Development of an engineered carbazole/thiazole orange conjugating probe for G-quadruplexes

Article abstract of DOI:10.1016/j.dyepig.2016.10.022

The development of selective and sensitive probes for sensing G-quadruplexes either in vitro or in cellulo has been the focus of investigation for a long time. Of those investigated, 3,6-bis(1-methyl-4-vinylpyridinium) carbazole diiodide (BMVC) is a promising fluorescent probe utilized in many studies investigating G-quadruplex structures. However, its shortcomings, including similar fluorescence responses for G-quadruplex and duplex DNA and green but not red fluorescent emission, might restrict the further application of BMVC. In this study, in order to improve the selectivity and optical properties of BMVC, we engineered a new probe (TO-CZ) by incorporating thiazole orange (TO) into the structure of BMVC. We next found that TO-CZ can act as a colorimetric and red-emitting fluorescent dual probe selective for G-quadruplexes without affecting the G-quadruplex topology. Further experiments showed that a 2:1 binding model involving the external binding of TO-CZ to both ends of the G-quadruplex is the most possible binding mode. Furthermore, we applied TO-CZ in sensing G-quadruplexes both in vitro and in cellulo, and found that TO-CZ can selectively and sensitively visualize G-quadruplexes. Notably, TO-CZ might be used to map DNA and RNA G-quadruplexes in cellulo, showing its great potential in investigating intracellular G-quadruplex structures.

Full text of DOI:10.1016/j.dyepig.2016.10.022

Accepted Manuscript
Development of an engineered carbazole/thiazole orange conjugating probe for G-
quadruplexes
Ming-Hao Hu, Rui-Jun Guo, Shuo-Bin Chen, Zhi-Shu Huang, Jia-Heng Tan
PII:
S0143-7208(16)30827-0
10.1016/j.dyepig.2016.10.022
DYPI 5538
DOI:
Reference:
To appear in: Dyes and Pigments
Received Date: 26 September 2016
Revised Date: 19 October 2016
Accepted Date: 19 October 2016
Please cite this article as: Hu M-H, Guo R-J, Chen S-B, Huang Z-S, Tan J-H, Development of an
engineered carbazole/thiazole orange conjugating probe for G-quadruplexes, Dyes and Pigments
(2016), doi: 10.1016/j.dyepig.2016.10.022.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Development of an engineered carbazole/thiazole orange conjugating
probe for G-quadruplexes
Ming-Hao Hu,† Rui-Jun Guo,† Shuo-Bin Chen, Zhi-Shu Huang and Jia-Heng Tan*
School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou, 510006, China
† These authors contributed equally.
* Corresponding author. Tel: +86-20-39943053; Email: tanjiah@mail.sysu.edu.cn.
Abstract
The development of selective and sensitive probes for sensing G-quadruplexes either in vitro
or in cellulo has been the focus of investigation for a long time. Of those investigated,
3,6-bis(1-methyl-4-vinylpyridinium) carbazole diiodide (BMVC) is a promising fluorescent
probe utilized in many studies investigating G-quadruplex structures. However, its
shortcomings, including similar fluorescence responses for G-quadruplex and duplex DNA and
green but not red fluorescent emission, might restrict the further application of BMVC. In this
study, in order to improve the selectivity and optical properties of BMVC, we engineered a new
probe (TO-CZ) by incorporating thiazole orange (TO) into the structure of BMVC. We next
found that TO-CZ can act as a colorimetric and red-emitting fluorescent dual probe selective
for G-quadruplexes without affecting the G-quadruplex topology. Further experiments showed
that a 2:1 binding model involving the external binding of TO-CZ to both ends of the
G-quadruplex is the most possible binding mode. Furthermore, we applied TO-CZ in sensing
G-quadruplexes both in vitro and in cellulo, and found that TO-CZ can selectively and
sensitively visualize G-quadruplexes. Notably, TO-CZ might be used to map DNA and RNA
G-quadruplexes in cellulo, showing its great potential in investigating intracellular
G-quadruplex structures.
Keywords
G-quadruplex, carbazole derivative, colorimetric probe, fluorescent probe, intracellular imaging
1

ACCEPTED MANUSCRIPT
1. Introduction
The development of highly selective and sensitive probes to detect nucleic acids is of profound
importance to the fields of biology and clinical diagnostics [1]. G-quadruplexes are unique
four-stranded structures formed by guanine-rich nucleic acid sequences. The basic structural
unit of this structure is the G-quartet, which is derived from the association of four guanines
into a cyclic Hoogsteen hydrogen-bonding arrangement [2]. G-quadruplex structures have
been identified in the human genome with enrichment in telomeres, rDNA, promoter regions,
untranslated regions of mRNA, and the first exons and first introns of many genes and are
known to be highly associated with human diseases [3, 4]. Accordingly, G-quadruplexes have
captivated much attention because of their biological significance in recent years [5-7]. The
ever-increasing interest in G-quadruplexes has resulted in the development of rapid and facile
approaches for the sensitive and selective detection of these structures. Notably,
G-quadruplex probes can be employed in the DNA-based sensing assays to detect specific
analytes. Such functions further expand their scope of application [8, 9]. Thus, visualizing
G-quadruplexes using chemical sensors is an extremely active area of research, and
significant progress has been made towards the development of colorimetric or fluorescent
probes for G-quadruplexes [10-12].
Carbazole derivatives exhibit a wide range of pharmacological activities. Recently, a few
carbazole derivatives have been reported to bind to G-quadruplexes [13-16]. Among them,
3,6-bis(1-methyl-4-vinylpyridinium) carbazole diiodide (BMVC) has attracted much attention. It
is not only an effective G-quadruplex stabilizer but also a valid fluorescent probe of
G-quadruplexes [17]. The fluorescent emission of BMVC is located at 570 nm upon interaction
with G-quadruplex structures, while it is located at approximately 545 nm in the presence of
duplex DNA. Thus, the different fluorescent colors of BMVC can be used to discriminate
between duplex and G-quadruplex DNA [13]. Such a probe has been applied in many studies
to reveal G-quadruplex structures either in vitro or in cellulo [18-24]. However, BMVC has very
similar affinities for duplex and G-quadruplex DNA, showing similar fluorescence
enhancements and emission bands upon binding to G-quadruplex and duplex DNA. This
2

ACCEPTED MANUSCRIPT
impedes the further application of BMVC, especially in selective sensing G-quadruplexes in
the presence of abundant double-stranded DNA both in vitro and in cellulo. Additionally, BMVC
exhibits green rather than red emission, which might again restrict its in cellulo and in vivo
imaging. It is notable that red-emitting fluorescent probes offer various advantages, including
minimum photodamage to biological samples, diminished scattering of light, deep tissue
penetration and minimum autofluorescence [25]. Thus, seeking a new carbazole probe
possessing better properties, such as high selectivity and red emission, for G-quadruplex is
important. Notably, one effective strategy for increasing the selectivity of a ligand for
G-quadruplexes over duplex DNA is to expand its aromatic system by incorporating another
aromatic moiety into the original molecule [26-30]. Such a modification might improve its
fluorescence emission property when a fluorophore is introduced [29, 30]. Therefore, we
attempt to introduce a promising fluorophore into the BMVC scaffold. Such a conjugation
strategy requires a suitable molecular framework to accommodate the BMVC scaffold.
Therefore, we payed attention to thiazole orange (TO), because TO and BMVC share the
same positive 1-methylpyridine moiety, making the two structures be easily fused into a new
molecule. Notably, TO is one of the most widely used fluorescent probes in nucleic acid
staining because of its high fluorescence quantum yield [31]. Although TO is non-selective for
double-stranded and G-quadruplex nucleic acids, it offers an attractive template and serves as
an outstanding fluorophore for the design of selective fluorescent probes for G-quadruplexes.
Several probes for sensing G-quadruplexes have been discovered based on TO, with much
better capabilities than their original molecules [26, 28, 32]. For all concerned, we are inspired
to assemble BMVC and TO motifs into a fusion scaffold that is very convenient to be
synthesized (Fig. 1). And we hope that such a modification would optimize the photophysical
properties and selectivity of BMVC for G-quadruplexes.
In this study, a new carbazole/thiazole orange conjugate named TO-CZ was designed and
synthesized. The detailed interactions of TO-CZ with various G-quadruplexes,
double-stranded and single-stranded nucleic acids were investigated using UV-vis
spectroscopic assays, fluorescence spectroscopic assays, circular dichroism spectroscopic
assays and molecular modeling studies. TO-CZ was found to be a colorimetric and
red-emitting fluorescent dual-output probe for G-quadruplexes with excellent selectivity. In
3

ACCEPTED MANUSCRIPT
addition, the detection abilities of the probe for G-quadruplexes were examined in both
solution and a gel matrix. Furthermore, its potential application in cell imaging was explored.
2. Material and methods
2.1. Synthesis and characterization
2.1.1. General methods
1H and 13C NMR spectra were recorded using TMS as the internal standard in DMSO-d6 with a
Bruker BioSpin GmbH spectrometer at 500 MHz and 126 MHz, respectively. Mass spectra
(MS) were recorded on a Shimadzu LCMS-2010A instrument with an ESI mass selective
detector and high resolution mass spectra (HRMS) were recorded on Shimadzu LCMS-IT-TOF.
Flash column chromatography was performed with silica gel (200–300 mesh) purchased from
Qingdao Haiyang Chemical Co. Ltd. The purity of synthesized compound was confirmed to be
higher than 95% by using analytical HPLC with a dual pump Shimadzu LC-20 AB system
equipped with a Ultimate XB-C18 column (4.6×250 mm, 5 ꢀm), and eluted with
methanol-water (80:20) containing 0.1% TFA at a flow rate of 1.0 mL/min. All chemicals were
purchased from commercial sources unless otherwise specified. All the solvents were of
analytical reagent grade and were used without further purification.
2.1.2. Synthesis of intermediates (1, 2 and 3)
The intermediates (1, 2, and 3) were prepared using the methods following the procedures
shown in Scheme 1 by our previous report [26, 33].
2.1.3.
Synthesis
of
(Z)-1,2-dimethyl-4-((3-methylbenzo[d]thiazol-2(3H)-ylidene)methyl)quinoli-nium iodide (4)
To a solution of 3 (0.54 g, 1.7 mmol) and 2,3-dimethylbenzothiazolium iodide (0.50 g, 1.7 mmol)
in 40 ml methanol, sodium bicarbonate (0.30 g, 3.4 mmol), dissolved in 10 mL water was
added. The reaction mixture was stirred at room temperature for 1 h and then reflux for 30 min.
After cooling, aqueous solution of excess potassium iodide was added into the mixture. The
precipitate was filtered, washed with water, acetone and dried under vacuum to afford the
product as a dark brown solid 4 (0.61 g, 80%). 1H NMR (500 MHz, DMSO-d6) δ 8.77 (d, J = 8.2
4

ACCEPTED MANUSCRIPT
Hz, 1H), 8.18 (d, J = 8.8 Hz, 1H), 8.03 – 7.96 (m, 2H), 7.78 – 7.72 (m, 2H), 7.60 (t, J = 7.8 Hz,
1H), 7.40 (t, J = 7.6 Hz, 1H), 7.36 (s, 1H), 6.86 (s, 1H), 4.07 (s, 3H), 3.99 (s, 3H), 2.88 (s, 3H).
13C NMR (126 MHz, DMSO-d6) δ 159.67, 154.77, 148.57, 141.06, 139.58, 133.63, 128.54,
126.87, 125.86, 124.73, 124.16, 123.97, 123.23, 118.79, 113.17, 111.11, 87.39, 37.56, 34.09,
23.23. ESI-MS m/z: 319.1 [M-I]+.
2.1.4.
Synthesis
of
2-((E)-2-(9-ethyl-9H-carbazol-3-yl)vinyl)-1-methyl-4-((Z)-(3-methylbenzo[d]-thiazol-2(3H)-ylide
ne)methyl)quinolinium iodide (TO-CZ)
A mixture of 4 (0.45 g, 1.0 mmol), N-ethyl-3-carbazolecarboxaldehyde (0.25 g, 1.1 mmol) and
3 drops of piperidine in 4 mL of n-butanol was heated at 80 °C overnight. After cooli ng, the
precipitate was filtered off, washed with n-butanol, and then purified on silica gel
chromatography (CH2Cl2:CH3OH=50:1) to afford the dark red product TO-CZ (0.40 g, 61%). 1H
NMR (500 MHz, DMSO-d6) δ 8.76 (s, 1H), 8.72 (d, J = 8.1 Hz, 1H), 8.23 (d, J = 7.7 Hz, 1H),
8.16 (d, J = 8.7 Hz, 1H), 8.10 – 8.06 (m, 2H), 7.98 – 7.94 (m, 1H), 7.88 (d, J = 15.7 Hz, 1H),
7.76 (d, J = 15.8 Hz, 1H), 7.74–7.65 (m, 5H), 7.56 (t, J = 8.3 Hz, 1H), 7.52 (t, J = 7.2 Hz, 1H),
7.36 (t, J = 7.8 Hz, 1H), 7.28 (t, J = 7.4 Hz, 1H), 6.85 (s, 1H), 4.49 (q, J = 7.0 Hz, 2H), 4.20 (s,
3H), 3.96 (s, 3H), 1.34 (t, J = 7.1 Hz, 3H). 13C NMR (126 MHz, DMSO-d6) δ 159.39, 153.04,
148.01, 143.19, 141.33, 141.13, 140.64, 139.61, 133.58, 128.48, 126.94, 126.88, 126.83,
126.78, 125.65, 124.51, 124.19, 124.04, 123.52, 123.13, 122.72, 122.34, 121.10, 120.06,
119.06, 118.32, 112.96, 110.21, 110.12, 108.28, 87.98, 38.51, 37.72, 34.04, 14.27. Purity: 95.6%
by HPLC. HRMS (ESI) m/z: calcd for (M-I)+ (C35H30N3S+) 524.2155, found 524.2168.
2.2. Oligonucleotides and compound
All oligonucleotides (Supplementary Table S1) used in this study were purchased from Sangon
(China). All oligonucleotides were dissolved in relevant buffers, and their concentrations were
determined from the absorbance at 260 nm using a NanoDrop 1000 Spectrophotometer
(Thermo Scientific, USA) based on the respective molar extinction coefficients. To obtain
G-quadruplex formation information, oligonucleotides were annealed in relevant buffers
containing 100 mM of KCl with heating at 95 °C for 5 minutes, followed by gradual cooling to
5

ACCEPTED MANUSCRIPT
room temperature. The oligonucleotides engaged in G-quadruplex formation, as determined
by circular dichroism (CD) measurements. Stock solutions of TO-CZ (10 mM) were dissolved
in DMSO and stored at -80 °C. Further dilutions of nucleic acid samples and TO-CZ to the
working concentrations were performed with the relevant buffer immediately prior to use.
2.3. UV-vis spectroscopic studies
UV-vis spectroscopic studies were performed on a UV-2450 spectrophotometer (Shimadzu,
Japan) using
1 cm path length quartz cuvettes. For the titration experiment, all
oligonucleotides were first prepared by heating at 95 °C for 5 min, followed by slow cooling to
room temperature. Small aliquots of a stock solution of oligonucleotide were added into the
solution containing TO-CZ at a fixed concentration (5 ꢀM) in Tris-HCl buffer (10 mM, pH = 7.2)
with 100 mM KCl. The final concentration of oligonucleotide was varied from 0 to 20 ꢀM. After
each addition of oligonucleotide, the reaction was stirred and allowed to equilibrate for at least
2 minutes, and the absorbance measurement was collected.
2.4. Fluorescence spectroscopic studies
Fluorescence studies were performed on a Fluoromax-4 luminescence spectrophotometer
(HORIBA, USA). A quartz cuvette with a 2 mm × 10 mm path length was used for spectral
recording at 5 nm excitation and emission slit widths unless otherwise specified. For titration
experiments, all oligonucleotides were first prepared by heating at 95 °C for 5 minutes followed
by slow cooling to room temperature. Small aliquots of a stock solution of oligonucleotides
were added into the solution containing TO-CZ at a fixed concentration (3 ꢀM) in Tris-HCl
buffer (10 mM, pH = 7.2) with 100 mM KCl. The final concentration of sample was varied from
0 to 12 ꢀM. After each addition of sample, the reaction was stirred and allowed to equilibrate
for at least 2 minute, and the fluorescence measurement was read when excited at 467 nm.
The fluorescence quantum yields (ΦF) of TO-CZ with various nucleic acids were calculated
relative to a standard solution of Nile Red in methanol (ΦF = 0.38) [34] and determined by
using the following formula: Φu = Φs (As/Au) × (Iu/Is), where Φ is the fluorescence quantum
yield, I is the measured integrated emission intensity, and A is the optical density (absorbance).
The u refers to the compound of unknown quantum yield, and s refers to the reference
6

ACCEPTED MANUSCRIPT
compound (Nile Red) of known quantum yield. The fluorescence spectra were recorded at 5
nm excitation and emission slit widths to determine Φ.
The LOD value of TO-CZ for G-quadruplex in solution was calculated on the basis of the
equation LOD = K × Sb/m. The K value is generally taken to be 3 according to the IUPAC
recommendation. The Sb value represents the standard deviation for multiple measurements
(n = 20) of blank solution. The m value is the slope of the calibration curve, which was derived
from the linear range of a TO-CZ fluorescence titration curve with G-quadruplex and standards
for the sensitivity of this method [35].
2.5. Circular dichroism spectroscopic studies
CD studies were performed on a Chirascan circular dichroism spectrophotometer (Applied
Photophysics, UK). A quartz cuvette with a 4 mm path length was used for recording of spectra
over a wavelength range of 230–330 nm with a 1 nm bandwidth, 1 nm step size and time of 0.5
s per point. CD spectroscopy was performed at a fixed concentration of G-quadruplex (3 ꢀM),
either with or without a fixed concentration (15 ꢀM) of TO-CZ in Tris-HCl buffer (10 mM, pH 7.2)
with 100 mM KCl.
2.6. Molecular modeling studies
The G-quadruplex structures of pu22 was generated based on their NMR structures (PDB ID:
2L7V) [36] using Discovery Studio 2.5 and optimized with AMBER10. The structure of TO-CZ
was constructed and optimized with Gaussian 03 using the HF/6-31G* basis set. Docking
analyses were performed with the AUTODOCK 4.2 program. The dimensions of the active site
box that was placed at the center of the G-quadruplex were set to 110 Å × 110 Å × 110 Å with a
grid spacing of 0.375 Å. Two hundred independent docking runs were performed for each
model using the Lamarckian genetic algorithm (LGA).
The calculation of molecular electrostatic potentials was performed in Gaussian03. Structure
optimizations for the small molecules and the single-point energy calculation were computed
at the B3LYP/6-31G* level of theory. Graphical representations of the electrostatic potential
surface maps were generated by mapping the ESP onto a molecular surface using Gaussian
View.
7

ACCEPTED MANUSCRIPT
2.7. Gel electrophoresis studies
Different concentrations (2.5 µM to 20 µM) of oligonucleotides were incubated with 20 µM of
TO-CZ, and then the mixtures were loaded onto a 20% bisacrylamide gel in 1× TBE buffer
containing 100 mM KCl and electrophoresed at 4 °C. After that, nucleic acid fragments were
visualized under UV light and photographed by using AlphaImager EC (ProteinSimple), and
then the same gel was stained by the commercial staining agent SYBR Green I (1×, 20
minutes).
2.8. Confocal laser scanning microscopy assays
SiHa cells grown on glass coverslips were fixed with 4% paraformaldehyde in PBS for 15 min,
and then with 0.1% Triton-X 100/PBS at 37 °C for 30 min. Cells were incubated with 100
units·mL-1 RNase-free DNase I, or 100 units·mL-1 DNase and Protease-free RNase A, or PBS
as control, at 37 °C for 10 h. Cells were stained w ith 10 ꢀM of TO-CZ and 0.5 ꢀg·mL-1 DAPI in
PBS for 30 min at 37 °C. The cells were imaged usin g a LSM 710 laser scanning confocal
microscope with a 63× objective lens. The excitation wavelengths were 405 nm for DAPI and
514 nm for TO-CZ, respectively. The fluorescence signals were collected via two channels.
One is from 409 nm to 533 nm for DAPI, and the other is from 539 nm to 753 nm for TO-CZ.
3. Results and discussion
3.1. Synthesis of the carbazole/thiazole orange conjugating probe TO-CZ
The desired probe, TO-CZ, was prepared through the condensation of a carbazole aldehyde
and a thiazole orange moiety, and its structure and purity were confirmed by 1H and 13C NMR
spectrometry, HRMS spectrometry, and HPLC analysis (Scheme 1 and Supplementary Fig.
S1–4).
3.2. UV-vis spectroscopic studies of TO-CZ upon interaction with nucleic acids
To investigate the interactions of TO-CZ with nucleic acids, an UV-vis spectroscopic titration
was performed. As shown in Fig. 2A, upon the addition of the G-quadruplex pu22, the intensity
of TO-CZ at 516 nm gradually increased and finally shifted to 546 nm, which was
8

ACCEPTED MANUSCRIPT
accompanied by a marked and vivid color change from pale purple to dark purple in ambient
light. In contrast, duplex DNA only caused a hypochromic effect at 516 nm (Supplementary Fig.
S5). The above results indicated that TO-CZ could be used as a colorimetric probe to
distinguish G-quadruplexes from other nucleic acid forms with high specificity.
3.3. Fluorescence spectroscopic studies of TO-CZ upon interaction with nucleic acids
The detailed fluorescence properties of TO-CZ with various G-quadruplexes and other nucleic
acids were explored by using fluorescence titration assays. As shown in Fig. 2B, TO-CZ alone
in buffer displayed weak fluorescence emission. With the gradual addition of the G-quadruplex
pu22, an emission peak at approximately 613 nm became significantly enhanced. The turn-on
fluorescence changes could be observed by the naked eye under UV light. It is also
noteworthy that the red emission of TO-CZ is very attractive for cellular imaging. In contrast,
negligible fluorescence enhancement was observed when titrating TO-CZ with duplex DNA
(Supplementary Fig. S6). Furthermore, we extended the test sequences (Supplementary Table
S1) and used the parallel G-quadruplex pu22, hybrid G-quadruplex (UpsB, tel26 and tel21),
antiparallel G-quadruplex HRAS, duplex DNA (hairpin and ds15), and single-stranded DNA
(T21 and A21) for titration experiments. As shown in Fig. 2C, all the G-quadruplexes triggered
notable fluorescence, while negligible fluorescence enhancement was observed when titrating
TO-CZ with either single-stranded DNA or duplex DNA. The corresponding fluorescence
quantum yield values of TO-CZ with different nucleic acids were obtained as well
(Supplementary Table S2). Such data were in agreement with the results of the titration
experiments, which showed that fluorescence enhancement was much more pronounced for
G-quadruplexes, confirming its utility as a specific probe for in vitro investigations of
G-quadruplex structures. Notably, TO-CZ had a much higher selectivity for G-quadruplexes
than BMVC in respect to fluorescence enhancement.
Furthermore, a competition titration experiment was performed to confirm the selective
fluorescence response of TO-CZ binding to G-quadruplexes. As shown in Fig. 2D, although
100 ꢀM ct-DNA led to a slight increase in the fluorescence signal, with the gradual addition of
pu22 into a solution containing TO-CZ and ct-DNA, the enhanced fluorescence trend was
similar to that observed in the experiment without ct-DNA. This finding revealed that the
9

ACCEPTED MANUSCRIPT
G-quadruplex pu22 is able to displace bound ct-DNA very effectively and create a much
stronger fluorescent adduct with TO-CZ. This result proved that TO-CZ has a high preference
for binding to G-quadruplexes under competitive biological assay conditions, again
demonstrating that TO-CZ has promising utility in the selective detection of G-quadruplexes.
3.4. Effects of TO-CZ on the topology of G-quadruplexes
An ideal probe should not affect the topology of the targeted G-quadruplexes. Thus, we
investigated the topology of G-quadruplexes upon the addition of TO-CZ using circular
dichroism (CD) spectroscopy. In the absence of TO-CZ, the parallel G-quadruplex pu22
exhibited a characteristic positive peak at 265 nm and a negative peak at 240 nm. The hybrid
G-quadruplex htg22 showed a positive peak at 295 nm and a characteristic shoulder peak at
265 nm. As shown in Fig. 3, TO-CZ had a negligible impact on the CD spectra of these two
G-quadruplexes. Even when the mixtures of the G-quadruplexes and TO-CZ were subjected
to extreme heat denaturation followed by co-annealing, the characteristic peaks of the
G-quadruplexes were only slightly affected. Collectively, TO-CZ can sense G-quadruplexes
without affecting the G-quadruplex topology.
3.5. Binding mode of TO-CZ to G-quadruplexes
To obtain additional details on the interaction of TO-CZ with the tested G-quadruplexes, we
further investigated the binding mode of TO-CZ to the G-quadruplexes. We first used Job plot
analysis to determine the binding stoichiometry of TO-CZ to the G-quadruplexes. As shown in
Fig. 4A, TO-CZ bound to the G-quadruplex pu22 at a stoichiometry of 2:1. It is noteworthy that
TO and BMVC preferred to end-stack on the G-quartets [37, 38]. Consequently, we assumed
that TO-CZ might stack on the two terminal G-quartet planes of pu22. To test this hypothesis,
we compared it to an end-stacking ligand, SYUIQ-05 [36]. We gradually added TO-CZ into the
solution containing 3 µM of pu22 incubated with or without SYUIQ-05, and then, the
fluorescence intensities at 613 nm were collected. As shown in Fig. 4B, we observed that the
fluorescence of TO-CZ was largely reduced when SYUIQ-05 co-existed with pu22 in the
titrated solution, demonstrating that TO-CZ and SYUIQ-05 might share the same binding sites
in the G-quadruplex.
10

ACCEPTED MANUSCRIPT
On the bases of the above results, a molecular docking study was performed using the
AUTODOCK 4.2 program to help understand the binding model of TO-CZ to the G-quadruplex.
The NMR G-quadruplex structure of pu22 (PDB ID: 2L7V) was used as the template [36]. As
shown in Fig. 4C and 4D, a 2:1 model of two TO-CZ molecules bound to one G-quadruplex in
the end-stacking model was obtained. TO-CZ stacked on the two terminal G-quartet planes of
pu22 to form a tight complex, and this binding orientation of TO-CZ could maximize π-π
stacking interactions with the G-quartets, which resulted in the significant fluorescence
enhancement.
3.6. Insights into the enhanced selectivity of TO-CZ for G-quadruplexes
The above results demonstrated that the selectivity of TO-CZ for G-quadruplexes is much
better than those of TO and BMVC. Then, we were interested in what led to the difference on
the selectivity by molecular modeling studies. On one hand, compared to TO and BMVC, the
aromatic system of TO-CZ was indeed expanded (Supplementary Fig. S7), which reduced the
interaction between TO-CZ and duplex DNA. This was in agreement with what we anticipated
when designing such a structure. On the other, electrostatic potential (ESP) surfaces were
mapped for TO-CZ, TO and BMVC, because previous study demonstrated the ESP surfaces
might contribute much to the difference on the selectivity [39]. As shown in Supplementary Fig.
S7, the ESP surfaces of TO-CZ, TO and BMVC were quite different. For BMVC, a more
positively charged surface was observed, and the most positively charged surfaces were
located near the two N-quaternized position, which would enhance the binding of BMVC to the
phosphate backbones of nucleic acids, including G-quadruplex and duplex DNA. Such
non-specific, strong electrostatic interactions might cause strong binding of BMVC to duplex
DNA, and thus weaken its selectivity for G-quadruplex over duplex DNA. In comparison,
TO-CZ displayed a much smaller ESP value, which could provide an appropriate fit for the
G-quartet and might reduce the non-specific electrostatic interactions with the phosphate
groups upon incubation with duplex DNA, leading to a better selectivity. Collectively, we
supposed that both the size and ESP surface of TO-CZ resulted in the excellent selectivity for
G-quadruplexes.
11

ACCEPTED MANUSCRIPT
3.7. Application of TO-CZ in sensing G-quadruplexes in vitro
Encouraged by the strong fluorescence emission enhancement and improved quantum yield
values of TO-CZ bound to the G-quadruplex, we then investigated the detection limit
associated with using TO-CZ to identify the G-quadruplex in solution. As shown in Fig. 5A, the
linear range of the fluorescence titration curve for TO-CZ with the G-quadruplex pu22 ranged
from 0 to 1,500 nM, and the corresponding limit of detection (LOD) value was calculated as 2.4
nM. Thus, TO-CZ can act as a highly sensitive probe for detecting G-quadruplexes in solution.
In addition to the LOD value of TO-CZ in solution, we were also interested in whether TO-CZ
could act as a G-quadruplex staining agent in electrophoresis gels. Fig. 5B shows the
fluorescence bands of TO-CZ in prestained gels with increasing concentrations of
G-quadruplex and duplex DNA. Clear dose-response fluorescence bands were observed for
G-quadruplex DNA, while the bands of duplex DNA were hardly detectable, even at the
highest concentration. On the basis of this dose-dependent staining experiment, we further
determined the LOD value of TO-CZ for G-quadruplex pu22 in a gel. Herein, the LOD was
defined as the amount of nucleic acid that forms an easily detectable clear band in the gel [40].
Thus, in this case, the LOD for pu22 in a gel was determined as 25 pmol. In contrast, as shown
in Fig. 5C, both G-quadruplex and duplex DNA bands were clearly observed when applying
SYBR Green I as a gel stain. Such results suggest the potential value of TO-CZ as a selective
gel stain for G-quadruplexes.
3.8. Application of TO-CZ in sensing G-quadruplexes in cellulo
The application of the dye as a selective staining agent for imaging G-quadruplexes in cancer
cells (SiHa cells) was investigated using confocal laser scanning microscopy. TO-CZ was able
to induce an obvious fluorescence response in fixed and permeabilized SiHa cells, including
the nucleus and certain regions of cytoplasm (Fig. 6). Notably, a distinct strong emission
response to the nucleoli was found in cells stained with TO-CZ. The nucleoli, which are the site
of ribosome synthesis and assembly, comprise tandemly organized GC-rich rDNA, rRNA and
proteins, which make them ideal targets for G-quadruplex-interacting agents [41]. To gain
insight into the nature of TO-CZ cellular targets, fixed and permeabilized SiHa cells were
12

ACCEPTED MANUSCRIPT
subjected to both DNase and RNase treatments before TO-CZ staining, and DAPI was used
as a fluorescent marker for DNA. As shown in Fig. 6, after DNase I treatment, on one hand, the
blue fluorescence of DAPI disappeared, showing that the DNA in the nucleus were
successfully digested. On the other, most of the red fluorescence signal of TO-CZ disappeared
in the nucleus, while the strong fluorescence foci in the nucleoli still existed. Additionally, the
red fluorescence signals in the cytoplasm were not erased by DNase I treatment. We
subsequently attributed such remaining fluorescence foci to the interaction of TO-CZ with RNA
G-quadruplexes because RNA is always found in the cytoplasm and nucleoli. In comparison,
RNase A treatment did not affect the cellular DNA structures, reflected in the remaining DAPI
fluorescence. However, RNase A treatment significantly diminished the fluorescence signals of
TO-CZ in both the nucleoli and the cytoplasm, whereas the red fluorescence signal in the
nucleus remained, which might correspond to the complex of TO-CZ with DNA
G-quadruplexes. Additionally, cells treated with RNase T1 exhibited no foci in the cytoplasm
and nucleoli, and these results were the same as those found with RNase A treatment
(Supplementary Fig. S8). Since RNase T1 splits RNA after guanosine residues, producing
3'-phosphorylated ends, the loss of TO-CZ fluorescence in the cytoplasm and nucleoli might
be a consequence of the breakage of RNA G-quadruplex structures. Furthermore, after both
DNase I and RNase A digestion, we observed little red fluorescence signals in whole cells (Fig.
6). Taken together, TO-CZ might sense DNA and RNA G-quadruplexes in cellulo.
4. Conclusion
G-quadruplexes are higher-order nucleic acid structures that exist in many crucial genomic
regions. An ever-increasing interest in G-quadruplexes has promoted the development of
selective and sensitive fluorescent probes as research tools. As a promising G-quadruplex
fluorescent probe, 3,6-bis(1-methyl-4-vinylpyridinium) carbazole diiodide (BMVC) has been
utilized in many studies investigating G-quadruplex structures. However, some drawbacks
were observed, including similar fluorescence enhancement upon binding G-quadruplexes
and duplex DNA and green fluorescent emission, both of which might restrict its further
application in the area of G-quadruplexes. To help develop a carbazole probe with optimized
photophysical properties and selectivity for G-quadruplexes, in this study, we decided to
13

ACCEPTED MANUSCRIPT
incorporate the TO fluorophore into the BMVC scaffold. We succeeded in engineering a new
carbazole/thiazole orange conjugating probe named TO-CZ for the detection of
G-quadruplexes. TO-CZ was found to display a distinct red-shift in its absorption band and
significant fluorescence enhancement in the red region upon its interaction with
G-quadruplexes, rather than non-quadruplex nucleic acids, which suggests that TO-CZ is a
colorimetric and red-emitting fluorescent dual probe selective for G-quadruplexes. Additionally,
the CD experiments demonstrated that TO-CZ is able to sense G-quadruplexes without
affecting their topology. Furthermore, we investigated the detailed interaction of TO-CZ with
G-quadruplexes using various experiments, revealing that TO-CZ might stack on the two
terminal G-quartets to form a tight, sandwich-like complex, and then realize the selective
fluorescence enhancement of TO-CZ. Considering the remarkable properties of this versatile
probe, we assessed the ability of TO-CZ to sense G-quadruplexes both in vitro and in cellulo.
In fact, our results suggested that TO-CZ acts as a highly sensitive and selective fluorescent
sensor for detecting G-quadruplexes either in solution or in a gel. Notably, TO-CZ might map
DNA and RNA G-quadruplexes in cellulo. This work provides a new promising probe specific
for G-quadruplexes, and this probe shows great potential in its application as a sensor of
intracellular DNA and RNA G-quadruplexes.
Acknowledgements
This work was supported by the National Natural Science Foundation of China [81330077 and
21272291], the Guangdong Natural Science Funds for Distinguished Young Scholars
[2015A030306004] and the Fundamental Research Funds for the Central Universities
[14ykpy09 and 2015ykzd06 to J.-H. T.].
Appendix A. Supplementary material
Supplementary data associated with this article can be found in the online version.
References
[1] Ma D-L, He H-Z, Leung K-H, Zhong H-J, Chan DS-H, Leung C-H. Label-free luminescent
oligonucleotide-based probes. Chem Soc Rev 2013;42:3427-40.
14

ACCEPTED MANUSCRIPT
[2] Burge S, Parkinson GN, Hazel P, Todd AK, Neidle S. Quadruplex DNA: sequence, topology
and structure. Nucleic Acids Res 2006;34(19):5402-15.
[3] Bochman ML, Paeschke K, Zakian VA. DNA secondary structures: stability and function of
G-quadruplex structures. Nat Rev Genet 2012;13(11):770-80.
[4] Chambers VS, Marsico G, Boutell JM, Di Antonio M, Smith GP, Balasubramanian S.
High-throughput sequencing of DNA G-quadruplex structures in the human genome. Nat
Biotechnol 2015;33:877-81.
[5] Livendahl M, Jamroskovic J, Ivanova S, Demirel P, Sabouri N, Chorell E. Design and
synthesis of 2,2'-diindolylmethanes to selectively target certain G-quadruplex DNA structures.
Chem - Eur J 2016;22(37):13004-9.
[6] Qin J-L, Qin Q-P, Wei Z-Z, Yu Y-C, Meng T, Wu C-X, et al. Stabilization of c-myc
G-quadruplex DNA, inhibition of telomerase activity, disruption of mitochondrial functions and
tumor cell apoptosis by platinum(II) complex with 9-amino-oxoisoaporphine. Eur J Med Chem
2016;124:417-27.
[7] Davis JT. G-quartets 40 years later: from 5'-GMP to molecular biology and supramolecular
chemistry. Angew Chem, Int Ed 2004;43:668-98.
[8] Lu L, Mao Z, Kang T-S, Leung C-H, Ma D-L. A versatile nanomachine for the sensitive
detection of platelet-derived growth factor-BB utilizing a G-quadruplex-selective iridium(III)
complex. Biosens Bioelectron 2016;85:300-9.
[9] Lu L, Wang W, Wang M, Kang T-S, Lu J-J, Chen X-P, et al. A luminescent
G-quadruplex-selective iridium(III) complex for the label-free detection of lysozyme. J Mater
Chem B 2016;4:2407-11.
[10] Ma D-L, Chan DS-H, Yang H, He H-Z, Leung C-H. Luminescent G-quadruplex probes.
Curr Pharm Des 2012;18:2058-75.
[11] Ma D-L, Zhang Z, Wang M, Lu L, Zhong H-J, Leung C-H. Recent developments in
G-quadruplex probes. Chem Biol 2015;22:812-28.
[12] Chen S-B, Hu M-H, Liu G-C, Wang J, Ou T-M, Gu L-Q, et al. Visualization of NRAS RNA
G-Quadruplex structures in cells with an engineered fluorogenic hybridization probe. J Am
Chem Soc 2016;138:10382-5.
[13] Chang C-C, Wu J-Y, Chien C-W, Wu W-S, Liu H, Kang C-C, et al. A fluorescent carbazole
15

ACCEPTED MANUSCRIPT
derivative: high sensitivity for quadruplex DNA. Anal Chem 2003;75:6177-83.
[14] Jin B, Zhang X, Zheng W, Liu X, Qi C, Wang F, et al. Fluorescence light-up probe for
parallel G-quadruplexes. Anal Chem 2014;86:943-52.
[15] Maji B, Kumar K, Kaulage M, Muniyappa K, Bhattacharya S. Design and synthesis of new
benzimidazole-carbazole conjugates for the stabilization of human telomeric DNA, telomerase
inhibition, and their selective action on cancer cells. J Med Chem 2014;57:6973-88.
[16] Panda D, Debnath M, Mandal S, Bessi I, Schwalbe H, Dash J. A nucleus-imaging probe
that selectively stabilizes a minor conformation of c-MYC G-quadruplex and down-regulates
c-MYC transcription in human cancer cells. Sci Rep 2015;5:13183.
[17] Tse T-Y, Chang C-C, Lin J-J, Chang T-C. A fluorescent anti-cancer agent,
3,6-bis(1-methyl-4-vinylpyridinium) carbazole diiodide, stains G-quadruplexes in cells and
inhibits tumor growth. Curr Top Med Chem 2015;15:1964-70.
[18] Chang C-C, Kuo I-C, Ling I-F, Chen C-T, Chen H-C, Lou P-J, et al. Detection of
quadruplex DNA structures in human telomeres by a fluorescent carbazole derivative. Anal
Chem 2004;76:4490-4.
[19] Chang C-C, Chien C-W, Lin Y-H, Kang C-C, Chang T-C. Investigation of spectral
conversion of d(TTAGGG)4 and d(TTAGGG)13 upon potassium titration by a G-quadruplex
recognizer BMVC molecule. Nucleic Acids Res 2007;35:2846-60.
[20] Huang F-C, Chang C-C, Lou P-J, Kuo I-C, Chien C-W, Chen C-T, et al. G-quadruplex
stabilizer 3,6-bis(1-methyl-4-vinylpyridinium)carbazole diiodide induces accelerated
senescence and inhibits tumorigenic properties in cancer cells. Mol Cancer Res
2008;6:955-64.
[21] Tsai Y-L, Wang Z-F, Chen W-W, Chang T-C. Emulsified BMVC derivative induced filtration
for G-quadruplex DNA structural separation. Nucleic Acids Res 2011;39:e114.
[22] Tseng T-Y, Wang Z-F, Chien C-H, Chang T-C. In-cell optical imaging of exogenous
G-quadruplex DNA by fluorogenic ligands. Nucleic Acids Res 2013;41:10605-18.
[23] Chen C-H, Hu T-H, Huang T-C, Chen Y-L, Chen Y-R, Cheng C-C, et al. Delineation of
G-quadruplex
alkylation
sites
mediated
by
3,6-bis(1-methyl-4-vinylpyridinium
iodide)carbazole-aniline mustard conjugates. Chem - Eur J 2015;21:17379-90.
[24] Huang W-C, Tseng T-Y, Chen Y-T, Chang C-C, Wang Z-F, Wang C-L, et al. Direct
16

ACCEPTED MANUSCRIPT
evidence of mitochondrial G-quadruplex DNA by using fluorescent anti-cancer agents. Nucleic
Acids Res 2015;43:10102-13.
[25] Yuan L, Lin W, Zheng K, He L, Huang W. Far-red to near infrared analyte-responsive
fluorescent probes based on organic fluorophore platforms for fluorescence imaging. Chem
Soc Rev 2013;42:622-61.
[26] Yan J-W, Ye W-J, Chen S-B, Wu W-B, Hou J-Q, Ou T-M, et al. Development of a universal
colorimetric indicator for G-quadruplex structures by the fusion of thiazole orange and
isaindigotone skeleton. Anal Chem 2012;84:6288-92.
[27] Hu M-H, Chen S-B, Wang Y-Q, Zeng Y-M, Ou T-M, Li D, et al. Accurate high-throughput
identification of parallel G-quadruplex topology by a new tetraaryl-substituted imidazole.
Biosens Bioelectron 2016;83:77-84.
[28] Lu Y-J, Deng Q, Hou J-Q, Hu D-P, Wang Z-Y, Zhang K, et al. Molecular engineering of
thiazole orange dye: change of fluorescent signaling from universal to specific upon binding
with nucleic acids in bioassay. ACS Chem Biol 2016;11:1019-29.
[29] Chen S-B, Wu W-B, Hu M-H, Ou T-M, Gu L-Q, Tan J-H, et al. Discovery of a new
fluorescent light-up probe specific to parallel G-quadruplexes. Chem Commun
2014;50:12173-6.
[30] Yan J-W, Chen S-B, Liu H-Y, Ye W-J, Ou T-M, Tan J-H, et al. Development of a new
colorimetric and red-emitting fluorescent dual probe for G-quadruplex nucleic acids. Chem
Commun 2014;50:6927-30.
[31] Nygren J, Svanvik N, Kubista M. The interactions between the fluorescent dye thiazole
orange and DNA. Biopolymers 1998;46:39-51.
[32] Lu Y-J, Yan S-C, Chan F-Y, Zou L, Chung W-H, Wong W-L, et al.
Benzothiazole-substituted benzofuroquinolinium dye: a selective switch-on fluorescent probe
for G-quadruplex. Chem Commun 2011;47:4971-3.
[33] Liu Z-Q, Zhuo S-T, Tan J-H, Ou T-M, Li D, Gu L-Q, et al. Facile syntheses of disubstituted
bis(vinylquinolinium)benzene derivatives as G-quadruplex DNA binders. Tetrahedron
2013;69(24):4922-32.
[34] Jose J, Burgess K. Syntheses and properties of water-soluble Nile Red derivatives. J Org
Chem 2006;71:7835-9.
17

ACCEPTED MANUSCRIPT
[35] Qiu B, Guo L, Wang W, Chen G. Synthesis of a novel fluorescent probe useful for DNA
detection. Biosens Bioelectron 2007;22:2629-35.
[36] Dai J, Carver M, Hurley LH, Yang D. Solution structure of a 2:1 quindoline-c-MYC
G-quadruplex: insights into G-quadruplex-interactive small molecule drug design. J Am Chem
Soc 2011;133:17673-80.
[37] Largy E, Hamon F, Teulade-Fichou MP. Development of a high-throughput G4-FID assay
for screening and evaluation of small molecules binding quadruplex nucleic acid structures.
Anal Bioanal Chem 2011;400:3419-27.
[38] Yang D-Y, Chang T-C, Sheu S-Y. Interaction between human telomere and a carbazole
derivative: a molecular dynamics simulation of a quadruplex stabilizer and telomerase inhibitor.
J Phys Chem A 2007;111:9224-32.
[39] Chen S-B, Shi Q-X, Peng D, Huang S-Y, Ou T-M, Li D, et al. The role of positive charges
on G-quadruplex binding small molecules: Learning from bisaryldiketene derivatives. Biochim
Biophys Acta 2013;1830(11):5006-13.
[40] Hu M-H, Chen S-B, Guo R-J, Ou T-M, Huang Z-S, Tan J-H. Development of a highly
sensitive fluorescent light-up probe for G-quadruplexes. Analyst 2015;140(13):4616-25.
[41] Laguerre A, Hukezalie K, Winckler P, Katranji F, Chanteloup G, Pirrotta M, et al.
Visualization of RNA-quadruplexes in live cells. J Am Chem Soc 2015;137:8521-5.
18

ACCEPTED MANUSCRIPT
FIGURE and SCHEME LEGENDS
Fig. 1. The structures of thiazole orange (TO), BMVC, and TO-CZ.
Fig. 2. Spectroscopic studies of TO-CZ upon binding with nucleic acids in 10 mM Tris-HCl
buffer, 100 mM KCl, pH=7.2. (A) The absorption spectra of 5 ꢀM TO-CZ titrated with pu22.
Arrow: concentrations of pu22 ranged from 0 to 20 ꢀM. (B) The fluorescence spectra of 3 ꢀM
TO-CZ titrated with pu22. Arrow: concentrations of pu22 ranged from 0 to 12 ꢀM, λex = 467 nm.
(C) The fluorescence intensity enhancement of 3 ꢀM TO-CZ at 613 nm against the ratio of
[DNA]/[TO-CZ], λex = 467 nm. (D) The fluorescence titration curves of 3 µM TO-CZ with the
stepwise addition of the G-quadruplex pu22 without and with 100 µM ct-DNA. The emission
was measured at 613 nm, λex = 467 nm.
Fig. 3. The effects of TO-CZ on the topology of the G-quadruplexes. The CD spectra of 3 µM
G-quadruplex pu22 and htg21 in 10 mM Tris-HCl buffer, 100 mM KCl, pH=7.2 with and without
15 µM TO-CZ under different experimental treatments.
Fig. 4. Interaction studies of TO-CZ with G-quadruplexes. (A) The Job plot analysis for the
binding stoichiometry of TO-CZ to the G-quadruplex pu22. The point of intersection of the best
fit line for the Job plot was near 0.67, showing a 2:1 stoichiometry of TO-CZ binding to the
G-quadruplex. (B) The fluorescence titrations of 3 ꢀM pu22 with the stepwise addition of
TO-CZ without and with 6 ꢀM SYUIQ-05 in 10 mM Tris-HCl buffer, 100 mM KCl, pH=7.2. The
emission was measured at 613 nm, λex = 467 nm. (C) Side view and (D) top view of the binding
model of TO-CZ with the G-quadruplex pu22 (PDB ID: 2L7V).
19

ACCEPTED MANUSCRIPT
Fig. 5. Application of TO-CZ in sensing G-quadruplex in vitro. (A) The linear fit equation for
calculating LOD value of TO-CZ for pu22. The concentration of TO-CZ was fixed at 3.0 ꢀM in
10 mM Tris-HCl buffer, 100 mM KCl, pH=7.2. Fluorescence intensities at 613 nm were
collected, λex = 467 nm. (B) The pre-stained gel of 20 µM TO-CZ incubated with various
concentrations of nucleic acids (pu22 and hairpin). (C) The same gel next stained by 1× SYBR
Green I solution. The conditions were as follows: 10 ꢀL of the mixtures were loaded onto a 20%
acrylamide gel in 1× TBE buffer containing 100 mM KCl. The nucleic acid concentration was
varied from 2.5 to 20 ꢀM.
Fig. 6. The application of TO-CZ in sensing G-quadruplexes in cellulo. Confocal fluorescence
images of fixed SiHa cells stained with 10 ꢀM of TO-CZ and 0.5 ꢀg·mL-1 of DAPI for 30 min
with or without DNase I or RNase A treatment. The cells were imaged using a LSM 710 laser
scanning confocal microscope. The excitation wavelengths were 405 nm for DAPI and 514 nm
for TO-CZ.
Scheme 1. Synthesis of the conjugating probe TO-CZ
20

ACCEPTED MANUSCRIPT
Fig. 1
21

ACCEPTED MANUSCRIPT
Fig. 2
22

ACCEPTED MANUSCRIPT
Fig. 3
23

ACCEPTED MANUSCRIPT
Fig. 4
24

ACCEPTED MANUSCRIPT
Fig. 5
25

ACCEPTED MANUSCRIPT
Fig. 6
26

ACCEPTED MANUSCRIPT
Scheme 1
27

ACCEPTED MANUSCRIPT
Highlights
A new colorimetric and fluorescent dual probe selective for G-quadruplex was developed.
Carbazole and thiazole orange frameworks were first assembled to construct the probe.
The probe could be further employed in sensing G-quadruplex either in vitro or in cellulo.