A simple structural modification to thiazole orange to improve the selective detection of G-quadruplexes

Article abstract of DOI:10.1016/j.dyepig.2015.11.010

Thiazole orange is a commonly used cyanine dye for binding to nucleic acids. Recently, it has been used for the detection of G-quadruplexes. However, thiazole orange is non-selective for G-quadruplex and other nucleic acids, thus hampering its further app

Full text of DOI:10.1016/j.dyepig.2015.11.010

Dyes and Pigments 126 (2016) 76e85
Contents lists available at ScienceDirect
Dyes and Pigments
journal homepage: www.elsevier.com/locate/dyepig
A simple structural modification to thiazole orange to improve the
selective detection of G-quadruplexes
Rui-Jun Guo a, Jin-Wu Yan b, Shuo-Bin Chen a, Lian-Quan Gu a, Zhi-Shu Huang a,
,
*
Jia-Heng Tan a
a School of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou 510006, China
b School of Bioscience and Bioengineering, South China University of Technology, Guangzhou 510006, China
a r t i c l e i n f o
a b s t r a c t
Article history:
Thiazole orange is a commonly used cyanine dye for binding to nucleic acids. Recently, it has been used
for the detection of G-quadruplexes. However, thiazole orange is non-selective for G-quadruplex and
other nucleic acids, thus hampering its further application. Herein, we designed and synthesized new
fluorescent probes by incorporating hydrocarbon rings into the chromophore of thiazole orange. This
simple modification dramatically improved selective binding to certain G-quadruplexes. The most
promising probe, the cyclopentane fused analogue, exhibited significant fluorescence enhancement
when treated with G-quadruplexes but retained weak fluorescence in the presence of double-stranded
and single-stranded DNA. The cyclopentane fused probe also displayed considerable selectivity for
parallel G-quadruplexes. These modifications reduced the quantum yield of thiazole orange. Further
study of the mechanism revealed that the introduction of a hydrocarbon ring altered the planarity of the
chromophore as well as the binding affinities for G-quadruplexes, and therefore, influenced the ability to
detect G-quadruplexes.
Received 23 July 2015
Received in revised form
12 November 2015
Accepted 14 November 2015
Available online 26 November 2015
Keywords:
G-quadruplex
Thiazole orange
Structural modification
Fluorescence probe
Selectivity
Binding mode
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction
antiparallel structures [9e11]. During the past two decades, G-
quadruplex structures have attracted considerable attention
Thiazole orange (TO, Fig. 1) is an exceptional asymmetric
cyanine dye. It is essentially non-fluorescent in aqueous solution
and obtains intense fluorescence when bonded to double-stranded
nucleic acids [1,2]. These unique properties make TO particularly
useful for the detection of double-stranded nucleic acids in a va-
riety of techniques. Examples include the detection of PCR products
in real time, staining double-stranded DNA in agarose gels and
capillary electrophoresis, and applications in fluorescent in situ
hybridization [3e6].
In addition to double-stranded nucleic acids, TO is used for the
detection of G-quadruplexes (G4s) [7,8]. G-quadruplexes are unique
four-stranded structures that are formed by guanine-rich nucleic
acid sequences. The building blocks of G-quadruplexes are G-
quartets, which stack up on top of each other to form secondary
DNA structures. G-quadruplexes are widely dispersed in eukaryotic
genomes and can be divided into two main topologies: parallel and
because of their biological significance and potential applications in
supramolecular chemistry [12e14]. The ever-increasing interest in
G-quadruplexes has promoted the development of rapid and sim-
ple approaches for the selective detection of these structures. Thus,
the discovery of selective fluorescent probes for G-quadruplexes
has become an extremely active area of research [15e25]. Notably,
unlike the commercially available dye thioflavin T (ThT) which is a
selective G-quadruplex probe [26e28], TO is non-selective for
double-stranded and G-quadruplex nucleic acids. However, it may
offer an attractive template for the design of selective fluorescent
probes for G-quadruplexes. Successful examples include the bis-
quinolinium/TO conjugate and the benzofuroquinolinium/TO con-
jugate [29,30].
As shown in Fig. 1, the conjugates includes the assembly of the
TO and the G-quadruplex binding small molecule all in a fusion
scaffold with the aim of combining their advantages. These fluo-
rescent probes exhibit distinct selectivity for G-quadruplexes over
double-stranded nucleic acids. However, the conjugation strategy
requires a suitable molecular framework to accommodate the TO
chromophore. One relevant example is our previous work on the
* Corresponding author.
E-mail address: tanjiah@mail.sysu.edu.cn (J.-H. Tan).
http://dx.doi.org/10.1016/j.dyepig.2015.11.010
0143-7208/© 2015 Elsevier Ltd. All rights reserved.

R.-J. Guo et al. / Dyes and Pigments 126 (2016) 76e85
77
Fig. 1. Structures of TO, the bisquinolinium/TO conjugate, the benzofuroquinolinium/TO conjugate, the isaindigotone/TO conjugate and the TO-derived compounds: T1, T2 and T3.
isaindigotone/TO conjugate [31]. This compound exhibits a high
selectivity for the G-quadruplex, but it is non-fluorescent upon
combination with the G-quadruplex. Incorporation of TO into the
isaindigotone framework surprisingly eliminates the fluorescent
properties of TO. Accordingly, it led us to reconsider the modifi-
cation strategy, and we developed a G-quadruplex-specific fluo-
rescent probe using a more simple structural modification to the TO
scaffold.
Based on our experience in developing selective G-quadruplex
binding ligands, we introduced some simple moieties to the TO
scaffold [32e34]. Herein, we present a new series of TO-derived
compounds: T1, T2 and T3. Compared with TO, these compounds
bear a hydrocarbon ring that has been suggested to be an important
factor in the binding of small molecules to a G-quadruplex. Their
photophysical properties were examined alone and in the presence
of different nucleic acids. In addition, molecular modeling ap-
proaches were employed to assist in understanding the differences
and the relevant mechanism.
were of analytical reagent grade and were used without further
purification. The 2,3-dimethylbenzothiazolium tosylate was syn-
thesized according to a previous report [35].
2.1.1. 9-Chloro-2,3-dihydro-1H-cyclopenta[b]quinolone (1)
To a stirred solution of 2-aminobenzoic acid (1.37 g, 10 mmol)
and cyclopentanone (1.5 mL, 16.9 mmol) cooled in an ice bath,
POCl3 (11 mL) was carefully added and the mixture was refluxed for
5 h. After cooling to room temperature, the mixture was concen-
trated to give a slurry. The residue was then diluted with EtOAc, and
neutralized with NaOH solution. The organic layer was dried over
anhydrous MgSO4. After concentration, the resulting residue was
purified by flash chromatography with EtOAc: petroleum ether
(1:3e1:5) to afford a white solid 1 (1.05 g, 52%). 1H NMR (400 MHz,
CDCl3)
d
8.17 (dd, J ¼ 8.3, 1.0 Hz, 1H), 8.06 (d, J ¼ 8.4 Hz, 1H),
7.72e7.66 (m, 1H), 7.61e7.55 (m, 1H), 3.26 (t, J ¼ 7.7 Hz, 2H), 3.18 (t,
J ¼ 7.5 Hz, 2H), 2.31e2.22 (m, 2H). ESI-MS m/z: 204.0 [MþH]þ.
2.1.2. 9-Chloro-1,2,3,4-tetrahydroacridine (2)
The method for the preparation of compound 1 was used by
replacing cyclopentanone with cyclohexanone. Compound 2 was
synthesized as a white solid (1.07 g, 49%). 1H NMR (400 MHz, CDCl3)
2. Experimental methods
2.1. Synthesis and characterization
d
8.16 (d, J ¼ 8.3 Hz, 1H), 7.98 (d, J ¼ 8.4 Hz, 1H), 7.66 (t, J ¼ 7.6 Hz,
1H and 13C NMR spectra were recorded using TMS as the internal
standard in CDCl3 or DMSO-d6 using a Bruker BioSpin GmbH
spectrometer at 400 MHz and 101 MHz, respectively. IR spectra
were determined on a Bruker Tensor 37 infrared spectrophotom-
eter. Mass spectra (MS) were recorded on a Shimadzu LCMS-2010A
instrument with an ESI or ACPI mass selective detector, and high
resolution mass spectra (HRMS) were obtained on a Shimadzu
LCMS-IT-TOF. Melting points (Mp) were determined using a SRS
OptiMelt automated melting point instrument without correction.
Flash column chromatography was performed with silica gel
(200e300 mesh) purchased from Qingdao Haiyang Chemical Co.
Ltd. The purity of the synthesized compound was confirmed to be
higher than 95% via analytical HPLC that was performed with a dual
pump Shimadzu LC-20 AB system equipped with a Ultimate XB-C18
1H), 7.53 (t, J ¼ 7.3 Hz, 1H), 3.13 (t, J ¼ 5.7 Hz, 2H), 3.02 (t, J ¼ 5.8 Hz,
2H), 2.05e1.91 (m, 4H). ESI-MS m/z: 218.1 [MþH]þ.
2.1.3. 11-Chloro-7,8,9,10-tetrahydro-6H-cyclohepta[b]quinoline (3)
The method for the preparation of compound 1 was used by
replacing cyclopentanone with cycloheptanone. Compound 3 was
synthesized as a white solid (1.12 g, 48%). 1H NMR (400 MHz, CDCl3)
d
8.17 (d, J ¼ 8.2 Hz, 1H), 7.99 (d, J ¼ 8.3 Hz, 1H), 7.67 (t, J ¼ 7.1 Hz,
1H), 7.56 (t, J ¼ 7.6 Hz, 1H), 3.31e3.18 (m, 4H), 1.95e1.87 (m, 2H),
1.86e1.71 (m, 4H). ESI-MS m/z: 232.1 [MþH]þ.
2.1.4. (Z)-4-methyl-9-((3-methylbenzo[d]thiazol-2(3H)-ylidene)
methyl)-2,3-dihydro-1H-cyclopenta[b]quinolin-4-ium tosylate (T1)
A mixture of 1 (0.41 g, 2.0 mmol), methyl tosylate (0.9 mL,
6.0 mmol), and toluene (1 mL) was heated at 110 ꢁC for 6 h. After
column (4.6 ꢀ 250 mm, 5
mm). All chemicals were purchased from
commercial sources unless otherwise specified. All of the solvents
cooling to room temperature,
a solution containing 2,3-

78
R.-J. Guo et al. / Dyes and Pigments 126 (2016) 76e85
dimethylbenzothiazolium tosylate (0.5 g, 1.5 mmol) in 1 mL ethanol
and KHCO3 (0.20 g, 2 mmol) in 1 mL water was added. Then, the
solution was stirred at reflux for another 1 h. After concentration,
the residue was diluted with CH2Cl2. The organic layer was dried
over anhydrous MgSO4. The CH2Cl2 was removed and the residue
was purified by flash chromatography with methanol: CH2Cl2
(1:20) to afford brownish black solid T1 (0.11 g, 11%). Mp:
2.2. Materials
All oligonucleotides (Table 1) used in this study were purchased
from Invitrogen (China), and their concentrations were determined
using the absorbance at 260 nm based on their respective molar
extinction coefficients using a NanoDrop 1000 Spectrophotometer
(Thermo Scientific, USA). To form the G-quadruplexes, oligonucle-
otides were annealed in a relevant buffer containing KCl by heating
to 95 ꢁC for 5 min, followed by gradual cooling to room tempera-
ture. The oligonucleotides were engaged in G-quadruplex forma-
tion, and their conformations were determined by circular
dichroism (CD) measurements (Fig. S1, Supplementary Data) [11].
Stock solutions of the compounds, TO, T1, T2 and T3 (10 mM), were
dissolved in DMSO and stored at ꢂ80 ꢁC. Further dilutions of the
compounds to working concentrations were prepared in a relevant
buffer immediately prior to use.
102e105 ꢁC; 1H NMR (400 MHz, DMSO-d6)
d
8.47 (d, J ¼ 8.1 Hz, 1H),
8.28 (d, J ¼ 8.6 Hz, 1H), 8.02 (t, J ¼ 7.6 Hz, 1H), 7.77 (t, J ¼ 7.3 Hz, 1H),
7.67 (d, J ¼ 7.4 Hz, 1H), 7.51e7.39 (m, 4H), 7.18 (t, J ¼ 7.1 Hz, 1H), 7.09
(d, J ¼ 7.4 Hz, 2H), 6.29 (s, 1H), 4.23 (s, 3H), 3.75 (s, 3H), 3.50 (t,
J ¼ 7.2 Hz, 2H), 3.06 (t, J ¼ 7.0 Hz, 2H), 2.33e2.20 (m, 5H). 13C NMR
(101 MHz, DMSO-d6) d 162.50, 156.66, 147.73, 145.95, 141.31, 137.95,
137.37, 132.82, 130.09, 128.01, 127.91, 127.18, 126.69, 125.44, 123.59,
122.95, 122.85, 122.07, 118.36, 111.10, 86.32, 33.87, 32.87, 31.93,
21.92, 20.70. NOESY experiment showed the correlation between
the alkene proton at 6.29 ppm and the N-methyl proton at
3.75 ppm of thiazole, thus the double bond was assigned as the Z
configuration. IR (cmꢂ1, KBr): 3060, 3019, 2923, 2853, 1648, 1587,
1504, 1372, 1120, 1061, 1033, 1010. Purity: 99.7% by HPLC. HRMS
(ESI): calcd for (M-TsO)þ (C31H30N3Sþ) 345.1420, found 345.1407.
2.3. UVeVis and fluorescence spectroscopic studies
The UVeVis spectra were obtained using a UV-2450 spectro-
photometer (Shimadzu, Japan) using a 1 cm path length quartz
cuvette. The fluorescence spectra were obtained using a LS-55
luminescence spectrophotometer (PerkineElmer, USA). A quartz
cuvette with a 2 mm ꢀ 10 mm path length was used for the spectra
recorded at 10 nm excitation and emission slit widths unless
otherwise specified.
For the titration experiment, small aliquots of a stock solution of
the samples (oligonucleotides) were added to the solution con-
taining T1 at a fixed concentration (5
(10 mM, pH 7.2) with 100 mM KCl. The final concentration of the
sample was varied from 0 to 30 M. After each sample addition, the
reaction was stirred and allowed to equilibrate for at least 1 min,
and the fluorescence measurement was obtained at an excitation
wavelength of 480 nm. In competition experiments, single-
stranded DNA mut-htg21 and double-stranded DNA hairpin were
used as the competitors.
2.1.5. (Z)-10-methyl-9-((3-methylbenzo[d]thiazol-2(3H)-ylidene)
methyl)-1,2,3,4-tetrahydroacridin-10-ium tosylate (T2)
The method for the preparation of compound T1 was used by
replacing compound 1 with compound 2. Compound T2 was syn-
thesized as a brownish black solid (0.18 g, 17%). Mp: 96e98 ꢁC; 1H
NMR (400 MHz, DMSO-d6)
d
8.39 (dd, J ¼ 8.4, 1.1 Hz, 1H), 8.36 (d,
J ¼ 8.9 Hz, 1H), 8.06e8.00 (m, 1H), 7.72 (t, J ¼ 6.3 Hz, 1H), 7.57 (d,
J ¼ 7.7 Hz, 1H), 7.46 (d, J ¼ 8.0 Hz, 2H), 7.45e7.37 (m, 2H), 7.16e7.07
(m, 3H), 6.11 (s, 1H), 4.23 (s, 3H), 3.74 (s, 3H), 3.25 (t, J ¼ 6.3 Hz, 2H),
2.93 (t, J ¼ 6.2 Hz, 2H), 2.28 (s, 3H), 1.95e1.80 (m, 4H). 13C NMR
mM) in TriseHCl buffer
m
(101 MHz, DMSO-d6) d 157.59, 154.20, 151.64, 145.80, 141.32, 138.27,
137.46, 133.61, 129.28, 127.95, 127.05, 126.56, 126.27, 125.42, 122.92,
122.27, 121.96, 121.57, 118.74, 110.81, 86.77, 37.25, 32.76, 29.39,
26.60, 21.35, 20.71, 20.54. NOESY experiment showed the correla-
tion between the alkene proton at 6.11 ppm and the N-methyl
proton at 3.74 ppm of thiazole, thus the double bond was assigned
as the Z configuration. IR (cmꢂ1, KBr): 3060, 3015, 2923, 2853, 1646,
1583, 1494, 1353, 1120, 1062, 1033, 1010. Purity: 97.3% by HPLC.
HRMS (ESI): calcd for (M-TsO)þ (C31H30N3Sþ) 359.1576, found
359.1567.
The fluorescence quantum yield (FF) of the compound was
calculated relative to a standard solution of rhodamine 123 in
ethanol (FF ¼ 0.90) and was determined using the following for-
mula: Fu
¼
Fs (As/Au) ꢀ (Iu/Is), where
F 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 an
unknown quantum yield, and s refers to the reference compound
(Rhodamine 123) of a known quantum yield. The fluorescence
spectra were recorded at 10 nm excitation and emission slit widths
2.1.6. (Z)-5-methyl-11-((3-methylbenzo[d]thiazol-2(3H)-ylidene)
methyl)-7,8,9,10-tetrahydro-6H-cyclohepta[b]quinolin-5-ium
tosylate (T3)
The method for the preparation of compound T1 was used by
replacing compound 1 with compound 3. Compound T3 was syn-
thesized as a brownish black solid (0.15 g, 14%). Mp: 92e94 ꢁC; 1H
for the determination of
F.
The LOD values of T1 for different nucleic acids in solution were
calculated on the basis of the equation LOD ¼ 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 mea-
surements (n ¼ 20) of blank solution. The m value is the slope of the
calibration curve, which was derived from the linear range of a T1
fluorescence titration curve with different nucleic acids and stan-
dards for the sensitivity of this method.
NMR (400 MHz, DMSO-d6)
d
8.46 (dd, J ¼ 8.3, 0.9 Hz, 1H), 8.41 (d,
J ¼ 8.8 Hz, 1H), 8.09e8.01 (m, 1H), 7.80 (t, J ¼ 6.0 Hz, 1H), 7.51 (d,
J ¼ 7.6 Hz, 1H), 7.47 (d, J ¼ 8.0 Hz, 2H), 7.36e7.29 (m, 2H), 7.10 (d,
J ¼ 7.9 Hz, 2H), 7.08e7.05 (m, 1H), 6.31 (s, 1H), 4.37 (s, 3H), 3.69 (s,
3H), 3.51e3.47 (m, 2H), 3.24e3.18 (m, 2H), 2.28 (s, 3H), 1.90e1.80
(m, 4H), 1.68e1.60 (m, 2H). 13C NMR (101 MHz, DMSO)
d 162.02,
154.87, 150.25, 145.90, 141.78, 137.86, 137.41, 133.21, 133.13, 128.52,
127.94, 127.61, 126.87, 125.45, 124.35, 122.20, 122.13, 121.75, 119.44,
110.13, 85.15, 32.33, 31.86, 29.58, 29.51, 26.63, 23.63, 20.71. NOESY
experiments showed the correlation between the alkene proton at
6.31 ppm and the N-methyl proton at 3.69 ppm of thiazole, thus the
double bond was assigned as the Z configuration. IR (cmꢂ1, KBr):
3057, 3022, 2922, 2853, 1636, 1584, 1503, 1352, 1119, 1061, 1032,
1010. Purity: 99.6% by HPLC. HRMS (ESI): calcd for (M-TsO)þ
(C31H30N3Sþ) 373.1733, found 373.1721.
2.4. Polyacrylamide gel electrophoresis (PAGE) studies
Different oligonucleotides were loaded onto a 20% bisacryla-
mide gel in 0.5 ꢀ TBE buffer containing 12.5 mM KCl and 12.5 mM
NaCl, and electrophoresed at 4 ꢁC. The oligonucleotides were
stained with T1 (30 mmol/L, 20 min), or by the commercial staining
agent GelRed (1ꢀ, 20 min). DNA fragments were visualized under
UV light and photographed by using AlphaImager EC
(ProteinSimple).

R.-J. Guo et al. / Dyes and Pigments 126 (2016) 76e85
79
Table 1
The DNA samples used in this study.
Name
Sequence
Structure in Kþ solution
Pu27
RET
Bcl-2
c-kit2
c-kit1
Htg21
TBA
ds26
Hairpin
mut-Pu27
mut-htg21
SPR-Pu27
SPR-htg21
SPR-dsDNA
SPR-ssDNA
50-d(TGGGGAGGGTGGGGAGGGTGGGGAAGG)-30
50-d(GGGGCGGGGCGGGGCGGGGG)-30
Parallel G4 DNA
Parallel G4 DNA
Parallel G4 DNA
Parallel G4 DNA
50-d(GGGCGGGCGCGGGAGGAAGGGGGCGGG)-30
50-d(CGGGCGGGCGCGAGGGAGGGT)-30
50-d(AGGGAGGGCGCTGGGAGGAGGG)-30
Parallel G4 DNA
50-d(GGGTTAGGGTTAGGGTTAGGG)-30
Antiparallel G4 DNA
Antiparallel G4 DNA
Double-Stranded DNA
Double-Stranded DNA
Single-Stranded DNA
Single-Stranded DNA
Parallel G4 DNA
Antiparallel G4 DNA
Double-Stranded DNA
Single-Stranded DNA
50-d(GGTTGGTGTGGTTGG)-30
50-d(CAATCGGATCGAATTCGATCCGATTG)-30
50-d(CGCGCGCGTTTTCGCGCGCG)-30
50-d(TGAAGAAGGTGAGGAAGATAGAGAAGG)-30
50-d(GAGTTAGAGTTAGAGTTAGAG)-30
50-biotin-d(ACGTACGTGGGGAGGGTGGGGAGGGTGGGGAAGGTGGGG)-30
50-biotin-d(GTTAGGGTTAGGGTTAGGGTTAGGGTTAGG)-30
50-biotin-d(TTCGCGCGCGTTTTCGCGCGCG)-30
50-biotin-d(AGTTAGAGTTAGAGTTAGAGTTAGAGTTAG)-30
2.5. Surface plasmon resonance (SPR) studies
respectively. Compounds T1, T2 and T3 were assigned as the Z
configuration on the basis of NOESY experiments.
The SPR measurements were performed on a ProteOn XPR36
Protein Interaction Array system (Bio-Rad Laboratories, Hercules,
CA) using a Neutravidin-coated GLH sensor chip. The biotinylated
oligonucleotides (SPR-Pu27, SPR-htg21, SPR-dsDNA, SPR-ssDNA)
were attached to the chip. In a typical experiment, biotinylated
DNA was folded in filtered and degassed running buffer (50 mM
TriseHCl, 100 mM KCl, pH 7.2). The DNA samples were then
captured (~1000 RU) in four flow cells, and one flow cell was used
as the blank. Solutions of the compounds, TO, T1, T2 and T3, were
prepared with running buffer through serial dilutions of the stock
solution. Five concentrations were injected simultaneously at a
3.2. Optical properties of the TO-derived compounds: T1, T2, and
T3
The optical properties of newly synthesized compounds, T1, T2,
and T3, were first investigated in different solvents. Compound TO
was used as the reference. As shown in Fig. 2, the absorbance of
these compounds was weak in aqueous solution but stronger in
organic solutions. Notably, the absorption spectrum of TO in water
exhibited a strong positive band at approximately 505 nm, a
shoulder peak at 480 nm, and a minor positive band at approxi-
mately 425 nm. This absorbance pattern was quite different from
those of compounds T1, T2, and T3, which exhibited only a major
positive band at approximately 515 nm. Similar results could be
found in other solvents. From the literature, the absorption bands at
approximately 480 and 425 nm can be assigned to the H-dimer and
higher H-aggregates of TO, respectively, in comparison with the
monomer band at approximately 505 nm [38,39]. The different
absorbance patterns indicated that T1, T2, and T3 exhibited weaker
propensity to aggregate in aqueous solution than TO.
flow rate of 50 mL/min for 400 s during the association phase, fol-
lowed with 400 s during the dissociation phase at 25 ꢁC. The GLH
sensor chip was regenerated with a short injection of 50 mM NaOH
between consecutive measurements. The final graphs were ob-
tained by subtracting blank sensorgrams from the different DNA
sensorgrams. The data were analyzed with ProteOn manager soft-
ware using the Langmuir model to fit the kinetic data.
2.6. Molecular modeling studies
With an increase in concentration, both TO and T1 showed
increased absorption intensity, whereas their band profiles were
quite different in the normalized absorption spectra at different
concentrations. As shown in Fig. 3A and B, absorption banks of TO
at approximately 480 and 425 nm gradually increased. However,
shape and position of absorption banks of T1 changed little. Be-
The structures of compounds TO, T1, T2 and T3 were con-
structed and optimized with Gaussian 03 using the HF/6e31G*
basis set. The parallel Pu27-derived NMR G-quadruplex structure,
Pu24I, with TMPyP4 and a antiparallel NMR G-quadruplex struc-
ture for the human telomeric with telomestatin were used as the
templates (PDB ID: 2A5R and 2MB3) for the docking studies [36,37].
The docking simulations were performed using Schrodinger soft-
ware for the binding site based on the reference ligands.
sides, variations of absorption intensity of T1 (0e100 mM) strictly
followed the BeereLambert law as compared to those of TO (Fig. 3C
and D). All of these results reinforce the conclusion that T1
exhibited weaker propensity to aggregate in buffer solution than
TO. Then, the fluorescence properties of compounds T1, T2, and T3
in different solvents were studied. Interestingly, their fluorescence
emissions were extremely week in both aqueous and organic so-
lutions. Thus, the maximum in the fluorescence emission spectrum
could not be determined.
3. Results and discussion
3.1. Synthesis of the TO-derived compounds: T1, T2 and T3
The TO-derived compounds, T1, T2 and T3, were synthesized
following the procedure reported previously by our group [31]. As
shown in Scheme 1, compound 1 was prepared via the condensa-
tion of 2-aminobenzoic acid and cyclopentanone in the presence of
POCl3. The treatment of 1 with methyl 4-methylbenzenesulfonate
yielded the N-methylquinolinium intermediate product. Without
further purification, the crude product was treated with 2,3-
dimethylbenzothiazolium tosylate to produce the final com-
pound, T1. The synthesis of T2 and T3 was accomplished following
the same procedures but from cyclohexanone and cycloheptanone,
3.3. Screening of the TO-derived compounds: T1, T2, and T3
To identify the most promising TO-derived fluorescent probe,
compounds T1, T2, and T3 were screened for their fluorescence
enhancement and selectivity for G-quadruplexes. Compound TO
was used as the reference. Three representative oligonucleotides,
including the single-stranded DNA mut-htg21, the double-
stranded DNA hairpin and the G-quadruplex DNA formed by the
oligonucleotide Pu27, were employed in the assays. UVeVis

80
R.-J. Guo et al. / Dyes and Pigments 126 (2016) 76e85
Scheme 1. Synthesis of compounds T1, T2 and T3.a
Fig. 2. Absorption spectra of 5 mM of TO (A), T1 (B), T2 (C) and T3 (D) in different solvents.
titration experiments were first performed to identify isosbestic
point at which fluorescence was generally excited (Fig. S2,
Supplementary Data). However, differences in wavelength be-
tween position of the isosbestic point and maximum of the
emission spectrum were too close. To reduce the interference
from excitation and ensure the integrity of fluorescence spectra,
fluorescence measurement was obtained at an excitation wave-
length of 480 nm. As shown in Fig. 4, the fluorescence emissions
of T1, T2, and T3 alone in buffer were weak, even in the presence
of single-stranded and double-stranded DNA. In contrast, all of
the compounds displayed fluorescence enhancement upon addi-
tion of G-quadruplex Pu27. T1 was the most promising compound
due to its much stronger fluorescence enhancement than T2 or T3
and more selective fluorescence response for the G-quadruplex
than TO. As
a result, T1 was chosen for further detailed
investigation.

R.-J. Guo et al. / Dyes and Pigments 126 (2016) 76e85
81
Fig. 3. Normalized absorption spectra of 1.25e20
mM of TO (A) and 5e100 mM of T1 (B), and their corresponding concentration-dependent absorbance at 505 nm for TO (C) and T1
(D) in buffer solution (10 mM TriseHCl buffer, 100 mM KCl, pH 7.2).
3.4. Fluorescence spectroscopic studies of T1 interactions with
nucleic acids
fluorescence emissions were practically identical to those in the
experiment without a competitor. Similar results could be found
when using double-stranded DNA hairpin instead of single-
stranded DNA. These results demonstrated the promising poten-
tial of T1 to serve as a promising G-quadruplex fluorescent probe
even in a competitive environment.
The detailed fluorescence properties of T1 with various G-
quadruplexes and other nucleic acids were explored via a fluores-
cence titration assay. As shown in Fig. 5A, T1 alone in buffer dis-
played a weak fluorescence emission. With the gradual addition of
G-quadruplex DNA Pu27, the emission peak at approximately
558 nm was significantly enhanced. Such resulting fluorescence
was stable for at least 1 h (Fig. S3, Supplementary Data). The sig-
nificant fluorescence was also observed when T1 was treated with
the G-quadruplexes, RET, bcl-2, c-kit2 and c-kit1, which have all
been determined to form a parallel structure (Fig. 5B). In contrast,
we observed a much weaker fluorescence enhancement for the
antiparallel G-quadruplexes htg21 and TBA under our experimental
conditions. Additionally,
a negligible fluorescence was also
observed when titrating T1 with double-stranded DNA (ds26 and
hairpin) and single-stranded DNA (mut-Pu27 and mut-htg21). This
trend in fluorescence intensities was observed by the naked eye
under UV light (Fig. S4, Supplementary Data).
To further ascertain the selective fluorescence response of T1,
we employed a competition experiment, in which the ability of T1
to retain an enhanced fluorescence intensity with the addition of G-
quadruplex Pu27 was challenged by single-stranded or double-
stranded DNA competitor. As shown in Fig. 6, when gradually
Fig. 4. Fluorescence responses of 0.5
lex M of T1, T2, and T3 to 20
lem ¼ 480/558 nm) in 10 mM of TriseHCl buffer and 100 mM of KCl at a
m
M of TO to 2
m
M of different oligonucleotides
(
/
lem ¼ 485/530 nm) and 5
m
mM of different oligonu-
adding the G-quadruplex Pu27 into the solution containing 5
mM T1
cleotides (lex
pH of 7.2.
/
and 25 single-stranded DNA mut-htg21, the enhanced
mM

82
R.-J. Guo et al. / Dyes and Pigments 126 (2016) 76e85
Fig. 5. (A) Fluorescence titration of 5
mM of T1 with stepwise addition of the G-quadruplex-forming oligonucleotide (Pu27, arrows: 0e6 mol equiv.) in 10 mM of TriseHCl buffer and
100 mM of KCl at a pH of 7.2. (B) The fluorescence intensity enhancement of 5
mM of T1 at 558 nm against the ratio of [Sample]/[T1], lex ¼ 480 nm.
Table 2
The fluorescence quantum yields of TO and T1
with different nucleic acids.
Sample
FF
T1Tris
T1RET
T1Pu27
T1bcl-2
T1c-kit2
T1c-kit1
T1htg21
T1TBA
0.001
0.047
0.046
0.036
0.020
0.032
0.004
0.002
0.002
0.002
0.002
0.001
0.004
0.604
0.219
0.153
0.327
0.203
0.197
T1ds26
T1hairpin
T1mut-Pu27
T1mut-htg21
TOTris
TOPu27
TOhtg21
TOTBA
TOds26
TO mut-Pu27
TO mut-htg21
Fig. 6. Fluorescence titration of 5
ruplex-forming oligonucleotide Pu27 without and with 25
m
M T1 with the stepwise addition of the G-quad-
M double-stranded DNA
m
hairpin or single-stranded DNA mut-htg21 in 10 mM of TriseHCl buffer and 100 mM of
KCl at a pH of 7.2.
In addition to the fluorescence quantum yield, we also investi-
gated the detection limits of T1 for different G-quadruplex. The LOD
values were calculated on the basis of the equation LOD ¼ K ꢀ Sb/m
[40]. The K value is generally taken to be 3 according to the Inter-
national Union of Pure and Applied Chemistry (IUPAC) recom-
mendation. The Sb value represents the standard deviation for
multiple measurements (n ¼ 20) of blank solution. The m value is
the slope of the calibration curve, which is derived from the linear
range of the T1 fluorescence titration curve with different nucleic
acids and represents the sensitivity of this method (Fig. S5,
Supplementary Data). The corresponding LOD values of T1 for G-
quadruplex Pu27, bcl-2, c-kit1, c-kit2 and RET were 15.2 nM,
16.7 nM, 31.8 nM, 21.8 nM and 16.0 nM, respectively.
The fluorescence quantum yield values of T1 with different
nucleic acids are summarized in Table 2. Notably, these values were
consistent with the results for fluorescence titration, showing that
the fluorescence emission enhancement was always more pro-
nounced for the parallel G-quadruplex structures (RET, Pu27, bcl-2,
c-kit2 and c-kit1). The quantum yield of T1 for the G-quadruplex
Pu27 was 0.046. This value was approximately 12 times higher than
the value for the antiparallel G-quadruplex htg21, approximately 23
times higher than that of the antiparallel G-quadruplex TBA,
approximately 23 times higher than that of the double-stranded
ds26, and approximately 46 times higher than that of the single-
stranded mut-htg21. In contrast, the quantum yields of TO for
different nucleic acids were similar. The quantum yield of TO for the
G-quadruplex Pu27 was 0.604, which was only approximately 2e4
times higher than the values for the antiparallel G-quadruplexes or
double-stranded and single-stranded DNA. All of these results
suggested that introduction of the hydrocarbon ring onto TO
significantly improved its selectivity for G-quadruplexes, especially
for the parallel structures. However, this modification reduced the
quantum yield of TO.
3.5. PAGE studies
Encouraged by the significant and selective fluorescence
enhancement of T1 binding with G-quadruplex DNA, we were
interested in the practical application of T1 and then investigated
its performance in staining G-quadruplex bands after gel electro-
phoresis. G-quadruplex DNA Pu27, double-stranded DNA hairpin
and single-stranded DNA mut-htg21 were employed in the native

R.-J. Guo et al. / Dyes and Pigments 126 (2016) 76e85
83
Table 3
PAGE experiments. After electrophoresis, the polyacrylamide gel
was immersed in 30 mol/L T1 staining solution for 20 min. As
Equilibrium dissociation constant (KD) determined by the SPR assay.
KD M)
Pu27
m
shown in Fig. 7, clear fluorescence band corresponding to G-
quadruplex Pu27 was observed, and the bands of double-stranded
DNA and single-stranded DNA were hardly detectable. Then, the
PAGE experiment was repeated by using commercial GelRed stain
as a benchmark. Compared with T1, GelRed was a stain for all types
of nucleic acids. Collectively, these results demonstrated the
feasibility of using T1 as selective fluorescent stain for G-
quadruplexes.
(m
htg21
dsDNA
ssDNA
TO
T1
T2
T3
0.138
2.35
2.38
3.45
0.525
2.00
11.9
0.996
ea
0.887
ea
ea
ea
ea
19.2
ea
a
No significant binding was found for addition of up to 20 mM ligand.
Table 4
3.6. SPR studies
Torsion angles of TO and the TO-derived compounds in the absence and in the
presence of the G-quadruplexes.
T1 exhibited a significantly improved selectivity for the G-
quadruplexes compared with TO. Therefore, it is important to
determine the mechanism of these interactions. Thus, we then
investigated the T1 interactions with the G-quadruplexes using
surface plasmon resonance (SPR) assays because the selectivity of
a compound for a G-quadruplex is closely related to the affinity.
The alternative biotinylated G-quadruplexes, Pu27, htg21 and
double-stranded dsDNA and single-stranded ssDNA, were
employed. As shown in Table 3, the binding affinities for the G-
quadruplexes occurred in the following order: TO > T1 > T2 > T3.
These findings were in agreement with the fluorescence
enhancement trend of TO and TO-derived compounds in the
presence of the G-quadruplexes. However, TO exhibited a high
affinity not only for the G-quadruplexes but also for double-
stranded and single-stranded DNA. In contrast, the binding of
T1 towards double-stranded and single-stranded DNA was
extremely weak. Thus, this discrepancy may explain why T1
could be used in the selective detection of G-quadruplexes.
Notably, T1 showed selective fluorescence response for the G-
quadruplex, Pu27, with a strong discrimination against htg21, but
their binding affinities were similar. Furthermore, the binding
affinities of T1, T2 and T3 for Pu27 were similar, but only T1
exhibited a significant fluorescence enhancement upon binding.
These results suggested that the selective fluorescence response
of T1 may arise not only from the binding selectivity but also
from other factors.
Torsion angle (ꢁ)
Alone
With G-quadruplex Pu24I
With G-quadruplex htg21
TO
T1
T2
T3
16.2
45.6
51.9
62.4
19.36
49.07
58.04
61.45
26.01
61.57
79.94
69.77
nearly planar. This locked planar conformation is the key factor that
is responsible for its significant fluorescence emission and further
twisting about the methine bridge beyond an interplanar angle of
60ꢁ would lead to its dark state [2,41]. In a previous study, we also
demonstrated that the introduction of a hydrocarbon ring affected
the planar state of an unfused aromatic chromophore [42]. Thus,
we predicted the possibility that the planarity of TO-derived
compounds upon their binding to G-quadruplexes may also impact
their fluorescence emissions. Therefore, modeling studies were
performed to understand the planarity of these compounds, and TO
was used as a reference.
The structures of TO and TO-derived compounds were first
constructed and optimized with GAUSSIAN using the HF/6e31G*
basis set. The planarity of these compounds could be defined by
measuring the torsion angle between the benzothiazole and qui-
nolinium rings. As shown in Table 4, the torsion angles of TO, T1, T2
and T3 were 16.2ꢁ, 45.6ꢁ, 51.9ꢁ, 62.4ꢁ, respectively, indicating that
increasing the hydrocarbon ring size resulted in a significantly
decreased planarity of their unfused aromatic core. These data
could explain the results observed in the self-aggregation mea-
surement and SPR assay because the non-planar conformations of
T1, T2 and T3 may not allow the self-aggregation and intercalative
binding in the duplex DNA to occur.
3.7. Molecular modeling studies
It has been reported that the fluorescence enhancement of TO
for nucleic acids may be caused by conformational changes in the
excited state of TO, most likely by the rotation restriction around
the methine-bridge that separates the benzothiazole and quinoli-
nium rings upon binding. In this case, the chromophore of TO is
Next, molecular models of TO and TO-derived compounds with
G-quadruplexes were generated by docking studies. The parallel
Fig. 7. Staining of G-quadruplex DNA Pu27, double stranded DNA hairpin and single-stranded DNA mut-htg21 by T1 (A) and GelRed (B).

84
R.-J. Guo et al. / Dyes and Pigments 126 (2016) 76e85
Fig. 8. Complex models of TO (A), T1 (B), T2 (C), T3 (D) with Pu24I and complex models of TO (E), T1 (F), T2 (G), T3 (H) with htg21.
Pu27-derived NMR G-quadruplex structure, Pu24I (PDB ID: 2A5R)
[36], and an antiparallel NMR G-quadruplex structure for htg21
(PDB ID: 2MB3) were used as the templates [37]. As shown in Fig. 8,
all of the compounds stacked on the G-quartet in a different
manner. Considering their torsion angles in the presence of the G-
quadruplexes (Table 4), TO was locked in an almost planar
conformation, but the TO-derived compounds remained non-
planar. Among the TO-derived compounds, the torsion angle of
T1 in the presence of the G-quadruplex, Pu24I, was much smaller
than in the presence of the G-quadruplex, htg21. Moreover, the
torsion angle was also much smaller than those of T2 and T3 in the
presence of the G-quadruplex, Pu24I. Notably, these findings were
in agreement with the trends observed in the fluorescence studies.
Because the binding affinities of T1, T2 and T3 for Pu27 and T1 for
htg21 were similar, the discrepancy of their planarity might be the
key factor responsible for the different fluorescence response upon
binding to the G-quadruplexes.
Acknowledgment
This work was financially supported by the National Natural
Science Foundation of China (Nos. 81330077 and 21272291) and
the Fundamental Research Funds for the Central Universities
(Grants 14ykpy09 to J.-H. Tan).
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.dyepig.2015.11.010.
References
[1] Rye HS, Yue S, Wemmer DE, Quesada MA, Haugland RP, Mathies RA, et al.
Stable fluorescent complexes of double-stranded DNA with bis-intercalating
asymmetric cyanine dyes: properties and applications. Nucleic Acids Res
1992;20:2803e12.
[2] Nygren J, Svanvik N, Kubista M. The interactions between the fluorescent dye
thiazole orange and DNA. Biopolymers 1998;46:39e51.
[3] Lee LG, Chen CH, Chiu LA. Thiazole orange: a new dye for reticulocyte analysis.
Cytometry 1986;7:508e17.
4. Conclusions
The development of highly sensitive fluorescent probes to
detect G-quadruplexes is of profound importance for a wide range
of investigations. Although TO is non-selective for double-
stranded and G-quadruplex nucleic acids, it may offer an attrac-
tive template for the design of selective G-quadruplex probes.
Herein, we designed and synthesized a series of new TO-derived
compounds by introducing a hydrocarbon ring. Among these
compounds, T1, which includes a five-member ring compared
with TO, was chosen as the most promising candidate due to its
highly selective fluorescence response to G-quadruplexes. This
simple structural modification to TO significantly improved its
selectivity. Further study of the mechanism revealed that intro-
duction of the hydrocarbon ring to TO significantly decreased the
planarity of the unfused aromatic chromophore. Such non-planar
conformations of T1 may not allow the intercalative binding to
duplex DNA. Notably, T1 displayed considerable selectivity for
parallel G-quadruplex structures with discrimination against some
antiparallel G-quadruplexes. Moreover, we found that the quan-
tum yields of T1 were lower than those of TO. A discrepancy in the
planarity upon its binding to different G-quadruplexes may also
play an important role. In summary, we have successfully
demonstrated that a simple structural modification to TO im-
proves the selective detection performance of G-quadruplexes.
The results will hopefully assist the search for a new generation of
selective probes for G-quadruplexes. Further investigations on
improving the sensitivity of T1 for G-quadruplexes are now
underway.
[4] Rye HS, Quesada MA, Peck K, Mathies RA, Glazer AN. High-sensitivity two-
color detection of double-stranded DNA with a confocal fluorescence gel
scanner using ethidium homodimer and thiazole orange. Nucleic Acids Res
1991;19:327e33.
[5] Zhu H, Clark SM, Benson SC, Rye HS, Glazer AN, Mathies RA. High-sensitivity
capillary electrophoresis of double-stranded DNA fragments using monomeric
and dimeric fluorescent intercalating dyes. Anal Chem 1994;66:1941e8.
[6] Hoevelmann F, Gaspar I, Ephrussi A, Seitz O. Brightness enhanced DNA FIT-
probes for wash-free RNA imaging in tissue. J Am Chem Soc 2013;135:
19025e32.
[7] Monchaud D, Allain C, Teulade-Fichou M-P. Development of a fluorescent
intercalator displacement assay (G4-FID) for establishing quadruplex-DNA
affinity and selectivity of putative ligands. Bioorg Med Chem Lett 2006;16:
4842e5.
[8] Monchaud D, Allain C, Teulade-Fichou MP. Thiazole orange: a useful probe for
fluorescence sensing of G-quadruplex-ligand interactions. Nucleosides Nu-
cleotides Nucleic Acids 2007;26:1585e8.
[9] Burge S, Parkinson GN, Hazel P, Todd AK, Neidle S, Quadruplex DNA. sequence,
topology and structure. Nucleic Acids Res 2006;34:5402e15.
[10] Bochman ML, Paeschke K, Zakian VA. DNA secondary structures: stability and
function of G-quadruplex structures. Nat Rev Genet 2012;13:770e80.
[11] Karsisiotis AI, Hessari NMa, Novellino E, Spada GP, Randazzo A, Webba da
Silva M. Topological characterization of nucleic acid G-quadruplexes by UV
absorption and circular dichroism. Angew Chem Int Ed 2011;50:10645e8.
[12] Davis JT. G-quartets 40 years later: from 50-GMP to molecular biology and
supramolecular chemistry. Angew Chem Int Ed 2004;43:668e98.
[13] Balasubramanian S, Hurley LH, Neidle S. Targeting G-quadruplexes in gene
promoters:
a novel anticancer strategy? Nat Rev Drug Discov 2011;10:
261e75.
[14] Xiong Y-X, Huang Z-S, Tan J-H. Targeting G-quadruplex nucleic acids with
heterocyclic alkaloids and their derivatives. Eur J Med Chem 2015;97:538e51.
[15] Ma D-L, Chan DS-H, Yang H, He H-Z, Leung C-H. Luminescent G-quadruplex
probes. Curr Pharm Des 2012;18:2058e75.
[16] Vummidi BR, Alzeer J, Luedtke NW. Fluorescent probes for G-quadruplex
structures. ChemBioChem 2013;14:540e58.

R.-J. Guo et al. / Dyes and Pigments 126 (2016) 76e85
85
[17] Largy E, Granzhan A, Hamon F, Verga D, Teulade-Fichou M-P. Visualizing the
quadruplex: from fluorescent ligands to light-up probes. Top Curr Chem
2013;330:111e78.
[18] Chan DS-H, Yang H, Kwan MH-T, Cheng Z, Lee P, Bai L-P, et al. Structure-based
optimization of FDA-approved drug methylene blue as a c-myc G-quadruplex
DNA stabilizer. Biochimie 2011;93:1055e64.
[19] He H-Z, Leung K-H, Wang W, Chan DS-H, Leung C-H, Ma D-L. Label-free
luminescence switch-on detection of T4 polynucleotide kinase activity using a
G-quadruplex-selective probe. Chem Commun 2014;50:5313e5.
[20] Laguerre A, Stefan L, Larrouy M, Genest D, Novotna J, Pirrotta M, et al. A twice-
as-smart synthetic G-quartet: PyroTASQ is both a smart quadruplex ligand
and a smart fluorescent probe. J Am Chem Soc 2014;136:12406e14.
[21] Lu L, Shiu-Hin Chan D, Kwong DWJ, He H-Z, Leung C-H, Ma D-L. Detection of
nicking endonuclease activity using a G-quadruplex-selective luminescent
switch-on probe. Chem Sci 2014;5:4561e8.
[30] Lu Y-J, Yan S-C, Chan F-Y, Zou L, Chung W-H, Wong W-L, et al. Benzothiazole-
substituted benzofuroquinolinium dye: selective switch-on fluorescent
a
probe for G-quadruplex. Chem Commun 2011;47:4971e3.
[31] 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;84:6288e92.
[32] Tan J-H, Ou T-M, Hou J-Q, Lu Y-J, Huang S-L, Luo H-B, et al. Isaindigotone
derivatives: a new class of highly selective ligands for telomeric G-quadruplex
DNA. J Med Chem 2009;52:2825e35.
[33] Hou J-Q, Chen S-B, Tan J-H, Ou T-M, Luo H-B, Li D, et al. New insights into the
structures of ligand-quadruplex complexes from molecular dynamics simu-
lations. J Phys Chem B 2010;114:15301e10.
[34] 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;50:6927e30.
[22] Zhang L, Er JC, Ghosh KK, Chung WJ, Yoo J, Xu W, et al. Discovery of a
structural-element specific G-quadruplex “light-up” probe. Sci Rep 2014;4:
3776.
[23] Bhasikuttan AC, Mohanty J. Targeting G-quadruplex structures with extrinsic
fluorogenic dyes: promising fluorescence sensors. Chem Commun 2015;51:
7581e97.
[24] Doria F, Oppi A, Manoli F, Botti S, Kandoth N, Grande V, et al. A naphthalene
diimide dyad for fluorescence switch-on detection of G-quadruplexes. Chem
Commun 2015;51:9105e8.
[25] 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:
8521e5.
[26] Gabelica V, Maeda R, Fujimoto T, Yaku H, Murashima T, Sugimoto N, et al.
Multiple and cooperative binding of fluorescence light-up probe thioflavin T
with human telomere DNA G-quadruplex. Biochemistry 2013;52:5620e8.
[27] Mohanty J, Barooah N, Dhamodharan V, Harikrishna S, Pradeepkumar PI,
Bhasikuttan AC. Thioflavin T as an efficient inducer and selective fluorescent
sensor for the human telomeric G-quadruplex DNA. J Am Chem Soc 2013;135:
367e76.
[28] Renaud de la Faverie A, Guedin A, Bedrat A, Yatsunyk LA, Mergny J-L. Thio-
flavin T as a fluorescence light-up probe for G4 formation. Nucleic Acids Res
2014;42:e65.
[35] Klochko OP, Fedyunyayeva IA, Khabuseva SU, Semenova OM, Terpetschnig EA,
Patsenker LD. Benzodipyrrolenine-based biscyanine dyes: synthesis, molec-
ular structure and spectroscopic characterization. Dyes Pigments 2010;85:
7e15.
[36] Phan AT, Kuryavyi V, Gaw HY, Patel DJ. Small-molecule interaction with a five-
guanine-tract G-quadruplex structure from the human MYC promoter. Nat
Chem Biol 2005;1:167e73.
[37] Chung WJ, Heddi B, Tera M, Iida K, Nagasawa K, Phan AT. Solution structure of
an intramolecular (3 þ 1) human telomeric G-quadruplex bound to a telo-
mestatin derivative. J Am Chem Soc 2013;135:13495e501.
[38] Lau V, Heyne B. Calix[4]arene sulfonate as a template for forming fluorescent
thiazole orange H-aggregates. Chem Commun 2010;46:3595e7.
[39] Dutta Choudhury S, Bhasikuttan AC, Pal H, Mohanty J. Surfactant-induced
aggregation patterns of thiazole orange: a photophysical study. Langmuir
2011;27:12312e21.
[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;140:4616e25.
[41] Silva GL, Ediz V, Yaron D, Armitage BA. Experimental and computational
investigation of unsymmetrical cyanine dyes: understanding torsionally
responsive fluorogenic dyes. J Am Chem Soc 2007;129:5710e8.
[42] Hou J-Q, Tan J-H, Wang X-X, Chen S-B, Huang S-Y, Yan J-W, et al. Impact of
planarity of unfused aromatic molecules on G-quadruplex binding: learning
from isaindigotone derivatives. Org Biomol Chem 2011;9:6422e36.
[29] Yang P, De Cian A, Teulade-Fichou M-P, Mergny J-L, Monchaud D. Engineering
bisquinolinium/thiazole orange conjugates for fluorescent sensing of
G-quadruplex DNA. Angew Chem Int Ed 2009;48:2188e91.