Neomycin-capped aromatic platforms: Quadruplex DNA recognition and telomerase inhibition

Article abstract of DOI:10.1039/b516378a

A series of aminoglycoside-capped macrocyclic structures 9-12 has been prepared using intramolecular bis-tethering of neomycin on three aromatic platforms (phenanthroline, acridine, quinacridine). Based on NMR and calculations studies, it was found that t

Full text of DOI:10.1039/b516378a

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Neomycin-capped aromatic platforms: quadruplex DNA recognition and
telomerase inhibition
Markus Kaiser,†a Anne De Cian,b Matthieu Sainlos,a Christian Renner,c Jean-Louis Mergnyb and
Marie-Paule Teulade-Fichou*a
Received 17th November 2005, Accepted 4th January 2006
First published as an Advance Article on the web 25th January 2006
DOI: 10.1039/b516378a
A series of aminoglycoside-capped macrocyclic structures 9–12 has been prepared using intramolecular
bis-tethering of neomycin on three aromatic platforms (phenanthroline, acridine, quinacridine). Based
on NMR and calculations studies, it was found that the cyclic compounds adopt a highly flexible
structure without conformational restriction of the aminoglycoside moiety. FRET-melting stabilization
measurements showed that the series displays moderate to high affinity for the G4-conformation of
human telomeric repeats, this effect being correlated with the size of the aromatic moiety. In addition, a
FRET competition assay evidenced the poor binding ability of all macrocycles for duplex DNA and a
clear binding preference for loop-containing intramolecular G4 structures compared to tetramolecular
parallel G4 DNA. Finally, TRAP experiments demonstrated that the best G4-binder (quinacridine 11)
is also a potent and selective telomerase inhibitor with an IC50 in the submicromolar range (200 nM).
available. In spite of several structural studies which provided valu-
able information about the interactions of small molecules with
Introduction
loops,10a–f a deeper understanding is still needed, and in particular
little is known about groove occupancy.10g,h Additionally, folded
quadruplexes show a highly dynamic and polymorphic structure,11
further complicating the drug design. As a consequence, the search
for G4-binders of high specificity has been proved difficult and
remains a challenging task.
Inhibition of telomerase or alteration of the telomere state are
both valuable concepts for inducing senescence and apoptosis
in cancer cells.1 A simultaneous targeting of telomeres and
the nucleoenzyme telomerase was recently demonstrated as a
promising approach for limiting cancer cell proliferation.2 The two
processes are intimately connected: modulation of the telomeric
structure impairs telomerase binding, resulting in inactivation of
both the catalytic and the maintenance activities of the enzyme.3
Thus, compounds able to disrupt the telomeric structure are
particularly interesting as potential telomere function modula-
tors and telomerase inhibitors.4 Structural perturbation of the
telomeres can be reached by inducing a folding of the G-rich
telomeric 3ꢀ overhang into a quadruplex conformation, which
can be achieved by binding of a highly selective G-quadruplex
ligand.4,5 Therefore, a reasonable strategy for identifying novel
anticancer drugs relies on the discovery of strong and selective
G4 ligands with consequent telomerase inhibition.6 In the past
decade, thousands of compounds were screened according to this
approach and three main classes of ligands could be established:
i) fused polycyclic intercalators,7 ii) macrocyclic compounds of
either natural or synthetic origin,8 and iii) polyaromatic unfused
systems.6a,9 However, structural data on the molecular interactions
between the ligands and their G4-DNA target are still scarce and
no general concept for the design of highly selective binders is
In the course of our studies on the design of G4-binders, we
developed the pentacyclic crescent-shaped quinacridine motif that
shows a high affinity for quadruplex DNA mainly due to strong
stacking interactions with G-quartets.7c,d Moreover, a dimeric
macrocyclic bisquinacridine was shown to elicit a high preference
for quadruplex over duplex DNA.8b Macrocyclic scaffolds are
particularly attractive for designing G4 ligands, as they show a
preferential binding to “non-standard” DNA conformations, due
to their sterically difficult intercalation between the base pairs
of “standard” B-DNA.12 In contrast, the external G-quartets
of quadruplexes constitute accessible planar sites of large area
which can accommodate large-sized molecules. This is remarkably
illustrated by the exquisite G4-binding specificity of the natural
compound telomestatin which is composed of seven oxazole rings
and a dihydro-thiazole moiety combined in a cyclic scaffold.8a,13
Aminoglycosides are natural antibiotics which have been widely
used to achieve selective recognition of various loop or bulge-
containing RNA structures.14 Despite exhibiting a low affinity
towards G4-quadruplex structures per se,15 aminoglycosides pos-
sess several ammonium centers able to establish multiple salt
bridges and H-bonding contacts with nucleic acids. In particular
neomycin has been shown to exhibit a high selectivity for DNA
triplexes.16 In addition, the 1,3-hydroxylamine motif commonly
found in aminoglycosides has been identified as a recognition
motif for the complexation of phosphate groups and of the
Hoogsteen face of guanine via a bidentate H-bonding/electrostatic
interaction.17 Altogether, and in view of the presence of short
loops in intramolecular G4 structures, we speculated that an
aLaboratoire de Chimie des Interactions Mole´culaires, Colle`ge de France,
CNRS UPR 285, 11, place Marcelin Berthelot, 75005 Paris, France. E-mail:
mp.teulade-fichou@college-de-france.fr; Fax: ++33 (0) 1 44 27 13 56
bLaboratoire de Biophysique, Muse´um National d’Histoire Naturelle USM
503, INSERM U 565, CNRS UMR 5153, 43 rue Cuvier, 75005 Paris,
France
cSchool of Biomedical and Natural Sciences, Nottingham Trent University,
Clifton Lane, Nottingham, United Kingdom NG11 8NS
† Current address: Chemical Genomics Centre, Otto-Hahn-Str. 15, 44227
Dortmund, Germany.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1049–1057 | 1049
©

View Online
appropriate derivatization of aminoglycosides with an intercalator
system could lead to high-affinity ligands through simultaneous
targeting of the G-quartet surface and of the loops phosphate or
residues. Along this line, we decided to assemble the two motifs in
a cyclic scaffold in the hope of compromising the duplex binding.
Hence, capping of various aromatic platforms with a neomycin
moiety has been investigated as an approach to the design of
cyclic “dome-shaped” scaffolds suitable for adapting the topology
of loop-containing quadruplexes.
be achieved via a benzoylation–Teoc-protection–hydrogenation
sequence (Scheme 1).19 Derivatization of the neomycin intermedi-
ate 4 with a Boc-protected lysine or aminocaproic acid building
block, led after deprotection to elongated neomycin diamine
derivatives 7 and 8, respectively. These were then condensed with
three different dicarboxaldehyde aromatic platforms (acridine,
quinacridine, phenanthroline) by reductive amination, yielding
the four macrocycles 9–12. Although Schiff base macrocyclisation
can lead to the formation of either [1 + 1] or [2 + 2] coupling
products,20 in our case the predominance of a [1 + 1] condensation
of the partners was evidenced by HPLC analysis, and appears
independent of the size of the aromatic system. A reasonable
explanation for the predominance of the intramolecular cy-
clisation is that pre-organization of the derivatized neomycin
moiety occurs, to a certain extent, during tethering. Although
the aminoglycoside scaffold is flexible, it is prone to adopt defined
conformations upon limitation of the rotation of the sugar units.
Therefore, it can be hypothesized that the aminoglycoside moi-
ety, after the first amination, adopts a conformation placing
the reactive groups favourably for internal ring closure. Hence,
entropic contribution due to optimal spatial fitting of the amino
and aldehyde functions, along with high dilution conditions,
could both participate to the predominance of the [1 + 1]
coupling.
Results and discussion
Synthesis
Condensation of a dialdehyde with a diamine is a well-known
strategy for constructing macrocyclic structures.18 This approach
has proven to be very high-yielding if performed under moderate to
high dilution conditions for minimising unwanted polymerisation.
The aminoglycoside neomycin features two reactive primary
amino groups linked to a methylene carbon whereas its other
amines are directly linked to sugar units and are thus sterically less
accessible for derivatization. Based on this structural peculiarity,
regioselective protection of the amino groups of neomycin can
Scheme 1 Reagents and conditions: (a) N-(Z)-5-norbornene-2,3-dicarboximide, TEA, DMSO–H2O (10 : 1), rt, 12 h, 51%; (b) Teoc-p-nitrophenyl
carbonate, TEA, dioxane–H2O (3 : 1), 55 ◦C, 48 h, 87%; (c) H2, Pd/C (10%), MeOH–H2O (9 : 1), rt, 2 h, 81%; (d) EDCI/HOAt, TEA, DMF
and Na-(Boc)–Lys(Z)–OH or N-(Z)-capronic acid, rt, 12 h, 92% for 5, 86% for 6; (e) H2, Pd/C (10%), MeOH–H2O (9 : 1), rt, 2 h, 92% for
7, 94% for 8; (f) i) TEA, DCM–MeOH (1 : 1) and 2,5-bis(dicarboxaldehyde)acridine or dibenzo[b,j][1,7]phenanthroline-2,10-dicarboxaldehyde or
2,9-bis(dicarboxaldehyde)-1,10-phenanthroline, rt, 4 d; ii) NaBH4, DCM–MeOH (1 : 1), rt, 2 h; iii) TFA–DCM (1 : 1), rt, 1 h, 29% for 9, 42% for 10, 37%
for 11, 12% for 12.
1050 | Org. Biomol. Chem., 2006, 4, 1049–1057
This journal is
The Royal Society of Chemistry 2006
©

View Online
NMR studies and calculations
In order to gain insight into the conformational flexibility of the
final capped macrocycles, the acridine derivative 9 was studied
by NMR. Two relatively well-defined conformers are found for
the pucker and relative orientations of the sugar rings (Fig. 1).
In one conformation the overall shape of the tetrasaccharide
resembles that of an “L” in which sugars B, C and D are
extended and form the long stem of the “L”. In the other
conformation rings A, B and C are arranged in a helix-like manner
leading to a V-shaped appearance. The aliphatic chain of Lys-
1 packs in both cases against sugar A, but the aromatic moiety
and Lys-2 exhibit variable orientations and seem quite flexible.
The fact that only one set of NMR resonances was observed
indicates fast conformational averaging on the NMR time-scale
of milliseconds. The two distinct conformations were detected in
the initial structure calculations with all NMR-derived distance
constraints. Because each conformation showed systematic and
characteristic violation of a few NMR distances, subsets of the
NMR constraints could be derived that corresponded exactly to
the two conformers. All NMR structures are quite compact with
the aromatic moiety fairly close to the sugar rings. In solution,
however, the molecules might also partly exist in more open forms
that cannot be detected by NMR due to the bias of the Overhauser
effect towards short distances and, therefore, compact structures.
Fig. 2 Concentration dependency of F21T stabilization (DT1/2 values, in
◦C). Data is presented on a semi-log scale for MMQ3 (crosses), 11 (orange
diamonds), 9 (green circles), 10 (red squares), 12 (blue triangles), MOP1
(light blue triangles) and neomycin (purple circles).
since the most active compound is the quinacridine derivative 11
(DT1/2 = 14.1
1.1 ◦C at 1 lM), whereas the acridine and
phenanthroline derivatives 10 and 12 show a much lower activity
(DT1/2 = 6.3 and 6.0 ◦C at 1 lM, respectively). Even at submi-
cromolar concentrations (0.2–0.5 lM), a significant stabilization
is found with compounds 9 and 11 (Fig. 2). As the three ligands
10–12 exhibit the same cationic charge, this result evidences the
strong p-stacking contribution to the binding since extension of
the surface area in contact is known to increase the attraction
between aromatic systems.22 The importance of electrostatics in
the stabilization of the G4 conformation is illustrated by the higher
effect of the acridine derivative 9 which bears two more cationic
amino groups than its analogue 10 [D(DT1/2
)
= + 4.8 ◦C].
9–10
Importantly, free neomycin has little or no effect on the melting of
F21T (DT1/2 = 2.1 ◦C at the highest drug concentration tested,
5 lM), which reveals that the high cationic charge (4+–6+) of
the free aminoglycoside23 is not sufficient to ensure an efficient
binding of the G4 structure. In addition, the acyclic tetraamino
phenanthroline derivative MOP1 (Fig. 3) has a very limited effect
on the melting (DT1/2 = 1.4 and 3.5 ◦C at 1 and 5 lM, respectively)
and the diamino acridine Mono◦A was previously found to be
completely ineffective (DT1/2 = 0 C at 1 lM).24
Fig. 1 (A) Superimposition of the two solvent conformations of 9 as
derived from NMR analysis (acridine system: light and dark green, lysine
linker: light and dark grey, neomycin moiety: blue, L shaped and red,
V-shaped). (B) Nomenclature of the structural elements of 9.
In conclusion, although the incorporation of aromatic systems
could have been expected to cause rigidity, rotation of the sugar
moieties in the final macrocycle seems not to be restrained by
the double tethering. This is likely attributable to the length and
flexibility of the lysine linkers. Thereby, compound 9 appears to be
a highly adaptable structure, and it is reasonable to assume that
macrocycles 10–12 adopt identical conformational behaviours.
Taken together, these data demonstrate that a synergistic effect
is obtained in combining the aromatic and aminoglycoside motifs
in the same scaffold, at least in the phenanthroline and acridine
series. Indeed, the additivity of the aromatic and electrostatic
contributions is not so obvious in the quinacridine series as the
acyclic control MMQ3 and ligand 11 display similar DT1/2 values
at all concentrations (Fig. 2). This might indicate that strong p-
stacking of the quinacridine moiety masks the other effects, or that
the relative weights of the various energetic contributions differ for
the two ligands.
FRET-melting stabilization assay
The interaction of the macrocycles with quadruplex DNA has
been investigated by a high-throughput FRET assay using a real-
time PCR apparatus and the doubly labelled F21T oligonucleotide
which mimics the human telomeric single-strand overhang.21
DT1/2 values were concentration dependent; as shown in Fig. 2.
Moderate to high stabilizations of F21T were observed in the
presence of macrocycles 9–12 at 0.2–5 lM (Fig. 2). Interestingly,
this effect is correlated with the size of the aromatic moiety
Competitive FRET-melting assay
A high binding selectivity is an essential criterion for the use of G4-
binders in complex environments. Hence the G4-selectivities of our
macrocycles were evaluated using a competitive FRET-melting
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1049–1057 | 1051
©

View Online
Fig. 3 Structures of the acyclic control compounds.
assay. To this end melting of F21T is performed in the presence
of various DNA competitors: a 26bp duplex (ds26) as described
previously,7c and two quadruplexes containing 5 guanine quartets:
a four-stranded parallel quadruplex [TG5T]4 and an intramolecu-
lar quadruplex (30AG) (Fig. 4).
As shown in Fig. 4A, the four macrocycles display a good
selectivity for the quadruplex over the control duplex, since
stabilization of F21T is only moderately affected at the highest
concentration of the duplex (10 lM). It is worth pointing out
that this is a stringent competition in terms of electrostatics and
quartet vs. base-pair binding, since ds26 is added in large excess
to F21T (respectively 15 and 50 molar eq.). On the other hand,
MMQ3 is displaced more easily from F21T, as seen from the
strong drop in DT1/2 observed even at the lowest concentration
of the duplex competitor (3 lM), reflecting likely intercalation of
the acyclic compound into the duplex DNA. In turn, the ability
of the macrocyclic series to distinguish between the quadruplex
and the duplex might be attributed to a poor insertion into
double-stranded DNA, supporting our initial design to achieve
quadruplex vs. duplex selectivity by cyclisation.
Fig. 4 Selectivity in FRET assay. Thermal denaturation of F21T was performed in the presence of the various compounds (1 lM) and in the presence of
various competitors: (A) double-stranded DNA ds26, (B) parallel tetramolecular G-quadruplex [TG5T]4, (C) intramolecular G-quadruplex 30AG. The
stabilization (DT1/2) is reported in ◦C for each compound and the following concentrations of competitors: no competitor; ds26 3 lM (dark blue), 10 lM
(light blue); (TG5T)4 1 lM (dark green), 3 lM (light green); 30AG 1 lM (dark red), 3 lM (light red). (D) Apparent melting temperature (T1/2) of F21T
in the presence of 1 lM of compound 11 with various concentrations of competitors (ds26 duplex: filled circles; 30AG: open cicles and [TG5T]4: crosses).
(E) Apparent melting temperature (T1/2) of F21T in the presence of 1 lM of compound MMQ3 with various concentrations of competitors (ds26 duplex:
filled circles; 30AG: open circles and [TG5T]4: crosses).
1052 | Org. Biomol. Chem., 2006, 4, 1049–1057
This journal is
The Royal Society of Chemistry 2006
©

View Online
In order to evaluate the possible interaction of the macrocyclic
compounds with the TTA loops of the G-quadruplexes, we com-
pared the decrease of DT1/2 obtained with the ds26 duplex and the
two types of G4 structure: the tetramolecular [TG5T]4 quadruplex,
which does not contain any loop, and the intramolecular G4
formed by 30AG. When pre-formed in favourable conditions,
[TG5T]4 cannot be denaturated in the conditions of the FRET
assay25 whereas ds26 and 30AG have a Tm of 70.5 ◦C and 74 ◦C,
respectively, under identical conditions (Na cacodylate 10 mM
pH 7.2, NaCl 100 mM) (data not shown). Melting of the 30AG is
dependent on the cation and the signature of the thermal difference
spectrum is consistent with a G-quadruplex.26 Despite the lack
of data on the exact structure or on the loop conformation,
it is likely that this oligonucleotide adopts a structure close to
that of F21T. Note that it is important to select competitors
which are thermally more stable that the F21T quadruplex;
otherwise they would be unfolded in the temperature range
chosen for the melting studies, and would therefore act as single-
stranded rather than double-stranded or quadruple-stranded
competitors.
contribute to the binding via direct contacts with the ligands or
indirectly via conformational constraint of the target. Quadru-
plex/duplex selectivity has been estimated using two independent
methods:
᭹ DT1/2 values in the presence of increasing concentrations
of duplexes or quadruplexes unambiguously indicate that an
intramolecular quadruplex (30AG) is a much better competitor
than a self-complementary duplex (ds26) or a parallel quadruplex
[TG5T]4 (Fig. 4D). The addition of 1.1 lM 30AG leads to a 50%
decrease in DT1/2, whereas >5 lM ds26 is required to obtain
the same DT1/2. These two-strand concentrations correspond to
5.5 lM quartets and 71 lM base pairs, respectively, demonstrating
that a much larger molar excess of base pairs is required to abolish
half of the ligand-induced stabilization of F21T. Even at 30 lM,
ds26 does not totally abolish stabilization by compound 11, in
constrast with 30AG, for which stabilization is completely lost at
10 lM.
᭹ An equilibrium dialysis experiment using a limited set of
compounds confirms a preference for 30AG over duplexes and
parallel quadruplexes (data not shown).
When the competition assay is carried out in the presence of 1
and 3 lM of the oligonucleotides forming quadruplex structures
(respectively 5 and 15 molar equivalents compared to F21T
but only 1 and 3 molar equivalents compared to the ligand),
(Fig. 4B,C) the drop in DT1/2 for the macrocyclic series is more
pronounced than for the duplex competitor. The comparison
of the destabilization obtained when adding 3 lM of each
competitor results in a decrease of 14 to 40% with ds26, of 53
to 73% with [TG5T]4 and the stabilization is fully abolished with
30AG. Interestingly, the results are somewhat different for MMQ3
as the addition of 3 lM of ds26 results in a decrease in DT1/2 of
In other words, both methods confirm that 11 has a preference
for intramolecular quadruplexes over tetramolecular quadru-
plexes and duplexes, in contrast with MMQ3 which exhibits little,
if any, selectivity (Fig. 4E).
Telomerase inhibition
Finally the compounds were examined for their ability to inhibit
human telomerase activity in vitro. Inhibition of telomerase was
measured by the TRAP assay with an internal standard to ensure
the validity of the test.27 The assay was performed at increasing
ligand concentrations; analysis by denaturating gel electrophoresis
is shown in Fig. 5A and quantitative analysis is provided in
Fig. 5B. As shown on the gel, the IC50 values of the macrocycles
rank in the low micromolar to submicromolar range, the most
potent effect being observed for compound 11 with an IC50 of
0.2 lM (Fig. 5C). In all cases, the internal control (ITAS) is only
affected at significantly higher concentrations compared to the
IC50, in agreement with the selective binding of the compounds.
Finally, it is worth noting that the IC50 values correlate well with
the G4-stabilization effects (DT1/2) determined by the FRET-
melting assay, again supporting the validity of the inhibition
strategy based on the conformational modification of the substrate
DNA.
◦
12.2 C (80%), whereas the addition of the same amount of G4
structure [TG5T]4 leads to a decrease of only 11 ◦C (72%). This is
again quite consistent with the role of the neomycin capping in the
preference of the macrocyclic series for quadruplex over duplex
DNA.
For [TG5T]4 at least 30% of the stabilization of F21T is
maintained in all cases, at the two concentrations of competitor
used. This result suggests that all the ligands tested exhibit a higher
affinity for the intramolecular quadruplex conformation of F21T
compared to the tetramolecular form of [TG5T]4. In contrast, when
adding the intramolecular G4 30AG, the competition pattern
shows a strong decrease in DT1/2 for all compounds and even
a complete loss of stabilization when 30AG is used at 3 lM
concentration (3 molar equivalent compared to ligand) (Fig. 4C).
The comparison with the competitor [TG5T]4 suggests that these
ligands are more easily displaced from F21T when the competitor
G4-DNA contains loops. Interestingly, MMQ3 seems to be
slightly less sensitive than 11 to the G4 competitor with loops,
mostly at 1 lM, and as emphasized earlier, it has more affinity
for the tetramolecular quadruplex than 11. It is worth noting that
the complex with the compound 11 is clearly less affected by the
tetramolecular G4 or the ds26 duplex (Fig. 4D) than the complexes
with the control MMQ3 (Fig. 4E).
Conclusion
The macrocyclic series described in the present study exhibits a
good to high affinity for intramolecular quadruplexes and a good
selectivity for DNA quadruplexes vs. duplexes. Given the poor
affinity and selectivity of the acyclic controls, the G4 preferential
binding of the macrocycles could be mediated by their particular
cyclic conformation. Moreover, the synergistic effects obtained
in the acridine and phenanthroline series strongly suggest that
the neomycin motif is likely to play a role in establishing specific
contacts with the G4-DNA target.
All together these data demonstrate that the neomycin-capped
macrocycles and in particular macrocycle 11 are able to dis-
criminate between loop-containing and tetramolecular parallel
quadruplexes, strongly suggesting a possible interaction of the
small-molecule binders with the loops. The presence of loops may
In addition, the competition assay with a parallel tetramolecular
quadruplex established that our compounds preferentially interact
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1049–1057 | 1053
©

View Online
Fig. 5 Telomerase inhibition by compounds 9–12 in a TRAP assay. (A) Increasing concentrations of compounds were added to the TRAP mixture in the
presence of an internal control (ITAS) and analysed by gel electrophoresis. TRAP activity was determined with 200 ng of a CHAPS extract of the A431
cell line. IC50s were determined by comparison with the telomerase activity with no compound. They are given with a 20% precision. (B) Quantitative
analysis of telomerase activity as a function of drug concentration. (C) IC50 values against telomerase found for the different compounds.
with loop-containing quadruplexes. This suggests the involvement
of loop motifs in the binding of the macrocycles, confirming
the structural importance of the loop for G4-DNA recognition.
Macrocycles 9–12 display sterically hindered conformations but
elicit a high plasticity, therefore, it would be of great interest
to further investigate their binding mode for understanding how
they adapt to the compact structure of G-quadruplex DNA, and
whether they are able to recognize the conformational diversity of
intramolecular quadruplexes.11g
Finally the TRAP assay demonstrated that macrocycle 11
inhibits telomerase in vitro in the submicromolar range, and thus
could serve as a lead for the design of a new series able to act as
telomere maintenance modulators. In addition, the concept of G4
sequestration of telomeric DNA by small molecules holds promise
for interfering with other telomere binding proteins such as POT1,
which actively participates in the prevention of the unfolding of the
3ꢀ-overhang and was recently shown to disrupt G4-quadruplexes.28
In conclusion, our neomycin-capped macrocycles display two
interesting features (i.e. quadruplex recognition and telomerase
inhibition) making them attractive structural scaffolds for further
developments aiming at the discovery of new and more selective
anticancer agents.
and gradients from 0.1% aqueous TFA to CH3CN containing 0.1%
TFA (flow rate: 3 mL min−1).
1H and 13C NMR spectra were recorded on a Bruker Avance300
spectrometer. For 1H and 13C, chemical shifts are reported in
ppm (d) downfield of tetramethylsilane (TMS) used as internal
standard.
LC HR MS measurements were performed on an Agilent 1100
series HPLC with an XTerra MS C8 3.5 lm column (2.1 × 100 mm
column dimension) coupled to a Bruker Daltonik microTOF mass
spectrometer (electrospray ionisation). The following gradient was
used: 5% aq. acetonitrile (0.05% trifluoroacetic acid) to 90% aq.
acetonitrile (0.05% trifluoroacetic acid) in 15 min at a flow of
250 lL min−1.
The three aromatic dialdehyde building blocks have already been
described29 and the synthesis of compounds 2–8 will be reported
elsewhere.
Neomycin bis-lysine acridine macrocycle (9). 2,5-Bis(dicarbox-
aldehyde)acridine (1.9 mg, 8 lmol) was dissolved in DCM–MeOH
(1 : 1, 150 mL) and a solution of the neomycin building block
7 (13.2 mg, 8 lmol) and TEA (3.4 lL, 2.4 mg, 24 lmol) in
DCM–MeOH (1 : 1, 25 mL) was slowly added. The resulting
solution was stirred for 4 d, filtrated through a celite pad and
evaporated to dryness. The residue was redissolved in DCM–
MeOH (1 : 1, 10 mL) and NaBH4 (1.8 mg, 48 lmol) was added.
After stirring at rt for 2 h, the reaction mixture was evaporated,
redissolved in CH2Cl2 and washed with 5% aq. NaHCO3 and
brine, dried over Na2SO4 and evaporated to dryness. The residue
of the conjugation–reduction procedure was dissolved in TFA–
CH2Cl2 (1 : 1, 2 mL) and stirred for 1 h. The reaction mixture was
evaporated and the residue was purified by reverse phase HPLC
(buffer A: 0.05% TFA in H2O, buffer B: 0.05% TFA in CH3CN;
0 min: 10% B, 10 min: 10% B, 15 min: 20% B, 65 min: 65% B).
Product-containing fractions were pooled and lyophilized to yield
9 as a slightly yellow powder (4.6 mg, 29%): HPLC (analytical) tR:
2.6 min. 1H NMR (300 MHz, D2O–CD3OD (10 : 1)) d: 9.77 (s, 1H,
Ar), 8.46 (s, 2H, Ar), 8.28 (d, 2H, J = 9.0 Hz, Ar), 8.18 (d, 2H,
Experimental
General methods
All commercially available chemicals were reagent grade and
were used without further purification. Flash chromatography
employed Merck silica gel [Kieselgel 60 (0.040–0.063 mm)]. Ana-
lytical TLC was performed with 0.2 mm silica-coated aluminium
sheets, visualization by UV light or by spraying either a solution
of ninhydrin (0.3% in weight in n-butanol containing 3% acetic
acid in volume) or an iodine solution (0.1 M in 10% sulfuric acid
aqueous solution). Preparative reversed-phase HPLC was carried
out on an Abimed-Gilson chromatograph using a Nucleodur 100
C18 ED 5l (250 × 10 mm) (Macherey & Nagel, Du¨ren, Germany)
1054 | Org. Biomol. Chem., 2006, 4, 1049–1057
This journal is
The Royal Society of Chemistry 2006
©

View Online
J = 9.6 Hz, Ar), 5.70 (d, 1H, J = 3.9 Hz, H-1ꢀ), 5.24 (d, 1H, J =
4.5 Hz, H-1ꢀꢀ), 5.00 (s, 1H, H-1ꢀꢀꢀ), 4.44 (bs, 4H, CH2Ar), 4.12–2.86
(m, 31H), 2.35–2.30 (m, 1H, H-2a), 1.81–1.55 (m, 9H), 1.35–1.20
(m, 4H). ES HR MS: (M + H) : 1074.5834, found: 1074.5820.
constraints were extracted from 2D NOESY32 and ROESY33
experiments with mixing times of 200 ms. Water suppression was
achieved with the WATERGATE scheme34 for samples containing
90% H2O and via presaturation in the case of D2O.
Neomycin bis-capronic acid acridine macrocycle (10). The title
compound was prepared in a manner analogous to the synthesis of
9, employing neomycin building block 8 (11.3 mg, 8 lmol). After
HPLC purification in the same conditions as described above, 10
(5.8 mg, 42%) was obtained as a slightly yellow powder. HPLC
(analytical) tR: 4.6 min. 1H NMR (300 MHz, D2O–CD3OD (10 :
1)) d: 9.82 (s, 1H, Ar), 8.47 (s, 2H, Ar), 8.29 (d, 2H, J = 9.0 Hz,
Ar), 8.19 (d, 2H, J = 9.3 Hz, Ar), 5.64 (d, 1H, J = 4.2 Hz, H-
1ꢀ), 5.20 (d, 1H, J = 3.0 Hz, H-1ꢀꢀ), 4.97 (s, 1H, H-1ꢀꢀꢀ), 4.44 (bs,
4H, CH2Ar), 4.14–2.84 (m, 29H), 2.33–2.29 (m, 1H, H-2a), 2.09–
2.01 (m, 4H, NHCOCH2), 1.71–1.66 (m, 1H, H-2e), 1.63–1.53 (m,
4H), 1.41–1.35 (m, 4H), 1.21–1.14 (m, 4H). ES HR MS: (M + H) :
1044.5616, found: 1044.5647.
Structure calculation
Structure calculations and evaluations were performed with the
INSIGHT II 2000 software package (Accelrys, San Diego, CA)
on Silicon Graphics O2 R5000 computers (SGI, Mountain View,
CA). A hundred structures were generated from the distance-
bound matrices. Triangle-bound smoothing was used. The NOE
intensities were converted into interproton distance constraints
˚
using the following classification: very strong (vs) 1.7–2.3 A, strong
˚
˚
˚
(s) 2.2–2.8 A, medium (m) 2.6–3.4 A, weak (w) 3.0–4.0 A, very
˚
weak (vw) 3.2–4.8 A, and the distances of the pseudo atoms
were corrected as described by Wu¨thrich.27 The structures were
generated in four dimensions, then reduced to three dimensions
with the EMBED algorithm and optimized with a simulated
annealing step according to the standard protocol of the DG II
package of INSIGHT II. All hundred structures were refined with
a short MD-SA protocol: after an initial minimisation, 5 ps at
300 K were simulated followed by exponential cooling to ∼0 K
during 10 ps. A time step of 1 fs was used with the CVFF force-
field while simulating the solvent H2O with a dielectric constant of
Neomycin bis-capronic acid quinacridine macrocycle (11). The
title compound was prepared under similar conditions as
for the synthesis of 9, employing neomycin building block
8 (11.3 mg, 8 lmol) and dibenzo[b,j][1,7]phenanthroline-2,10-
dicarboxaldehyde (2.7 mg, 8 lmol). After HPLC purification in
the same conditions as described above, 11 (3.5 mg, 25%) was
obtained as a yellow powder. HPLC (analytical) tR: 8.02 min. 1H
NMR (300 MHz, D2O–CD3OD (10 : 1)) d: 10.44 (s, 1H, Ar), 9.03
(s, 1H, Ar), 8.51 (s, 1H, Ar), 8.38 (t, 2H, J = 8.1 Hz, Ar), 8.29–8.26
(m, 2H, Ar), 8.06 (d, 1H, J = 8.1 Hz, Ar), 7.96–7.91 (m, 2H, Ar),
5.60 (d, 1H, J = 3.6 Hz, H-1ꢀ), 5.21 (d, 1H, J = 3.9 Hz, H-1ꢀꢀ), 4.96
(s, 1H, H-1ꢀꢀꢀ), 4.46 (dapp, 4H, J = 13.2 Hz, CH2Ar), 4.09–2.84 (m,
29H), 2.33–2.29 (m, 1H, H-2a), 2.07–1.99 (m, 4H, NHCOCH2),
1.73–1.46 (m, 5H), 1.41–1.26 (m, 4H), 1.20–1.08 (m, 4H). ES HR
MS: (M + H) : 1145.5883, found: 1145.5859.
80.0. The experimental distance constraints were applied at every
stage of the calculation with 50 kcal mol−1 A .
−2
˚
After simulated annealing with DISCOVER the structures were
sorted according to their final energies and the structures with
the lowest energies were analyzed. In the first calculations with
all 49 NMR-derived distance constraints two conformational
families were obtained, both with four characteristic violations.
Two subsets of NMR constraints were constructed by removing
once the four constraints that were persistently violated in the first
conformational family and once the other four constraints that
could not be fulfilled by the second conformational family. All
low-energy structures that resulted from calculations with one of
the subsets belonged to the corresponding conformational family
and exhibited no significant violations of distance constraints.
Neomycin bis-capronic acid phenanthroline macrocycle (12).
The title compound was prepared analogous to the synthesis of
9, employing neomycin building block 8 (11.3 mg, 8 lmol) and
2,9-bis(dicarboxaldehyde)-1,10-phenanthroline (1.9 mg, 8 lmol).
After HPLC purification in the same conditions as described
above, 12 (5.1 mg, 37%) was obtained as a slightly pink powder.
HPLC (analytical) tR: 7.0 min. 1H NMR (300 MHz, D2O–CD3OD
(10 : 1)) d: 8.61 (d, 2H, J = 8.1 Hz, Ar), 8.06 (s, 2H, Ar), 7.87 (dd,
2H, J = 2.1 Hz, J = 8.4 Hz, Ar), 5.93 (d, 1H, J = 3.9 Hz, H-1ꢀ),
5.21 (d, 1H, J = 3.9 Hz, H-1ꢀꢀ), 4.96 (s, 1H, H-1ꢀꢀꢀ), 4.46 (dapp, 4H,
J = 13.2 Hz, CH2Ar), 4.09–2.84 (m, 29H), 2.33–2.29 (m, 1H, H-
2a), 2.07–1.99 (m, 4H, NHCOCH2), 1.96–1.81 (m, 5H), 1.72–1.61
(m, 4H), 1.54–1.41 (m, 4H). ES HR MS: (M + H) : 1045.5569,
found: 1045.5570.
Oligonucleotides
All oligonucleotides were synthesized and purified by Eurogentec
(Belgium). The parallel quadruplex (TGGGGGT)4 was obtained
after incubation of the monomer at 500 lM in a 10 mM lithium
cacodylate pH 7.2 buffer containing 100 mM NaCl for at least one
night at 4 ◦C. Further dilutions were made in the same buffer.
FRET-melting assay
NMR analysis
Denaturation of the oligonuceotide F21T (fluorescein-3ꢀ-
GGGTTAGGGTTAGGGTTAGGG-5ꢀ-TAMRA) to probe the
interaction of a ligand with G-quadruplex DNA was described
elsewere.21a In the experiments presented here, a real-time PCR
apparatus (Mx3000P, Stratagene) was used, allowing the simul-
taneous recording of 96 samples. Fluorescence measurements
with the F21T oligonucleotide (0.2 lM) were studied in 10 mM
lithium cacodylate pH 7.2 buffer containing 100 mM NaCl. The
melting of the G-quadruplex was monitored alone or in the
presence of 1 lM of compound, by measuring the fluorescence
NMR spectra were recorded at 10 ◦C on a Bruker DRX 500
and an AV900 spectrometer equipped with pulsed-field-gradient
(PFG) accessories and a cryoprobe in the case of the 900 MHz
spectrometer. The sample was dissolved in a 9 : 1 H2O–D2O (v/v)
mixture or in >99.9% D2O at 5 mM concentrations resulting in
an (uncorrected) pH of 4. Resonance assignments were performed
according to the method of Wu¨thrich.30 The 2D TOCSY was
recorded with a spin-lock period of 70 ms using the MLEV-17
sequence for isotropic mixing.31 The 49 experimental distance
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1049–1057 | 1055
©

View Online
of fluorescein. To test the binding selectivity of the compound
to the quadruplex structure, we added various concentrations of
competitors: double-stranded DNA (self-complementary oligo-
nucleotide ds26: 5ꢀ-GTTAGCCTAGCTTAAGCTAGGCTAAC-
3ꢀ), tetramolecular G-quadruplex [TG5T]4 or intramolecular G-
quadruplex (30AG: 5ꢀ-AGGGGGTTAGGGGGTTAGGGGG-
TTAGGGGG-3ꢀ). Emission of fluorescein was normalized be-
tween 0 and 1, and T1/2 was defined as the temperature for which
the normalized emission is 0.5. T1/2 and DT1/2 are the mean of at
least 2–4 values standard deviation.
Smirnov, Biopolymers, 2001, 56, 209; (e) J.-L. Mergny, P. Mailliet, F.
Lavelle, J.-F. Riou, A. Laoui and C. He´le`ne, Anti-Cancer Drug Des.,
1999, 14, 327.
6 (a) G. Pennarun, C. Granotier, L. R. Gauthier, D. Gomez, F. Hoffschir,
E. Mandine, J.-F. Riou, J.-L. Mergny, P. Mailliet and F. D. Boussin,
Oncogene, 2005, 24, 1; (b) D. Gomez, R. Paterski, T. Lemarteleur,
K. Shin-ya, J.-L. Mergny and J.-F. Riou, J. Biol. Chem., 2004, 279,
41487; (c) M. Sumi, T. Tauchi, G. Sashida, A. Nakajima, A. Gotoh,
K. Shin-Ya, J. H. Ohyashiki and K. Ohyashiki, Int. J. Oncol., 2004, 24,
1481; (d) A. M. Burger, F. Dai, C. M. Schultes, A. P. Reszka, M. J.
Moore, J. A. Double and S. Neidle, Cancer Res., 2005, 65, 1489; (e) C.
Leonetti, S. Amodei, C. D’Angelo, A. Rizzo, B. Benassi, A. Antonelli,
R. Elli, M. F. G. Stevens, M. D’Incalci, G. Zupi and A. Biroccio, Mol.
Pharmacol., 2004, 66, 1138.
7 (a) M. Read, R. J. Harrison, B. Romagnoli, F. A. Tanious, S. H.
Gowan, A. P. Reska, W. D. Wison, L. R. Kelland and S. Neidle, Proc.
Natl. Acad. Sci. U. S. A., 2001, 98, 4844; (b) P. J. Perry, A. P. Reska,
A. A. Wood, M. A. Read, S. M. Gowan, H. S. Dosanjh, J. O. Trent,
T. C. Jenkins, L. R. Kelland and S. Neidle, J. Med. Chem., 1998, 41,
4873; (c) J.-L. Mergny, L. Lacroix, M.-P. Teulade-Fichou, C. Hounsou,
L. Guittat, M. Hoarau, P. B. Arimondo, J.-P. Vigneron, J.-M. Lehn,
TRAP assay
The TRAP reaction was performed in a 20 mM Tris HCl pH 8.3
buffer containing 63 mM KCl, 1.5 mM MgCl2, 1 mM EDTA,
0.005% Tween 20, 0.1 mg ml−1 BSA, 50 lM dTTP, dGTP and
dATP, 5 lM dCTP, and the oligonucleotides TS (5ꢀ-AATCCG-
TCGAGCAGAGTT-3ꢀ) (0.4 lM), ACX (5ꢀ-GCGCGGCTTA-
CCCTTACCCTTACCCTAACC-3ꢀ) (0.4 lM), NT (5ꢀ-ATCGC–
TTCTCGGCCTTTT-3ꢀ) (0.4 lM) and TSNT (5ꢀ-ATTCCGT–
CGAGCAGAGTTAAAAGGCCGAGAAGCGAT-3ꢀ) (20 nM),
2 units of Taq polymerase, 0.02 mCi mL−1 of [a32P]-dCTP and
200 ng of A431 CHAPS extracts. After telomerase elongation for
15 minutes at 30 ◦C, 30 cycles of PCR were performed (94 ◦C,
30 s; 50 ◦C, 30 s; and 72 ◦C for 90 s). Telomerase extension
products were then analysed on a denaturing 6% polyacrylamide,
7 M urea 1X Tris Borate EDTA (TBE) vertical gel. Extension
products were quantitated using a Phosphorimager apparatus;
telomerase relative activity was plotted against each compound
concentration.
J.-F. Riou, T. Garestier and C. H
e´le`ne, Proc. Natl. Acad. Sci. U. S. A.,
2001, 98, 3062; (d) M.-P. Teulade-Fichou, C. Hounsou, L. Guittat,
J.-L. Mergny, P. Alberti, C. Carrasco, C. Bailly, J.-M. Lehn and W. D.
Wilson, Nucleosides, Nucleotides Nucleic Acids, 2003, 22, 1483; (e) F.
Keoppel, J.-F. Riou, A. Laoui, P. Mailliet, P. B. Arimondo, D. Labit,
O. Petigenet, C. He´le`ne and J.-L. Mergny, Nucleic Acids Res., 2001, 29,
1087; (f) R. A. Heald, C. Modi, J. C. Cookson, I. Hutchinson, C. A.
Laughton, S. M. Gowan, L. R. Kelland and M. F. G. Stevens, J. Med.
Chem., 2002, 45, 590; (g) O. Y. Feodoroff, M. Salazar, H. Han, V. V.
Chemeris, S. M. Kerwin and L. H. Hurley, Biochemistry, 1998, 37,
12367.
8 (a) M.-Y. Kim, H. Vankayalapati, K. Shin-ya, K. Wierzba and L. H.
Hurley, J. Am. Chem. Soc., 2002, 124, 2098; (b) M.-P. Teulade-Fichou,
C. Carrasco, L. Guittat, C. Bailly, P. Alberti, J.-L. Mergny, A. David,
J.-M. Lehn and W. D. Wilson, J. Am. Chem. Soc., 2003, 125, 4732;
(c) E. M. Rezler, J. Seenisamy, S. Bashyam, M.-Y. Kim, E. White,
W. D. Wilson and L. H. Hurley, J. Am. Chem. Soc., 2005, 127,
9439.
9 (a) J.-F. Riou, L. Guittat, P. Mailliet, A. Laoui, E. Renou, O. Petitgenet,
F. Me´gnin-Chanet, C. He´le`ne and J.-L. Mergny, Proc. Natl. Acad. Sci.
U. S. A., 2002, 99, 2672; (b) J. A. Shouten, S. Ladame, S. J. Mason,
M. A. Cooper and S. Balasubramanian, J. Am. Chem. Soc., 2003, 125,
5594; (c) P. M. Murphy, V. A. Philips, S. A. Jennings, N. C. Garbett,
J. B. Chaires, T. C. Jenkins and R. T. Wheelhouse, Chem. Commun.,
2003, 1160; (d) C. Granotier, G. Pennarun, L. Riou, F. Hoffschir, L. R.
Gautier, A. De Cian, D. Gomez, E. Mandine, J.-F. Riou, J.-L. Mergny,
P. Mailliet, B. Dutrillaux and F. D. Boussin, Nucleic Acids Res., 2005,
33, 4182.
10 (a) S. M. Haider, G. N. Parkinson and S. Neidle, J. Mol. Biol., 2003, 326,
117; (b) E. Gavathiotis, R. A. Heald, M. F. G. Stevens and M. S. Searle,
J. Mol. Biol., 2003, 334, 25; (c) Q. Chen, I. Kuntz and R. H. Shafer,
Proc. Natl. Acad. Sci. U. S. A., 1996, 93, 2635; (d) J. A. Schouten, S.
Ladame, S. J. Mason, M. A. Cooper and S. Balasubramanian, J. Am.
Chem. Soc., 2003, 125, 5594; (e) I. M. Dixon, F. Lopez, J.-P. Este`ve,
A. M. Tejera, M. A. Blasco, G. Pratviel and B. Meunier, ChemBioChem,
2005, 6, 123; (f) A. T. Phan, V. Kuryavyi, H. Y. Gaw and D. Patel, Nat.
Chem. Biol., 2005, 1, 167; (g) M. A. Read and S. Neidle, Biochemistry,
2000, 39, 13422; (h) R. J. Harrison, J. Cuesta, G. Chessari, M. A. Read,
S. K. Basra, A. P. Reszka, J. Morrell, S. M. Gowan, C. M. Incles, F. A.
Tanious, W. D. Wilson, L. R. Kelland and S. Neidle, J. Med. Chem.,
2003, 46, 4463.
11 (a) Y. Wang and D. J. Patel, Structure, 1993, 1, 263; (b) G. N. Parkinson,
M. P. Lee and S. Neidle, Nature, 2002, 417, 876; (c) L. Ying, J. J. Green,
H. Li, D. Klenerman and S. Balasubramanian, Proc. Natl. Acad. Sci.
U. S. A., 2003, 100, 14629; (d) S. Redon, S. Bombard, M.-A. Elizondo-
Riojas and J. C. Chottard, Nucleic Acids Res., 2003, 31, 1605; (e) A. T.
Phan and D. J. Patel, J. Am. Chem. Soc., 2003, 125, 15021; (f) J. Qi and
R. H. Shafer, Nucleic Acids Res., 2005, 33, 3185 and references cited
therein; (g) J. Li, J. J. Correia, L. Wang, J. O. Trent and J. B. Chaires,
Nucleic Acids Res., 2005, 33, 4649; (h) M. Vorlickova, J. Chladkova, I.
Kejnovska, M. Fialova and J. Kypr, Nucleic Acids Res., 2005, 33, 5851;
(i) J. Qi and R. H. Shafer, Nucleic Acids Res., 2005, 33, 3185; (j) D.
Miyoshi, H. Karimata and N. Sugimoto, Angew. Chem., Int. Ed., 2005,
44, 3740.
Acknowledgements
M. K. thanks the European Union for a postdoctoral Marie Curie
Intra-European Fellowship. This study has been supported by
L’Association pour la Recherche contre le Cancer (grant #3482
to MPTF and #3365 to JLM) and an EU FP6 “MolCancerMed”
grant to JLM. The help of E. Weyher-Stingl (MPI of Biochemistry,
Martinsried, Germany) for the LC HR MS measurements is
greatly acknowledged. David Monchaud (Laboratoire de Chimie
des Interactions Mole´culaires, College de France) is greatly
acknowledged for his help with the graphics.
References
1 (a) S. A. Stewart, I. Ben-Porath, V. J. Carey, B. F. O’Connor, W. C.
Hahn and R. A. Weinberg, Nat. Genet., 2003, 33, 492; (b) J. Karlseder,
A. Smogorzewska and T. de Lange, Science, 2002, 295, 2446; (c) X.
Zhang, V. Mar, W. Zhou, L. Harrington and M. O. Robinson, Genes
Dev., 1999, 13, 2388; (d) S. Neidle and D. E. Thurston, Nat. Rev. Cancer,
2005, 5, 285.
2 (a) Y. Mo, Y. Gan, S. Song, J. Johnston, X. Xiao, M. G. Wientjes and
J. L.-S. Au, Cancer Res., 2003, 63, 579; (b) M. A. Shammas, R. J. S.
Reis, C. Li, H. Koley, L. H. Hurley, K. C. Anderson and N. C. Munshi,
Clin. Cancer Res., 2004, 10, 770.
3 (a) T. R. Cech, Cell, 2004, 116, 273; (b) J. W. Shay, Cancer Res., 2005,
65, 3513.
4 (a) E. M. Rezler, D. J. Bearss and L. H. Hurley, Annu. Rev. Pharmacol.
Toxicol., 2003, 43, 359; (b) S. Neidle and G. Parkinson, Nat. Rev. Drug
Discovery, 2002, 1, 383.
5 (a) S. Neidle and G. N. Parkinson, Curr. Opin. Struct. Biol., 2003,
13, 275; (b) L. H. Hurley, Nat. Rev. Cancer, 2002, 2, 188; (c) T.
Simonsson, Biol. Chem., 2001, 382, 621; (d) R. H. Shafer and I.
1056 | Org. Biomol. Chem., 2006, 4, 1049–1057
This journal is
The Royal Society of Chemistry 2006
©

View Online
12 (a) M.-P. Teulade-Fichou and J.-P. Vigneron, in Small molecule DNA
and RNA binders: From Synthesis to Nucleic Acid Complexes, ed.
M. Demeunynck, C. Bailly and W. D. Wilson, Wiley-VCH, Weinheim,
2003, vol. 1, ch. 11, pp. 278–311 and references cited therein; (b) J.
Seenisamy, S. Bashyam, V. Gokhale, H. Vankayalapati, D. Sun, A.
Siddiqui-Jain, N. Streiner, K. Shin-ya, E. White, W. D. Wilson and
L. H. Hurley, J. Am. Chem. Soc., 2004, 127, 2944.
23 pKa’s of the free neomycin are: 5.7, 7.6, 7.6, 8.1 (NH2 attached to the
sugar rings); 8.6, 8.7 (NH2 attached to the two primary carbons).16a
Thus, the free neomycin has a global charge of 4+–6+ depending on
the experimental pH (7.2 for FRET-melting, 8.3 for TRAP). In the
macrocyclic series, the two benzylic nitrogens have pKa’s of around 7–8,
and the ring nitrogens of the three aromatics are not protonated, neither
are the two basic NH2 residues of neomycin engaged in the amide
bound. Hence, the global charge of the macrocycles at experimental
13 F. Rosu, V. Gabelica, K. Shin-Ya and E. De Pauwn, Chem. Commun.,
2003, 2702.
+
pH (7.2) is estimated to be around 4+ (amino caproic linker) and 6
14 (a) Y. Tor, ChemBioChem, 2003, 4, 998; (b) F. Walter, Q. Vicens and E.
Westhof, Curr. Opin. Chem. Biol., 1999, 3, 694; (c) M. Zacharias, Curr.
Med. Chem., 2003, 2, 161.
(lysine linker).
24 P. Alberti, J. Ren, M.-P. Teulade-Fichou, L. Guittat, J.-F. Riou, J. B.
Chaires, C. He´le`ne, J.-P. Vigneron, J.-M. Lehn and J.-L. Mergny,
J. Biomol. Struct. Dyn., 2001, 19, 505.
15 J.-L. Mergny, et al., unpublished work.
16 (a) D. P. Arya, R. L. Coffee, B. Willis and A. L. Abramovitch, J. Am.
Chem. Soc., 2001, 123, 5385; (b) D. P. Arya, L. Micovic, I. Charles,
R. L. Coffee, B. Willis and L. Xue, J. Am. Chem. Soc., 2003, 13, 3733.
17 (a) M. Hendrix, P. B. Alper, E. S. Priestley and C.-H. Wong, Angew.
Chem., Int. Ed. Engl., 1997, 36, 95; (b) D. Fourmy, M. I. Recht, S. C.
Blanchard and J. D. Puglisi, Science, 1996, 274, 1367.
18 (a) S. M. Nelson, Pure Appl. Chem., 1980, 52, 2461; (b) D. Chen and
A. E. Martell, Tetrahedron, 1991, 47, 6895; (c) J. Jazwinski, J.-M. Lehn,
R. Me´ric, J.-P. Vigneron, M. Cesario, J. Guilhem and C. Pascard,
Tetrahedron Lett., 1987, 28, 3489.
19 M. Sainlos, P. Belmont, J.-P. Vigneron, P. Lehn and J.-M. Lehn,
Eur. J. Org. Chem., 2003, 2764.
20 J. L. Sessler and A. K. Burell, in Macrocycles, Topics in Current
Chemistry, ed. E. Weber and F. Vo¨gtle, Springer-Verlag, Berlin, 1992,
vol. 161, p. 179.
25 J.-L. Mergny, A. De Cian, A. Ghelab, B. Sacca` and L. Lacroix, Nucleic
Acids Res., 2005, 33, 81.
26 J.-L. Mergny, J. Li, L. Lacroix, S. Amrane and J. B. Chaires, Nucleic
Acids Res., 2005, 33, e138.
27 D. Gomez, J.-L. Mergny and J.-F. Riou, Cancer Res., 2002, 62,
3365.
28 A. J. Zaug, E. R. Podell and T. R. Cech, Proc. Natl. Acad. Sci. U. S. A.,
2005, 102, 10864.
29 (a) M.-P. Teulade-Fichou, J.-P. Vigneron and J.-M. Lehn, J. Supramol.
Chem., 1995, 5, 139; (b) O. Baudoin, M.-P. Teulade-Fichou, J.-P.
Vigneron and J.-M. Lehn, J. Org. Chem., 1997, 62, 5454; (c) C. J.
Chandler, L. W. Deady and J. A. Reiss, J. Heterocycl. Chem., 1981,
18, 599.
30 K. Wu¨thrich, NMR of Proteins and Nucleic Acids, John Wiley, New
York, 1986.
21 (a) J.-L. Mergny and J.-C. Maurizot, ChemBioChem, 2001, 2, 124; (b) F.
Rosu, E. De Pauwn, L. Guittat, P. Alberti, L. Lacroix, P. Mailliet, J.-F.
Riou and J.-L. Mergny, Biochemistry, 2003, 42, 10361.
22 (a) J. M. Sanders and C. Hunter, J. Am. Chem. Soc., 1990, 112, 5525;
(b) E. A. Meyer, R. K. Castellano and F. Diederich, Angew. Chem., Int.
Ed., 2003, 42, 1210.
31 A. Bax and D. G. Davis, J. Magn. Reson., 1985, 65, 355.
32 J. Jeener, B. H. Meier, P. Bachman and R. R. Ernst, J. Chem. Phys.,
1979, 71, 4546.
33 A. Bax and D. G. Davis, J. Magn. Reson., 1985, 63, 207.
34 V. Sklenar, M. Piotto, R. Leppik and V. Saudek, J. Magn. Reson., 1993,
102, 241.
This journal is
The Royal Society of Chemistry 2006
Org. Biomol. Chem., 2006, 4, 1049–1057 | 1057
©