showed detectable binding to double-
stranded DNA, even at high ligand concentration (300 µM), thus
this class of molecules is highly selective for quadruplex as
compared to duplex. The enantiomers 1a,b were found to bind
hTelo with comparable dissociation constants (KD) of 15 and 12
µM, respectively. The SPR data showed a 1:1 stoichiometry for
ligand-quadruplex binding, which is suggestive of a specific
ligand-quadruplex interaction with a single guanine tetrad as
proposed for other classes of quadruplex binding ligands.7 1a,b each
contain three butylamine side chains exclusively on one face of
the macrocycle, leaving the opposite face unhindered and therefore
accessible for π-π interactions with the top tetrad of the quadru-
plex. The comparable binding affinities for 1a,b suggest that an
inversion of all three stereocenters does not affect homochiral
quadruplex structure recognition. However, inversion of only one
stereocenter presents side chains on both faces of the macrocycle,
which led to a 4- to 6-fold drop in binding affinity for diastereo-
isomer 1c, as compared to 1a and 1b, for hTelo and c-kit
quadruplexes (Table 1). This result is also consistent with tetrad
based quadruplex recognition by 1a,b. Moreover, it is noteworthy
that for the macrocycles 1a,b a 3-fold selectivity for c-kit vs hTelo
was observed.
Primary amines are protonated at physiological pH and may be
involved in stabilizing interactions with the grooves and loops of
the quadruplex as well as the negatively charged phosphate
backbone. To evaluate the role of the amine side chains, we
prepared macrocycle 1d10 with shorter methylamine side chains.
Compared to 1a, macrocycle 1d showed a reduced binding affinity
for hTelo (4-fold), but an unchanged equilibrium binding to c-kit.
This immediately suggests a different potential for the side chains
to recognize each quadruplex.
To strengthen and complement the SPR binding studies, we
examined the ability of the macrocycles 1a-d to stabilize G-
quadruplex DNA by a fluorescence resonance energy transfer
(FRET) assay11 (Figure 2) using dual-labeled hTelo and c-kit
quadruplexes, in addition to a double stranded DNA (see Supporting
Information). It is noteworthy that FRET melting analysis measures
the ligand-induced stabilization of a folded quadruplex, rather than
Macrocycles of general structure 1 were synthesized by assembly
of three amino acid building blocks (2, Figure 1) (see Supporting
Information). Their potential as G-quadruplex ligands was first
investigated using surface plasmon resonance (SPR), which has
proven to be a valuable method to examine DNA-small molecule
interactions.7c-d,9 The experiments were performed using three
different immobilized DNA targets: two quadruplexes formed from
either the human telomeric sequence d(biotin-[GT2A(G3T2A)4G2])
(hTelo) or d(biotin-[C3G3CG3CGCGAG3AG4AG2]) (c-kit) and a
9
13662
J. AM. CHEM. SOC. 2006, 128, 13662-13663
10.1021/ja064713e CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Dissociation Constants (KD) Determined by SPR and
Stabilization Potential Determined by FRET Melting
significant increase of the band at 265 nm characteristic of a parallel
conformation. This effect was much more pronounced for 1a,b,
which have been found to be the tightest binders by SPR, than for
the weaker binders 1c,d. Addition of the macrocycle to prefolded
hTelo quadruplex did not induce significant change in quadruplex
conformation, even after 3 h, suggesting that conformational
interconversion is slow. This indicates a thermodynamic preference
of macrocycles 1a-d for binding to a parallel G-quadruplex
structure as compared to an antiparallel conformation. The c-kit
quadruplex already exists as a predominantly parallel structure in
K+-containing buffer and showed no significant change in the CD
spectrum upon inclusion of ligands 1a-d (Figure 3, right). These
results are consistent with the preference of the macrocycles 1 for
the predominantly parallel c-kit quadruplex3 as compared to hTelo
which is highly polymorphic. This new class of quadruplex binding
ligands represents a versatile and promising scaffold. Further
optimization of such structures and an evaluation of biological
activity are in progress.
DNA
1a
1b
1c
1d
K
D [µM]
hTelo
c-kit
hTelo
c-kit
15 ( 1
5 ( 2
5.2
10.8
0.0
12 ( 1
4 ( 2
6.4
8.7
0.0
56 ( 11
31 ( 5
4.5
6.1
0.0
53 ( 8
8 ( 3
0.5
1.2
0.7
∆Tm
at 1 µM
[°C]
ds DNA
Figure 2. (left) SPR binding curve for macrocycle 1a to hTelo (9), c-kit
(2), and ds DNA (f); running-buffer, 50 mM Tris‚HCl pH 7.4, 100 mM
KCl. (right) FRET assay for hTelo (9), c-kit (2), and ds DNA (f) in the
presence of 1a; buffer, 60 mM potassium cacodylate pH 7.4.
Acknowledgment. We thank the Gates Cambridge Trust for a
studentship (K.J.) and the Cancer Research U.K. for program
funding (R.R. and S.L.) and the BBSRC for funding (P.S.). S.B. is
a BBSRC Career Development Research Fellow. We also thank
Dr. A. Bugaut for careful reading of the manuscript.
Supporting Information Available: Experimental procedures for
the synthesis of the macrocycles 1, SPR binding curves, experimental
details for CD and FRET. This material is available free of charge via
the Internet at http://pubs.acs.org.
References
Figure 3. CD spectra of hTelo (left) and c-kit (right) in K+ containing
buffer (50 mM Tris‚HCl pH 7.4, 100 mM KCl): without ligand (black);
folded in the presence of 1a (red), 1b (blue), 1c (green), and 1d (cyan).
(1) Zahler, A. M.; Williamson, J. R.; Cech, T. R.; Prescott, D. M. Nature
1991, 350, 717.
(2) (a) Huppert, J. L.; Balasubramanian, S. Nucleic Acids Res. 2005, 33, 2908.
(b) Todd, A. K.; Johnston, M.; Neidle, S. Nucleic Acids Res. 2005, 33,
2901.
equilibrium binding. For all ligands 1a-d no stabilization of duplex
DNA was observed, which is consistent with the SPR observations
and confirms the high selectivity for G-quadruplex DNA. For
ligands 1a-c the trend for G-quadruplex stabilization effects were
consistent with the KD values obtained by SPR, the tightest binder
inducing the strongest thermal stabilization (Table 1). However,
macrocycle 1d, which binds to hTelo and c-kit with affinities
comparable to that of 1c and 1a, did not induce a significant thermal
stabilization, even at high ligand concentration (up to 50 µM). This
suggests that while 1d binds to c-kit with an affinity comparable
to those of ligands 1a-c, its mode of binding does not stabilize
(or destabilize) the folded c-kit quadruplex structure.
To provide further insight into the mode of binding we have
used circular dichroism (CD) spectroscopy to elucidate ligand
effects on the folded conformations of hTelo. In the absence of the
ligands, the hTelo oligonucleotide coexists in K+-containing buffer
as a mixture of parallel and antiparallel G-quadruplex conformations
exhibiting two maxima at 265 and 291 nm as well as a minimum
at 243 nm (Figure 3, left).12 When hTelo is folded in the presence
of a 10-fold excess of the respective macrocycle 1, a loss of the
antiparallel signal (291 nm) was observed simultaneously with a
(3) Fernando, H.; Reszka, A. P.; Huppert, J.; Ladame, S.; Rankin, S.;
Venkitaraman, A. R.; Neidle, S.; Balasubramanian, S. Biochemistry 2006,
45, 7854.
(4) Wang, Y.; Patel, D. J. J. Mol. Biol. 1993, 234, 1171.
(5) Parkinson, G. N.; Lee, M. P. H.; Neidle, S. Nature (London) 2002, 417,
876.
(6) Ambrus, A.; Chen, D.; Dai, J.; Bialis, T.; Jones, R. A.; Yang, D. Nucleic
Acids Res. 2006, 34, 2723.
(7) For recent examples see (a) Dixon, I. M.; Lopez, F.; Este`ve, J.-P.; Tejera,
A. M.; Blasco, M. A.; Pratviel, G.; Meunier, B. ChemBioChem 2005, 6,
123. (b) Seenisamy, J.; Bashyan, S.; Gokhale, V.; Vankayalapati, H.; Sun,
D.; Siddiqui-Jain, A.; Steiner, A. N.; Shin-ya, K.; White, E.; Wilson, W.
D.; Hurley, L. H. J. Am. Chem. Soc. 2005, 127, 2944. (c) Teulade-Fichou,
M.-P.; Carrasco, C.; Guittat, L.; Bailly, C.; Alberti, P.; Mergny, J.-L.;
David, A.; Lehn, J.-M.; Wilson, W. D. J. Am. Chem. Soc. 2003, 125,
4732. (d) Schouten, J. A.; Ladame, S.; Mason, S. J.; Cooper, M. A.;
Balasubramanian, S. J. Am. Chem. Soc. 2003, 125, 5594.
(8) Lucke, A. J.; Tyndall, J. D. A.; Singh, Y.; Fairlie, D. P. J. Mol. Graphics
Modell. 2003, 21, 341.
(9) (a) Moore, M. J. B.; Schultes, C. M.; Cuesta, J.; Cuenca, F.; Gunaratnam,
M.; Tanious, F. A.; Wilson, W. D.; Neidle, S. J. Med. Chem. 2006, 49,
582. (b) Ladame, S.; Schouten, J. A.; Stuart, J.; Roldan, J.; Neidle S.;
Balasubramanian, S. Org. Biomol. Chem. 2004, 2, 2925.
(10) Mink, D.; Mecozzi, S.; Rebek, J., Jr. Tetrahedron Lett. 1998, 39, 5709.
(11) Mergny, J.-L.; Maurizot, J.-C. ChemBioChem 2001, 2, 124.
(12) Rezler, E. M.; Seenisamy, J.; Bashyam, S.; Kim, M.-Y.; White, E.; Wilson,
W. D.; Hurley, L. H. J. Am. Chem. Soc. 2005, 127, 9439.
JA064713E
9
J. AM. CHEM. SOC. VOL. 128, NO. 42, 2006 13663