Halogenated pentamethine cyanine dyes exhibiting high fidelity for G-quadruplex DNA

Article abstract of DOI:10.1016/j.bmc.2012.10.008

Design and optimization of quadruplex-specific small molecules is developing into an attractive strategy for anti-cancer therapeutics with some promising candidates in clinical trials. A number of therapeutically favorable features of cyanine molecules can be effectively exploited to develop them as promising quadruplex-targeting agents. Herein, the design, synthesis and evaluation of a series of dimethylindolenine cyanine dyes with varying halogen substitutions are reported. Their interactions with telomeric and c-myc quadruplexes as well as a reference duplex sequence have been evaluated using thermal melting, biosensor-surface plasmon resonance, circular dichroism, isothermal titration calorimetry and mass spectrometry. Thermal melting analysis indicates that these ligands exhibit significant quadruplex stabilization and a very low duplex binding, with the dimethyl incorporation of paramount importance for decreased duplex affinity. Circular dichroism studies showed that the interaction of cyanines with quadruplex structures are primarily through stacking at one or both ends of the terminal tetrads with the two (trimethylammonium)propyl groups interacting in the accessible quadruplex grooves. Surface plasmon resonance and mass spectral studies shows the formation of an initial strong 1:1 complex followed by a significantly weaker secondary binding. Isothermal calorimetry studies show that the interaction of cyanines is predominantly entropy driven. In line with the design principles, this work provides new insights for further developing potent, highly selective cyanines as promising quadruplex-specific agents.

Full text of DOI:10.1016/j.bmc.2012.10.008

Bioorganic & Medicinal Chemistry 20 (2012) 7002–7011
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Halogenated pentamethine cyanine dyes exhibiting high fidelity
for G-quadruplex DNA
Rupesh Nanjunda , Eric A. Owens , Leah Mickelson, Sergey Alyabyev, Nancy Kilpatrick, Siming Wang,
⇑
⇑
Maged Henary , W. David Wilson
Department of Chemistry, Georgia State University, Atlanta, GA 30303, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Design and optimization of quadruplex-specific small molecules is developing into an attractive strategy
for anti-cancer therapeutics with some promising candidates in clinical trials. A number of therapeuti-
cally favorable features of cyanine molecules can be effectively exploited to develop them as promising
quadruplex-targeting agents. Herein, the design, synthesis and evaluation of a series of dimethylindoli-
ene cyanine dyes with varying halogen substitutions are reported. Their interactions with telomeric
and c-myc quadruplexes as well as a reference duplex sequence have been evaluated using thermal melt-
ing, biosensor-surface plasmon resonance, circular dichroism, isothermal titration calorimetry and mass
spectrometry. Thermal melting analysis indicates that these ligands exhibit significant quadruplex stabil-
ization and a very low duplex binding, with the dimethyl incorporation of paramount importance for
decreased duplex affinity. Circular dichroism studies showed that the interaction of cyanines with quad-
ruplex structures are primarily through stacking at one or both ends of the terminal tetrads with the two
(trimethylammonium)propyl groups interacting in the accessible quadruplex grooves. Surface plasmon
resonance and mass spectral studies shows the formation of an initial strong 1:1 complex followed by
a significantly weaker secondary binding. Isothermal calorimetry studies show that the interaction of
cyanines is predominantly entropy driven. In line with the design principles, this work provides new
insights for further developing potent, highly selective cyanines as promising quadruplex-specific agents.
Published by Elsevier Ltd.
Received 21 August 2012
Revised 5 October 2012
Accepted 10 October 2012
Available online 16 October 2012
Keywords:
Cyanines
G-quadruplex
halogenation
Telomere stabilization
Surface plasmon resonance
Isothermal calorimetry
Mass spectrometry
1. Introduction
of the quadruplex and the flexible cationic group(s) in one or more
of the quadruplex grooves. With this model in mind, a systematic
It has now been convincingly demonstrated that small mole-
cules that target quadruplexes can have cellular anticancer activity
and this area is promising for new drug development.1–4 One com-
pound (CX-3543, Quarfloxin) has entered clinical trials with
encouraging initial results against neuroendocrine cancer.5 In spite
of this success, additional quadruplex targeting compounds that
display appealing pharmacodynamics with selective quadruplex
binding remain highly desirable. A review of quadruplex-targeting
structures illustrates a dominance of planar aromatic substituents
and an overall cationic nature.6–8 The charged groups are fre-
quently located on a flexible chain that is linked to the aromatic
system. A common binding mode for compounds displaying this
motif is stacking of the aromatic moiety on the terminal tetrads
set of new cationic carbocyanines (Fig. 1) were designed and syn-
thesized as potentially selective ligands for DNA G-quadruplexes.
Cyanine dyes have extensive applications including cancer imag-
ing, nucleic acid detection, photographic processes, and nonlinear
optical materials that utilize their outstanding photophysical prop-
erties.9,10 Synthetic cyanines have characteristics, such as low toxic-
ity and stability towards chemical decomposition, that make their
cellular use extremely appealing. Additionally, this class of com-
pounds displays excellent biocompatibility while being amenable
to diverse chemical modifications.9 With multiple synthetically ac-
tive sites, there exist a multitude of carbocyanine dyes that can be
designed and tested. Various cyanine supramolecular assemblies
have been shown to probe the presence of the G-quadruplex struc-
ture in human telomere sequences.11 Cyanine dyes have been ex-
plored as fluorescent tags to observe the three-dimensional
folding properties of G-quadruplex DNA structures and dyes based
on the benzothiazole heterocyclic moiety have been shown to inhi-
bit telomerase via a suggested G-quadruplex stabilization mecha-
nism.12 The benzothiazole cyanines, however, generally have very
strong DNA duplex minor groove interactions; therefore, a system-
atic set of dimethylindolenine-based pentamethine cyanine dyes
⇑
Corresponding authors. Addresses: Georgia State University, Department of
Chemistry, 100 Piedmont Avenue NE, Petit Science Center Office 315, Atlanta, GA
30303, USA. Tel.: +1 404 413 5566; fax: +1 404 413 5505 (M.H.); Department of
Chemistry, Georgia State University, 50 Decatur Street, Room 540 Natural Science
Center, Atlanta, GA 30303, USA. Tel.: +1 404 413 5503; fax: +1 404 413 5505 (W.D.W.).
E-mail addresses: mhenary1@gsu.edu (M. Henary), chewdw@panther.gsu.edu,
wdw@gsu.edu (W. David Wilson).
Both authors contributed equally to this work.
0968-0896/$ - see front matter Published by Elsevier Ltd.
http://dx.doi.org/10.1016/j.bmc.2012.10.008

R. Nanjunda et al. / Bioorg. Med. Chem. 20 (2012) 7002–7011
7003
R
R
R
MeCN, Reflux
O
N
Br
AcOH, Reflux
N
Br
N
HN
Br
NH2
HCl
N
Br
1: R = H
2: R = Cl
3: R = Br
4: R = H
5:
6:
R = Cl
R = Br
7:
8:
9:
R = H
R = Cl
R = Br
Cl
R
R
Ph
Ph
N
N
H
H
X
10:
11: X = Cl
12: X = Br
X = H
N
N
Br
Br
X
EtOH
NaOAc
Br
N
Br
N
14:
17:
20:
15:
18:
21:
13:
R = H; X = Cl
R = Br; X = H
R = Cl; X = Cl
R = H; X = Br
R = Cl; X = Br
R = Br; X = Br
R = H; X = H
16: R = Cl; X = H
19: R = Br; X = Cl
Figure 1. The synthesis of the cyanine structures used in developing G-quadruplex binding agents.
were synthesized to induce steric hindrance by the dimethyl moie-
ties towards insertion of the compound into the duplex minor
groove. Ammonium cations were introduced to the heterocyclic
nitrogen to enhance solubility and increase quadruplex affinity by
electrostatic interactions. Further structural diversity was achieved
by halogenation at the 5-position and the meso-carbon of the
polymethine-bridge. The interactions of these cyanine analogs with
quadruplex and duplex DNA were evaluated using a variety of pow-
erful biophysical techniques, including: DNA thermal melting
screening, biosensor-surface plasmon resonance, circular dichro-
ism, isothermal titration calorimetry and mass spectrometry. These
methods reveal encouraging results as these cyanine structures
bind strongly to quadruplexes with excellent selectivity over duplex
DNA.
melting studies as a screening assay. Thermal melting studies were
done only with the telomeric G-quadruplex because the melting
temperature of c-myc quadruplex is too high under the currently
studied salt conditions (50 mM KCl) and the changes in Tm for
most of the compounds go beyond the measurable limit. The
D
Tm results at ratios of cyanine to DNA ranging from 2:1 to 6:1
are shown in Table 1. The parent cyanine, compound 13, shows a
relatively low Tm for Tel22. Adding chlorine (14) or bromine
(15) atom to the central carbon of the pentamethine chain resulted
in a large increase in Tm with no significant difference between
D
D
the Cl or Br compounds. Pentamethine cyanines with heterocyclic
halogenation were then tested. Di-halogenated compounds (with
terminal Cl and Br, compounds 16 and 17, respectively) do not
show any significant increase in
DTm over the mono-substituted
derivatives. The tricationic pentamethine structure was then
synthesized with the incorporation of three halogen substituents
2. Results
(18–21) and these showed a substantial increase in the
DTm with
no significant difference between the Cl and Br derivatives. In sum-
mary, pentamethine cyanines with multiple halogen substituents
bind to the telomeric quadruplex much more strongly than the
unsubstituted parent and with very high selectivity over duplex
DNA. The selectivity of these compounds is clearly very promising.
2.1. Chemistry
In order to optimize the binding specificity and strength, we uti-
lized the general 2,3,3-trimethylindoleline heterocyclic scaffold with
heterocyclichalogenation and a 3-(trimethylammonium)propylbro-
mide substituent on the indolenine nitrogen. The synthetic pathway
for obtaining thefinalpentamethine structuresbeginswith Fisher-in-
dole synthesis utilizing hydrazine derivatives 1–3 with 3-methyl-2-
butanone to form modified Fischer-bases 4–6 (Fig. 1). These obtained
indolenine derivatives 4–6 were alkylated with (3-bromopropyl)
trimethylammonium bromide in anhydrous acetonitrile yielding
dicationic salts 7–9. Reagent 10was commercially obtained and com-
pounds 11 and 12 were prepared by decarboxylation of either muco-
chloric or mucobromic acid by aniline. Salts 7–9 were individually
reacted with 10–12 in absolute ethanol under basic conditions to
yield the nine corresponding tricationic pentamethine carboycyanine
dyes 13–21 with varied halogen substituents.
2.3. Circular dichroism (CD)
CD experiments were conducted to qualitatively evaluate the
possible binding modes of the cyanine analogs with the quadru-
plex sequences. Achiral ligands in an unbound state do not exhibit
any optical activity, however, upon complex formation with
biomolecules such as DNA, they can exhibit an induced CD (ICD)
signal that can be qualitatively evaluated to rationalize their bind-
ing modes.13 As observed in duplex DNA, ligands that bind in the
grooves of quadruplex motifs exhibit strong ICD signals whereas
the end-stacking ligands exhibit weak ICD relative to the intensity
of the DNA signal.8,14,15 Figure 2 shows the CD plots of the parent
cyanine (13) and the brominated analogs (17 and 21). The high
extinction coefficients of the cyanine molecules precludes obtain-
ing CD spectra in the compound absorption region beyond certain
ligand–DNA ratios for some cyanines analogs and therefore CD
2.2. Thermal melting (Tm)
An initial comparison of the interaction of all cyanines with
Tel22 quadruplex and duplex DNA was done using UV thermal

7004
R. Nanjunda et al. / Bioorg. Med. Chem. 20 (2012) 7002–7011
(K1 = 1.7 ꢀ 106 Mꢁ1
)
relative
to
the
parent
compound
Table 1
(K1 = 1.3 ꢀ 106 Mꢁ1), whereas, the chloro substitution (14) resulted
in a decrease in the affinity with the Tel22 quadruplex. Among the
symmetric, tri-substituted and mixed halogen derivatives, no
improvement in the binding was observed relative to the parent
compound (Table 2). The lack of an increase in affinity for the hal-
ogens, relative to the parent, as expected from the Tm results, is
due to some adsorption of the halogen derivatives to the sensor sur-
face and resulting decrease in the compound concentration. Also,
dyes such as cyanines tend to aggregate in solution and this affects
and reduces SPR binding affinities since the aggregated compounds
must first dissociate to bind to DNA. SPR experiments were con-
ducted in different buffer conditions to test for reproducibility
and in each trial very similar value were obtained. The listed bind-
ing affinities (Table 2) are the average of two different experiments.
Interestingly, all the compounds exhibited a similar trend in the
relative binding affinities with the c-myc quadruplex but with a
significantly stronger binding (ca. 10- to 25-fold higher than the
telomere) and a similar weak secondary binding. These results
show a high selectivity for quadruplex sequences and confirm
the selectivity observed in thermal melting studies.
Thermal melting studies of the cyanine analogs with telomeric quadruplex (Tel22)
and a control duplex (AATT)
Compound Tel22 (
ratios)
D
Tm, °C) (2:1, 4:1, 6:1
Duplex (
ratio)
DTm, °C) (6:1
13
14
15
16
17
18
19
20
21
2.1, 4.6, 8.7
0
4.6, 12.2, 13.2
4.6, 11.6, 15.6
3.1, 8.9, 13.2
2.5, 15.6, —a
8.0, 17.0, 20.0
9.0, 16.0, 21.0
7.0, 15.0, 19.0
8.0, 16.2, 20.3
2.8
0.1
0.5
1.1
1.8
2.5
2.3
1.2
The thermal melting values reported are an average of two independent trials and
are reproducible within 0.5 °C.
a
Could not be accurately determined due to complex aggregation.
spectra only for ratios up to 3:1 are shown. All the compounds
showed very weak ICD signals with Tel22 sequence as is commonly
observed for ligands that stack on the terminal tetrads of quadru-
plex structures.16,17 The addition of halogens did not have much ef-
fect on the magnitude and the pattern of the ICD signals suggesting
a very similar end-stacking binding mode for the ligands. No
changes in the DNA CD signal pattern were observed upon ligand
binding suggesting the hybrid conformation of Tel22 sequence as
the preferred structure for these ligands. The lack of any ICD signal
with the c-myc sequence is also consistent with the end-stacking
binding mode of the ligands (data not shown).
2.5. Isothermal titration calorimetry (ITC)
ITC experiments were conducted by titrating all of the bromo
derivatives (compounds 15, 17 and 21) into Tel22 and c-myc
quadruplexes as well as the duplex model system. No significant
heat change was observed in the duplex titration (not shown) in
agreement with the negligible binding observed in Tm and SPR
experiments. Experimental thermograms for the mono-bromo
compound (15), are shown as examples in Figure 4 with both
Tel22 and c-myc quadruplexes. Integrated heat versus molar ratios
is also shown with the experimental results in Figure 4. The results
for the c-myc sequence could be adequately fit with a single-site
binding model over the range that could be covered in the ITC
experiment. The binding with Tel22 was significantly weaker and
no adequate fit to the heat versus molar ratio plot could be done
over the accessible concentration range. However, it is qualita-
tively clear from the plot of compound 15 with Tel22 (Fig. 4A) that
the binding enthalpy for Tel22 is slightly more negative than for
c-myc. The fit values for binding enthalpy and equilibrium binding
constant are given for the c-myc quadruplex in Table 3. From these
values, the Gibbs free energy and entropy of binding were calcu-
lated and are also included in Table 3. Similar ITC results were ob-
tained with di and tri-bromo substituted derivatives (compounds
17 and 21) with c-myc quadruplex (Table 3). These results clearly
2.4. Surface plasmon resonance (SPR)
Biosensor-SPR studies were conducted with biotinylated, sen-
sor-surface captured quadruplex and duplex sequences to quantita-
tively evaluate the binding affinity, selectivity and stoichiometry of
the cyanine analogs. Figure 3 shows the SPR sensorgrams and the
steady-state affinity plots for the parent compound (13) with the
two quadruplex and reference duplex DNA sequences, and the
binding affinities of all cyanine derivatives are summarized in Table
2. All the compounds exhibited an initial strong binding followed by
a weaker secondary binding (ca. 5- to 50-fold) for both quadruplex
sequences. The duplex binding in all the cases was too low to detect
in the concentration range that was tested. The binding kinetics
was rapid under the experimental conditions with both association
and dissociation reactions occurring near the instrument injection
mixing limit. Among the mono-substituted analogs, the bromo sub-
stitution (15) resulted in a slight increase in the binding affinity
Compound 17 - Tel22
Compound 21 - Tel22
Compound 13 - Tel22
15
0
1
2
3
0
1
2
3
0
1
2
3
10
5
0
300
400
500
600
300
400
500
600
300
400
500
600
Wavelength (nm)
Wavelength (nm)
Wavelength (nm)
Figure 2. CD spectra of Tel22 with non-halogenated dye 13, di-brominated (17) and tri-brominated analog (21) at 25 °C in 50 mM KCl. Single strand quadruplex
concentration is 4 M. The inset shows the ligand:quadruplex ratios.
l

R. Nanjunda et al. / Bioorg. Med. Chem. 20 (2012) 7002–7011
7005
A
c-myc
Tel22
AATT
125
100
75
50
25
0
0
200
400
600
0
200
400
600
0
200
400
600
Time (seconds)
Time (seconds)
Time (seconds)
B
150
Compound 13
Compound 15
Tel22
c-myc
AATT
100
50
0
0
2 10-7 4 10-7 6 10-7 8 10-7 1 10-6
[compound conc]
0
2 10-7 4 10-7 6 10-7 8 10-7 1 10-6
[compound conc]
Figure 3. (A) SPR sensorgram of compound 13 with the two quadruplex as well as a reference duplex sequence. (B) Steady-state equilibrium binding plots of analogs 13 and
15. The solid colored lines are the fits obtained using a two-site model. In all cases, the steady-state response with duplex sequence was too low to fit reasonably.
Table 2
that the heat is quite low for the weaker binding site, and the sec-
ond site is not easily detected.
Equilibrium binding constants of the cyanine analogs with quadruplex and duplex
sequences obtained from SPR studies
Compound
Tel22 (K1, K2) ꢀ 106 Mꢁ1
c-myc (K1, K2) ꢀ 106 Mꢁ1
Duplex
2.6. Mass spectrometry (MS)
13
14
15
16
17
18
19
20
21
1.3, 0.17
11.9, 0.07
9.7, 0.60
13.0, 0.07
No binding
5.5, 3.6
7.2, 0.25
4.8, 0.20
8.0, 0.23
9.7, 2.0
<105
<105
<105
<105
<105
<105
<105
<105
<105
0.56, 0.08
1.7, 0.74
To quantitatively and directly evaluate complex stoichiometry,
mass spectrometry was done on the complexes at several com-
pound:DNA ratios to obtain an independent determination of
binding stoichiometry. Experiments were conducted with all three
bromo compounds and results for compounds 13 and 15 are
shown in Figure 5. For free DNA, the -4 and -5 charge species dom-
inate and are of similar magnitude. On addition of the compound, a
1:1 complex peak can be seen. As the ratio is increased, the 1:1
peak grows in intensity and a smaller 2:1 peak can be seen at high-
er ratios. It is clear from the mass spectrometry results that the
binding is a two site non-cooperative interaction in agreement
with the SPR analysis.
No bindinga
0.79, 0.59
1.0, 0.13
0.35, 0.05
0.59, 0.08
0.39, 0.06
The steady-state response obtained as a function of free ligand concentration was
fitted to a two-site model to obtain K1 and K2. In all the cases, no duplex binding
was observed suggesting a high selectivity of the ligands.
No binding: no sensorgrams were observed in the concentration range that was
tested. This might be due to the adsorption of significant amount of ligand to the
sensor surface.
a
3. Discussion
show that the primary driving component for binding of cyanines
to c-myc is the entropy. As noted above the binding enthalpy is
The discovery of G-quadruplex structures and their potential
involvement in important biological pathways as well as cancer
development has significantly extended our understanding of the
importance of non-B-form DNA structures in cellular function. In
addition, such non-B structures can provide an alternate platform
for developing structure-specific small molecules as part of a
new therapeutic approach. To use the DNA quadruplex structure
more negative for Tel22 and with a less negative
DG of binding,
the driving components for Tel22 binding will be more balanced
between enthalpy and entropy.
Although ITC data was best fit with only one binding site, the
SPR results suggest a weaker, second binding site. The binding heat
from the primary binding site is low and the fitting results suggest

7006
R. Nanjunda et al. / Bioorg. Med. Chem. 20 (2012) 7002–7011
Time (min)
75 100 125 150
Time (min)
25
50
75 100 125 150
25
50
0.00
-0.05
-0.10
-0.15
-0.20
0
-1
-2
-3
A
2.0
B
2.0
0.5
1.0
1.5
0.5
1.0
1.5
Molar Ratio
Molar Ratio
Figure 4. ITC thermograms of the mono-bromine compound 15 with Tel22 (A) and c-myc (B) sequences. Raw heats are displayed in top panels, while the corrected,
integrated plots are in the bottom panel. The solid blue lines represent the fit obtained using a single-site model, and the thermodynamic parameters are shown in Table 3.
The initial screening with thermal melting showed a very high
selectivity by all the cyanine analogs for the quadruplex over du-
Table 3
Thermodynamic parameters for the bromine analogs with c-myc quadruplex
plex structures. This very promising finding is consistent with
the design principles aimed at the development of strong quadru-
plex and weaker duplex binding. The thermal stability of the quad-
ruplex sequence further increased with the addition of halogens
and the tri-halogenated analogs inducing the highest stability.
The introduction of either chlorine or bromine clearly increased
the stabilization potential as compared to the parent molecule,
however, no significant difference in the thermal melting values
were found between the Cl or Br analogs as they both bound
strongly to quadruplexes. The presence of polar atoms such as
chlorine and bromine clearly has a positive and an additive effect
on the binding to the quadruplex sequence. The lack of significant
selectivity between Cl and Br suggests that the unfavorable steric
contributions are minimal for these substitutions. Furthermore,
the presence of two bulky methyl groups on both aromatic rings
provides a steric hindrance to minimize binding to the duplex
sequence.
Circular dichroism analysis indicates predominant end-stacking
and much weaker groove interactions for all the ligands with quad-
ruplex structures, based on the presence of weak ICD signals in the
compound absorbance region upon complex formation. Weak ICD
is a commonly observed pattern for ligands that either intercalate
with duplex sequences24,25 or stack on the terminal tetrads of
quadruplex motifs with weak groove interactions.14 The two 3-
(trimethylammonium)propyl substituents on these cyanine sys-
tems (Fig. 1) can favorably interact with the accessible grooves of
the quadruplex. The secondary binding observed in SPR might be
due to the weak interactions of a second ligand in the quadruplex
grooves resulting in ICD signals of low magnitude. The lack of any
CD changes in the DNA region of the spectra further indicates the
native conformation as the preferred quadruplex binding con-
former for the cyanine analogs. Human telomeric quadruplex can
adopt different quadruplex folds under solution conditions and
these different folds have been shown to exist in equilibrium due
to a low energy barrier between them.24–26 Various small mole-
cules have been designed and developed that can target different
sequence derived from ITC and SPR studies
Compound
D
H (kcal/mol)
D
G (kcal/mol)
T
D
S (kcal/mol)
n
15
17
21
ꢁ2.1
ꢁ1.9
ꢁ3.6
ꢁ9.8
ꢁ9.3
ꢁ9.6
7.7
7.3
6.0
1.2
0.98
0.91
The intergrated heats were fit to a single-site model using the primary binding
constant values (K1) from SPR to obtain the enthalpy values. G was calculated
using S using G = HꢁT S.
G = ꢁRTlnK1 and T
D
D
D
D
D
D
as a platform for anticancer drug development, it is essential to de-
sign and discover new agents that can discriminate between quad-
ruplex and duplex structures with very high selectivity.
Furthermore, the diversity of quadruplex conformations found in
the genome also makes the development of ligands that can dis-
criminate between different quadruplex folds quite attrac-
tive.2,18,19 The dimethylindolenine cyanine scaffold was selected
based on the concepts that it could not fit into the minor groove,
as with planar cyanines, but could stack on the terminal tetrads
of quadruplex structure with cationic chains in the quadruplex
grooves. A survey of all the approved drugs that are currently being
used, as well as drug candidates that are in clinical development,
shows that a majority of the drugs have halogen atoms incorpo-
rated to increase the pharmacological properties of the drug mole-
cule.20–22 While fluorine and chlorine are the most prevalent
halogens used, brominated drugs can also increase the desired
pharmacological features.23 With this in mind for an initial test
series, coupled with a number of features of cyanine molecules
that are considered to be therapeutically favorable, a series of pen-
tamethine-bridged cyanine analogs with various combinations of
chlorine and bromine substitutions on the core aromatic system
as well as on the meso-carbon of the pentamethine chain were
synthesized. Their quadruplex binding potential and selectivity
over duplex binding were evaluated using a complementary series
of biophysical techniques.

R. Nanjunda et al. / Bioorg. Med. Chem. 20 (2012) 7002–7011
7007
DNA only
11:05:03 02-Feb-2012
RUPESH_CMYC_NEG_WILSON_020212
TOF MS ES-
82.6
original solution
11:03:50 14-Apr-2011
TOF MS ES-
28 (0.524) Sm (SG, 3x3.00); Cm (28:14A4)
1217.8
LEAH_DNA_ESI_NEG_WILSON_041411 474 (8.794) Sm (SG, 3x3.00); Cm (349:582)
100
1526.8
114
B
100
1217.8
1522.8
1187.6
1152.7
2044.9
2080.1
1440.4
1440.3
1151.8
1337.8
1304.7
1300
0
m/z
0
1100
m/z
1200
1400
1500
1600
1700
1800
1900
2000
1200
1300
1400
1500
1600
1700
1800
1900
2000
DNA :E46=1:1
11:21:18 02-Feb-2012
RUPESH_CMYC_E46_1TO1_NEG_WILSON_020212 333 (6.188) Sm (SG, 3x3.00); Cm (268:376)
TOF MS ES-
167
original DNA:ligand solution=1:2
15:00:11 21-Apr-2011
1217.8
100
LEAH_DNA_LIG_1TO2_ESI_NEG_WILSON_042111 126 (2.343) Sm (SG, 3x3.00); Cm (114:435)
TOF MS ES-
250
1526.9
100
1522.7
1217.9
1684.9
1660.9
1328.5
1344.1
1440.3
1489.3
1187.6
1598.7
1600
1838.7
1799.0
1440.2
2041.8
1152.2
1750.1
1152.0
1283.8
1937.7
1279.7
1300
0
m/z
0
1100
m/z
1200
1300
1400
1500
1600
1700
1800
1900
2000
1200
1400
1500
1700
1800
1900
2000
DNA :E46=1:2
11:37:32 02-Feb-2012
original DNA:ligand solution=1:4
15:09:26 21-Apr-2011
RUPESH_CMYC_E46_1TO2_NEG_WILSON_020212 470 (8.730) Sm (SG, 3x3.00); Cm (384:492)
TOF MS ES-
153
LEAH_DNA_LIG_1TO4_ESI_NEG_WILSON_042111 289 (5.376) Sm (SG, 3x3.00); Cm (169:495)
TOF MS ES-
362
1665.0
100
1684.8
100
1217.7
1522.4
1328.4
1526.9
1344.1
1217.8
1799.1
1842.7
1602.7
1440.2
1756.6
1438.7
1187.5
1200
1582.9
1600
2041.6
m/z
1921.1
1983.4
1152.2
1200
2047.4
m/z
2000
1275.2
1300
0
1100
0
1400
1500
1600
1700
1800
1900
2000
1300
1400
1500
1700
1800
1900
DNA :E46=1:4
11:49:32 02-Feb-2012
original DNA:ligand solution=1:6
15:19:39 21-Apr-2011
RUPESH_CMYC_E46_1TO4_NEG_WILSON_020212 397 (7.377) Sm (SG, 3x3.00); Cm (354:462)
TOF MS ES-
LEAH_DNA_LIG_1TO6_ESI_NEG_WILSON_042111 303 (5.640) Sm (SG, 3x3.00); Cm (201:532)
TOF MS ES-
352
1660.7
1684.9
100
100
1328.2
1799.0
1842.8
1217.9
1522.4
1344.1
1526.7
1470.7
1440.3
1438.8
1404.9
1217.8
1602.4
1600
1760.1
1930.1
1996.2
2000
1131.6
2097.0
m/z
1582.7
1600
1278.5
1300
1872.7
1900
1155.5
1200
0
0
m/z
1100
1200
1400
1500
1700
1800
1900
1300
1400
1500
1700
1800
2000
Figure 5. Mass spectra of c-myc quadruplex with 0:1, 1:1, 2:1, 4:1 ratios (top to bottom) of parent dye 13 (A) and mono-brominated compound 15 (B).
folds of telomeric quadruplex as well as small molecules that can
facilitate the interconversion between different conformers upon
binding.14,17,27–29 In almost all cases, ligand induced conforma-
tional changes upon binding is a commonly observed phenomena.
The different conformations of the telomeric quadruplex does not
necessarily alter the ligand binding mode because of the very small
energy required to induce the interconversion between different
folds without changing the mode of binding for the ligand. From
the CD results, the lack of any significant changes in the DNA spec-
tral region indicates that the native hybrid conformation of the
telomeric quadruplex is the preferred quadruplex conformer for
the cyanine analogs.
CD results are further supported by the presence of an initial
strong binding site detected by SPR studies. The two-site model
predicted an initial strong binding that can be attributed to end-
stacking and a significantly weaker secondary binding. Interest-
ingly, all the compounds exhibited a 10-fold higher affinity for
the c-myc quadruplex over the telomeric quadruplex. This can be
due to the structural differences between the quadruplex folds
from c-myc and telomeric sequence. The c-myc sequence folds into
a single, highly stable parallel architecture where all the strands
are parallel to each other and the connecting loops run diagonally
along the grooves of the quadruplex. In such a parallel architecture,
both the terminal tetrads of the quadruplex are completely ex-
posed without any steric hindrance from the loop nucleobases,
and therefore the exposed surface of terminal tetrads are highly
conducive for favorable ligand stacking. Telomeric quadruplex,
however, can fold into a multitude of conformers in solution with
the hybrid conformation as the predominant fold. The two lateral
TTA loops of the hybrid fold run across the terminal tetrads on both
ends, and therefore the highly dynamic nucleobases of the loop
render some steric hindrance for effective ligand stacking resulting
in decreased binding affinities for telomeric quadruplex as com-
pared to c-myc.
The stoichiometry analysis was strongly supported by the mass
spectral results where the formation of a predominant 1:1 complex
species and a weaker 2:1 complex was observed. ITC thermody-
namic analysis of cyanine–quadruplex interactions revealed a pre-
dominantly entropy driven complex formation with c-myc
quadruplex. Although the binding enthalpy largely remained the
same for the mono- and di-bromo additions, the tri-bromo analog
showed a favorable increase in enthalpy (ca. twofold), owing to the

7008
R. Nanjunda et al. / Bioorg. Med. Chem. 20 (2012) 7002–7011
larger van der Waals interactions due to the additional bromine. It
also had a corresponding small decrease in entropy, probably due
to the conformational restriction imposed by the bulky bromine
addition upon complex formation.
2,3,3-Trimethyl-1-(3-(trimethylammonio)propyl)-3 H-indol-1-
ium bromide (7) was synthesized by our group and is previously
reported.7
5.1.1.1 5-Chloro-2,3,3-trimethyl-1-(3-(trimethylammonio)
propyl)-3H-indol-1-ium bromide (8). Yield 82%; mp 215–217 °C;
4. Conclusions
1H NMR (400 MHz, DMSO-d6), d: 1.65 (s, 6H), 2.56 (t, J = 8.0 Hz, 2H),
3.25 (s, 9H), 3.73 (tt, J = 8.0 Hz, J = 8.0 Hz, 2H), 4.64 (t, J = 8.0 Hz, 2H),
4.80 (s, 3H), 7.69 (dd, J = 8.8 Hz, J = 2.0 Hz, 1H), 7.84–7.92 (m, 2H);
13C NMR (100 MHz, DMSO-d6): d 12.2, 20.0, 20.7, 43.3, 52.1, 53.2,
61.5, 115.9, 123.3, 128.8, 134.3, 138.2, 142.3, 197.9. Composition in
theory: C(44.91%), H(5.99%), N(6.16%); Composition found:
C(45.01%), H(6.29%), N(6.08%); HRMS (ESI) calculated for
[C17H27ClN2]2+ m/z 147.4356, found m/z 146.9909, 148.0421
In conclusion, a series of halogenated cyanine dyes were sys-
tematically designed, synthesized and tested for their interaction
potential with quadruplex sequences using complementary lines
of biophysical techniques. In conjunction with the design
principles, to enhance quadruplex specificity with low duplex
affinity, several of the cyanine analogs in this study showed ele-
vated binding to both telomeric and c-myc quadruplex structures
while maintaining a very low duplex selectivity. The dimethyl
incorporation on both indolenine rings provide the necessary steric
hindrance for the cyanine moiety to decrease duplex groove inter-
actions, while the more polar halogen substitutions enhanced
quadruplex binding. Thermal melting studies showed a relative in-
crease in quadruplex stability as the number of halogen substitu-
tions was increased. The small ICD signal observed upon complex
formation indicates that the cyanines are interacting with the
quadruplex structure primarily through end-stacking mode. The
formation of a strong 1:1 complex as observed in SPR and mass
spectral studies further supports the end-stacking binding mode
of the ligands. The currently designed ligands provide the neces-
sary platform in developing further potent cyanine molecules as
promising quadruplex-specific agents.
5.1.1.2. 5-Bromo-2,3,3-trimethyl-1-(3-(trimethylammonio)
propyl)-3H-indol-1-ium bromide (9). Yield 81%; Mp 225–227 °C;
1H NMR (400 MHz, DMSO-d6), d: 1.57 (s, 6H), 2.27 (s, 9H), 3.02 (s, 3H),
3.83 (t, J = 8.0 Hz, 2H), 4.57 (t, J = 8.0 Hz, 2H), 7.61 (m, 2H), 7.88 (t,
J = 5.6 Hz, 1H). 13C NMR (100 MHz, DMSO-d6): d 15.2, 22.0, 25.5,
30.6, 52.3, 54.3, 63.9, 115.6, 123.5, 128.8, 129.3, 140.9, 141.7, 197.7.
Composition in theory: C(40.91%), H(5.45%), N(5.61%); Composition
found: C(40.84%), H(5.48%), N(5.94%); HRMS (ESI) calculated for
[C17H27BrN2]2+ m/z 169.6611, found m/z 169.7002.
5.1.2. General synthetic procedure for the preparation of
methine bridge reagent
(E)-N-((Z)-2-Chloro-3-(phenylamino)allylidene) benzenaminium
chloride (11) was synthesized using the reported procedure.30
88% yield; Mp 214–216 °C.
5. Experiments
5.1.2.1. (E)-N-((Z)-2-Bromo-3-(phenylamino)allylidene) benze-
naminium chloride (12). Mucobromic acid (5.940 g, 23.04 mmol)
was dissolved in ethanol (40 mL) and aniline (4.286 g, 4.2 mL,
46.1 mmol), diluted with ethanol (20 mL), was added dropwise
over a 10-min period. The reaction vessel was under vigorous stir-
ring and was gently heated to 40 °C. After the addition of aniline
was finished, CO2(g) was further produced indicating the reaction
was not complete. Upon completion of CO2(g) evolution, the golden
brown reaction mixture was cooled in an ice bath. Diethyl ether
(50 mL) was added slowly at reduced temperature under heavy
stirring until a bright yellow solid precipitated. The resulting solid
was filtered, washed with ether, dried and was used without puri-
fication in subsequent reactions. Yield 75%; Mp 189–191 °C; 1H
NMR (400 MHz, DMSO-d6): d 7.330 (t, J = 7.2 Hz, 2H), 7.496 (t,
J = 8 Hz, 4H), 7.708 (d, J = 8 Hz, 4H), 9.559 (s, 2H), 11.513 (s, 2H);
13C NMR (100 MHz, DMSO-d6): d 89.3, 119.5, 123.9, 126.0, 129.2,
129.6, 138.5, 157.4.
5.1. Pentamethine dye synthesis
Sigma–Aldrich or TCI America were the commercial sources for
the starting materials utilized in the presented synthesis and the
reagents were used without purification. The 1H NMR and 13C
NMR spectra were recorded on a Bruker Avance (400 MHz) spec-
trometer using DMSO-d6 or MeOD-d4 containing tetramethylsilane
(TMS) as an internal calibration standard. UV–Vis/NIR absorption
spectra were recorded on a Varian Cary 50 spectrophotometer.
High-resolution accurate mass spectra (HRMS) were obtained
either at the Georgia State University Mass Spectrometry Facility
using a Waters Q-TOF micro (ESI-Q-TOF) mass spectrometer or uti-
lizing a Waters Micromass LCT TOF ES+ Premier Mass Spectrome-
ter. Liquid chromatography (LC) utilized a Waters 2487 single
wavelength absorption detector with wavelengths set between
640 and 700 nm depending on the dye’s photophysical properties.
The column used in LC was a Waters Delta-Pak 5 lM 100A
3.9 ꢀ 150 mm reversed phase C18 column. Evaporative light scat-
tering detection analyzes trace impurities that cannot be observed
by alternate methods; a SEDEX 75 ELSD was utilized in tandem
with liquid chromatography to observe and confirm purity.
5-Chloro-2,3,3-trimethyl-3H-indole (5) and 5-Bromo-2,3,3-trimethyl-
3H-indole (6) were prepared using the procedure published by Zim-
mermann et al.6 They were obtained as reddish brown oils in 71%
and 90% yield, respectively.
5.1.3. General procedure for obtaining tricationic
pentacarbocyanines 13–21
A mixture of individual salts 7–9 (2 equivalents), either malon-
aldehyde bisphenylimine monohydrochloride 10, chlorinated ana-
log 11 or brominated analog 12 (1 equiv) and sodium acetate
(3 equiv) were refluxed for 6 hrs in anhydrous ethanol under nitro-
gen atmosphere and then cooled to room temperature. The reac-
tion mixture was concentrated in vacuo. The crude product was
dissolved in deionized water and the target compounds 9–12 were
5.1.1. General synthetic procedure for preparation of
heterocyclic salts
isolated using silica gel 40–63 lm 90 Å C18 reversed phase column
chromatography using 20% methanol in water as eluting solvent.
A mixtureof corresponding indolederivative 4–6 (31.4 mmol) and
3-bromopropyl trimethylammonium bromide (9.0 g, 34.5 mmol)
was refluxed for 72 h in anhydrous acetonitrile (100 mL) under nitro-
gen. The reaction was cooled to room temperature and solid product
precipitated after 4 h. The precipitate was filtered and was washed
with acetone and diethyl ether.
5.1.3.1. 2-((1E,3E,5E)-5-(3,3-Dimethyl-1-(3-(trimethylammonio)-
propyl)indolin-2-ylidene)penta-1,3-dien-1-yl)-3,3-dimethyl-1-
(3-(trimethylammonio)propyl)-3H-indol-1-ium bromide (13). Yield
71%; Mp 218–221 °C; H NMR (400 MHz, MeOD-d4): d 1.781 (s, 12H),
2.359–2.380 (m, 4H), 3.248 (s, 18H), 3.750 (t, J = 8.4 Hz, 4H), 4.254 (t,
1

R. Nanjunda et al. / Bioorg. Med. Chem. 20 (2012) 7002–7011
7009
J = 7.6 Hz, 4H), 6.628 (d, J = 13.6 Hz, 2H), 7.140 (t, J = 12.4 Hz, 1H),
7.287–7.327 (m, 2H), 7.436–7.486 (m, 4H), 7.540 (d, J = 7.6 Hz, 2H),
8.345 (t, J = 13.2 Hz, 2H); 13C NMR (400 MHz, MeOD-d4): d 20.93,
26.56, 40.26, 49.22, 52.49, 63.08, 103.72, 110.41, 122.20, 125.07,
127.11, 128.50, 141.23, 141.76, 154.80, 173.47. High-resolution electro-
spray ionization (ESI) accurate mass spectra calculated m/z for
[C37H55N4Br2]+ 713.2793, found 713.2767.
(400 MHz, DMSO-d6): d 1.77 (s, 12H), 2.19 (t, J = 7.6 Hz, 4H), 3.13 (s,
18H), 3.63 (tt, J = 7.6 Hz, J = 7.8 Hz, 4H), 4.25 (t, J = 7.8 Hz, 4H), 6.34
(d, J = 13.4 Hz, 2H), 7.53 (d, J = 8.0 Hz, 2H), 7.68 (d, J = 8.0 Hz, 2H),
7.94 (s, 2H), 8.61 (d, J = 13.4 Hz, 2H); 13C NMR (400 MHz, DMSO-d6),
d: 18.5, 20.6, 26.4, 41.4, 49.8, 52.3, 62.2, 102.7, 113.3, 117.0, 123.2,
128.3, 130.1, 140.5, 143.4, 150.3, 174.4. Composition in theory:
(x3H2O): C (44.56%), H (5.86%), N (5.62%); Composition found: C
(44.54%),
H (5.78%), N (5.56%); HRMS (ESI) calculated for
5.1.3.2. 2-((1E,3Z,5E)-3-Chloro-5-(3,3-dimethyl-1-(3-(trimethyl-
ammonio)propyl)indolin-2-ylidene)penta-1,3-dien-1-yl)-3,3-
dimethyl-1-(3-(trimethylammonio)propyl)-3H-indol-1-ium
bromide (14). Yield 46%; Mp 225–227 °C; 1H NMR (400 MHz,
DMSO-d6): d 1.81 (s, 12H), 2.40 (t, J = 7.4 Hz, 4H), 3.26 (s, 18H),
3.71–3.79 (m, 4H), 4.38 (t, J = 7.6 Hz, 4H), 6.55 (d, J = 13.4 Hz, 2H),
7.36 (t, J = 7.2 Hz, 2H), 7.50 (t, J = 7.6 Hz, 2H), 7.58 (d, J = 7.6 Hz,
2H), 7.63 (d, J = 7.6 Hz, 2H), 8.50 (d, J = 13.4 Hz, 2H); 13C NMR
(100 MHz, DMSO-d6), d: 20.9, 26.0, 40.7, 49.4, 52.5, 63.0, 100.9,
111.9, 122.2, 123.2, 125.6, 128.6, 141.4, 141.7, 148.2, 175.1. Compo-
sition in theory: (x3H2O): C (50.27%), H (6.84%), N (6.34%); Compo-
[C37H52Br2Cl2N4] + m/z 780.1936, found 780.1916.
5.1.3.7. 5-Bromo-2-((1E,3Z,5E)-5-(5-bromo-3,3-dimethyl-1-(3-(tri
methylammonio)propyl)indolin-2-ylidene)-3-chloropenta-1,3-di
en-1-yl)-3,3-dimethyl-1-(3-(trimethylammonio)propyl)-3H-indol-
1-ium bromide (19). Yield 80%; Mp 227–229 °C; 1H NMR (400 MHz,
DMSO-d6): d 1.76 (s, 12H), 2.17 (t, J = 7.0 Hz, 4H), 3.12 (s, 18H), 3.58–
3.68 (m, 4H), 4.26 (t, J = 6.8 Hz, 4H), 6.35 (d, J = 13.4 Hz, 2H), 7.61 (d,
J = 8.6 Hz, 2H), 7.65 (d, J = 8.6 Hz, 2H), 8.05 (s, 2H), 8.56 (d,
J = 13.4 Hz, 2H); 13C NMR (400 MHz, DMSO-d6), d: 20.7, 26.4, 41.2,
49.7, 52.3, 62.2, 100.3, 113.7, 118.1, 123.5, 126.0, 131.3, 140.4,
143.5, 148.7, 174.9. Composition in theory: (x3H2O): C (42.65%), H
(5.61%), N (5.38%); Composition found: C (42.09%), H (5.80%), N
(5.26%); HRMS (ESI) calculated for [C37H52Br2ClN4]+ m/z 745.2247,
found m/z 745.2255.
sition found:
C (50.16%), H (6.78%), N (6.26%); HRMS (ESI)
calculated for [C37H54ClN4]3+ m/z 197.1331, found m/z 197.1337.
5.1.3.3. 2-((1E,3Z,5E)-3-Bromo-5-(3,3-dimethyl-1-(3-(trimethyl-
ammonio)propyl)indolin-2-ylidene)penta-1,3-dien-1-yl)-3,3-
dimethyl-1-(3-(trimethylammonio)propyl)-3H-indol-1-ium
bromide (15). Yield 66%; Mp 174–176 °C; 1H NMR (400 MHz,
DMSO-d6): d 1.77 (s, 12H), 2.31 (t, J = 7.1 Hz, 4H), 3.13 (s, 18H),
3.58–3.68 (m, 4H), 4.34 (t, J = 7.0 Hz, 4H), 6.40 (d, J = 13.2 Hz,
2H), 7.61 (d, J = 8.6 Hz, 2H), 7.76 (d, J = 8.6 Hz, 2H), 8.11 (s, 2H),
8.61 (d, J = 13.2 Hz); 13C NMR (100 MHz, DMSO-d6), d: 22.4, 27.6,
42.2, 51.8, 54.0, 64.2, 112.3, 118.3, 123.2, 127.1, 130.8, 143.6,
151.3, 176.1. Composition in theory: C (50.82), H (6.22), N (6.41);
Composition found: C (50.48), H (6.75), N (6.45). High-resolution
electrospray ionization (ESI) accurate mass spectra calculated m/z
for C37H54N4Br2 712.2715, found 712.2722.
5.1.3.8. 5-Chloro-2-((1E,3Z,5E)-3-chloro-5-(5-chloro-3,3-dimethyl-
1-(3-(trimethylammonio)propyl)indolin-2-ylidene)penta-1,
3-dien-1-yl)-3,3-dimethyl-1-(3-(trimethylammonio)propyl)-
3H-indol-1-ium bromide (20). Yield 77%; Mp 222–224 °C; 1H
NMR (400 MHz, DMSO-d6): d 1.76 (s, 12H), 2.17 (t, J = 7.2 Hz, 4H),
3.13 (s, 18 H), 3.65 (tt, J = 7.6 Hz, J = 7.2 Hz, 4H), 4.27 (t,
J = 7.6 Hz, 4H), 6.34 (d, J = 13.6 Hz, 2H), 7.51 (d, J = 8.4 Hz, 2H),
7.70 (d, J = 8.4 Hz, 2H), 7.93 (s, 2H), 8.73 (d, J = 13.6 Hz, 2H); 13C
NMR (100 MHz, DMSO-d6), d: 20.7, 26.4, 41.3, 49.7, 52.3, 62.1,
100.3, 113.3, 123.2, 123.5, 128.3, 130.0, 140.5, 143.4, 148.4,
174.3. Composition in theory: (x3H2O): C (46.63%), H (6.13%), N
(5.88%); Composition found: C (46.22%), H (6.12%), N (5.78%);
HRMS (ESI) calculated for [C37H52Cl3BrN4]+ m/z 736.2441, found
m/z 736.2468.
5.1.3.4. 5-Chloro-2-((1E,3E,5E)-5-(5-chloro-3,3-dimethyl-1-
(3-(trimethylammonio)propyl)indolin-2-ylidene)penta-1,
3-dien-1-yl)-3,3-dimethyl-1-(3-(trimethylammonio)propyl)-
3H-indol-1-ium bromide (16). Yield 88%; 1H NMR (400 MHz,
MeOD-d4): d 1.78 (s, 12H), 2.32 (s, 4H), 3.25 (s, 18H), 3.75 (t,
J = 8.4 Hz, 4H), 4.24 (t, J = 7.6 Hz, 4H), 6.60 (d, J = 13.2 Hz, 2H),
7.11 (t, J = 12.4 Hz, 1H), 7.50–7.43 (m, 4H), 7.58 (s, 2H), 8.35 (t,
J = 13.2 Hz, 2H). 13C NMR (400 MHz, MeOD-d4): d 20.89, 26.50,
40.58, 49.35, 52.57, 63.05, 104.20, 111.92, 122.72, 127.75, 128.49,
130.54, 140.59, 143.15, 155.05, 173.28. High-resolution electro-
spray ionization (ESI) accurate mass spectra calculated m/z for
[C37H53N4Cl2]3+ 207.7877 found 207.7755.
5.1.3.9. 5-Bromo-2-((1E,3Z,5E)-3-bromo-5-(5-bromo-3,3-dimethyl
-1-(3-(trimethylammonio)propyl)indolin-2-ylidene)penta-1,3-di
en-1-yl)-3,3-dimethyl-1-(3-(trimethylammonio)propyl)-3H-indol-
1-ium bromide (21).
Yield 77.6%; Mp 232–234 °C; 1H NMR
(400 MHz, DMSO-d6): d 1.76 (s, 12H), 2.18 (t, J = 7.1 Hz, 4H), 3.12 (s,
18H), 3.55–3.65 (m, 4H), 4.24 (t, J = 7.0 Hz, 4H), 6.34 (d, J = 13.2 Hz,
2H), 7.60 (d, J = 8.6 Hz, 2H), 7.66 (d, J = 8.6 Hz, 2H), 8.06 (s, 2H), 8.60
(d, J = 13.2 Hz, 2H); 13C NMR (400 MHz, DMSO-d6), d: 20.66, 26.46,
41.37, 49.79, 52.35, 62.25, 102.70, 113.74, 116.99, 118.22, 126.03,
131.24, 140.93, 143.72, 150.42, 174.34. Composition in theory:
(3 ꢀ H2O): C (40.91) H (5.38), N (5.16); Composition found: C (40.56),
H (5.37), N (5.07). High-resolution electrospray ionization (ESI) accu-
rate mass spectra calculated m/z for [C37H52N4Br4]+ 868.0925, found
868.0930.
5.1.3.5. 5-Bromo-2-((1E,3E,5E)-5-(5-bromo-3,3-dimethyl-1-(3-(tri
methylammonio)propyl)indolin-2-ylidene)penta-1,3-dien-1-yl)-3,
3-dimethyl-1-(3-(trimethylammonio)propyl)-3H-indol-1-ium
bromide (17). Yield 62%; Mp 237–240 °C; 1H NMR (400 MHz, MeOD-
d4):d1.782 (s, 12H), 2.318 (t, J = 7.1 Hz, 4H), 3.241 (s, 18H), 3.707–3.750
(m, 4H), 4.224 (t, J = 7.2 Hz, 4H), 6.642 (d, J = 13.6 Hz, 2H), 7.176 (t,
J = 12.8 Hz, 1H), 7.384 (d, J = 8.4 Hz, 2H), 7.611 (d, J = 8.4 Hz, 2H),
7.735 (s, 2H), 8.362 (t, J = 12.8 Hz, 2H); 13C NMR (400 MHz, MeOD-
d4): d 20.89, 26.06, 26.50, 40.55, 49.32, 52.60, 63.03, 104.27, 112.25,
117.93, 125.63, 131.45, 141.04, 143.44, 155.07, 173.11 High-resolution
electrospray ionization (ESI) accurate mass spectra calculated m/z for
[C37H52N4Br2]+ 710.2559, found 710.2552.
5.2. DNA and compound preparation
DNA oligonucleotides: Tel22, d[AGGG(TTAGGG)3]; c-myc,
d[(AGGGTGGGG)2A]; control duplex, d[CGAATTCGTTTTCGAATTC
G] with and without 50-biotin labels were purchased from Inte-
grated DNA Technologies (Coralville, IA) with HPLC purification.
NMR and mass spectral analysis performed on these sequences fur-
ther confirmed the purity. The concentration of oligonucleotides
was determined from absorbance at 260 nm with extinction coef-
ficients calculated by the nearest-neighbor method.31 Appropriate
stock solutions of the compounds were prepared in double deion-
ized water and diluted with the experimental buffer prior to use.
5.1.3.6. 2-((1E,3Z,5E)-3-Bromo-5-(5-chloro-3,3-dimethyl-1-(3-(tri
methylammonio)propyl)indolin-2-ylidene)penta-1,3-dien-1-yl)-
5-chloro-3,3-dimethyl-1-(3-(trimethylammonio)propyl)-3H-
indol-1-ium bromide (18). Yield 88%; Mp 238–240 °C; 1H NMR

7010
R. Nanjunda et al. / Bioorg. Med. Chem. 20 (2012) 7002–7011
Experiments were performed in 10 mM HEPES buffer containing
50 mM KCl and 3 mM EDTA adjusted to pH 7.3 using 1 N HCl.
two-site interaction model, RUmax=ligandðK1Cfree þ 2K1K2C2freeÞ=ð1þ
K1Cfree þ K1K2C2freeÞ using nonlinear least-squares optimization to
obtain an optimal fit with BIAevaluation (BIAcore Inc.) and Kaleida-
graph (Synergy Software) softwares. K1 and K2 are equilibrium
binding constants for two types of binding sites.
5.3. UV-thermal melting studies
Thermal denaturation studies were conducted on a Cary 300
BIO UV–visible spectrophotometer in quartz cuvettes of 1 cm
pathlength. DNA solutions were prepared at single-strand concen-
5.6. Isothermal titration calorimetry
trations in the range of 2–3
l
M in HEPES buffer. Samples of com-
ITC experiments were performed with a VP-ITC instrument
(MicroCal, Northampton, MA). Calorimetric titrations were con-
pound to DNA ratios ranging from 0:1 to 6:1 was prepared. The
absorbance of the oligonucleotides and the complexes was mon-
itored at the recommended wavelength of 295 nm for Tel22 and
260 nm for duplex DNA sequences as a function of temperature.
Cuvettes were mounted in a thermal block, and the solution tem-
peratures were monitored by a thermistor in a reference cuvette
with computer-controlled heating and cooling rates of 0.5 °C/min.
A total of four scans were collected (two heating and two cooling
ramps). Data were analyzed and plotted using Kaleidagraph 4.0
software.
ducted by injecting 10
lL of the ligand (200
lM in Tris/K+ buffer)
into the quadruplex solutions (20
l
M in Tris/K+ buffer) every
300 s for a total of 29 injections. Similar experiments were per-
formed to determine the heats of dilution of the ligand with buffer.
The heat produced for each injection of compound into DNA or buf-
fer was obtained by integration of the area under each peak of the
titration plots with respect to time. The heats of reaction were
obtained by subtraction of the integrated heats of dilution of the
compounds from the heats corresponding to the injection of com-
pound into DNA. Averaged subtraction was applied to ITC titration
data. Data corresponding to the first injection were discarded. The
5.4. Circular dichroism
binding enthalpy (DH) for each titration was obtained by fitting the
CD spectra were recorded using a Jasco J-810 spectrophotome-
ter with a 1-cm pathlength quartz cuvette at a scan speed of
50 nm/min and response time of 1 s over the wavelength range
results of heat per mole as a function of total molar ratio using a
single-site binding model. The data was analyzed by VPViewer
2000 and Origin 7.0 software.
of 230 nm to 700 nm. Quadruplex–DNA solutions (4–5 lM) were
annealed overnight prior to the collection of spectra. Appropriate
amount of ligands were sequentially titrated from the stock solu-
tion into the DNA solution in the cuvette until the desired mole ra-
tios of compound to quadruplex were obtained. The spectra were
averaged over four scans. A buffer scan was collected in the same
cuvette and subtracted from the average scan of each ratio. Data
were processed and plotted using Kaleidagraph 4.0 software.
5.7. Mass spectrometry
The mass spectrometry analysis was performed on a Waters
Q-TOF micromass spectrometer equipped with electrospray ioniza-
tion source (ESI) in negative mode (Waters Corporate, Milford, MA).
The mass range was from 500–3000 Da (displayed range in Fig. 5 is
from 1100–2100 m/z). Since mass spectrometry generally requires
DNA sequences dissolved in a volatile salt solution, all the se-
quences tested with mass spectrometry were dialyzed several
times with a 1000 Da cut-off membrane (Spectrum Laboratories
Inc., Rancho Dominguez, CA, USA) in 50 mM ammonium acetate
buffer at pH 7.2 in order to remove any low molecular weight impu-
rities from DNA. The final concentrations of the dialyzed DNA were
5.5. Surface plasmon resonance
Biosensor-SPR experiments were performed with
a four-
channel Biacore 2000 optical biosensor system (GE Healthcare,
Sweden). The streptavidin-derivatized gold chip (SA chip from
BIAcore) was prepared for use by conditioning with a series of
60 s injections of 1 M NaCl in 50 mM NaOH followed by extensive
washing with buffer. Biotinylated DNA samples (25–50 nM) in
HEPES buffer were immobilized on the flow cell surface by non-
covalent capture as previously described.32 Flow cell 1 was left
blank as a reference, while flow cells 2–4 were immobilized with
50-biotinylated DNA by manual injection of DNA stock solutions
determined as above (Section 5.2). Samples containing 25–30 lM
c-myc sequence and appropriate amounts of ligands in the desired
molar ratios to DNA were prepared in 50 mM ammonium acetate
buffer with the final volume of 100
l
L. The samples were intro-
duced into the ion source through direct infusion at 5
ll/min flow
rate. The instrument parameters were as follows: capillary voltage
of 2200 V, sample cone voltage of 30 V, extraction cone voltage of
2.5 V, desolvation temperature of 70 °C and source temperature of
100 °C. Nitrogen was used as nebulizing and drying gas. Spectra
were collected for 10 min and the last two minutes of the scan were
used for data analysis. Data analysis and interpretation were per-
formed using MassLynx 4.1 software.
(flow rate of 1
units was obtained (350–400 RU). Typically, a series of different
ligand concentrations (1 nM to 1 M) were injected onto the chip
(flow rate of 25 L/min, 5 min) until a constant steady-state
lL/min) until the desired amount of DNA response
l
l
response was obtained followed by a dissociation period (running
buffer, 5 min). After every cycle, the chip surface was regenerated
with 10 mM glycine solution (pH 2.5 for 30 s) followed by
multiple buffer injections to obtain a stable baseline for following
cycles.
The observed response (RUobs) in the steady-state region is
proportional to the amount of bound drug and was typically deter-
mined by linear averaging over 10–20 s or longer time span. The
Acknowledgments
The biophysical studies on cyanine-DNA interactions were sup-
ported by NIH NIAID Grant AI-064200 (WDW). The compound syn-
thesis for this study was supported by a Georgia Research Alliance
grant (MH). EAO was supported through internal funds from the
Research Initiation and Mentor Grants at GSU and MH appreciates
the Georgia State University Center for Diagnostics and Therapeu-
tics for their support.
predicted maximum response per bound compound (RUmax/ligand
)
was calculated from the DNA molecular weight, the compound
molecular weight, the amount of DNA immobilized on the flow
cell, and the refractive index gradient ratio of the compound and
DNA, as previously described.32,33 The RUmax/ligand value is required
to convert RUobs to determine the moles of drug bound to DNA. The
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmc.2012.10.008.
RUobs was plotted as a function of free ligand concentration (Cfree
and the equilibrium binding constants were determined using a
)

R. Nanjunda et al. / Bioorg. Med. Chem. 20 (2012) 7002–7011
16. Paramasivan, S.; Rujan, I.; Bolton, P. H. Methods 2007, 43, 324.
7011
References and notes
17. Rezler, E. M.; Seenisamy, J.; Bashyam, S.; Kim, M.-Y.; White, E.; Wilson, W. D.;
Hurley, L. H. J. Am. Chem. Soc. 2005, 127, 9439.
1. Neidle, S.; Balasubramanian, S. Quadruplex nucleic acids; RSC Pub: Cambridge,
2006.
18. Phan, A. T.; Kuryavyi, V.; Patel, D. J. Curr. Opin. Struct. Biol. 2006, 16, 288.
19. Hurley, L.; Wheelhouse, R.; Sun, D.; Kerwin, S.; Salazar, M.; Fedoroff, O.; Han, F.;
Han, H.; Izbicka, E.; Von Hoff, D. Pharmacol. Ther. 2000, 85, 141.
20. Buchini, S.; Buschiazzo, A.; Withers, S. G. Angew. Chem., Int. Ed. 2008, 47, 2700.
21. Maehr, H. Bioorg. Med. Chem. 1997, 5, 473.
22. Chessari, G.; Woodhead, A. J. Drug Discovery Today 2009, 14, 668.
23. Hernandes, M. Z.; Cavalcanti, S. M.; Moreira, D. R.; Azevedo, F.; Leite, A. C. Curr.
Drug Targets 2010, 11, 303.
24. Gray, R. D.; Li, J.; Chaires, J. B. J. Phys. Chem. B. 2009, 113, 2676.
25. Petraccone, L.; Garbett, N. C.; Chaires, J. B.; Trent, J. O. Biopolymers 2010, 93,
533.
26. Dai, J.; Carver, M.; Yang, D. Biochimie 2008, 90, 1172.
27. Garner, T. P.; Williams, H. E. L.; Gluszyk, K. I.; Roe, S.; Oldham, N. J.; Stevens, M.
F. G.; Moses, J. E.; Searle, M. S. Org. Biomol. Chem. 2009, 7, 4194.
28. Kim, M.-Y.; Gleason-Guzman, M.; Izbicka, E.; Nishioka, D.; Hurley, L. H. Cancer
Res. 2003, 63, 3247.
29. Debray, J.; Zeghida, W.; Jourdan, M.; Monchaud, D.; Dheu-Andries, M.-L.;
Dumy, P.; Teulade-Fichou, M.-P.; Demeunynck, M. Org. Biomol. Chem. 2009, 7,
5219.
30. Peng, X.; Yang, Z.; Wang, J.; Fan, J.; He, Y.; Song, F.; Wang, B.; Sun, S.; Qu, J.; Qi,
J.; Yan, M. J. Am. Chem. Soc. 2011, 133, 6626.
31. Fasman, G. D. Handbook of Biochemistry and Molecular Biology, Nucleic Acids;
CRC Press: Cleveland, 1975.
32. Nanjunda, R.; Munde, M.; Liu, Y.; Wilson, W. In Methods for Studying Nucleic
Acid/Drug Interactions; Tor, Y., Wanunu, M., Eds.; CRC Press: Boca Raton, 2011;
p 91.
2. Neidle, S.; Parkinson, G. N. Biochimie 2008, 90, 1184.
3. Neidle, S.; Thurston, D. E. Nat. Rev. Cancer 2005, 5, 285.
4. Han, H.; Hurley, L. Trends Pharmacol. Sci. 2000, 21, 136.
5. Drygin, D.; Siddiqui-Jain, A.; O’Brien, S.; Schwaebe, M.; Lin, A.; Bliesath, J.; Ho, C.
B.; Proffitt, C.; Trent, K.; Whitten, J. P.; Lim, J. K. C.; Von Hoff, D.; Anderes, K.;
Rice, W. G. Cancer Res. 2009, 69, 7653.
6. Monchaud, D. Org. Biomol. Chem. 2008, 6, 627.
7. Shi, D. F.; Wheelhouse, R. T.; Sun, D.; Hurley, L. H. J. Med. Chem. 2001, 44, 4509.
8. White, E. W.; Tanious, F.; Ismail, M. A.; Reszka, A. P.; Neidle, S.; Boykin, D. W.;
Wilson, W. D. Biophys. Chem. 2007, 126, 140.
9. Henary, M.; Mojzych, M. In Heterocyclic Polymethine Dyes: Synthesis, Properties
and Applications; Strekowski, L., Ed.; Springer: Berlin, 2008; Vol. 14, p 221.
10. Mojzych, M.; Henary, M. In Heterocyclic Polymethine Dyes: Synthesis, Properties
and Applications; Strekowski, L., Ed.; Springer: Berlin, 2008; Vol. 14, p 1.
11. Yang, Q.; Xiang, J.; Yang, S.; Zhou, Q.; Li, Q.; Tang, Y.; Xu, G. Chem. Commun.
2009, 1103.
12. Kerwin, S. M.; Sun, D.; Kern, J. T.; Rangan, A.; Thomas, P. W. Bioorg. Med. Chem.
Lett. 2001, 11, 2411.
13. Rodger, A.; Norden, B. Circular Dichroism and Linear Dichroism; University Press:
New York, Oxford, 1997.
14. Nanjunda, R.; Musetti, C.; Kumar, A.; Ismail, M. A.; Farahat, A. A.; Wang, S.;
Sissi, C.; Palumbo, M.; Boykin, D. W.; Wilson, W. Curr. Pharm. Des. 1934, 2012,
18.
15. Munde, M.; Kumar, A.; Nhili, R.; Depauw, S.; David-Cordonnier, M.-H.; Ismail,
M. A.; Stephens, C. E.; Farahat, A. A.; Batista-Parra, A.; Boykin, D. W.; Wilson, W.
D. J. Mol. Biol. 2010, 402, 847.
33. Davis, T. M.; Wilson, W. D. Anal. Biochem. 2000, 284, 348.