Synthesis and spectroscopic studies of the aminoglycoside (Neomycin)-perylene conjugate binding to human telomeric DNA

Article abstract of DOI:10.1021/bi1017304

Synthesis of a novel perylene-neomycin conjugate (3) and the properties of its binding to human telomeric G-quadruplex DNA, 5′-d[AG ₃(T₂AG₃)₃] (4), are reported. Various spectroscopic techniques were employed to characterize the binding of conjugate 3 to 4. A competition dialysis assay revealed that 3 preferentially binds to 4, in the presence of other nucleic acids, including DNA, RNA, DNA-RNA hybrids, and other higher-order structures (single strands, duplexes, triplexes, other G-quadruplexes, and the i-motif). UV thermal denaturation studies showed that thermal stabilization of 4 increases as a function of the increasing concentration of 3. The fluorescence intercalator displacement (FID) assay displayed a significantly tighter binding of 3 with 4 as compared to its parent constituents [220-fold stronger than neomycin (1) and 4.5-fold stronger than perylene diamine (2), respectively]. The binding of 3 with 4 resulted in pronounced changes in the molar ellipticity of the DNA absorption region as confirmed by circular dichroism. The UV-vis absorption studies of the binding of 3 to 4 resulted in a red shift in the spectrum of 3 as well as a marked hypochromic change in the perylene absorption region, suggesting that the ligand-quadruplex interaction involves stacking of the perylene moiety. Docking studies suggest that the perylene moiety serves as a bridge that end stacks on 4, making contacts with two thymine bases in the loop, while the two neomycin moieties branch into the grooves of 4.(Figure Presented)

Full text of DOI:10.1021/bi1017304

ARTICLE
pubs.acs.org/biochemistry
Synthesis and Spectroscopic Studies of the Aminoglycoside
(Neomycin)-Perylene Conjugate Binding to Human Telomeric DNA
Liang Xue,† Nihar Ranjan, and Dev P. Arya*
Laboratories of Medicinal Chemistry, Clemson University, Clemson, South Carolina 29634, United States
S Supporting Information
b
ABSTRACT: Synthesis of a novel perylene-neomycin conjugate (3) and the
properties of its binding to human telomeric G-quadruplex DNA, 50-d[AG3-
(T2AG3)3] (4), are reported. Various spectroscopic techniques were employed to
characterize the binding of conjugate 3 to 4. A competition dialysis assay revealed
that 3 preferentially binds to 4, in the presence of other nucleic acids, including
DNA, RNA, DNA-RNA hybrids, and other higher-order structures (single
strands, duplexes, triplexes, other G-quadruplexes, and the i-motif). UV thermal
denaturation studies showed that thermal stabilization of 4 increases as a function
of the increasing concentration of 3. The fluorescence intercalator displacement
(FID) assay displayed a significantly tighter binding of 3 with 4 as compared to its
parent constituents [220-fold stronger than neomycin (1) and 4.5-fold stronger
than perylene diamine (2), respectively]. The binding of 3 with 4 resulted in
pronounced changes in the molar ellipticity of the DNA absorption region as
confirmed by circular dichroism. The UV-vis absorption studies of the binding of
3 to 4 resulted in a red shift in the spectrum of 3 as well as a marked hypochromic
change in the perylene absorption region, suggesting that the ligand-quadruplex
interaction involves stacking of the perylene moiety. Docking studies suggest that
the perylene moiety serves as a bridge that end stacks on 4, making contacts with two thymine bases in the loop, while the two
neomycin moieties branch into the grooves of 4.
ver the past few decades, a large number of human and
other eukaroytic chromosomal end sequences (also known
to play an important role in the interaction of telomerase at the
ends of chromosomes.5 Intervention of telomerase interaction at
the telomeric ends has become one of the popular approaches to
control cancer proliferation.
O
as telomeric sequences) have been identified.1 These telomeric
ends are uniquely characterized with the abundance of repetitive
guanine bases. It is now well established that telomeric DNAs,
under physiological conditions, can adopt unique structures
called G-quadruplexes.2 G-Quadruplexes are formed by stacking
G-quartets. The edges of guanine bases can serve as H-bonding
donors (N2 and N1) or acceptors (O6 and N7), and a coplanar
arrangement (G-quartet) results among four guanines via
Hoogsteen H-bonds (Figure 1) in an inter- or intramolecular
arrangement. In humans, the telomeric ends consisting of a
tandem hexameric 50-d(TTAGGG) repeat unit are mostly
duplex in nature, yet the extreme 30-ends are single-stranded
overhangs. The 30-end of the chromosomal DNA, apart from
capping the chromosal ends and prohibiting the end to end
fusion processes,3 also plays the role of primer sequence for the
reverse transcriptase activity of an enzyme called telomerase.4
Telomerase, which remains dormant in normal somatic cells,
assumes the role of an “evil enzyme” by conducting unwanted
repair of telomeric ends in cancer cells, thereby making them
immortal. Both the duplex and single-stranded regions of the
telomers, in conjunction with telomere end binding proteins
such as TFRF1, TRF2, POT1, and other assisting proteins,
form the so-called “telosome” complex and have been proposed
Formation of G-quadruplexes has been shown to inhibit the
activity of telomerase, which plays a pivotal role in the proliferation
of tumor cells6 and has become a potential therapeutic target in
oncology. For more than a decade, small molecules have received
wide attention because of their ability to interact and stabilize
G-quadruplex DNA structures as well as their potential to inhibit
telomerase activity. Ligands that facilitate G-quadruplex formation
and subsequently inhibit telomerase activity could be potential
anticancer drugs.7 A large number of small molecule G-quadruplex
binders have emerged since the early reports of telomerase inhibi-
tion from the laboratories of Hurley and Neidle.8,9 One such small
molecule, telemostatin, has been reported to have minimal inhibi-
tory concentration for telomerase inhibition as low as 5 nM.10
Various other ligands that contain fused aromatic platforms con-
taining heteroatoms in most cases have been reported to interact
with G-quadruplex structures. For instance, quinacridines,11 BRA-
CO-19,12 porphyrins,13 and telemostatin-based macrocycles14,15
Received: October 28, 2010
Revised:
February 3, 2011
Published: February 17, 2011
r
2011 American Chemical Society
2838
dx.doi.org/10.1021/bi1017304 Biochemistry 2011, 50, 2838–2849
|

Biochemistry
ARTICLE
sequence of Oxytricha nova) with moderate affinities (Ka ∼ 105
M-1).41 Neomycin binds with this G-quadruplex DNA in a 1:1
stoichiometry and forms a single ion pair with the quadruplex, as
determined by salt-dependent studies. Our studies revealed that
neomycin binds to both parallel and antiparallel G-quadruplexes
with differing affinities, and the binding of neomycin to the
antiparallel quadruplex occurs in the wide groove. A recent study
has shown that aminoglycosides can inhibit telomerase activity.42
In light of these findings, we herein report the development of a
perylene-neomycin conjugate (3) (Figure 2) and its ability to
specifically recognize human telomeric G-quadruplex DNA (4)
using spectroscopic techniques.
Human telomeric DNA is highly polymorphic in nature and
can adopt various G-quadruplex confomations.43 For instance, in
the presence of sodium ions, 50-d[AGGG(TTAGGG)3] (4), a
fragment of human telomeric DNA, adopts an antiparallel
structure that contains both lateral and diagonal loops.44 Differ-
ent structures of the quadruplex have been reported on the basis
of the technique (NMR or X-ray),44,45 the salt used (Naþ vs Kþ),
or the flanking bases.46-50 Nevertheless, because of the impor-
tance of developing inhibtors of human telomeric DNA, ligand
binding to this structure continues to be examined51 and sub-
micromolar inhibition of telomerase by a macrocycle-capped
neomycin has been reported.52 In this report, we show that by
covalently linking two neomycin units to a perylene diimide unit,
a ligand that specifically binds to 4 is obtained and its binding to 4
is better than that of each of its constituent units.
Figure 1. Hydrogen bonding interactions (shown by dashed lines)
present in a G-quartet. The guanine bases use the Hoogsteen face for
hydrogen bonding. The circle (colored black) in the middle of G-quartet
represents a metal cation.
are a few examples of molecules that preferentially bind G-quad-
ruplexes. A recent review by Monchaud et al. describes the
current state of G-quadruplex binding ligands.16
Most ligands that bind to G-quadruplexes have been proposed
to end stack or form interactions in the loops.17,18 On the other
hand, molecules that selectively recognize G-quadruplex grooves
are much more limited, and features that allow ligands to be
G-quadruplex groove specific are poorly understood.19 Discovery
of G-quadruplex groove specific binders is, thus, an important
endeavor for two prime reasons: (i) to develop ligands that advance
the knowledge of recognition principles of quadruplex grooves and
(ii) to generate conjugates with mulitple moieties that could bind
simultaneously through groove recognition and stacking interac-
tions, thus leading to ligands with higher affinities and specificities.
Dual recognition allows us to conveniently develop potent ligands
by exploiting the noncompetitive binding20 of individual pharma-
cophores for the nucleic acid target. We have recently reported
that using such a design, specific and high-affinity ligands can be
developed. These include pyrene-neomycin,21 BQQ-neomycin,22
and anthraquinone-neomycin20 conjugates for binding DNA
triplexes and a methidium-neomcyin conjugate designed to
target hybrid nucleic structures with subnanomolar affinities.23
Aminoglycosides are aminosugars, long known for their
effective antiobiotic action. They have been largely known for
interacting with bacterial rRNA (A-site). They have, however,
been shown to bind other nucleic acid targets.24,25 Recent studies
have shown that apart from its euabacterial RNA target, neomy-
cin (and other aminoglycosides) also binds to various nucleic
acids, including DNA triplexes,25-29 RNA triplexes,26,27,29 hy-
brid nucleic acids,23,26 and single strands.30 We have shown that
aminoglycosides prefer to bind targets that share characteristics
of A-form nucleic acids31 irrespective of the parent nucleic acid
(DNA or RNA). Recently, we have developed various neomycin
conjugates by tethering neomycin with other DNA binding
moieties or oligonucleotides, because neomycin also aids lipid-
mediated oligonucleotide delivery.32 Such conjugates recognize
different nucleic acid structures21,22,33-40 with binding affinities
much higher than those of their parent constituents.20 Recent
studies have shown that neomycin and paromomycin bind in the
wide groove of G-quadruplex DNA (formed from the telomeric
’ MATERIALS AND METHODS
General Methods. Unless otherwise specified, chemicals were
purchased from Aldrich (St. Louis, MO) or Fisher Scientific
(Pittsburgh, PA) and used without further purification. Neomycin
B trisulfate was purchased from MP Biomedicals (Solon, OH). Di-
tert-butyl dicarbonate (Boc anhydride) was purchased from Ad-
vanced ChemTech (Louisville, KY). All solvents were purchased
from VWR (West Chester, PA). Silica gel (32-65 μM mesh size)
was purchased from Sorbtech (Atlanta, GA). Reactions were con-
ducted under N2 using dry solvent, unless otherwise noted. 1HNMR
spectra were recorded on a JEOL (Tokyo, Japan) ECA 500 MHz
FT-NMR spectrometer or Bruker Avance-500 spectrometer. MS
(MALDI-TOF) spectra were recorded using a Kratos analytical
(Columbia, MD) KOMPACT SEQ mass spectrometer or Bruker
Daltonics/Omniflex MALDI-TOF spectrometer. UV spectra were
recorded on a Varian (Walnut Creek, CA) Cary 100 Bio UV-vis
spectrophotometer equipped with a thermoelectrically controlled
12-cell holder. Circular dichroism spectra were recorded on a JASCO
(Easton, MD) J-810 spectropolarimeter equipped with a thermo-
electrically controlled cell holder. Fluorescence spectra were re-
corded on a Fluomax-3 (Jobin Yvon, Edison, NJ) or Photon
Technology International (Lawrenceville, NJ) instrument.
Nucleic Acids. The concentrations of nucleotide solutions
were determined using the extinction coefficients (per mole
of nucleotide) calculated according to the nearest neighbor
method. The concentrations of all the polymer solutions were
determined spectrophotometrically using the following extinc-
tion coefficients (in units of moles of nucleotide per liter per
centimeter): ε265 = 9000 for poly(dT), ε260 = 6000 for poly-
(dA) poly(dT), ε253 = 7400 for poly(dG) poly(dC), and ε262
=
3
3
6600 for poly(dA-dT) poly(dA-dT). In all cases where men-
3
tioned, the term rdb refers to the drug:base molar ratio. Nucleic
acid solutions were prepared in either phosphate buffer [8 mM
2839
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
Figure 2. Structures of ligands used in this study.
Na2HPO4, 2 mM NaH2PO4, 1 mM Na2EDTA, and 185 mM
NaCl (pH 7.0)] or cacodylate buffer [10 mM sodium cacodylate,
0.5 mM EDTA, and 100 mM NaCl (pH 7.0)]. The quadruplex
was formed by heating the human telomeric DNA at 95 °C for 20
min followed by slow cooling to room temperature. The stock
solution was then kept at 4 °C for at least 2 days before being used.
The formation of the quadruplex was checked by CD spectroscopy.
The numbering adopted for the synthesis of compounds given
below is shown in Scheme 1 (6). In a 25 mL round-bottom flask, N-
tert-butoxycarbonylbutyl-1,4-diamine (0.92 g, 4.9 mmol) and zinc
acetate dihydrate (1.0 g) were added to a suspension of 5 (707 mg,
1.8 mmol) in anhydrous pyridine (15.0 mL), and the mixture was
refluxed overnight under an atmosphere of N2. Thin layer chroma-
tography (TLC) using silica gel showed the completion of the
reaction. The reaction mixture was then concentrated. Flash chro-
matography of the residue [silica gel, 5% (v/v) CH3OH in CH2Cl2]
yielded the desired product 6 as a dark red solid (200 mg): Rf = 0.71
[silica gel, 10% (v/v) CH3OH in CH2Cl2]; MS (MALDI-TOF)
m/z calcd for C40H40N4O8Na (M þ Naþ) 755.8, found 756.5. This
product was taken to the next step without further characterization.
Method A (2). 6 (20 mg, 0.028 mmol) was dissolved in TFA
(2.0 mL) and the solution stirred for 3 h at room temperature.
The reaction mixture was concentrated, and the residue was
portioned between deionized (DI) water (30 mL) and ether
(30 mL). The aqueous layer was washed twice with ether
(30 mL) and concentrated to dryness to yield the desired
product 2 in quantitative yield (21.2 mg, 15% for two steps).
Method B (2). To a suspension of 3,4,9,10-perylenetetracar-
boxylic dianhydride (250 mg, 0.64 mmol) in benzene (10 mL)
was added putrescine (500 mg, 5.66 mmol). The reaction
mixture was refluxed for 3 h, filtered, and washed with benzene.
The solid obtained was added to KOH (5 M, 10 mL) and the
mixture stirred at room temperature for 8 h. The reaction mixture
was filtered, washed with water, and dried. The resulting solid
was dissolved in HCl (12 M, 10 mL) and filtered. The filtrate was
added to ethanol (200 mL), and the resulting solid was collected
by centrifugation and washed with ethanol (50 mL) to yield a
dark brown solid (254 mg, 75%): mp >320 °C dec; IR (KBr)
2925, 1682, 1337, 1140 cm-1; 1H NMR (500 MHz, CF3COOD)
δ 8.78-8.76 (d, 4H, J = 7.8 Hz), 8.73-8.72 (d, 4H, J = 7.8 Hz),
6.71 (s, br, NH2, 4H), 4.36-4.28 (t, br, NCH2CH2, 4H),
3.34-3.26 (t, br, CH2CH2NH3þ, 4H), 1.96-1.88 (m,
CH2CH2CH2CH2, 8H); MS (MALDI-TOF) m/z calcd for
C32H29N4O4 ([M þ H]þ) 533.59, found 533.54.
2840
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
Scheme 1. Synthesis of Perylene-Neomycin Conjugate 3a
a Reagents and conditions: (a) NH2(CH2)4NHBoc, pyridine, reflux, 5 h; (b) TFA, 3 h, room temperature, quantitative; (c) 2, DMSO/pyridine, 60 °C,
overnight; (d) TFA/CH2Cl2 (50:50), room temperature, 3 h (32.5% overall yield for steps c and d).
Method A (3). In a 25 mL round-bottom flask, 2 (4.2 mg, 5.5
μmol) and 4-dimethylaminopyridine (catalytic amount) were
added to a DMSO/pyridine [4 mL, 1:1 (v/v) DMSO:pyridine
ratio] solution of 7 (21 mg, 15.9 μmol), and the mixture was
stirred overnight under an atmosphere of N2 at 60 °C. Comple-
tion of the reaction was monitored by TLC using silica gel. The
reaction mixture was then concentrated. Flash chromatography
of the residue [silica gel, 6% (v/v) CH3OH in CH2Cl2] yielded
the desired product 8 as a red solid (7.1 mg). The product
obtained was taken to the deprotection step without further
characterization. Compound 8 (7.1 mg) was dissolved in a TFA/
CH2Cl2 mixture [2 mL, 1:1 (v/v)] and stirred for 3 h at room
temperature. The reaction mixture was concentrated, and the
residue was partitioned into DI water (20 mL) and ether
(20 mL). The aqueous layer was washed twice with ether
(20 mL) and concentrated by lyophilization to yield the desired
product 3 (3.9 mg, 32.5% overall yield for two steps).
(2.0 mL) followed by addition of a catalytic amount of 4-di-
methylaminopyridine in CH2Cl2 (0.5 mL). The mixture was
heated at 80 °C overnight. TLC using silica gel indicated
formation of the product [Rf = 0.55 in a 9:1 (v/v) CH2Cl2/
CH3OH]. Volatiles were removed under a stream of dry nitro-
gen, and the crude product thus obtained was purified on a silica
gel column using a gradient of CH2Cl2 and CH3OH as eluent.
The desired product 8 was obtained as a red solid (15 mg). This
product was taken to the deprotection step without further
characterization. Compound 8 (15 mg 4.7 μmol) was dissolved
in a TFA/CH2Cl2 mixture [3 mL, 1:1 (v/v)] and stirred at room
temperature for 4 h. The reaction mixture was concentrated, and
the residue was dissolved in DI water (4 mL) and washed three
times with ether (5 mL). The aqueous layer was lyophilized to
yield the desired product 3 (8.1 mg, 78% yield for two steps): UV
λ = 220, 293, 500, 535 nm; ε (500 nm) = 36283.8 M-1 cm-1; 1H
NMR (500 MHz, CF3COOD) δ 8.37-8.34 (br, 2H), 8.17-8.16
(d, 2H, J = 6.5 Hz), 7.44-7.41 (br, 4H), 6.44-6.22 (br, 2H),
5.72-5.69 (br, 4H), 4.67-4.40 (20H), 4.29-3.55 (24H),
3.38-3.32 (m, 6H), 3.20-3.11 (4H), 3.08-2.88 (6H),
2.66-2.58 (d, br, 2H), 1.72-1.56 (2H), 1.47-1.44 (t, 8H, J
= 7.35 Hz), 1.35-1.32 (t, 6H, J = 7.35 Hz), 1.04-1.02 (m, 2H);
Method B (3). To a solution of 2 (4.0 mg, 6.66 μmol) in a
mixture of DMSO and DMF [12 mL, 2:1 (v/v)] was added
triethylamine (60 μL), and the mixture was stirred at room
temperature for 15 min under an atmosphere of argon. To this
mixture was added a solution of 7 (20 mg, 15.2 μmol) in DMSO
2841
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
1H NMR (500 MHz, D2O) δ 7.79-7.76 (br, 2H), 7.60-7.59 (d,
2H, J = 7.3 Hz), 6.71-6.67 (br, 4H), 6.02-5.95 (br, 2H), 5.36-
5.31 (br, 2H), 5.23 (s, br, 2H), 4.78-4.60 (peaks masked by the
residual HDO peak), 4.40-4.21 (8H), 4.15 (s, br, 2H), 4.03 (t,
2H, J = 9.1 Hz), 3.96-3.90 (t, 2H, J = 9.6 Hz), 3.88-3.80 (4H),
3.75 (s, br, 4H), 3.68-3.59 (2H), 3.56-3.18 (20H), 3.15-
3.05 (6H), 2.80-2.60 (6H), 2.46-2.38 (2H), 1.89-1.79
(m, 2H), 1.22-1.03 (6H), 0.85-0.79 (m, 2H); MS (MALDI-
TOF) m/z calcd for C84H126N18O28S4 ([M]þ) 1961.70, found
1959.83.
Ultraviolet (UV) Spectroscopy. Spectrophotometer stability
and wavelength alignment were checked prior to initiation of
each melting point experiment. For all experiments, the samples
were prepared by dilution of a stock sample that was formed by
heating the DNA to 95 °C followed by slow cooling back to room
temperature. The melting of DNA with and without the ligand
was conducted at a heating rate of 0.2 °C/min. Samples were
brought back to 20 °C after each run. All UV melting experiments
were monitored at 260 nm. For the Tm determinations, deriva-
tives were used. Data were recorded every 1.0 °C. The UV
melting experiments were conducted in phosphate buffer [8 mM
Na2HPO4, 2 mM NaH2PO4, 1 mM Na2EDTA, and 185 mM
NaCl (pH 7.0)].
Competition Dialysis. Competition dialysis experiments
were performed in a MINI dialysis flotation device. For each
assay, 180 μL of different nucleic acids (75 μM per monomeric
unit of each polymer) was placed in the dialysis units and then
dialyzed against 400 mL of a 1 μM ligand solution in phosphate
buffer for 24 h at ambient temperature (20-22 °C). At the end of
the experiment, the nucleic acid samples were carefully removed
to microfuge tubes and were taken to a final sodium dodecyl
sulfate (SDS) concentration of 1% (w/v). Each mixture was
allowed to equilibrate for 2 h. The concentration of the ligand
after dialysis was determined by fluorescence spectroscopy. An
appropriate correction was made to account for volume changes.
The amount of bound drug was determined by difference (Cb =
Ct - Cf, where Cf is the concentration of bulk drug solution, Ct is
the total drug concentration that enters the membrane, and Cb is
the concentration of drug bound to the nucleic acids). The data
were plotted as a bar graph using Kaleidagraph version 3.5
(Synergy Software). A calibration curve was made before each
experiment by plotting the fluorescence intensities at character-
istic ligand wavelengths versus corresponding ligand concentra-
tions. Ct and Cf were determined on the basis of a calibration curve.
Circular Dichroism (CD) Spectroscopy. All CD experiments
were conducted at 20 °C in cacodylate buffer [10 mM sodium
cacodylate, 0.5 mM EDTA, and 100 mM NaCl (pH 7.0)]. We
observed that less perylene aggregation was observed in cacody-
late buffer than in phosphate buffer. The CD spectra were
recorded as a function of wavelength (200-350 nm) and are
averages of 25 scans. For CD titrations, aliquots of a stock
solution of 3 were added to the preformed quadruplex of 4 (10
μM) to reach the desired concentrations. After each addition, the
solution was mixed by magnetic stirring and then allowed to
equilibrate for 5 min prior to a scan. Data processing was
conducted using Kaleidagraph version 3.5.
3 were added to the 4/TO solution to reach the desired
concentrations and allowed to equilibrate for 4 min prior to
the fluorescence recording (excitation at 501 nm and emission at
510-650 nm). The slit width used was 3.0 mm. The G4DC50
values were determined by fitting the experimental data using
a dose-response curve (Origin version 5.0). G4DC50 represents
the amount of ligand required to displace 50% of bound thiazole
orange from a G-quadruplex target.
Fluorescence Titrations. The titration assays were performed
in a 3 mL quartz cuvette at room temperature (22-24 °C) in
phosphate buffer [8 mM Na2HPO4, 2 mM NaH2PO4, 1 mM
Na2EDTA, and 185 mM NaCl (pH 7.0)] or cacodylate buffer
[10 mM sodium cacodylate, 0.5 mM EDTA, and 100 mM NaCl
(pH 7.0)]. Aliquots of a stock solution of G-quadruplex 4 were
added to a solution of 3 (33 nM in phosphate buffer or 0.25 μM
in cacodylate buffer) and allowed to equilibrate for 5 min before
the fluorescence was recorded (excitation at 500 nm for phos-
phate buffer and 496 nm for cacodylate buffer and emission at
530-750 nm). The data were then processed using Kaleidagraph
version 3.5.
Molecular Modeling. All dockings were performed as blind
dockings (blind docking refers to the use of a grid box that is large
enough to encompass any possible ligand-receptor complex)
using Autodock Vina version 1.0. All rotatable bonds within the
ligand were allowed to rotate freely, and the receptor was
considered rigid. The structure of G-quadruplex 4 was obtained
from Protein Data Bank entry 143D. All ligand structures were
created using Discovery Studio Visualizer version 2.5 and
energetically minimized with the Vega ZZ program53 using a
conjugate gradient method with an SP4 force field. The opti-
mized ligand and G-quadruplex structures were then converted
into proper file formats for AutoDock Vina using Autodock
Tools version 1.5.4.54 The docking validation (correctly identify-
ing the binding site and scoring the receptor-ligand inter-
actions) was confirmed by using AutoDock Vina to accurately
predict the binding of different ligands to their previously
established targets.
’ RESULTS AND DISCUSSION
Ligands that can stabilize G-quadruplexes are currently under
intense investigation because of their possible roles in the
inhibition of telomerase activity. While a number of ligands that
interact with different G-quadruplexes have been discovered in
the past decade, ligands with high selectivity for G-quadruplex
grooves are limited in number.55 Most of the G-quadruplex
ligands contain planar aromatic moieties and stack to G-quad-
ruplexes with moderate binding affinities. Ligands that can bind
the quadruplex in the grooves are scant,56 and not much is known
about their biophysical, biochemical, and structural properties.
Herein, we report a G-quadruplex ligand 3 containing both a
groove binder and an intercalating unit that selectively recognizes
human telomeric DNA 4. The groove widths and depths present
in the quadruplex vary significantly from the groove widths of a
DNA or RNA duplex. Therefore, a higher selectivity for quad-
ruplex structures can be envisioned by targeting the quadruplex
grooves. When this selectivity is combined with the differences in
base stacking surface areas of a G-quadruplex versus a DNA
duplex, ligands with much higher selectivity for the G-quadruplex
can be designed.
Fluorescent Intercalator Displacement (FID) Titrations. All
fluorescence titration experiments were performed at 20 °C in
cacodylate buffer [10 mM sodium cacodylate, 0.5 mM EDTA,
and 100 mM NaCl (pH 7.0)]. The DNA solutions (2.0 mL) were
prepared by mixing G-quadruplex 4 (0.25 μM/strand) and
thiazole orange (TO, 0.50 μM). Aliquots of a stock solution of
Design and Synthesis of the Neomycin-Perylene Con-
jugate (3). Perylene diimide was chosen as an intercalating unit
2842
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
Figure 3. (A) Competition dialysis of various nucleic acids with 1 μM perylene-neomycin conjugate. (B) Difference plot for the perylene-neomycin
conjugate. Different nucleic acids (180 μL, 75 μM per monomeric unit of each polymer) were dialyzed with an aqueous solution (400 mL) of 3 (1 μM) in
8 mM Na2HPO4, 2 mM NaH2PO4, 1 mM Na2EDTA, and 185 mM NaCl (pH 7.0) for 24 h. The accumulated ligand in each dialysis tube was analyzed
using fluorescence, and the bound concentrations were determined with the help of a calibration curve.
because its derivatives are known to bind human telomeric
sequences such as 4 and inhibit telomerase activity.8,57-59
that represent a wide variety of structures were screened against
conjugate 3. The nucleic acids (whose formation was checked
using CD and UV spectroscopy) were taken in dialysis tubes and
then immersed in a common dialysate solution of 3. The ligands
were allowed to slowly diffuse into the dialysis tubes over a period
of 24 h. The subsequent determinations of bound ligand con-
centrations (after the surfactant treatment) were conducted
using fluorescence. The results from competition dialysis are
presented in Figure 3. As seen in panels A and B, ligand 3 exhibits
the strongest preference (bound concentration of 4 μM) for
human telomeric sequence {AG3 quadruplex = 50-d[AGGG-
(TTAGGG)3], 4} among all the nucleic acids. 3 binds to poly-
A
recent study showed that perylene diimide-containing polya-
mines bind to G-quadruplex DNA with polyamines residing in
the grooves.60 Perylene diimide was also chosen as a stacking
moiety because side chains can be conveniently introduced into
the scaffold without perturbing perylene’s steric and electronic
properties to a great extent. In this work, neomycin, an amino
sugar, was attached to perylene diimide, making the resulting
molecule 3 more water-soluble (hydrophilic) and thus easier to
study. Because neomycin possesses moderate binding affinity for
G-quadruplex DNA,41 conjugation of neomycin with perylene
diimide should yield a more potent G-quadruplex binding ligand
that has a dual recognition mode. We also envisioned that
conjugation of neomycin to perylene diimide could alter the
known aggregation properties of perylene in aqueous solutions.61
In addition, 3, containing two neomycin units, is also expected to
incorporate quadruplex groove binding. The synthesis of the
desired conjugate was performed in a manner where both
neomycin and perylene moieties were modified independently
and then linked together using a thiourea linkage (Scheme 1).
Compound 2 was synthesized by reacting 3,4,9,10-perylenete-
tracarboxylicdianhydride 5 with BocNH(CH2)4NH2 in pyridine
under reflux followed by removal of Boc protecting groups using
trifluoroacetic acid (TFA). Neomycin isothiocyanate 7 was
prepared using previously established procedures.38 Conjugate 3
was synthesized by coupling 7 with 2 followed by removal of Boc
groups using a TFA/CH2Cl2 mixture in moderate yields. Alter-
natively, compound 2 can be synthesized in one step from 5,
followed by coupling of 2 to 7 to give intermediate 8, which can
be then deprotected to give target ligand 3 in better yields
(Scheme S1 of the Supporting Information).
nucleotide duplex and triplex RNA structures [poly(rA) poly-
3
r(U) and polyr(A) 2polyr(U)] with observed bound concentrations
3
of 2.4 and 1.6 μM, respectively. The binding of 3 to single-
stranded DNA and RNA was not observed. Conjugate 3 does
bind to native DNA duplexes (Clostridium perfringens and
Micrococcus lysodeikticus), a hybrid duplex [poly(dA) poly-
3
(rU)], and 16S A-site rRNA; however, the binding preference
of 3 (e1 μM) for these nucleic acids is much weaker than the
binding of 3 to 4. Even more significantly, 3 shows an only weak
preference for two of the other intermolecular G-quadruplexes
(GTG quadruplex 50-d[G10T4G10] and TGT quadruplex 50-
d[T10G4T10]), even though neomycin has been shown by us
to have a weak affinity for 4.41 It is also noteworthy that 3 also
binds with weak affinity (∼1.2 μM) to the i-motif, a quadruplex
structure formed between cytosines.
The competition dialysis assay clearly indicates preferential
binding of conjugate 3 to human telomeric G-quadruplex DNA
4. Moderate affinities of 3 for duplex and triplex RNA sequences
can be attributed to the neomycin moiety, which is a strong
binder of ribonucleic acids (A-form structures).31 Nevertheless,
conjugate 3 stands out for its predilection for human telomeric
intramolecular G-quadruplex DNA 4. It is noteworthy that 3
binds to intermolecular G-quadruplexes with different structures
such as TGT quadruplex (parallel) and GTG quadruplex
(antiparallel) approximately 40-fold less than to 4, making it
specific for the recognition of human telomeric quadruplex DNA.
The relative trend of preference for human telomeric DNA by
3 is consistent with the similar competition dialysis findings in
Competition Dialysis Assay. Competition dialysis, an exten-
sion of equilibrium dialysis, has been developed by Chaires for
the rapid screening of a ligand for its nucleic acid structure
selectivity.62 While a more generalized protocol was developed a
decade ago,63 a recent report has also highlighted its applications
in sorting out G-quadruplex specific ligands.64 To promptly
assess the binding preference of conjugate 3, competition dialysis
experiments were conducted. In our experiment, 20 nucleic acids
2843
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
Figure 4. FID titration plots for ligands (A) 1, (B) 2, and (C) 3 in 10 mM sodium cacodylate, 0.5 mM EDTA, and 100 mM NaCl (pH 7.0). For each
titration, human telomeric DNA (0.25 μM) was mixed with thiazole orange (0.50 μM) in a 1:2 ratio. The bound thiazole orange was then gradually
displaced with test ligand until no more fluorescence change was observed. The obtained raw data were fitted with the dose-response curve using Origin
version 5.0. All experiments were conducted at 20 °C. Thiazole orange excitation was achieved at 501 nm, and the resulting emission spectra were
recorded between 510 and 650 nm.
Table 1. G4DC50 Values for 1-3 Derived from the FID
the Chaires laboratory where another perylene derivative, PI-
PER, was found to have the strongest preference for the human
Titrations As Shown in Figure 4
telomeric DNA.64
ligand
G4DC50 (μM)
Fluorescent Intercalator Displacement Assay. Competi-
tion dialysis described above unambiguously revealed the bind-
ing preference of 3 for human telomeric DNA 4. However, the
contribution of neomycin could not be realized from this
technique given the lack of a chromophore in neomycin. We,
therefore, utilized a fluorescent intercalator displacement (FID)
assay65 to compare the binding of 3 to 4 with its two parent
constituents, neomcyin (1) and perylene diamine (2). The FID
assay allowed us to gauge the contributions of each pharmaco-
phore in conjugate 3. The assay was originally used for identify-
ing DNA duplex binding ligands66 but has recently also been
extended for screening quadruplex specific ligands.67 In our
experiments, thiazole orange (TO) was first bound to G-quad-
ruplex 4, and the prebound TO was successively displaced from 4
1
2
3
9.48 ( 0.75
0.18 ( 0.02
0.04 ( 0.01
by addition of the ligand of interest with increasing amounts
(Figure 4). The amount of ligand needed to displace 50% of the
prebound TO is defined as G4DC50. The G4DC50 values for 1-3
are listed in Table 1.
The G4DC50 value of conjugate 3 determined by the FID assay
was 4.1-fold lower than that of 2 and 220-fold lower than that of
neomycin, suggesting that the strong binding preference of 3 for
G-quadruplex 4 predominantly results from the affinity of the
perylene diimide moiety. A direct comparison between affinities
2844
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
Figure 6. Absorption spectra of solutions of the perylene-neomycin
conjugate in 8 mM Na2HPO4, 2 mM NaH2PO4, 1 mM Na2EDTA, and
185 mM NaCl (pH 7.0): no quadruplex (black line) or in the presence of
quadruplex 50-d[AG3(T2AG3)3]-30 (blue line). The spectrum was
recorded at room temperature and corrected with buffer absorption.
Figure 5. UV thermal denaturation profiles of G-quadruplex 4 at
260 nm in 8 mM Na2HPO4, 2 mM NaH2PO4, 1 mM Na2EDTA, and
185 mM NaCl (pH 7.0): (a) no ligand, (b) 3 (2 μM), (c) 3 (10 μM), and
(d) 3 (1 μM). The samples were heated to 95 °C and cooled to 20 °C at
a rate of 0.2 °C/min. All UV melting experiments were monitored under
260 nm, and data points were recorded every 1.0 °C. For the Tm
determinations, derivatives were used.
and G4DC50 values cannot be made because the ligands have
different stoichiometries of binding to the quadruplex. To
exclude the possibility that the observed fluorescence decrease
results from quenching by polyamines, we have previously shown
that nonspecific binding of polyamines to a G-quadruplex does
not result in a fluorescence change.41
The results obtained from FID assays were also consistent
with the observation that both perylene amine 2 and conjugate 3
have a strong preference for binding to human telomeric DNA
(Table 1). This also suggests that low G4DC50 values obtained for
ligand 3 are also suggestive of a large contribution from the
perylene moiety of the ligand to human telomeric DNA binding.
The high concentrations required by neomycin to displace the
probe are consistent with its low affinity for human telomeric
DNA as also determined by ITC experiments.41
UV Thermal Denaturation Studies. The effect of 3 on the
stabilization of G-quadruplex 4 was also investigated by UV
thermal denaturation. The UV thermal denaturation curves of
G-quadruplex 4 were recorded in the absence and presence of 3
(Figure 5; also see Figure S2 of the Supporting Information for
derivative plots). The temperature at which 4 dissociates into
random coils was noted as 64 °C in the absence of ligand. The
denaturation temperature of 4 increased as a function of an
increasing concentration of 3. For instance, increments of 6, 12,
and 25 °C were observed in the presence of 1, 2, and 10 μM 3,
respectively. The observed thermal stabilization of 4 by 3, in
addition to the binding studies, suggests strong binding of 3 with
G-quadruplex 4.
UV-Vis Absorption Studies. The UV-vis absorption spec-
tra of conjugate 3 in the absence or presence of 4 were recorded.
In the presence of 4, the absorption spectrum of 3 shows a slight
red shift (∼5 nm) of λmax at 504 nm, while the other major
absorption peak at 550 nm remains unchanged (Figure 6). The
peak at 504 nm also undergoes a marked hypochromic change
(∼14%). Such changes in the absorption spectra of perylene
derivatives upon binding to G-quadruplex DNA have also been
reported by other groups,59,60,68,69 indicating the interactions
Figure 7. CD titration of G-quadruplex DNA 4 with 3 in 10 mM
sodium cacodylate, 0.5 mM EDTA, and 100 mM NaCl (pH 7.0). Ligand
was successively added to the DNA solution and the mixture stirred
with a magnetic stirrer followed by equilibration for 5 min. Each CD
measurement represents an average of 25 scans. The titration was
performed at 20 °C.
between 3 and 4 occur through an intercalative or stacking
fashion. The interaction of conjugate 3 with 4 is also time-
dependent. The absorbance spectrum of 3 undergoes a gradual
change after the addition of 4, typically requiring 2 h to achieve
equilibrium.
The UV absorption studies show a marked hypochromic shift
in the perylene absorption at 500 nm, similar to previous
studies,60 but in contrast with other results.68 Such hypochromic
changes in absorbance have been seen in drug-quadruplex
stacking interactions. For instance, numerous porphyrin deriva-
tives that bind to G-quadruplexes by stacking interactions have
pronounced hypochromicities upon ligand binding.70-72 The
apparent differences in the absorption behavior between differ-
ent perylene ligands can be attributed to the different ligand
aggregation properties, concentrations of ligands used, and
2845
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
Figure 8. Fluorescence titration of 3 (33 nM) with 4 in 8 mM Na2HPO4, 2 mM NaH2PO4, 1 mM Na2EDTA, and 185 mM NaCl (pH 7.0). The
concentration of the quadruplex DNA stock solution was 60 μM per strand for titration. The concentration of DNA in the solution varied with each
addition of DNA, ranging from 10-9 to 10-5 M. Ligand 3 was excited at a wavelength of 500 nm, and the resulting emission was recorded from 530 to
750 nm. The raw data were plotted using Kaleidagraph version 4.0. The titration was performed at room temperature (∼22 °C).
solution conditions employed. Unlike the perylene ligands
previously reported,68 3 can also be classified as a more “rigid”
perylene derivative because the movement of the perylene is
restricted by groove binding of neomycin. The perylene deriva-
tives previously reported have much smaller linkers to reach the
qudruplex grooves. In contrast, conjugate 3 has a longer linker
that allows binding of two neomycins (the two neomycin units
contribute to a total of 12 amino groups). A recent report aimed
at studying the effect of straight chain polyamine side chains has
proposed a similar groove binding of polyamine ends.60
Circular Dichroism (CD) Spectroscopic Studies. To further
characterize the binding of 3 with G-quadruplex 4, CD titration
experiments were conducted. The CD spectra of 4 were recorded
after aliquots of 3 were successively added to a solution of 4. In
the absence of 3, the CD spectrum of 4 shows a large positive
band at 295 nm and a negative band at 260 nm, suggesting an
antiparallel quadruplex structure in the presence of Naþ, which is
consistent with previous studies73 (Figure 7). The absence of any
shoulder peaks at ∼280 nm is also indicative of single predomi-
nant antiparallel quadruplex species in the solution. Addition of 3
to the solution resulted in a decrease in the CD intensity of the
band at 295 nm and an increase in the CD intensity at 260 nm,
indicating interaction of 3 with 4. The binding of 3 also resulted
in a slight red shift of the band at 295 nm in the spectra of 4.
Taken together, the changes in the CD intensity of the 295 and
260 nm bands as well the slight red shift in the band at 295 nm
suggest the complexation of 3 with 4. Cacodylate buffer was used
in the CD titration experiments because the observed aggrega-
tion of 3 was less severe in cacodylate buffer than in phosphate
buffer. Conjugate 3 was observed to coagulate into dots of red
colored solids at high concentrations. The results presented here
reflect the CD scans of the ligand-DNA complex free from such
coagulation effects. Aggregation is a commonly observed prop-
erty of perylene-based compounds and has also been proposed to
mediate G-quadruplex binding.69
overall conformation of G-quadruplex 4 as the antiparallel
structure is maintained throughout the titration. We, however,
did not observe strong induced CD signals in the perylene
absorption region. This observation could be attributed to the
following. (1) It could result from the low concentration (e5
μM) of 3 used in the CD titration experiments. Precipitation of
the ligand-quadruplex complex was observed, and consequently,
measurements become more difficult once the concentration of 3
is >5 μM. At such a low concentration, the observed CD
absorption of the perylene diimide moiety was very weak and
had a high N/S (noise-to-signal) ratio, and reliable readings were
therefore not possible in the perylene absorption region. Pre-
vious reports of induced CD effects of perylene diimide deriva-
tives used a concentration of 20 μM.60 Hurley and co-workers
also reported a very weak induced CD signal of a perylene
derivative upon binding to G-quadruplexes.8 (2) Binding of 3
with 4 may not significantly alter its achirality. A previous report
also suggests that quadruplex intercalation and/or stacking
may not necessarily lead to induced CD changes in the ligand
absorption region, and likely depends on the experimental
conditions employed.71
Fluorescence Titration Studies. It has been reported that the
fluorescence of perylene derivatives can be quenched upon the
addition of nucleic acids, which indicates binding.68,74 Therefore,
the fluorescence of 3 was recorded after addition of aliquots of a
stock solution of 4 at 20 °C. The fluorescence of 3 decreased as a
function of the increasing concentration of 4, suggesting an
interaction between 3 and 4. The fluorescence of 3 was almost
completely quenched at a very low DNA-to-ligand ratio. The
saturation of 4 with 3 was assumed when the fluorescence of 3
did not change with a change in the DNA-to-ligand ratios
(Figure 8). Attempts to fit the fluorescence titration data to
obtain a binding constant were unsuccessful.
The binding of 3 with 4 resulted in rather rapid quenching of
the fluorescence of 3. Such quenching results from the transfer of
energy of the perylene moiety to the stacked bases of DNA.74
The fluorescence signal was completely quenched at a very low
ligand-to-G-quadruplex ratio, indicating the binding of approxi-
mately three ligands per quadruplex (Figure S3 of the Supporting
Information). A Scatchard plot of the quenching data yields a
With an increase in the concentration of 3, a continuous
decrease in the CD ellipticity was observed at 295 nm. This
change in intensity, accompanied by an increase in ellipticity at
260 nm but without any significant red or blue shifts in the
spectrum, suggests that the binding of 3 does not perturb the
2846
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
’ CONCLUSIONS
In this study, we report that a novel perylene-neomycin
conjugate 3 has been synthesized by covalently tethering neo-
mycin with perylene diamine 2. Conjugate 3 exhibited a con-
siderable preference for human telomeric DNA as opposed to
various other nucleic acids. The UV thermal melting and
absorption binding studies clearly showed the binding of 3 to
human telomeric G-quadruplex 4, and the binding event likely
takes place via a dual recognition mode in which the perylene
portion of 3 binds via quadruplex stacking. FID experiments
showed that 3 supersedes the binding of individual units,
indicating the conjugation yields a much improved G-quadruplex
binding ligand. CD experiments also revealed the interaction of 3
with 4; however, the exact mode of binding could not be
determined by this method because of the absence of induced
CD changes. The fluorescence titrations affirmed the binding of
the perylene moiety to the quadruplex. Ligand aggregation at
higher concentrations did not allow any direct measurement of
association constants using calorimetric techniques. The docking
studies indicated stacking of the perylene moiety of 3 with the
thymine bases, while the two neomycin units occupy the wide
and one of the medium-sized grooves. Further structural studies
will be needed to completely characterize the binding, but the
severe aggregation and precipitation at higher concentrations will
certainly make the structural studies a very challenging endeavor.
In summary, our results indicate that dual recognition conjugates
(base stacking and groove recognition) can be designed to
increase ligand specificity for nucleic acids such as the DNA
quadruplex. This approach, as outlined here for the development
of G-quadruplex ligands, can be applied to the design of ligands
for selective targeting of any conceivable nucleic acid structure
where base stacking and groove formation are present.
Figure 9. Computer model of 3 docked into human telomeric DNA 4.
Conjugate 3 is colored green.
nonlinear plot. Unfortunately, an association constant could not
be determined using this analysis.
Docking Studies. Docking studies were conducted using
newly introduced Autodock Vina.75 The DNA structure was
retrieved from the Protein Data Bank (entry 143D). The
structure of conjugate 3 was energetically minimized with
VegaZZ53 using the conjugate gradient method. The possible
mode of binding was determined by molecular modeling.
The perylene moiety of 3 makes stacking interactions with the
flipped thymine base (T17) and also makes van der Waals
interactions with another thymine base (T5) of G-quadruplex
4 (Figure 9 and Figure S4 of the Supporting Information).
The neomycin moieties reside in two grooves. One fits into a
larger part of the wide groove, while the other interacts with one
of the medium grooves falling between the narrow and wide
grooves. This is a conceivable binding mode because neomycin
is unlikely to have stacking interactions given its nonplanar
structure.
’ ASSOCIATED CONTENT
S
Supporting Information. UV thermal denaturation pro-
b
files and UV/NMR/MALDI spectra of new compounds. This
material is available free of charge via the Internet at http://
pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: dparya@clemson.edu. Phone: (864) 656-1106. Fax:
(864) 656-6613.
The docking results suggest that neomycin occupies two
quadruplex grooves (wide and medium-sized) while the perylene
moiety of 3 acts as bridge between these two grooves. The
bridging perylene moiety forms a stacking interaction with
thymine (T17) and makes contacts with another thymine (T5)
via van der Waals interactions. The neomycin binding in the wide
groove showed more contacts than the one binding in the
medium-sized groove. Interestingly, docking of neomycin alone
to human telomeric quadruplex 4 shows that neomycin’s most
favorable binding occurs in the wide groove.41 A similar observa-
tion has been seen for binding of neomycin to an antiparallel
Present Addresses
†Department of Chemistry, University of the Pacific, 3601 Pacific
Ave., Stockton, CA 95211.
Funding Sources
We thank the National Institutes of Health (R15CA125724) for
financial support.
’ ACKNOWLEDGMENT
We thank David Hyde-Volpe for helpful discussions about
modeling studies.
ꢀ
quadruplex resulting from the telomeric sequence of Oxytricha
nova.41 Further structural studies are needed to determine the
actual mode of binding of 3 with 4. The weak stacking interac-
tions present between the aromatic platform of the perylene part
of the conjugate and the thymine base (T17) may also contribute
to the observed absence of a sharp induced CD.
’ ABBREVIATIONS
FID, fluorescence intercalator displacement; UV, ultraviolet;
CD, circular dichroism; DI, deionized; TFA, trifluoroacetic acid.
2847
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
’ REFERENCES
(23) Shaw, N. N., Xi, H., and Arya, D. P. (2008) Molecular
Recognition of a DNA:RNA Hybrid: Sub-Nanomolar Binding by a
Neomycin-Methidium Conjugate. Bioorg. Med. Chem. Lett. 18, 4142–
4145.
(1) Burge, S., Parkinson, G. N., Hazel, P., Todd, A. K., and Neidle, S.
(2006) Quadruplex DNA: Sequence, Topology and Structure. Nucleic
Acids Res. 34, 5402–5415.
(24) Arya, D. P., Shaw, N., and Xi, H. (2007) Novel Targets for
Aminoglycosides. In Aminoglycoside Antibiotics: From chemical biology to
drug discovery (Arya, D. P., Ed.) pp 289-314, Wiley, New York.
(25) Arya, D. P. (2005) Aminoglycoside-Nucleic Acid Interactions:
The Case for Neomycin. Top. Curr. Chem. 253, 149–178.
(26) Arya, D. P., Coffee, R. L., Jr., and Charles, I. (2001) Neomycin-
Induced Hybrid Triplex Formation. J. Am. Chem. Soc. 123, 11093–
11094.
(27) Arya, D. P., Micovic, L., Charles, I., Coffee, R. L., Jr., Willis, B.,
and Xue, L. (2003) Neomycin Binding to Watson-Hoogsteen (W-H)
DNA Triplex Groove: A Model. J. Am. Chem. Soc. 125, 3733–3744.
(28) Arya, D. P., Coffee, R. L., Jr., Willis, B., and Abramovitch, A. I.
(2001) Aminoglycoside-Nucleic Acid Interactions: Remarkable Stabili-
zation of DNA and RNA Triple Helices by Neomycin. J. Am. Chem. Soc.
123, 5385–5395.
(29) Arya, D. P., and Coffee, R. L., Jr. (2000) DNA Triple Helix
Stabilization by Aminoglycoside Antibiotics. Bioorg. Med. Chem. Lett.
10, 1897–1899.
(30) Xi, H., Gray, D., Kumar, S., and Arya, D. P. (2009) Molecular
Recognition of Single-Stranded RNA: Neomycin Binding to Poly(A).
FEBS Lett. 583, 2269–2275.
(31) Arya, D. P., Xue, L., and Willis, B. (2003) Aminoglycoside
(Neomycin) Preference is for A-Form Nucleic Acids, Not just RNA:
Results from a Competition Dialysis Study. J. Am. Chem. Soc. 125,
10148–10149.
(2) Keniry, M. A. (2000) Quadruplex Structures in Nucleic Acids.
Biopolymers 56, 123–146.
(3) Blackburn, E. H. (2001) Switching and Signaling at the Telo-
mere. Cell 106, 661–673.
(4) Autexier, C., and Lue, N. F. (2006) The Structure and Function
of Telomerase Reverse Transcriptase. Annu. Rev. Biochem. 75, 493–517.
(5) De Cian, A., Lacroix, L., Douarre, C., Temime-Smaali, N.,
Trentesaux, C., Riou, J. F., and Mergny, J. L. (2008) Targeting
Telomeres and Telomerase. Biochimie 90, 131–155.
(6) Zahler, A. M., Williamson, J. R., Cech, T. R., and Prescott, D. M.
(1991) Inhibition of Telomerase by G-Quartet DMA Structures. Nature
350, 718–720.
(7) Mergny, J. L., and Helene, C. (1998) G-Quadruplex DNA: A
Target for Drug Design. Nat. Med. 4, 1366–1367.
(8) Fedoroff, O. Y., Salazar, M., Han, H., Chemeris, V. V., Kerwin, S. M.,
and Hurley, L. H. (1998) NMR-Based Model of a Telomerase-Inhibiting
Compound Bound to G-Quadruplex DNA. Biochemistry 37, 12367–12374.
(9) Sun, D., Thompson, B., Cathers, B. E., Salazar, M., Kerwin, S. M.,
Trent, J. O., Jenkins, T. C., Neidle, S., and Hurley, L. H. (1997)
Inhibition of Human Telomerase by a G-Quadruplex-Interactive Com-
pound. J. Med. Chem. 40, 2113–2116.
(10) Shin-ya, K., Wierzba, K., Matsuo, K., Ohtani, T., Yamada, Y.,
Furihata, K., Hayakawa, Y., and Seto, H. (2001) Telomestatin, a Novel
Telomerase Inhibitor from Streptomyces anulatus. J. Am. Chem. Soc.
123, 1262–1263.
(32) Napoli, S., Carbone, G. M., Catapano, C. V., Shaw, N., and Arya,
D. P. (2005) Neomycin Improves Cationic Lipid-Mediated Transfec-
tion of DNA in Human Cells. Bioorg. Med. Chem. Lett. 15, 3467–3469.
(33) Arya, D. P., Coffee, R. L., Jr., and Xue, L. (2004) From Triplex
to B-Form Duplex Stabilization: Reversal of Target Selectivity by
Aminoglycoside Dimers. Bioorg. Med. Chem. Lett. 14, 4643–4646.
(34) Arya, D. P., and Willis, B. (2003) Reaching into the Major
Groove of B-DNA: Synthesis and Nucleic Acid Binding of a Neomycin-
Hoechst 33258 Conjugate. J. Am. Chem. Soc. 125, 12398–12399.
(35) Willis, B., and Arya, D. P. (2010) Triple Recognition of B-DNA
by a Neomycin-Hoechst 33258-Pyrene Conjugate. Biochemistry 49,
452–469.
(11) Hounsou, C., Guittat, L., Monchaud, D., Jourdan, M., Saettel,
N., Mergny, J. L., and Teulade Fichou, M. P. (2007) G-Quadruplex
Recognition by Quinacridines: A SAR, NMR, and Biological Study.
ChemMedChem 2, 655–666.
(12) Burger, A. M., Dai, F., Schultes, C. M., Reszka, A. P., Moore, M. J.,
Double, J. A., and Neidle, S. (2005) The G-Quadruplex-Interactive Molecule
BRACO-19 Inhibits Tumor Growth, Consistent with Telomere Targeting
and Interference with Telomerase Function. Cancer Res. 65, 1489–1496.
(13) Anantha, N. V., Azam, M., and Sheardy, R. D. (1998) Porphyrin
Binding to Quadruplexed T4G4. Biochemistry 37, 2709–2714.
(14) Jantos, K., Rodriguez, R., Ladame, S., Shirude, P. S., and
Balasubramanian, S. (2006) Oxazole-Based Peptide Macrocycles: A
New Class of G-Quadruplex Binding Ligands. J. Am. Chem. Soc.
128, 13662–13663.
(36) Charles, I., and Arya, D. P. (2005) Synthesis of Neomycin-
DNA/peptide Nucleic Acid Conjugates. J. Carbohydr. Chem. 24,
145–160.
(15) Pilch, D. S., Barbieri, C. M., Rzuczek, S. G., LaVoie, E. J., and Rice,
J. E. (2008) Targeting Human Telomeric G-Quadruplex DNA with
Oxazole-Containing Macrocyclic Compounds. Biochimie 90, 1233–1249.
(16) Monchaud, D., and Teulade Fichou, M. P. (2008) A Hitchhi-
ker’s Guide to G-Quadruplex Ligands. Org. Biomol. Chem. 6, 627–636.
(17) Campbell, N. H., Patel, M., Tofa, A. B., Ghosh, R., Parkinson,
G. N., and Neidle, S. (2009) Selectivity in Ligand Recognition of
G-Quadruplex Loops. Biochemistry 48, 1675–1680.
(18) Haider, S. M., Parkinson, G. N., and Neidle, S. (2003) Structure
of a G-quadruplex-Ligand Complex. J. Mol. Biol. 326, 117–125.
(19) Cosconati, S., Marinelli, L., Trotta, R., Virno, A., Mayol, L.,
Novellino, E., Olson, A. J., and Randazzo, A. (2009) Tandem Application
of Virtual Screening and NMR Experiments in the Discovery of Brand New
DNA Quadruplex Groove Binders. J. Am. Chem. Soc. 131, 16336–16337.
(20) Xue, L., Xi, H., Kumar, S., Gray, D., Davis, E., Hamilton, P.,
Skriba, M., and Arya, D. P. (2010) Probing the Recognition Surface of a
DNA Triplex: Binding Studies with Intercalator Neomycin Conjugates.
Biochemistry 49, 5540–5552.
(37) Charles, I., Xi, H., and Arya, D. P. (2007) Sequence-Specific
Targeting of RNA with an Oligonucleotide-Neomycin Conjugate.
Bioconjugate Chem. 18, 160–169.
(38) Charles, I., Xue, L., and Arya, D. P. (2002) Synthesis of
Aminoglycoside-DNA Conjugates. Bioorg. Med. Chem. Lett. 12,
1259–1262.
(39) Arya Dev, P. (2011) New Approaches toward Recognition of
Nucleic Acid Triple Helices. Acc. Chem. Res. 44, 134–146.
(40) Willis, B., and Arya, D. P. (2006) An Expanding View of
Aminoglycoside-Nucleic Acid Recognition. Adv. Carbohydr. Chem.
Biochem. 60, 251–302.
(41) Ranjan, N., Andreasen, K. F., Kumar, S., Hyde-Volpe, D., and
Arya, D. P. (2010) Aminoglycoside Binding to Oxytricha Nova Telo-
meric DNA. Biochemistry 49, 9891–9903.
(42) Dominick, P. K., Keppler, B. R., Legassie, J. D., Moon, I. K., and
Jarstfer, M. B. (2004) Nucleic Acid-Binding Ligands Identify New Mecha-
nisms to Inhibit Telomerase. Bioorg. Med. Chem. Lett. 14, 3467–3471.
(43) Dai, J., Carver, M., and Yang, D. (2008) Polymorphism of
Human Telomeric Quadruplex Structures. Biochimie 90, 1172–1183.
(44) Wang, Y., and Patel, D. J. (1993) Solution Structure of the Human
Telomeric Repeat d[AG3(T2AG3)3] G-Tetraplex. Structure 1, 263–282.
(45) Parkinson, G. N., Lee, M. P. H., and Neidle, S. (2002) Crystal
Structure of Parallel Quadruplexes from Human Telomeric DNA.
Nature 417, 876–880.
(21) Xue, L., Charles, I., and Arya, D. P. (2002) Pyrene-Neomycin
Conjugate: Dual Recognition of a DNA Triple Helix. Chem. Commun.
1, 70–71.
(22) Arya, D. P., Xue, L., and Tennant, P. (2003) Combining the
Best in Triplex Recognition: Synthesis and Nucleic Acid Binding of a
BQQ-Neomycin Conjugate. J. Am. Chem. Soc. 125, 8070–8071.
2848
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849

Biochemistry
ARTICLE
(46) Dai, J., Punchihewa, C., Ambrus, A., Chen, D., Jones, R. A., and
Yang, D. (2007) Structure of the Intramolecular Human Telomeric
G-Quadruplex in Potassium Solution: A Novel Adenine Triple Forma-
tion. Nucleic Acids Res. 35, 2440–2450.
(65) Tse, W. C., and Boger, D. L. (2004) A Fluorescent Intercalator
Displacement Assay for Establishing DNA Binding Selectivity and
Affinity. Acc. Chem. Res. 37, 61–69.
(66) Boger, D. L., Fink, B. E., Brunette, S. R., Tse, W. C., and
Hedrick, M. P. (2001) A Simple, High-Resolution Method for Establish-
ing DNA Binding Affinity and Sequence Selectivity. J. Am. Chem. Soc.
123, 5878–5891.
(67) Monchaud, D., Allain, C., and Teulade-Fichou, M. (2006)
Development of a Fluorescent Intercalator Displacement Assay (G4-
FID) for Establishing Quadruplex-DNA Affinity and Selectivity of
Putative Ligands. Bioorg. Med. Chem. Lett. 16, 4842–4845.
(68) Kern, J. T., and Kerwin, S. M. (2002) The Aggregation and
G-Quadruplex DNA Selectivity of Charged 3,4,9,10-Perylenetetracar-
boxylic Acid Diimides. Bioorg. Med. Chem. Lett. 12, 3395–3398.
(69) Kerwin, S. M., Chen, G., Kern, J. T., and Thomas, P. W. (2002)
Perylene Diimide G-Quadruplex DNA Binding Selectivity is Mediated
by Ligand Aggregation. Bioorg. Med. Chem. Lett. 12, 447–450.
(70) Wei, C., Jia, G., Yuan, J., Feng, Z., and Li, C. (2006) A
Spectroscopic Study on the Interactions of Porphyrin with G-Quad-
ruplex DNAs. Biochemistry 45, 6681–6691.
(71) Freyer, M. W., Buscaglia, R., Kaplan, K., Cashman, D., Hurley,
L. H., and Lewis, E. A. (2007) Biophysical Studies of the c-MYC NHE
III1 Promoter: Model Quadruplex Interactions with a Cationic Por-
phyrin. Biophys. J. 92, 2007–2015.
(72) Haq, I., Trent, J. O., Chowdhry, B. Z., and Jenkins, T. C. (1999)
Intercalative G-Tetraplex Stabilization of Telomeric DNA by a Cationic
Porphyrin. J. Am. Chem. Soc. 121, 1768–1779.
(73) Vorlíꢁckovꢂa, M., Chlꢂadkovꢂa, J., Kejnovskꢂa, I., Fialovꢂa, M., and
Kypr, J. (2005) Guanine Tetraplex Topology of Human Telomere DNA
is Governed by the Number of (TTAGGG) Repeats. Nucleic Acids Res.
33, 5851–5860.
(74) Bevers, S., Schutte, S., and McLaughlin, L. W. (2000) Naphtha-
lene- and Perylene-Based Linkers for the Stabilization of Hairpin
Triplexes. J. Am. Chem. Soc. 122, 5905–5915.
(47) Luu, K. N., Phan, A. T., Kuryavyi, V., Lacroix, L., and Patel, D. J.
(2006) Structure of the Human Telomere in Kþ Solution: An Intra-
molecular (3þ1) G-Quadruplex Scaffold. J. Am. Chem. Soc. 128,
9963–9970.
ꢁ
(48) Renꢁciuk, D., Kejnovskꢂa, I., Skolꢂakovꢂa, P., Bednꢂaꢁrovꢂa, K.,
Motlovꢂa, J., and Vorlíꢁckovꢂa, M. (2009) Arrangements of Human
Telomere DNA Quadruplex in Physiologically Relevant Kþ Solutions.
Nucleic Acids Res. 37, 6625–6634.
(49) Xu, Y., Noguchi, Y., and Sugiyama, H. (2006) The New Models
of the Human Telomere d[AGGG(TTAGGG)3] in Kþ Solution. Bioorg.
Med. Chem. 14, 5584–5591.
(50) Li, J., Correia, J., Wang, L., Trent, J. O., and Chaires, C. B.
(2005) Not so Crystal Clear: The Structure of the Human Telomere
G-Quadruplex in Solution Differs from that Present in a Crystal. Nucleic
Acids Res. 33, 4649–4659.
(51) Hudson, J. S., Brooks, S. C., and Graves, D. E. (2009)
Interactions of Actinomycin D with Human Telomeric G-Quadruplex
DNA. Biochemistry 48, 4440–4447.
(52) Kaiser, M., Cian, A. D., Sainlos, M., Renner, C., Mergny, J., and
Teulade-Fichou, M. (2006) Neomycin-Capped Aromatic Platforms:
Quadruplex DNA Recognition and Telomerase Inhibition. Org. Biomol.
Chem. 4, 1049–1057.
(53) Pedretti, A., Villa, L., and Vistoli, G. (2004) VEGA-an Open
Platform to Develop Chemo-Bio-Informatics Applications, using Plug-
in Architecture and Script Programming. J. Comput.-Aided Mol. Des.
18, 167–173.
(54) Sanner, M. F. (1999) Python: A Programming Language for
Software Integration and Development. J. Mol. Graphics Modell.
17, 57–61.
(55) Dixon, I. M., Lopez, F., Tejera, A. M., Estꢀeve, J., Blasco, M. A.,
Pratviel, G., and Meunier, B. (2007) A G-Quadruplex Ligand with 10000-
Fold Selectivity Over Duplex DNA. J. Am. Chem. Soc. 129, 1502–1503.
(56) Cosconati, S., Marinelli, L., Trotta, R., Virno, A., De Tito, S.,
Romagnoli, R., Pagano, B., Limongelli, V., Giancola, C., Baraldi, P. G.,
Mayol, L., Novellino, E., and Randazzo, A. (2010) Structural and
Conformational Requisites in DNA Quadruplex Groove Binding: An-
other Piece to the Puzzle. J. Am. Chem. Soc. 132, 6425–6433.
(57) Tuntiwechapikul, W., Lee, J. T., and Salazar, M. (2001) Design
and Synthesis of the G-Quadruplex-Specific Cleaving Reagent Perylene-
(75) Trott, O., and Olson, A. J. (2010) AutoDock Vina: Improving
the Speed and Accuracy of Docking with a New Scoring Function,
Efficient Optimization, and Multithreading. J. Comput. Chem. 31,
455–461.
EDTA Iron(II). J. Am. Chem. Soc. 123, 5606–5607.
3
(58) Tuntiwechapikul, W., and Salazar, M. (2001) Cleavage of
Telomeric G-Quadruplex DNA with Perylene-EDTA Fe(II). Biochem-
3
istry 40, 13652–13658.
(59) Rossetti, L., Franceschin, M., Bianco, A., Ortaggi, G., and
Savino, M. (2002) Perylene Diimides with Different Side Chains are
Selective in Inducing Different G-Quadruplex DNA Structures and in
Inhibiting Telomerase. Bioorg. Med. Chem. Lett. 12, 2527–2533.
(60) Micheli, E., Lombardo, C. M., D’Ambrosio, D., Franceschin,
M., Neidle, S., and Savino, M. (2009) Selective G-Quadruplex Ligands:
The Significant Role of Side Chain Charge Density in a Series of
Perylene Derivatives. Bioorg. Med. Chem. Lett. 19, 3903–3908.
(61) Franceschin, M., Pascucci, E., Alvino, A., D’Ambrosio, D.,
Bianco, A., Ortaggi, G., and Savino, M. (2007) New Highly Hydro-
soluble and Not Self-Aggregated Perylene Derivatives with Three and
Four Polar Side-Chains as G-Quadruplex Telomere Targeting Agents
and Telomerase Inhibitors. Bioorg. Med. Chem. Lett. 17, 2515–2522.
(62) Chaires, J. B. (2005) Competition Dialysis: An Assay to
Measure the Structural Selectivity of Drug-Nucleic Acid Interactions.
Curr. Med. Chem.: Anti-Cancer Agents 5, 339–352.
(63) Ren, J., and Chaires, J. B. (1999) Sequence and Structural
Selectivity of Nucleic Acid Binding Ligands. Biochemistry 38, 16067–16075.
(64) Ragazzon, P., and Chaires, J. B. (2007) Use of Competition
Dialysis in the Discovery of G-Quadruplex Selective Ligands. Methods
43, 313–323.
2849
dx.doi.org/10.1021/bi1017304 |Biochemistry 2011, 50, 2838–2849