Carbazole-based fluorescent probes for G-quadruplex DNA targeting with superior selectivity and low cytotoxicity

Article abstract of DOI:10.1016/j.bmc.2020.115641

G-quadruplex DNA plays a very important role in clinical diagnosis and fluorescence analysis has attracted extensive attention. A class of carbazole-based fluorescent probes for the detection of G-quadruplex DNA was established in this work. In this system, the installation of an oligo(ethylene glycol) chain on the scaffold will improve the water-solubility and biocompatibility. The presence of styrene-like different side groups could tune the selectivity toward G-quadruplex DNA binding. Results revealed that the substitution pattern and position gave a great influence on the ability for the discrimination of the G-quadruplex from other DNA structures. Especially, probe E1 bound to G-quadruplex DNA with superior selectivity, which exhibiting almost no fluorescence response in the presence of non-G-quadruplex DNA structures. Comprehensive analyses revealed that E1 could bind both ends of the G-quadruplex, resulting in a significant increase of fluorescence emission intensity. Cellular uptake assay suggested that E1 could pass through membrane and enter living cells with low cytotoxicity.

Full text of DOI:10.1016/j.bmc.2020.115641

Bioorganic&MedicinalChemistry28(2020)115641
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Carbazole-based fluorescent probes for G-quadruplex DNA targeting with
superior selectivity and low cytotoxicity
Quan-Qi Yu, Ming-Qi Wang⁎
School of Pharmacy, Jiangsu University, Zhenjiang 212013, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
G-quadruplex DNA plays a very important role in clinical diagnosis and fluorescence analysis has attracted
extensive attention. A class of carbazole-based fluorescent probes for the detection of G-quadruplex DNA was
established in this work. In this system, the installation of an oligo(ethylene glycol) chain on the scaffold will
improve the water-solubility and biocompatibility. The presence of styrene-like different side groups could tune
the selectivity toward G-quadruplex DNA binding. Results revealed that the substitution pattern and position
gave a great influence on the ability for the discrimination of the G-quadruplex from other DNA structures.
Especially, probe E1 bound to G-quadruplex DNA with superior selectivity, which exhibiting almost no fluor-
escence response in the presence of non-G-quadruplex DNA structures. Comprehensive analyses revealed that E1
could bind both ends of the G-quadruplex, resulting in a significant increase of fluorescence emission intensity.
Cellular uptake assay suggested that E1 could pass through membrane and enter living cells with low cyto-
toxicity.
G-quadruplex DNA
Fluorescent probe
Carbazole
Cell imaging
Cytotoxicity
1. Introduction
significant contributions due to overwhelming advantages such as ra-
pidity, high selectivity, and sensitivity, low cost, non-invasive detection
Targeting of non-canonical DNA secondary structures such as G-
quadruplexes is now considered as an appealing opportunity for clinical
diagnosis and development of new chemotherapeutic agents, especially
because accumulating evidence has suggested the existence of G-
quadruplexes in vivo1–4. G-quadruplexes, with distinct and various
structural features, are four-stranded nucleic acid structures formed by
the stacking of two or more G-quartets in the presence of metal ions
such as K+ and Na+5–7. Bio-informatics suggests that putative G-
quadruplex-forming sequences are abundantly present across the
chromosomes and transcriptomes8. As an important secondary nucleic
acid structure, G-quadruplex has an essential role in protecting the
chromosome from fusion, maintaining the stability of genome and
as well as real-time detectability15–18
.
Recently, several classes of small-molecule fluorescent probes with
high selectivity for the G-quadruplex over duplex DNA have been de-
veloped19–24. Carbazole derivatives are the widely used fluorescence
molecules for G-quadruplex DNA targeting probably due to their at-
tractive structural properties as well as high binding affinities25–31
.
Substitution of the subunits at a suitable position of carbazole core is
found to exert excellent optical properties, like BMVC25, CZ-BT26 TO-
CZ27, BPBC28, 9E PBIC29 and others. They show significant fluorescence
enhancements and higher binding preference to some G-quadruplexes.
However, few of them are able to give almost no fluorescence response
when binding with duplex DNAs. In addition, intrinsic toxicity also
limits their in vivo application. Despite much progress, fluorescent
probes for the detection of G-quadruplex DNA structures with superior
selectivity are rare and still in highly demand.
controlling the expression of genes at the transcriptional level2,9–11
.
Therefore, selective detection of G-quadruplex structures would be very
important in the understanding of their biological processes. One of the
major challenges in targeting G-quadruplexes is the high specificity
toward G-quadruplexes over duplex DNAs, because the sequences of G-
quadruplex are engulfed in the soupy embrace of duplexes. To address
the challenge, much of the recent efforts are aimed towards the de-
velopment of biophysical tools and techniques12–14. In particular,
fluorescence probes, based on the small organic molecules have made
Oligo (ethylene glycol) (OEG) chains functionalized with fluor-
escent probes have received significant interest in considering their
flexibility, chemical stability, good biocompatibility and limited toxi-
city properties32. Based on our experience on the design fluorescent
probes for specific G-quadruplex DNA33,34, in this work, to obtain ef-
ficient, superior selectivity and low toxicity fluorescent probes that can
⁎ Corresponding author.
E-mail address: wmq3415@163.com (M.-Q. Wang).
https://doi.org/10.1016/j.bmc.2020.115641
Received 23 May 2020; Received in revised form 1 July 2020; Accepted 5 July 2020
Availableonline10July2020
0968-0896/©2020ElsevierLtd.Allrightsreserved.

Q.-Q. Yu and M.-Q. Wang
Bioorganic&MedicinalChemistry28(2020)115641
be used for G-quadruplex DNA sensing, a series of new fluorescent
probes (E1, E2 and E3) which are designed on the combination of a
triethylene glycol mono methyl side chain in the center of carbazole
scaffold and styrene-like different side groups at its 3-position have
been developed. Their fluorescence performance on various DNA forms
was investigated. The resulting probes have turn-on fluorescent signal
enhancements towards G-quadruplex DNAs, which exhibited remark-
able differentiation with respect to its structural position and type of
the side groups. More importantly, probe E1 exhibited virtually no
fluorescence response to various duplex DNAs and can penetrate into
living cell with low cytotoxicity. The detailed binding properties for G-
quadruplex DNA were comprehensively assessed through both experi-
mental and modeling studies.
8.71 (s, 1H), 8.53 (d, J = 8.4 Hz, 1H), 8.42 (t, J = 7.9 Hz, 1H), 8.18
(dd, J = 20.4, 11.8 Hz, 2H), 8.03–7.91 (m, 1H), 7.79 (t, J = 6.8 Hz,
1H), 7.74 (d, J = 8.6 Hz, 1H), 7.67 (d, J = 8.2 Hz, 1H), 7.61 (s, 1H),
7.57 (s, 1H), 7.49 (t, J = 7.7 Hz, 1H), 7.28 (t, J = 7.5 Hz, 1H), 4.60 (t,
J = 5.0 Hz, 2H), 4.39 (s, 3H), 3.45 (dd, J = 5.7, 3.6 Hz, 2H), 3.41–3.19
(m, 7H), 3.13 (d, J = 6.5 Hz, 3H). 13C NMR (100 MHz, DMSO‑d6) δ:
154.1, 153.4, 147.0, 146.0, 145.0, 144.4, 144.1, 142.4, 141.6, 141.3,
141.2, 141.1, 129.1, 127.5, 127.3, 126.8, 126.6, 126.5, 126.1, 124.9,
124.6, 124.4, 123.1, 122.7, 122.6, 122.4, 122.2, 121.9, 120.9, 120.7,
120.2, 119.8, 117.8, 114.2, 110.8, 110.7, 110.4, 110.4, 71.6, 70.5,
70.4, 70.1, 70.0, 69.3, 58.4, 46.6, 43.4, 43.2; HRMS: (positive mode, m/
z) calculated 431.2329, found 431.2326 for [M−I]+
2.2.3. Preparation of probe E2
;
2. Experimental methods
This complex was synthesized in a manner identical to that de-
scribed for complex E1, with 1,4-dimethylpyridin-1-ium iodide in place
of 1,2-dimethylpyridin-1-ium iodide. The residue after rotary eva-
poration was purified by column chromatography on silica gel eluting
with CH2Cl2 /CH3OH to afford E2 (55.2%, 325 mg) as a yellow solid. 1H
NMR (400 MHz, DMSO‑d6) δ: 8.80 (d, J = 6.8 Hz, 2H), 8.56 (s, 1H),
8.19 (t, J = 10.8 Hz, 4H), 7.86 (dd, J = 8.7, 1.3 Hz, 1H), 7.70 (dd,
J = 23.6, 8.4 Hz, 2H), 7.50 (dd, J = 18.2, 12.1 Hz, 2H), 7.27 (t,
J = 7.4 Hz, 1H), 4.59 (t, J = 5.2 Hz, 2H), 4.22 (s, 3H), 3.44 (dd,
J = 5.8, 3.6 Hz, 2H), 3.42–3.27 (m, 7H), 3.13 (s, 3H). 13C NMR
(100 MHz, DMSO‑d6) δ: 153.5, 145.1, 142.7, 142.2, 141.3, 126.7,
126.7, 126.6, 123.2, 123.1, 122.5, 121.4, 120.7, 120.4, 120.1, 110.9,
110.6, 71.6, 70.5, 70.1, 70.0, 69.3, 58.4, 47.0, 43.3; HRMS: (positive
2.1. Materials and instrumentation
The starting precursors used for synthesis of fluorescent probes were
supplied by commercial companies. The chemicals were used as re-
ceived and no purification process was performed. All oligonucleotides
were purchased from Sangon Biotechnology Co., Ltd. (Shanghai, China)
and were dissolved in 10 mM Tris-HCl buffer (containing 60 mM KCl,
pH 7.4). The sequences were listed in Table S1. DNA secondary
structures were confirmed by CD spectra (Fig. S1). High resolution
mass spectra (HRMS) were recorded on a Shimazu LCMS-IT-TOF in-
strument with an ESI detector. TEM were carried out on a JEM 2100F
field emission transmission electron microscope. Stock solution of
probes E1, E2 and E3 (1 mM) were prepared in DMSO and stored at 4
℃.
mode, m/z) calculated 431.2329, found 431.2326 for [M−I]+
2.2.4. Preparation of probe E3
.
Probe E3 was prepared by the same method as above but with 1,2-
dimethylquinolin-1-ium iodide to give an orange solid (52.4%,
314 mg). 1H NMR (400 MHz, DMSO) δ: 9.09–8.80 (m, 2H), 8.73–8.38
(m, 3H), 8.24 (dd, J = 16.2, 7.8 Hz, 2H), 8.09 (t, J = 9.2 Hz, 2H),
8.03–7.80 (m, 2H), 7.70 (dd, J = 32.5, 8.4 Hz, 2H), 7.49 (t, J = 7.4 Hz,
1H), 7.29 (t, J = 7.3 Hz, 1H), 4.57 (d, J = 11.5 Hz, 5H), 3.45 (d,
2.2. Synthesis and characterization of probes
The synthetic route of the final probes was depicted in Scheme S1.
2.2.1. Preparation of 9-(2-(2-(2-methoxyethoxy)ethoxy)ethyl)-9H-
carbazole-3-carbaldehyde (intermediate compound 4)35
J = 4.8 Hz, 2H), 3.4–3.28 (m, 6H), 3.29–3.19 (m, 2H), 3.14 (s, 3H). 13
C
POCl3 (0.89 mL, 1 mmol) was added into DMF (7 mL, 90.8 mmol)
under the condition of ice water and the mixture was stirred vigorously
for 1 h. Then 9-(2-(2-(2-methoxyethoxy)ethoxy)ethyl)-9H-carbazole
(3.1 g, 3 mmol) dissolved in 1, 2-dichloroethane was added into the
mixture refluxed overnight under the protection of nitrogen. After
cooling to room temperature, Sodium bicarbonate solution was poured
into the solution to neutralize the acidity and then extracted with di-
chloromethane. The solvent was dried with anhydrous sodium sulfate
and evaporated on a rotary evaporation. The crude product was pur-
ified on a silica gel column eluting with CH2Cl2/CH3OH yielding 0.8 g
(yield 32%) of the desired product. 1H NMR (400 MHz, CDCl3) δ: 10.06
(s, 1H), 8.56 (d, J = 1.2 Hz, 1H), 8.37–7.86 (m, 3H), 7.75–7.10 (m,
4H), 4.50 (t, J = 5.8 Hz, 2H), 3.87 (t, J = 5.8 Hz, 2H), 3.62–3.32 (m,
9H), 3.30 (s, 3H).
NMR (100 MHz, DMSO‑d6) δ: 156.5, 149.5, 145.4, 143.2, 142.7, 141.2,
139.2, 134.6, 130.1, 128.6, 128.3, 127.4, 126.8, 126.5, 123.1, 122.7,
122.5, 121.0, 120.8, 120.3, 119.2, 115.7, 110.7, 71.6, 70.5, 70.2, 70.0,
69.2, 58.4, 43.3; HRMS: (positive mode, m/z) calculated 481.2486,
found 481.2483 for [M−I]+
.
2.3. Measurements and methodology
The instruments and the experiments used were identical to the one
used in the article we reported earlier36. The 2-Ap titration experiment
was carried out as following: probe E1 or hemin were gradually added
into the solution containing 2-Ap labeled c-myc G-quadruplex DNA at
fixed concentration (2 μM) in 10 mM Tris-HCl buffer, pH 7.4 in the
presence of 60 mM KCl. The fluorescence measurement was taken as
Ex = 305 nm.
2.2.2. Preparation of probe E1
A mixture of intermediate 9-(2-(2-(2-methoxyethoxy)ethoxy)ethyl)-
9H-carbazole-3-carbaldehyde (4, 350 mg, 1 mmol), 1,2-dimethylpyr-
idin-1-ium iodide (6, 240 mg, 1 mmol) and 40 mL anhydrous ethanol,
followed by 5 drops catalytic piperidine was refluxed for 12 h under
nitrogen with stirring. After cooling to room temperature, the solvent
was evaporated under reduced pressure and the residue was re-
crystallized from dichloromethane to afford E1 (45.8%, 270 mg) as a
yellow solid. 1H NMR (400 MHz, DMSO‑d6) δ: 8.83 (d, J = 6.2 Hz, 1H),
3. Results and discussion
3.1. Probes design and synthesis
The synthesis of final probes was accomplished by the functionali-
zation of carbazole with a 1-methylpyridinium/1-methylquinolinium
moiety through vinylic linkages, using a previously reported standard
procedure31. The conjugated with different side groups at different
2

Q.-Q. Yu and M.-Q. Wang
Bioorganic&MedicinalChemistry28(2020)115641
positions were introduced for the investigation of the selectivity to-
wards G-quadruplex DNA structures. The incorporation of the posi-
tively charged methylpyridine and methylquinoline could enhance the
solubility in buffer solutions. Moreover, the presence of OEG chain in
probes facilitates their aqueous compatibility and cell permeability for
the imaging in live cells. The detailed synthetic procedure was given in
Scheme S1. Experimental data including 1H NMR, 13C NMR and HRMS
supported their structures (see the Supplementary Information).
vinylic-bridge upon its G-quadruplex binding, as clearly shown by the
emission enhancement in a viscous medium (Fig. S5).
To further ascertain the enhancement of fluorescent signal was due
to the specific binding of E1 with G-quadruplex structure, a competitive
assay was carried out in the presence of large amounts of single- and
double-stranded DNA with c-myc G-quadruplex DNA (Fig. 2B). The
enhancement of fluorescent intensity after addition of c-myc G-quad-
ruplex revealed that c-myc was able to displace bound excess ss- and ds-
DNA very effectively. Furthermore, to mimic the cellular environment
in vivo, we used BSA and HSA proteins as interfering biomolecules (Fig.
S6). The increase of BSA and HSA concentration caused negligible
fluorescence change. These findings again demonstrated that E1 has the
promising utility for selective detection of G-quadruplex in the com-
petitive biological environment.
3.2. Selectivity study of synthesized probes towards G-quadruplex DNAs
against other nucleic acids
Initially, the spectral properties of the final probes E1-E3 were listed
in Table S2. The probes exhibit an absorption maximum in the visible
region between 400 and 445 nm and have a large Stokes shift in Tris -
HCl buffer solution at pH = 7.4, which indicated that the probes pos-
sessed good anti-interference effect and less energy loss property.
Compared with chemical structures of the three probes, featured with
the incorporation of quinoline ring on the carbazole core, the corre-
sponding wavelength of absorption maximum was red-shifted and the
quantum yield was decreased (Φ < 0.01). This observation could be
mainly assigned to the more efficient intramolecular charge transfer
(ICT) process31. The sensing selectivity of the three probes towards G-
quadruplex DNAs was then investigated and evaluated by using fluor-
escence titrations with different DNA structures including G-quad-
ruplexes, single- and double-stranded DNAs. The changes of the fluor-
escent spectra obtained during the titration experiments were shown in
Figs. S2-4. Generally, the new probes induced different levels of fluor-
escence enhancement upon addition of the DNA structures. Obviously,
E1 that methylpyridine ring connected at its position 2 was perfectly
capable of distinguishing G-quadruplex from non-G-quadruplex struc-
tures (Fig. 1). The enhanced fluorescence intensity of E1 was found to
be remarkable upon binding with different G-quadruplex DNAs (par-
allel, anti-parallel and hybrid-type), while single- and double-stranded
DNAs showed almost no fluorescence enhancement. Interestingly, when
the methylpyridine ring connected at its position 4 (E2), it responded to
most of these DNAs, which gave poor selectivity to G-quadruplex DNA.
On the other hand, we found that E3 which bearing a 2-substituted
methylquinoline group instead of methylpyridine moiety also exhibited
certain fluorescence response for binding with non-G-quadruplex DNAs.
By comparing the fluorescence properties of the three probes, it seems
that the structures and positions of substituted group are very crucial to
fit the probes with the G-quadruplex DNA structure. Therefore, with the
preliminary data from fluorescence titration assays, E1 displayed the
best performance on discriminating G-quadruplex from other single-
and double stranded DNAs and was suitable for the research of binding
mechanism and cell applications.
Then, we determined the fluorescence lifetime of E1 in the absence
and presence of the c-myc G-quadruplex DNA (λex = 400 nm and
λem = 545 nm) (Fig. 2C). The fluorescence lifetime of free E1 was
calculated as 0.513 ns. In the presence of the c-myc G-quadruplex, the
lifetime obviously increased to 3.871 ns. Apparently, the probe E1 was
a G-quadruplex-sensitive probe.
Titration of E1 with c-myc G-quadruplex was also followed by
fluorescence to determine the binding affinity and stoichiometry
(Fig. 2D). The binding constant was determined as 0.48 × 106 M−1 by
an independent-site model37, which also implied the formation of a
complex with 1:1 stoichiometry of E1 and c-myc. Moreover, a Job’s plot
experiment (Fig. S7), which exhibits a maximum at 0.5 fraction of E1,
further indicated that a 1:1 complex was formed. The detection limit of
E1 for the detection was also an important parameter. The fluorescence
titration curve revealed that the fluorescence intensity of E1 at 545 nm
increased linearly with the amount of c-myc G-quadruplex DNA in the
0–0.8 μM range (R2 = 0.99, Fig. 2D Inset). Thus, the detection limit
was calculated to be 0.18 μM based on the equation LOD = 3σ/k38
.
Taken together, such results suggested that E1 had a potential quanti-
tative sensing ability.
3.4. The effect of E1 on the conformation and stability of G-quadruplex
DNA
The circular dichroism (CD) spectroscopic studies were carried out
to assess the structural changes of G-quadruplex DNA upon interaction
with probe E1. In the presence of K+ ions, c-myc exhibited a parallel
structure, with a positive band at 265 nm and a negative band at
242 nm (Fig. 3A). The addition of E1 to c-myc DNA solution failed to
change the locations and intensities of two signal peaks, demonstrating
that E1 did not affect the conformation of c-myc G-quadruplex DNA
structure.
Having established that probe E1 could sense and interact with G-
quadruplex DNA structures, we further investigated whether such probe
could induce G-rich sequence c-myc into G-quadruplex structure and
present obstacles to DNA synthesis by DNA polymerases. For this pur-
pose, the primer extension reaction was initiated by the addition of Taq
DNA polymerase to a mixture of 24-mer c-myc G-quadruplex forming
sequence and its partial complementary sequence (21-mer). As ex-
pected, the reaction performed in the absence of any added ligand
generated a full-length product (Fig. 3B, lane 1). In theory, if a ligand
could transform the G-rich sequence c-myc into G-Quadruplex structure,
the primer extension by Taq DNA polymerase would be blocked. It was
observed that the inhibitory effect becomes more obvious at high
concentration of E1 (Fig. 3B), indicating that E1 could enhance Taq
DNA polymerase suspend the DNA polymerization process by stabi-
lizing the G-quadruplex structure. To further confirm the inhibitory
3.3. Binding affinities of E1 with G-quadruplex DNA
We investigated the changes in fluorescence of probe E1 in the
presence of a larger panel of quadruplex DNA structures, as well as
single- and double-stranded DNAs, and the representative titration with
c-myc G-quadruplex was shown in Fig. 2A. In a dilute solution of E1, it
showed weak fluorescent signal at 547 nm (Φ = 0.04). Upon adding c-
myc G-quadruplex DNA, a significantly enhanced emission peak ap-
peared and the signal increased gradually as the concentration of c-myc
was increased in the solution (Φ = 0.24). This fluorescent property was
also observed when treated with other G-quadruplex DNAs. The fluor-
escence enhancement might be caused by conformational changes in
the excited state of E1, most likely by the rotation restriction around the
3

Q.-Q. Yu and M.-Q. Wang
Bioorganic&MedicinalChemistry28(2020)115641
Fig. 1. Comparison of enhancements of emission intensity of E1, E2 and E3 with single-, and double-stranded DNAs; G-quadruplexes DNAs in 10 mM Tris-HCl buffer
(pH 7.4) containing 60 mM KCl. The concentration of each probe was 1 μM.
effect, a control oligomer, Mut c-myc, which cannot form the G-quad-
ruplex structure, was conducted. Under the same conditions, the control
oligomer demonstrated no inhibition (Fig. S8). These results further
indicated that E1 was able to stabilize the structures of c-myc G-quad-
ruplex DNA, and inhibit in vitro DNA synthesis.
In order to get a deep insight into the nature binding of E1 to G-
quadruplex DNA, an elementary and rapid technique, G4/hemin per-
oxidase inhibition assay39,40, was exploited in this study. The binding
modes between G-quadruplex and small molecule usually contain
groove binding and end-stacking. Hemin considered as binding to G-
quadruplex through end-stacking, could result catalytic activity on the
oxidation of ABTS2− by H2O2. If E1, the candidate ligand, competes
with hemin for G-quadruplex binding, will lead to diminish of the
catalytic activity of G4/hemin peroxidase, then could be confirmed that
it has the same reaction site with G-quadruplex as hemin. As shown in
3.5. Study on the binding mechanism of E1 with c-myc G-quadruplex
The above experimental results showed that the probe E1 was
giving excellent fluorescent signal discrimination to G-quadruplex DNA.
4

Q.-Q. Yu and M.-Q. Wang
Bioorganic&MedicinalChemistry28(2020)115641
Fig. 2. (A) Fluorescence titration of 1 μM of E1 with
stepwise addition of the c-myc G-quadruplex DNA,
in 10 mM of Tris-HCl buffer and 60 mM of KCl at a
pH of 7.4 (arrows: 0–3 μM). (B) Fluorescence titra-
tion of 1 μM of E1with the stepwise addition of c-
myc G-quadruplex DNA with 20 μM ds26 in 10 mM
of Tris-HCl buffer and 60 mM of KCl at a pH of 7.4.
(C) The fluorescence lifetime decay of E1 without
and with c-myc G-quadruplex in 10 mM of Tris-HCl
buffer and 60 mM of KCl at a pH of 7.4. (D)
Fluorescence titration, presented as a relative in-
crease of the integral fluorescence (F/F0) of c-myc G-
quadruplex DNA with probe E1 (1 μM) in 10 mM of
Tris-HCl buffer and 60 mM of KCl at a pH of 7.4. The
lines represent fitting to the independent-site model.
Inset: Normalized response of the fluorescence
signal to changing c-myc G-quadruplex DNA con-
centrations.
Fig. 4, with addition of different concentrations of E1, the absorbance at
415 nm (oxidized product of ABTS2−) was slightly restrained, in-
dicating that E1 could not compete with hemin by the end-stacking
model. Given that, there were two possible conclusions:1) E1 bound
with G-quadruplex DNA with the loops/grooves; 2) E1 gave a low
binding affinity at the same site as hemin.
Molecular docking models were generated by using Autodock Vina
modeling tool41,42 to investigate the best binding mode between G-
quadruplex and ligand. The NMR G-quadruplex structure of c-myc (PDB
217v) was used as the template. Probe E1 could undergo a conforma-
tional change upon interaction with G-quadruplex, as it has the rota-
tional flexibility due to vinyl bond between carbazole and pyridyl un-
like previous reported structure with large rigid conjugated aromatic
groups. From the blind docking results, two possible binding models
(end-stacking at the site of 5′ terminal and 3′ terminal) were selected on
the basis of their low energy (Fig. 5). In these models, E1 presented a
stretching conformation and covered a large surface area of G-tetrad. In
addition, the carbazole moiety exhibited π-π stacking interaction with
G-quartet planes to form a complex, allowing effective interaction of
triethylene glycol monomethyl ether group and pyridine ring into G-
quadruplex groove region, which helped E1 to anchor in the binding
To shed light on the binding mode, mutated c-myc G-quadruplex
DNA sequences with different loops were prepared. A fluorescence ti-
tration experiment was performed by adding these mutated c-myc se-
quences to the solution of E1, so as to investigate the influence of these
loop regions. As shown in Table 1 and Fig. S9, the extension and
shortening of three propeller loop and two flanking loop regions has
almost no effect on fluorescence intensity. This result provided evi-
dence that E1 bound to c-myc G-quadruplex DNA may through end-
stacking mode rather than groove binding mode.
Fig. 3. (A) CD titration spectra of c-myc G-quad-
ruplex DNA(5 μM) with 0–2 equiv E1 in 10 mM Tris-
HCl buffer, 60 mM KCl, pH 7.4. (B) Effects of E1 on
the PCR-stop assay with c-myc. E1 (0–20 μM) was
added to the reaction mixture containing 1x PCR
buffer, 10 pmol c-myc, 10 pmol C-myc rev, 0.2 mM
dNTPs and 1.5 units of Taq polymerase. The PCR
products were separated on 15% non-denaturing
polyacrylamide gels in 1 × TBE.
5

Q.-Q. Yu and M.-Q. Wang
Bioorganic&MedicinalChemistry28(2020)115641
site of the G-quadruplex DNA. The binding energy in 5′ terminal
(−6.7 kcal/mol) was equivalent to that in 3′ terminal (−6.8 kcal/mol),
so the occupation of E1 to each binding site contributes equally.
Modification of 2-aminopurine (2-Ap) in different loops has been
widely accepted as a method to study the binding sites between G-
quadruplex and binding ligand43. In order to further determine whether
the binding site of probe E1 to c-myc G-quadruplex was at the 3′ or 5′
terminal, we monitored the impact of E1 binding on the modified c-myc
G-quadruplex DNA with 2-Ap substituted at position of 4th (Ap4),13th
(Ap13) and 22rd (Ap22) from the 5′-terminus of the c-myc sequence,
which were located at 5′ terminal, propeller loop region and 3′ term-
inal, respectively. With the resulting emission spectra being shown in
Fig. 6, fluorescence intensities of Ap4, Ap13 and Ap22 were all dis-
turbed significantly, indicating that closer interaction occurred among
these bases. We further used hemin which is in favor of both 3′ and 5′-
end stacking of G-quadruplex DNA for comparison39. Under the same
experimental conditions, the fluorescence intensities of Ap4, Ap13 and
Ap22 in c-myc were also obviously disturbed by hemin. Combinded
with above results, it seems that E1 had a weak competition with hemin
at the same site for G-quadruplex binding. In order to validate the
binding mechanism, we carried out a FID assay, in which the ability of
hemin to displace the fluorescent probe of E1 from G-quadruplex DNA
system. Hemin having the same binding site will compete with E1 for G-
Quadruplex binding, resulting in the large decrease of the emission. The
data shown in Fig. S10 illustrated that with addition of 1 equiv hemin
to the E1-G-Quadruplex DNA system, the almost fully quenched in
fluorescence emission was observed, which confirmed the competitive
binding of hemin with E1.
Fig. 4. Inhibition activity of E1 on c-myc G4/hemin complexes in 10 mM Tri-
HCl buffer, 60 mM KCl, pH 7.4. Conditions: [ABTS2−
] = 0.18 mM;
[H2O2] = 0.06 mM; [c-myc] = 0.4 μM; [hemin] = 0.8 μM; The curves re-
present the change of the product concentration (absorbance at 415 nm) of G4/
hemin peroxidase within 30 min.
Table 1
Relative fluorescence intensities of E1 (1 μM) in the presence of c-myc G-
Quadruplexes (3 μM) with different flanking loops in 10 mM Tris-HCl buffer,
60 mM KCl, pH 7.4.
Name
Sequence (5′-3′)
Relative FI
c-myc
TTGA GGG T GGG TA GGG T GGG TAAA
TTG AATGGG TGGGT AGGGT GGGT AAA
TTG AGGG TAAGGGT AGGGT GGGT AAA
TTG AGGG TGGGT AATGGGT GGGT AAA
TTG AGGG TGGGT AGGGT ATGGGT AAA
TTG AGGG TGGGT AGGGT GGGT AAATA
AGGG TGGGT AGGGT GGGT AAA
695
702
910
632
570
852
846
861
c-myc-1
c-myc-2
c-myc-3
c-myc-4
c-myc-5
c-myc-6
c-myc-7
3.6. Confocal fluorescence imaging
The most important advantage of fluorescent probes would be in-
tracellular detection. And low cytotoxicity is also one of the most cri-
tical requirements for fluorescent probes. Taken the prior results into
consideration, it seems that probe E1 has the potential to be applied
into the further study of living cell staining and imaging due to its se-
lective enhancement of fluorescence upon binding to G-quadruplex
DNA. First, cell viability assays have been performed using CCK-8
method to evaluate the cytotoxicity of E1 towards three cell lines, in-
cluding two cholangiocarcinoma cell lines (HCCC-9810, RBE) and a
non-malignant human cell line (293 T). After three cells exposed to
different concentrations of E1 from 0 to 100 μM for 72 h, the viable
cells were quantified, as shown in Fig. S11. It was noteworthy that
growth of these cells was not significantly inhibited at a low con-
centration (IC50 > 40 μM), indicating that this probe was safe for bio-
imaging in living cells. Thus, the application of E1 as a selective
staining agent for the detection and imaging of G-quadruplexes in
HCCC-9810 was investigated by using confocal fluorescence micro-
scopy. As for comparison, probes E2 and E3 were also studied. The
confocal imaging showed that all of these three probes could enter
living cells (Fig. 7). E2 and E3 exhibited higher cellular uptake than E1
based on the fluorescence intensity in cells. The co-staining with nu-
cleus-targeted dye DAPI showed that E1 and E2 with pyridinium arms
mainly located in cytoplasm, and E3 with quinolinium arm mainly lo-
cated in nucleoli. The results of experiment in vitro was not in corre-
spondence with that of experiments in vivo, which may be due to some
reasons:1) G-quadruplex formation in gene occurs transitorily when the
double-stranded DNA is actively denatured during transcription and
replication, the fluorescence of complex upon binding to G-quadruplex
might be hardly observed; 2) their diverse affinities to the constituents
of the nucleus as well as differences in their photophysical properties;
3) these probes located in a viscosity environment just like they dis-
solved in ethylene glycol solution. Further we will focus on the mod-
ification of these probes and for enhancing their ability of visualizing G-
quadruplex DNA structure in living cells.
TTG AGGG TGGGT AGGGT GGGT
Fig. 5. End-stacking binding modes of E1 to c-myc G-quadruplex DNA by
molecular docking. (A) Side view and (B) Top view of E1 binding to 5′-end; (C)
Side view and (D) Top view of E1 binding to 3′-end (PDB: 217v).
6

Q.-Q. Yu and M.-Q. Wang
Bioorganic&MedicinalChemistry28(2020)115641
Fig. 6. Plot of (F0-F)/F0 of 2-Ap individually labelled c-myc G-quadruplex DNA versus binding ratio of [E1]/[c-myc] (A) and [Hemin]/[c-myc] (B), λex = 305 nm,
λem = 368 nm. Inset: structures model of the parallel G-quadruplex c-myc and hemin. The positions of 2-Ap bases were marked.
Fig. 7. Confocal fluorescence images of living HCCC-9810 cells stained with 20 μM E1, E2, E3 and 5 mg/mL DAPI for 0.5 h. Scale bar is 50 μm.
4. Conclusions
Validation, Writing - review & editing.
In this work, the knoevenagel type chemistry has been used to ob-
tain a series of carbazole-based fluorescence probe for the detection of
G-quadruplex DNA. We demonstrated that the significance of pyr-
idinium side group and its position when conjugating with the carba-
zole scaffold through an ethylene bridge in molecular probe design. The
fluorescent probe, E1, possessed an excellent selectivity towards G-
quadruplex DNA over duplex DNA and other biologically-relevant
molecules with fluorescent turn-on signal. The comprehensive binding
mechanism studies, together with the molecular modeling analysis in-
dicated that E1 bound with G-Quadruplex by the end-stacking model.
Furthermore, this probe can enter into living cells and mainly locate in
cytoplasma with low cytotoxicity. In fact, there is still a strong need for
novel probes that are capable of responding to G-quadruplex DNA
structures. The current study highlights the potential of the molecular
scaffold of E1 in G-quadruplex selective recognition and gave some
crucial factors on developing of effective probes for G-quadruplex DNA
applications.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
Acknowledgements
This work was financially supported by the National Natural Science
Foundation of China (21907042), Natural Science Foundation of
Jiangsu Province (BK20180857).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmc.2020.115641.
References
CRediT authorship contribution statement
1. Biffi G, Tannahill D, McCafferty J, Balasubramanian S. Quantitative visualization of
DNA G-quadruplex structures in human cells. Nat Chem. 2013;5:182–186.
2. Rhodes D, Lipps HJ. G-quadruplexes and their regulatory roles in biology. Nucleic
Acids Res. 2015;43:8627–8637.
Quan-Qi Yu: Data curation, Investigation, Formal analysis, Writing
- original draft. Ming-Qi Wang: Conceptualization, Funding acquisi-
tion, Methodology, Project administration, Resources, Supervision,
3. Henderson A, Wu YL, Huang YC, et al. Lansdorp, Detection of G-quadruplex DNA in
7

Q.-Q. Yu and M.-Q. Wang
Bioorganic&MedicinalChemistry28(2020)115641
mammalian cells. Nucleic Acids Res. 2014;42:860–869.
telomeres fluorescent carbazole derivative. Anal Chem. 2004;76:4490–4494.
26. Gao FL, Cao SH, Sun W, Long SR, Fan JL, Peng XJ. Development of a two-photon
carbazole derivative probe for fluorescent visualization of G-quadruplex DNA in
cells. Dyes Pigments. 2019;171.
4. Murat P, Balasubramanian S. Existence and consequences of G-quadruplex structures
in DNA. Curr Opin Genet Dev. 2014;25:22–29.
5. Gellert M, Lipsett MN, Davies DR. Helix formation by guanylic acid. PNAS.
1962;48:2013–2018.
27. Lu Y-J, Deng Q, Hou J-Q, 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;11:1019–1029.
6. Burge S, Parkinson GN, Hazel P, Todd AK, Neidle S. Quadruplex DNA: Sequence,
topology and structure. Nucleic Acids Res. 2006;34:5402–5415.
7. Mendoza O, Bourdoncle A, Boule JB, Brosh RM, Mergny JL. G-quadruplexes and
helicases. Nucleic Acids Res. 2016;44:1989–2006.
28. Wei Y, Zhang X, Wang L, et al. Interaction of bisbenzimidazole-substituted carbazole
derivatives with G-quadruplexes and living cells. RSC Adv. 2015;5:75911–75917.
29. Lin D, Fei X, Gu Y, et al. A benzindole substituted carbazole cyanine dye: A no-
veltargeting fluorescent probe for parallel c-myc G-quadruplexes. Analyst.
2015;140:5772–5780.
8. Stegle O, Payet L, Mergny JL, MacKay DC, Huppert JL. Predicting and understanding
the stability of G-quadruplexes. Bioinformatics. 2009;25:374–382.
9. Tian T, Chen YQ, Wang SR, Zhou X. G-quadruplex: a regulator of gene expression and
its chemical targeting. Chemistry. 2018;4:1314–1344.
30. Hao GY, Sun JY, Wei CY. Studies on interactions of carbazole derivatives with DNA,
cell image, and cytotoxicity. Bioorg Med Chem. 2018;26:285–294.
31. Wang M-Q, Ren G-Y, Zhao S, et al. Development of a carbazole-based fluorescence
probe for G-quadruplex DNA: The importance of side-group effect on binding spe-
cificity. Spectrochim Acta A. 2018;199:441–447.
10. Kuzmova E. The Role of G-quadruplexes in Transcriptional Regulation. Chem Listy.
2019;113:320–327.
11. C. Monsen, Robert, Trent, O. John, G-quadruplex virtual drug screening: A review,
Biochimie, 152; 2018:134-148.
12. Schaffitzel C, Berger I, Postberg J, Hanes J, Lipps HJ, Pluckthun A. In vitro generated
antibodies specific for telomeric guanine-quadruplex DNA react with Stylonychia
lemnae macronuclei. PNAS. 2001;98:8572–8577.
32. Li J, Cui Y-C, Bi C-X, et al. Oligo(ethylene glycol)-Functionalized Ratiometric
Fluorescent Probe for the Detection of Hydrazine in Vitro and in Vivo. Anal Chem.
2019;91:7360–7365.
13. Liu M, Liu Y, Wu F, Du Y, Zhou X. Specific stabilization of DNA G-quadruplex
structures with a chemically modified complementary probe. Bioorg Med Chem.
2019;27:1962–1965.
33. Wang M-Q, Xu J, Zhang L, et al. Tuning the selectivity of N-alkylated styr-
ylquinolinium dyes for sensing of G-quadruplex DNA. Bioorg Med Chem.
2019;27:552–559.
14. Duan N, Gong W, Wu S, Wang Z. Selection and Application of ssDNA Aptamers
against Clenbuterol Hydrochloride Based on ssDNA Library Immobilized SELEX. J
Agr Food Chem. 2017;65:1771–1777.
34. Wang M-Q, Liao Y-F, Zhang S-H, Yu Q-Q, Huang J-Q. Synthesis, G-Quadruplex DNA
binding and cytotoxic properties of naphthalimide substituted styryl dyes. Bioorg Med
Chem. 2020;51:115325.
15. Pan W, Zhao J, Chen Q. Fabricating Upconversion Fluorescent Probes for Rapidly
Sensing Foodborne Pathogens. J Agr Food Chem. 2015;63:8068–8074.
16. Zhang J, Chai X, He X-P, Kim H-J, Yoon J, Tian H. Fluorogenic probes for disease-
relevant enzymes. Chem Soc Rev. 2019;48:683–722.
35. Divya KP, Sreejith S, Ashokkumar P, et al. A ratiometric fluorescent molecular probe
with enhanced two-photon response upon Zn2+ binding for in vitro and in vivo
bioimaging. Chem Sci. 2014;5:3469–3474.
36. Wang M-Q, Zhang Y, Zeng X-Y, et al. A benzo(f)quinolinium fused chromophore-
based fluorescent probe for selective detection of c-myc G-Quadruplex DNA with a
red emission and a large Stocks shift. Dyes Pigments. 2019;168:334–340.
37. Stootman FH, Fisher DM, Rodger A, Aldrich-Wright JR. Improved curve fitting
procedures to determine equilibrium binding constants. Analyst.
17. Long L, Cao S, Jin B, Yuan X, Han Y, Wang K. Construction of a Novel Fluorescent
Probe for On-site Measuring Hydrogen Sulfide Levels in Food Samples. Food Anal
Method. 2019;12:852–858.
18. Liu HW, Chen LL, Xu CY, et al. Recent progresses in small-molecule enzymatic
fluorescent probes for cancer imaging. Chem Soc Rev. 2018;47:7140–7180.
19. Dhamodharan V, Pradeepkumar P. Specific Recognition of Promoter G-Quadruplex
DNAs by Small Molecule Ligands and Light-up Probes. ACS Chem Biol.
2019;14:2102–2114.
2006;131:1145–1151.
38. Shortreed M, Kopelman R, Kuhn M, Hoyland B. Fluorescent fiber-optic calcium
sensor for physiological measurements. Anal Chem. 1996;68:1414–1418.
39. Cheng XH, Liu XJ, Bing T, Cao ZH, Shangguan DH. General peroxidase activity of G-
Quadruplex-hemin complexes and its application in ligand screening. Biochemistry.
2009;48:7817–7823.
20. Pandith A, Siddappa RG, Seo YJ. Recent developments in novel blue/green/red/NIR
small fluorescent probes for in cellulo tracking of RNA/DNA G-quadruplexes. J
Photoch Photobio C. 2019;40:81–116.
40. Lu J, Wang J, Hu X, et al. Electrochemical Biosensor Based on Tetrahedral DNA
Nanostructures and G-Quadruplex-Hemin Conformation for the ltrasensitive
Detection of MicroRNA-21 in Serum. Anal Chem. 2019;91:7353–7359.
41. Trott O, Olson AJ. Software News and Update AutoDock Vina, Improving the Speed
and Accuracy of Docking with a New Scoring Function, Efficient Optimization, and
Multithreading. J Comput Chem. 2010;31:455–461.
21. Yuan, J.-H. Shao, W. Chen, S.-B., Huang Z.-S., Tan, J.-H. Recent advances in fluor-
escent probes for G-quadruplex nucleic acids, Biochem. Bioph. Res. Co. In press;
2020.
22. Ma DL, Zhang ZH, Wang MD, Lu LH, Zhong HJ, Leung CH. Recent Developments in
G-Quadruplex Probes. Chem Biol. 2015;22:812–828.
23. Manna S, Srivatsan SG. Fluorescence-based tools to probe G-quadruplexes in cell-free
and cellular environments. RSC Adv. 2018;8:25673–25694.
42. Jiang MZ, Yan H, He RH, Ma YK. Purification and a molecular docking study of α-
glucosidase-inhibitory peptides from a soybean protein hydrolysate with ultrasonic
pretreatment. Eur Food Res Technol. 2018;244:1995–2005.
24. Zhang P-L, Wang Z-K, Chen Q-Y, Du X, Gao J. Biocompatible G-Quadruplex/BODIPY
assembly for cancer cell imaging and the attenuation of mitochondria. Bioorg Med
Chem Lett. 2019;29:1943–1947.
43. Barbieri CM, Srinivasan AR, Rzuczek SG, Rice JE, LaVoie EJ, Pilch DS. Defining the
mode, energetics and specificity with which a macrocyclic hexaoxazole binds to
human telomeric G-quadruplex DNA. Nucleic Acids Res. 2007;35:3272–3286.
25. Chang CC, Kuo IC, Ling IF, et al. Detection quadruplex DNA structures human
8