Topology-Selective, Fluorescent “Light-Up” Probes for G-Quadruplex DNA Based on Photoinduced Electron Transfer

Article abstract of DOI:10.1002/chem.201801701

Six novel probes were prepared by covalent attachment of a G4-DNA ligand (bis(quinolinium) pyridodicarboxamide; PDC) to various coumarin or pyrene fluorophores. In the absence of DNA, the fluorescence of all probes is quenched due to intramolecular photoinduced electron transfer (PET), as evidenced by photophysical and electrochemical studies, molecular modeling, and DFT calculations. All probes demonstrate similarly high thermal stabilization of various G4-DNA substrates belonging to different folding topologies, as assessed by fluorescence melting experiments; however, their fluorimetric response is strongly heterogeneous with respect to the structures of the probes and G4-DNA targets. Thus, the probes containing the 7-diethylaminocoumarin fluorophore demonstrate significant fluorescence enhancement in the presence of G4-DNA, with the strongest “light-up” response (20- to 180-fold) observed for antiparallel G4 structures as well as for hybrid G4 structures, formed by the variants of human telomeric sequence and capable of a conformation change to the antiparallel isoform. These results shed light on the influence of the linker and electronic properties of fluorophores on the efficiency of G4-DNA “light-up” probes operating via PET.

Full text of DOI:10.1002/chem.201801701

A Journal of
Accepted Article
Title: Topology-Selective Fluorescent "Light-Up" Probes for G-
Quadruplex DNA Based on Photoinduced Electron Transfer
Authors: Xiao Xie, Oksana Reznichenko, Ludovic Chaput, Pascal
Martin, Marie-Paule Teulade-Fichou, and Anton Granzhan
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201801701
Link to VoR: http://dx.doi.org/10.1002/chem.201801701
Supported by

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
Topology-Selective Fluorescent “Light-Up” Probes for
G-Quadruplex DNA Based on Photoinduced Electron Transfer
Xiao Xie,[a,b] Oksana Reznichenko,[a,b] Ludovic Chaput,[a,b,c] Pascal Martin,[d] Marie-Paule Teulade-
Fichou,[a,b] and Anton Granzhan*[a,b]
Abstract: Six novel probes were prepared by covalent attachment of
a G4-DNA ligand (PDC) to various coumarin or pyrene fluorophores.
In the absence of DNA, the fluorescence of all probes is quenched
due to intramolecular photoinduced electron transfer (PET)
evidenced by photophysical and electrochemical studies, molecular
modeling and DFT calculations. All probes demonstrate similarly
high thermal stabilization of various G4-DNA substrates belonging to
different folding topologies, as assessed by fluorescence melting
experiments; however, their fluorimetric response is strongly hetero-
geneous with respect to structures of the probes and G4-DNA
targets. Thus, the probes containing the 7-diethylaminocoumarin
fluorophore demonstrate significant fluorescence enhancement in
the presence of G4-DNA, with the strongest “light-up” response (20-
to 180-fold) observed for antiparallel G4 structures as well as for
hybrid G4 structures, formed by the variants of human telomeric
sequence and capable of a conformation change to the antiparallel
isoform. These results shed light on the influence of the linker and
electronic properties of fluorophores on the efficiency of G4-DNA
“light-up” probes operating via PET.
some attempts, the stability and folding topology of G4-DNA and
G4-RNA still hardly can be predicted purely on the basis of
sequence information.[8–10] A growing body of structural and
biophysical data demonstrates that these structures frequently
escape the consensus motif (i.e., Gn-Ni-Gn-Nj-Gn-Nk-Gn, where n
= 2 to 4; i, j, k = 1 to 7; and N = any base), as evidenced by
snap-back, G-vacant and bulged G4 structures which hardly can
be predicted by bioinformatics algorithms.[11–14] In this context,
experimental probing of formation and topology of G4 structures,
both in vitro and in cellular environment, is crucial for
understanding of their persistence and biological roles, as well
as for improvement of bioinformatics algorithms.[15–18]
Along these lines, a promising approach to assess the
formation and topology of G4 structures relies on the use of
small-molecule fluorescent “light-up” probes, whose emission is
triggered by interaction with the substrate. A number of fluo-
rescent probes for G4-DNA have been described to date, as
summarized in several recent reviews.[19–21] However, in most
cases, the mechanism underlying the quadruplex-promoted
fluorescence enhancement relies on the suppression of non-
radiative excited-state deactivation of the probe, for example
through restriction of intramolecular rotation (e.g., in the probes
belonging to cyanine,[22–25] stryryl[26–32] and biaryl[33–38] dye
families). Since intramolecular movements of the probe may
also be restricted by events other than binding to the DNA target
(e.g., in a viscous microenvironment or upon binding to
proteins)[25,38,39] leading to a false positive “light-up” output,
probes based on other photophysical mechanisms are highly
desired. Along these lines, “smart” fluorescence probes such as
PyroTASQ[40,41] or coumarin–NDI dyads,[42,43] which operate
through substantial conformational changes upon binding to
G-quadruplex target, hold a particular promise. In the absence of
the substrate, the fluorescence of these probes is quenched due
to photoinduced electron transfer (PET) or formation of a non-
fluorescent intramolecular charge-transfer complex between the
fluorophore and a G-quadruplex-recognizing moiety (e.g., a G4
ligand, or a self-assembling G-quartet in PyroTASQ). Binding to
the target disrupts the intramolecular interaction, leading to
fluorescence enhancement.
Introduction
G-quadruplex (G4) structures, readily adopted by guanine-rich
nucleic acid sequences in vitro, are believed to play important
roles as regulatory elements in DNA- and RNA-related
processes such as recombination, replication, transcription and
genome maintenance.[1–3] Thus, whole-genome sequencing
experiments and bioinformatics studies predict the formation of
numerous G4 structures in the genome.[4–7] However, in spite of
[a]
Dr. X. Xie, O. Reznichenko, Dr. M.-P. Teulade-Fichou, Dr. L. Chaput,
Dr. A. Granzhan
CNRS UMR9187, INSERM U1196
Institut Curie, PSL Research University
F-91405, Orsay (France)
E-mail: anton.granzhan@curie.fr
Dr. X. Xie, O. Reznichenko, Dr. M.-P. Teulade-Fichou, Dr. L. Chaput,
Dr. A. Granzhan
[b]
CNRS UMR9187, INSERM U1196
Université Paris Sud, Université Paris-Saclay
F-91405, Orsay (France)
Dr. L. Chaput
CNRS UPR2301
Institut de Chimie des Substances Naturelles (ICSN)
F-91198, Gif-sur-Yvette (France)
Dr. P. Martin
ITODYS, CNRS UMR7086
Université Paris Diderot
Aiming at the systematic development of “smart” probes
operating through conformational change upon binding to G4-
DNA, we concentrated our efforts on the fluorophore conjugates
of G-quadruplex ligands. In this context, bis(quinolinium)
pyridinedicarboxamide (PDC) derivatives represent one of the
most appealing and well-established families of G4-DNA ligands
due to their high affinity and selectivity as well as facile synthetic
access. Ligands belonging to this series, such as PDC 360A
(herein PDC, Figure 1, a) localize to telomeres and exhibit anti-
cancer activity.[44,45] These properties of PDC derivatives were
exploited, for example, for G4-selective photo-crosslinking and
[c]
[d]
F-75205, Paris (France)
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
photo-cleavage of DNA[46,47] as well as for telomere targeting of
anti-cancer platinum(II) complexes.[48] Moreover, the PDC
scaffold has already been exploited in the design of fluorescent
probes for G4-DNA via covalent linking of a DNA-sensitive probe
(Thiazole Orange); however, the resulting conjugate (PDC-L-
TO) appeared to lose the selectivity to G4-DNA due to
significant non-specific DNA binding of the fluorophore moiety.[49]
obtained through systematic variation of the fluorophore (FL)
and the linker (Ln) parts of PDC-fluorophore conjugates.
Results
Synthesis
The synthesis of PDC-fluorophore conjugates (PDC-Ln-FL,
where L1–L3 = linker and FL = fluorophore) is presented on
Scheme 1. The key intermediate 1 was obtained in a 79% yield
through a one-pot, two-step reaction of chelidamic acid with
3-aminoquinoline, which represents an improvement of the
previously established three-step procedure.[48] Subsequent
substitution of the chlorine atom in 1 with aliphatic diamines in
pyridine solvent gave three linker-substituted derivatives
endowed with terminal primary amino groups (2-Ln-NH2; L1–L3:
see Scheme 1), in a yield of 57–70%. These were made to react
with activated (NHS) esters of substituted coumarin-3-carboxylic
acids (C1–C3) or of pyrene-1-carboxylic acid, to give the PDC-
fluorophore conjugates (2-Ln-Cm and 2-L2-PY, respectively) in
good yields. Finally, the endocyclic nitrogen atoms of quinoline
residues of the conjugates were methylated using excess
iodomethane in DMF to give five cationic, water-soluble
derivatives PDC-Ln-Cm as well as one pyrene analogue,
PDC-L2-PY.
Figure 1. a) Structure of bis(quinolinium) pyridodicarboxamide 360A (PDC); b)
general structure and suggested mode of operation of “light-up” PDC-
fluorophore conjugates; Ln = linker (L1 = −(CH2)3−, L2 = −(CH2)2O(CH2)2−, L3
= −CH2(CH2OCH2)2CH2−); FL = fluorophore.
In order to study the mutual influence of the PDC and the
fluorophore moieties on the photophysical and G4-DNA-binding
properties of the conjugates, we also designed model water-
soluble compounds bearing a 2-methoxyethyl substituent (L) as
a linker surrogate. Towards this end, the PDC derivative PDC-L
was prepared using a similar reaction sequence. In parallel, N-
(2-methoxyethyl)coumarin-3-carboxamides C1-L, C2-L and
C3-L, as well as the pyrene analogue PY-L, were prepared
In view of these data, we designed and prepared a novel
series of conjugates through covalent linking of PDC moiety to
electron-rich fluorophores a priori devoid of significant DNA
affinity, such as coumarins or pyrene. We anticipated that
fluorescence of these conjugates could be intrinsically quenched
due to PET or formation of an intramolecular charge-transfer
complex between the fluorophore and the electron-deficient
PDC moiety.[50–52] In this case, binding to G4-DNA accompanied
by a conformational change of the probe could, at least partly,
restore its fluorescence (Figure 1, b), similarly to what is
observed with “smart” G4-probes. We report herein the
synthesis and G4-DNA-probing studies of this series of probes,
through
a reaction of NHS esters of the corresponding
carboxylic acids with 2-methoxyethylamine (Scheme 1). The
1
identity and purity of all new compounds were confirmed by H
and 13C NMR spectroscopy, mass-spectrometry and micro-
analysis data; in addition, the purity of all compounds used in
photophysical and DNA-binding studies was assessed by
LC/MS analysis and proved to be >96% in all cases.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
Scheme 1. Synthesis of PDC-fluorophore conjugates (PDC-Ln-FL) and model compounds (PDC-L, FL-L); n = 1 to 3, FL = fluorophore. Reagents and conditions:
(i) SOCl2, DMF, reflux, 24 h; (ii) NEt3, DMF, 0 °C, 4 h; (iii) DIPEA, pyridine, 90 °C, 24–72 h; (iv) N-hydroxysuccinimide, EDCI, DMF, room temp., 18 h; (v) pyridine,
room temp., 20 h; (vi) MeI, DMF, 40 °C; (vii) THF, room temp., 1 h.
Table 1. Photophysical properties of PDC-fluorophore conjugates and model compounds in acetonitrile and in aqueous buffer solution.[a]
Compound
Acetonitrile
Aqueous buffer solution[b]
λabs / nm [c]
347
ε / cm−1 M−1
36350
H / % [d]
λem / nm [e]
402
φ / %
0.17[f]
4.5[g]
4.2[g]
3.9[g]
28[g]
λabs / nm [c]
347
ε / cm−1 M−1
24500
H / % [d]
28
λem / nm [e]
401
φ / %
PDC-L2-C1
PDC-L1-C2
PDC-L2-C2
PDC-L3-C2
PDC-L2-C3
PDC-L2-PY
6
0.21[f]
415
40500
8
461
403
21800
39
476
0.05[h]
415
40500
8
462
430
28650
32
476
0.02[h] (1.6[i])
0.01[h] (2.1[i])
0.45[h] (7.3[i])
0.07[f]
415
40950
4
460
432
33550
22
476
433
32500
18
8
478
446
19950
42
494
343
330
35300
28250
380
400
0.06[f]
345
24450
22
385
399
PDC-L
C1-L
C2-L
C3-L
PY-L
351
344
414
431
17100
23650
45200
41250
n/d [j]
398
460
476
n/d
346
348
428
446
14650
24550
46000
38500
n/d
n/d
68[f]
14[g]
103[g]
18[f]
403
478
493
73[f]
1.5[h]
78[h]
75[f]
340
326
31750
22750
380
399, 421
339
326
28400
20100
383
403
[a] For details of quantum yield measurements, cf. Supporting Information, Table S1. [b] K-100 buffer (10 mM LiAsO2Me2, 100 mM KCl, pH 7.2). [c] Long-wavelength
absorption maxima (at c = 20 µM). [d] Hypochromism of the long-wavelength absorption band [H = 1 – ƒPDC-Ln-FL / (ƒFL-L + ƒPDC-L), where ƒ is oscillator strength,
[f]
calculated through integration of long-wavelength absorption bands]. [e] Fluorescence emission maxima.
λ
= 340 nm, reference: quinine sulfate in 0.1 M aq.
ex
H2SO4 (φ = 59%). [g] λex = 420 nm, reference: Coumarin 153 in EtOH (φ = 38%). [h] λex = 440 nm, reference: Coumarin 153 in EtOH (φ = 38%). [i] In the presence of
a saturating concertation of 22AG (cprobe ≤ 5 µM, c22AG = 10 µM). [j] Not determined.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
Photophysical properties
of H-type dimers (or higher intermolecular aggregates) of PDC-
L1-C2 in the aqueous medium since, at lower concentrations (c
≤ 5 µM), this band completely shifted to ~430 nm (Supporting
Information, Figure S3), which is similar to other C2 derivatives
(cf. Table 1). The absorption spectra of other conjugates were
not affected by concentration changes, suggesting that they do
not form intermolecular aggregates in the working concentration
range (c < 40 µM).
Absorption spectra: The photophysical properties of PDC-
fluorophore conjugates and of model compounds were studied
in acetonitrile and in an aqueous buffer solution of high ionic
strength (10 mM Li+, 100 mM K+), compatible with native G4-
DNA structures (Table 1). In acetonitrile, the absorption spectra
of PDC-fluorophore conjugates (Figure 2) resemble the
arithmetical sum of the spectra of the corresponding model
compounds, i.e. PDC-L and Cm-L (or PY-L). However, we
observed a small decrease of the intensity of the long-
wavelength absorption band of the conjugates with respect to
In contrast to amino-substituted coumarin derivatives, the
absorption spectra of C1-L and PY-L, as well as their PDC
conjugates, showed only slight or no positive solvatochromism.
the sum of absorption of model compounds (hypochromic effect), This fact may be attributed to a significantly less polar excited
indicative of intramolecular interactions between PDC and
fluorophore moieties.[53–55] This effect is quantitatively expressed
by percent hypochromism H (PDC-L2-C3: H = 18%, other
conjugates: H < 10% in acetonitrile, Table 1). In aqueous buffer
solution, the hypochromic effect was much more pronounced
(up to H ≈ 40% for PDC-L1-C2 and PDC-L2-C3), giving
evidence of significantly more efficient intramolecular stacking
interactions.
Upon change from acetonitrile to aqueous buffer solution,
the absorption bands of coumarins C2-L and C3-L, as well as
the ones of the corresponding PDC conjugates, undergo batho-
chromic shift by ~15 nm (Supporting Information, Figures S1 and
S2), as normally would be expected for coumarin derivatives
showing positive solvatochromism.[56–58] A notable exception was
observed in the case of PDC-L1-C2, whose long-wavelength
band, in addition to a strong hypochromism, undergoes a
hypsochromic shift from by 12 nm, namely from 415 nm in
acetonitrile to 403 nm in K-100 buffer (at c = 20 µM, Figure S2,
b). This particular behaviour could be attributed to the formation
state of C1 and pyrene fluorophores, compared with amino-sub-
stituted coumarin derivatives whose excited state demonstrates
a strong intramolecular charge-transfer character.[59]
Fluorescence properties: Model coumarin and pyrene
carboxamide derivatives are strongly fluorescent compounds,
with quantum yield of fluorescence approaching unity in the case
of C3-L in MeCN (Supporting Information, Figures S4–S6 and
Table S1). The fluorescence quantum yield of C2-L was reduced
almost 10-fold in aqueous buffer solution, comparing with
acetonitrile (from 14% to 1.5%), whereas those of C1-L and
C3-L were much less affected. In contrast, the fluorescence
quantum yield of PY-L was higher in aqueous buffer (75%) than
in acetonitrile (18%), without significant changes of emission
wavelengths (Table 1). However, in all PDC-fluorophore
conjugates the fluorescence was strongly quenched. In aceto-
nitrile, the quantum yield of C2- and C3-derivatives of PDC was
reduced about 3.5-fold with respect to the corresponding model
compounds (PDC-Ln-C2: φ ≈ 4%, PDC-L2-C3: φ = 28%),
whereas PDC-L2-C1 and PDC-L2-PY were much more dras-
tically affected (φ = 0.17% and 0.06%, respectively). In aqueous
buffer, all derivatives showed very weak fluorescence, with C2
derivatives being particularly affected (φ ≤ 0.05%). At the same
time, the positions of emission bands of PDC conjugates were
almost unchanged with respect to model fluorophores, giving
evidence that fluorescence originates from the fluorophore
moiety and not from a charge-transfer complex. Altogether,
these data demonstrate an efficient quenching of fluorescence
of all coumarin and pyrene chromophores by the linked PDC
moiety, which is particularly reinforced in aqueous medium.
a)
b)
C2-L
PDC-L1-C2
PDC-L2-C2
PDC-L3-C2
1.5
1.0
0.5
0.0
1.5
1.0
0.5
0.0
C1-L
PDC-L2-C1
PDC-L
250
300
350
400
450
500
250
300
350
400
450
500
c)
d)
1.5
1.0
0.5
0.0
1.5
1.0
0.5
0.0
C3-L
PDC-L2-C3
PY-L
PDC-L2-PY
Electrochemical properties
Electrochemical properties of model compounds C1-L, C2-L,
C3-L, PY-L and PDC-L were investigated by cyclic voltammetry
in acetonitrile solutions (Supporting Information, Figure S7). The
redox potentials were extracted from the reversible oxidation or
reduction peaks (slow electron transfer) corresponding to
oxidation or reduction of the fluorophore or the PDC core.[50,60]
The redox potentials of coumarin-3-carboxamide derivatives
(Table 2) demonstrate that both one-electron oxidation
potentials and one-electron reduction potentials decrease in the
series C1-L / C2-L / C3-L (Eox: from 1.30 to 0.87 V; Ered: from
−1.31 to −1.55 V vs. Ag/AgCl), in agreement with the increasing
electron-donating strength of the substituent in the position 7 of
coumarin residue; thus, C3-L is most easily oxidized, and C1-L
250
300
350
400
450
500
250
300
350
400
450
500
Wavelength / nm
Wavelength / nm
Figure 2. Absorption spectra of PDC-fluorophore conjugates (blue lines) and
model compounds (black lines) in acetonitrile (c = 20 µM). a) PDC-L2-C1 and
C1-L; b) PDC-Ln-C2 (n = 1–3) and C2-L; c) PDC-L2-C3 and C3-L; d) PDC-
L2-PY and PY-L. The spectrum of PDC-L (black dotted line) is shown in all
panels.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
is most easily reduced coumarin derivative in this family. PY-L
has the lowest reduction potential in the series (Ered = −1.86 V).
A comparison with parent, previously reported coumarin and
pyrene derivatives (Supporting Information, Table S2) demon-
strates that introduction of the N-alkylcarboxamide group into
coumarin or pyrene chromophores leads to increased tendency
to reduction, as characterized by a systematic increase of
reduction potential by 0.3–0.4 V; this is consistent with the
electron-deficient nature of the substituent. The electrochemical
properties of PDC and PDC-L were also assessed; interestingly,
one-electron oxidation potentials of both compounds were close,
while Ered of PDC-L was significantly higher that of PDC (−0.80
and −1.40 V, respectively), demonstrating a higher propensity of
PDC-L to serve as electron acceptor despite the presence of
electron-donating amine substituent in the pyridine ring.
where Eox is the oxidation potential of the fluorophore (electron
donor), Ered is the reduction potential of electron acceptor
(PDC-L), E00 is vertical excitation energy (determined as a mid-
point between the long-wavelength absorption and emission
maxima of the fluorophore), and e²/εd is the coulombic term
(~0.06 V in acetonitrile). The computed ∆GET,
values (Table
PDC-L
1) demonstrate that PET quenching is a highly exergonic
process for all four fluorophores, with PY-L and C1-L being most
efficiently quenched by PDC-L (∆GET = −34.3 and −30.5 kcal
mol−1, respectively). This observation is in agreement with the
particularly low fluorescence quantum yields of PDC-L2-PY and
PDC-L2-C1 (φ = 0.06% and 0.17% in acetonitrile, respectively).
In the presence of DNA, fluorophores can also participate in
a PET reaction with nucleosides, which can serve as either
electron donors or electron acceptors. Among all nucleosides,
guanosine, with its highest oxidation potential (Eox (dG) = +1.47 V
vs. NHE in acetonitrile),[62] is prone to act as electron donor in
the reductive electron-transfer fluorescence quenching (3):
Table 2. Redox potentials for model compounds in acetonitrile solutions
and computed free energy changes for photoinduced electron transfer
reactions.
FL* + dG → FL•− + dG•+,
(3)
[a]
[a]
[b]
[c]
[d]
Comp.
Eox
/ V
Ered
/ V
E00
∆GET PDC-L
,
∆GET,dG
/ kcal
mol−1
/ kcal mol−1
whose efficiency is given by eq. (4).
∆GET,dG = Eox (dG) – Ered (FL) – E00 − e²/εd
/ eV
(4)
C1-L
1.30
1.09
0.87
1.23
1.18
1.17
−1.31
−1.47
−1.55
−1.86
−0.80
−1.40
3.36
2.84
2.74
3.46
−30.5
−23.4
−26.0
−34.3
−19.4
−3.8
0.4
The computed ∆GET,dG values (Table 1) demonstrate that
C1-L, PY-L and, to a small extent, C2-L are prone to quenching
by guanosine residues in DNA (∆GET,dG = −19.4, −8.9 and −3.8
kcal mol−1, respectively) in contrast to C3-L which is not
expected to participate in PET reactions with dG. A similar trend
was observed with related coumarin dyes[62] and pyrene
C2-L
C3-L
PY-L
−8.9
PDC-L [e]
PDC [f]
derivatives.[52,63] Note that the ∆GET,
and ∆GET,dG values
PDC-L
have not been corrected for the effects of aqueous medium,[62]
and are merely intended for comparison of relative
susceptibilities of the four fluorophores to electron-transfer
quenching by PDC and guanosine residues, respectively.
[a]
In MeCN solution, supporting electrolyte: 0.1
M LiClO4, reference
electrode: Ag/AgCl (satd. KCl). [b] Energy of the 0–0 transition, calculated
as a midpoint between long-wavelength absorption and short-wavelength
emission maxima in MeCN. [c] Computed Gibbs free energy change for the
PET reaction of fluorophore with PDC-L as electron acceptor (in MeCN). [d]
Idem for the PET reaction with dG as electron donor (in MeCN). [e] Counter-
Molecular modeling
Molecular dynamics and conformational analysis: To get
insight into the molecular structure and conformational stability
of PDC-fluorophore conjugates, we performed a molecular
dynamics (MD) simulation with a representative derivative PDC-
L2-C2. Towards this end, the structure of the dication was
initially minimized using OPLS_2005 force field[64] and placed in
a cuboid box filled with water molecules (TIP3P model) and
chloride anions for charge neutralization; the periodic boundary
conditions were applied to the system. After equilibration, a
12-ns simulation was performed, generating 2000 MD snapshots
(Figure 3, a–b). We observed that, in the course of the
simulation, the initially “unstacked” conformation of the probe
(i.e., with a stretched linker and the coumarin fluorophore in a
distal position with respect to the PDC moiety) converted, after
~1.3 ns, into a “stacked” one, with the fluorophore located in a
close contact with the PDC moiety, and mostly remained so until
the end of the simulation (Figure 3, c). This demonstrates the
significant stability of the “stacked” conformation of the ligand in
aqueous medium (81% of MD snapshots).
ion: BF4 . .
[f] Structure: cf. Figure 1, counter-ion: PF6
−
−
With the knowledge of redox potentials, the efficiencies of
photoinduced electron transfer reactions can be estimated using
the Gibbs free energy change, ∆GET. In the absence in DNA,
the photo-excited fluorophore (FL*) can interact with the PDC
moiety, leading to oxidative electron-transfer fluorescence
quenching (1):
[61]
FL* + PDC2+ → FL•+ + PDC•+
(1)
The thermodynamic driving force of this process (∆GET,PDC-L) is
given by eq. (2):
∆GET,PDC-L = Eox (FL) – Ered (
) – E00 − e²/εd
(2)
PDC-L
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
demonstrates the significant flexibility of the PDC moiety in
aqueous environment, in contrast to the exclusively “closed”
conformation experimentally observed in an organic solvent.[65]
Nevertheless, the significant proportion of the “closed”
conformation demonstrates the high degree of preorganization
and is, likely, responsible for the high G-quadruplex DNA affinity
of this class of compounds, which were shown to bind to G4-
DNA in the “closed” form.[66,67]
Quantum chemical calculations: Electron properties of PDC-
L2-C2 were additionally investigated by means of DFT
calculations with a hybrid dispersion-corrected functional B97D,
well-suited for studies of long-range π-π interactions.[68,69] Full
geometry optimization of the final snapshot of MD simulation
using a B97D/6-31G*/PCM (water) level of theory confirmed the
stability of the stacked conformation of the conjugate (Figure 4).
Moreover, the analysis of frontier molecular orbitals speaks in
favor of the supposed PET process between the coumarin
fluorophore and the PDC moiety. Thus, the HOMO of the
conjugate is predominantly located on the coumarin moiety,
while two lowest unoccupied molecular orbitals are located on
both quinolinium fragments of the PDC moiety (note that these
orbitals are non-degenerate due to the effect of the hovering
coumarin group). This demonstrates that PET between the
chromophore and one of quinolinium groups in the stacked
conformation of the conjugate is feasible.
HOMO
LUMO
LUMO + 1
(E = −0.182 eV)
(E = −0.132 eV)
(E = −0.128 eV)
Figure 3. Conformational analysis of PDC-L2-C2 from MD simulations. a)
Side and b) top views of an overlay of 41 snapshots from MD calculations,
aligned with respect to the pyridine ring. c) Analysis of the mutual arrangement
(stacked / unstacked) of the fluorophore and PDC moieties, as characterized
by the N1(pyridine)–C4a(coumarin) distance. d) Conformation analysis of the
PDC moiety, as characterized by dihedral angles θ(N1–C2-C(O)–NH); two
sets of data (black and red dots) correspond to two amide groups. e)
Structures of “closed” and “open” conformations of PDC moiety encountered
during the MD simulation; the rotatable bonds are shown bold.
Figure 4. Excitation-related molecular orbitals of PDC-L2-C2 (isovalue =
0.025) from ground-state DFT calculations, plotted on the energy-minimized
structure (B97D/6-31G*/PCM in water). Wave function sign is arbitrary.
DNA-binding properties
Circular dichroism spectra: Binding of ligands to G-quadruplex
structures may induce conformation changes of the latter, due to
a particular ligand’s affinity to one or another quadruplex fold.
The most prominent example of this transformation is the shift
from hybrid conformations adopted by the human telomeric
sequence (22AG) in K+-rich conditions to an antiparallel form,
induced by PDC and other ligands belonging to the dicarbox-
amide series.[70,71] In order to assess this phenomenon, we
compared CD spectra of several G4-DNA structures belonging
to different topological groups in the absence and in the
presence of two molar equivalents of PDC-L, PDC-L2-PY, and
three PDC-coumarin conjugates differing in linker length (PDC-
L1-C2, PDC-L2-C2, PDC-L3-C2). Specifically, we employed four
hybrid (22AG: mixture of hybrid isoforms, 24TTG: hybrid-1,
25TAG and bcl2Mid: hybrid-2), two antiparallel (22CTA and
The conformational dynamics of the PDC moiety was also
analyzed (Figure 3, d). This part turned out to be relatively
flexible due to the rotation of both C2(py)–C(O) and C3(quino)–
NH bonds (py = pyridine, quino = quinoline); as a result, each
amide group can adopt two “closed” (with amide hydrogens
pointing inwards the axis of pyridine ring) and two “open”
conformations (with amide hydrogens pointing outwards the
pyridine axis; cf. Figure 3, e). As a consequence, three following
combinations were encountered: (i) both amide groups in one of
“closed” conformations (46% of MD snapshots), (ii) one amide in
a “closed”, the other one in an “open” conformation (53%), and
(iii) both amides in one of “open” conformations (1%). This result
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
a) 22AG
15
b) 25TAG
c) 24TTG
d) bcl2Mid
15
10
5
15
10
5
15
10
5
10
5
0
0
0
0
-5
-5
-5
-5
240
260
280
300
320
240
260
280
300
320
240
260
280
300
320
240
260
280
300
320
15
10
5
15
10
5
e) hras-1
f) 22CTA
g) c-kit87up
h) c-myc
PDC-L
PDC-L1-C2
PDC-L2-C2
PDC-L3-C2
PDC-L2-PY
no ligand
20
10
0
20
10
0
0
0
-5
-5
-10
-10
-10
-10
240
260
280
300
320
240
260
280
300
320
240
260
280
300
320
240
260
280
300
320
Wavelength / nm
Wavelength / nm
Wavelength / nm
Wavelength / nm
Figure 5. CD spectra of G4-DNA samples belonging to different topology groups: hybrid (a–d), antiparallel (e–f), parallel (g–h), in the absence of ligands (DMSO
control, black lines) and after incubation for 1 h in the presence of selected ligands, as indicated in the legend. The insets schematically depict the folding topology
of G4-DNA in the absence of ligands (except for 22AG, presenting a mixture of two hybrid folds), and the arrows indicate the ligand-induced changes in CD
spectra. Conditions: c(G4-DNA) = 5 µM; c(ligand) = 10 µM in K-100 buffer, T = 20 °C.
hras-1) and two parallel G4-DNA structures (c-kit87up: snap-
back parallel, c-myc: simple parallel), whose sequences are
provided in the Supporting Information (Table S3). At the
employed conditions (100 mM K+), all substrates showed CD
spectra typical of their topology (Figure 5). Notably, in the case
of the G4 structures adopted by the variants of the human telo-
meric sequence (22AG, 25TAG, 24TTG), addition of PDC-L led
to strong changes in CD spectra, namely an increase of ellipti-
city at 245 and 290 nm and a decrease at 265 nm, indicative of
the aforementioned conversion to an antiparallel form (Figure 5,
a–c). Compared with PDC-L, the effect induced by various PDC-
fluorophore conjugates was also clearly observed but less
pronounced. With all ligands, the magnitude of the ligand-
induced spectral changes was decreasing in the series 22AG >
25TAG > 24TTG; at the same time, no significant changes were
observed in CD spectra of bcl2Mid forming to the same hybrid-2
fold as 25TAG (Figure 5, d). In the case of hras-1, small
increase of CD bands at 260 and 290 nm was observed in the
presence of all tested ligands (Figure 5, e); however, these
changes are rather indicative of ligand-induced stiffening of the
antiparallel G-quadruplex structure than of a conformational
change. Finally, neither PDC-L nor PDC–fluorophore conjugates
induced significant changes in CD spectra of 22CTA or the two
parallel G4-DNA structures (Figure 5, f–h). Altogether, these
results indicate that PDC–fluorophore conjugates, similarly to
other PDC derivatives, induce conformation changes to G4
structures formed by the polymorphic human telomeric
sequence, namely a shift form hybrid into antiparallel forms. This
change does not seem to affect other (non-polymorphic) G4
folds. Obviously, these data need to be taken into account for
interpretation of the fluorimetric response of the ligands towards
different G4-DNA substrates.
Thermal denaturation experiments: The binding of PDC-
fluorophore conjugates and two “parent” ligands (PDC and
PDC-L) to G4-DNA structures were evaluated by means of
fluorescence-monitored melting experiments performed with
several quadruplex-forming oligonucleotides (hybrid: 21G,
25TAG, 24TTG; antiparallel: hras-1, 22CTA; parallel: c-kit87up,
c-myc), double-labeled with a fluorophore (5′, 6-FAM) and a
quencher (3′, TAMRA).[72–74] The conditions of these
experiments, namely the K+ content of the buffer, were chosen
such that all G4-DNA substrates demonstrated comparable
0
stability if the absence of ligands (Tm ≈ 60 °C, Supporting
Information, Table S4). In addition to relative ranking of ligands
with respect to their binding to G4-DNA, as semi-quantitatively
characterized by the ligand-induced thermal stabilization (∆Tm),
fluorescence melting experiments can provide information on
quadruplex-vs.-duplex selectivity of ligands through analysis of
the drop of ∆Tm values observed upon addition of unlabeled
double-stranded competitor (ds26, 15 or 50 molar equivalents).
The results of melting experiments are presented in Figure 6.
As a general trend, the extent of ligand-induced stabilization of
G4-DNA substrates was following: hras-1 < 22CTA ≈ c-myc <
24TTG < 25TAG < 21G ≈ c-kit87up. Thus, antiparallel G4-DNA
structures (22CTA and, particularly, hras-1) were less prone to
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
30
20
10
stabilization by PDC derivatives. Due to the choice of buffer
conditions, this trend should not reflect the intrinsic
thermodynamic stability of the substrates, but rather a lower
affinity of PDC derivatives to antiparallel G4-DNA substrates;
moreover, a benchmark ligand PhenDC3 induced comparable
stabilization of all substrates, including hras-1 (∆Tm > 30 °C, data
not shown).
21G
0
More specifically, these data demonstrate that the PDC
conjugates with coumarin fluorophores (PDC-Ln-Cm) provided
significant stabilization of all G4-DNA substrates, with ∆Tm
values ranging from about 10 °C (for hras-1) to 20–25 °C (for
hybrid and parallel substrates). Notably, this effect was
systematically, by 5 to 10 °C, lower with respect to PDC and
PDC-L, demonstrating the adverse effect of coumarin
fluorophores on the interaction with G4-DNA. At the same time,
we were not able to infer any systematic influence of the length
of the linker (in the series PDC-Ln-C2) or the structure of the
coumarin unit (in the series PDC-L2-Cm) on the G4-DNA
stabilization efficiency of the probes. Conversely, the ∆Tm values
obtained with PDC-L2-PY were almost as high as the ones
obtained with the “parent” ligands, indicating that the impact of
pyrene fluorophore on the G4-DNA affinity was less pronounced.
Finally, most PDC-fluorophore conjugates conserved the
significant selectivity with respect to double-stranded DNA, since
their ∆Tm values were almost unaffected by the presence of the
duplex competitor. A notable exception was observed in the
case of PDC-L1-C2, whose stabilization effect was strongly
diminished (by up to 8–10 °C, with most substrates) in the
presence of ds26.
25TAG
30
20
10
0
30
24TTG
20
10
0
hras-1
22CTA
10
0
30
20
10
The effect of model fluorophore derivatives (Cm-L and PY-L)
on thermal denaturation of G4-DNA was also evaluated
(Supporting Information, Figure S8). None of these compounds
demonstrated significant (> 3 °C) stabilization of any G4-DNA
substrate, confirming our assumption that coumarin and pyrene
fluorophores do not exhibit significant G4-DNA affinity.
0
c-kit87up
30
20
10
Fluorimetric response towards various G4-DNA structures:
The fluorimetric response of PDC-fluorophore conjugates was
initially investigated using a panel of 20 representative DNA
substrates including 13 G4-DNA structures (four hybrid, four
anti-parallel, and five parallel folds), three double-stranded ones,
and four single strands (Supporting Information, Table S2).
These experiments were performed in high ionic-strength, K+-
rich conditions (100 mM K+), using probe concentration of 5 µM
and a DNA (strand)-to-probe ratio of 2:1 (or an equivalent
nucleotide concentration in the case of double-stranded
substrates). The fluorescence intensity of probe–DNA
complexes was measured using fixed combinations of excitation
and emission filters, chosen on the basis of excitation and
emission spectra of representative samples (Supporting
Information, Figure S9). The results, presented as relative
fluorescence intensity enhancement (RFE, Figure 7)
demonstrated significant differences in the fluorimetric response
of various probes towards DNA substrates. Thus, two
conjugates, namely PDC-L2-C2 and PDC-L3-C2, showed
significant fluorescence enhancement (20- to 160-fold) in the
presence of certain G4-DNA targets, in particular the ones
0
30
c-myc
20
10
0
-L
PDC
PDC
-L2-C1 -L1-C2 -L2-C2 -L3-C2 -L2-C3
-L2-PY
PDC
PDC
PDC
PDC
PDC
PDC
Figure 6. Ligand-induced changes of melting temperature (∆Tm / °C) of
various G4-DNA substrates (F-G4-T), assessed by fluorescence melting
experiments in the absence (dark green bars) or in the presence of double-
stranded competitor ds26 (green: 15, light green: 50 molar equiv.) Conditions:
c(F-G4-T) = 0.2 µM, c(ligand) = 1 µM in K-100 (21G, 25TAG, 24TTG,
c-kit87up), K-10 (22CTA, hras-1) or K-1 (c-myc) buffer. Error bars: standard
deviation from three technical replicates.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
belonging to hybrid (22AG, 25TAG) and anti-parallel (hras-1,
22CTA) topology groups. Thus, in the presence of 22AG, the
fluorescence quantum yield reached 1.6% and 2.1% for PDC-
L2-C2 and PDC-L3-C2, respectively (cf. Table 1). Notably, both
probes demonstrated the same order of fluorimetric response
towards hybrid G4 structures (22AG > 25TAG > 24TTG >
bcl2Mid), which follows the susceptibility of these substrates to
ligand-induced conformational change into antiparallel folds, as
monitored by CD spectroscopy. At the same time, both probes
gave much lower fluorescence increase (< 20-fold) in the
presence of parallel G4 folds, two antiparallel structures (TBA, c-
kit*) and double-stranded substrates; notably, a significant
fluorescence enhancement (up to 45-fold, in the case of PDC-
L3-C2) was also observed in the presence of certain single-
stranded substrates. Two conjugates (PDC-L1-C2 and PDC-L2-
C3) demonstrated a significantly lower degree of fluorescence
enhancement (PDC-L1-C2: up to 26-fold, with hras-1, PDC-L2-
C3: up to 14, with 22AG and 22CTA). Notably, PDC-L2-C3 also
demonstrated preferential “light-up” effect with the variants of
human telomeric sequence (22AG, 25TAG, 24TTG, 22CTA),
with fluorescence quantum yield reaching 7.3% in the presence
of saturation concentrations of 22AG (cf. Table 1). Conversely,
PDC-L1-C2 showed a poor discrimination of different G4-folding
topologies. Finally, two conjugates (PDC-L2-C1 and PDC-L2-
PY) showed almost no fluorescence enhancement in the
presence of all tested DNA substrates (RFE
≤
1.4),
demonstrating their incapacity to serve as G4-DNA probes.
To evaluate the temporal behavior of fluorescence intensity
of the probes upon interaction with their substrates, the
fluorescence intensity of the derivative PDC-L2-C2 was
monitored, in a separate experiment, during 60 min after
addition of several representative G4-DNA structures. In most
cases, the increase of fluorescence intensity upon addition of an
aliquot of G4-DNA (c-kit87up, bcl2Mid, 24TTG, 22CTA, hras-1)
was rapid, and a steady intensity could be achieved in about
5 min (Supporting Information, Figure S10). However, in the
case of 22AG, a rapid initial increase of fluorescence (up to 55-
fold) was followed by a slow decrease to a 35-fold value, which
was close to completion only after about 60 min. This behavior
speaks in favor of a slow conformational rearrangement of the
initially formed fluorescent DNA–ligand complex in favor of a
more stable but less fluorescent structure.
Lastly, in order to verify that coumarin and pyrene fluoro-
phores do not interact with DNA substrates per se, we also
accessed the fluorimetric response of model derivatives (Cm-L
and PY-L) towards the same panel of DNA substrates. In the
case of derivatives C1-L, C3-L and PY-L, the fluorescence
intensity was essentially unchanged in the presence of DNA
structures, speaking in favor of absence of interaction
(Supporting Information, Figure S11). Only in the case of C2-L,
2
PDC-L2-C1
1
0
30
PDC-L1-C2
15
0
120
PDC-L2-C2
60
0
we observed
a
slight, but consistent, enhancement of
200
PDC-L3-C2
fluorescence intensity in the presence of parallel G4-DNA
substrates such as 26CEB (1.6-fold), c-myc (1.5-fold) and c-kit2-
T12T21 (1.7-fold). These results may point out to a (weak)
interaction of coumarin C2 with parallel G4-DNA, leading to a
decrease of molecular flexibility (hindered rotation about the
C−NEt2 bond upon binding to G4-DNA) and reduction of a
radiationless excited-state deactivation pathway.
100
0
20
PDC-L2-C3
10
0
2
Fluorimetric titrations: The interaction of the four most
promising probes, namely PDC-Ln-C2 (n = 1–3) and PDC-L2-
C3, with four representative G4-DNA (22AG, hras-1, 22CTA,
c-kit87up) was additionally studied by fluorimetric titrations. In
order to minimize the possible artifacts, the excitation wave-
length in these experiments was set as to minimize the changes
in absorbance resulting from interaction of the probes with DNA
(λex = 410 nm, cf. Supporting Information, Figure S12). The
binding isotherms, corresponding to the relative increase of
integral fluorescence intensity in the presence of G4-DNA are
presented in Figure 8. In most cases, the observed fluorescence
increase was somewhat lower, as compared with single-
wavelength measurements (cf. Figure 7); thus, the highest
relative fluorescence enhancement (RFE) was observed with
PDC-L3-C2 (~70, with 22AG and hras-1) and PDC-L2-C2 (~40,
with 22AG). This points out to the significant bias inherent to
PDC-L2-PY
1
0
T
A
G
G
A
yc
TA
c-kit*
EB
c-m
N
ss 4ss 6ss 7
ds26
ex 3
TB
Pu24
c-kit87up
2Mid
hras-1
dT26
D
22AG
24TT
25TA
22C
26C
ct
Bcl
dupl
c-kit2-T12T21
Figure 7. Relative emission enhancement (RFE) of PDC–fluorophore
conjugates in the presence of G4 (hybrid: violet, antiparallel: blue, parallel:
light blue), double-stranded (red) and single-stranded (orange bars) DNA
structures (cf. Supporting Information, Table S2). Error bars represent the
standard deviation from two technical replicates. Conditions: c(probe) = 5 µM,
c(DNA)
= 10 µM (strand concentration for oligonucleotides) or 230 µM
phosphate (ct DNA) in K-100 buffer; incubation time: 1 h. Excitation / emission
wavelengths: PDC-L2-C1 and PDC-L2-PY, 355 / 405 nm; PDC-Ln-C2, 440 /
485 nm; PDC-L2-C3: 440 / 520 nm, respectively.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
single-wavelength measurements, due to the shifts of absorption
and fluorescence spectra in the presence of DNA. However, the
G4-DNA selectivity profile was mostly conserved, with hras-1
and 22CTA giving the highest fluorescence enhancement, and
c-kit87up the lowest.
In order to evaluate the binding constants of the dyes, the
data from fluorimetric titrations were fitted to the independent-
site model.[75] In most cases, the data indicated fractional
stoichiometry of binding (0.8 to 2 probe molecules per G4-DNA
substrate), as confirmed by the results of continuous-variation
titrations (Job plots: Supporting Information, Figure S13);
however, for the sake of simplicity, we used simple 1:1 or 2:1
models provided the quality of fit was satisfactory. The results
(Table 3) indicate that, among the four probes, PDC-L1-C2
displayed lowest binding constants to all G4-DNA substrates (Ka
= 0.12–0.76 × 106 M−1) and PDC-L3-C2 the highest ones (Ka of
up to 107 M−1, with 22AG). Of the four substrates, c-kit87up
systematically demonstrated the lowest apparent affinity and
22AG the highest one (taking into account the formation of 2:1
complexes with PDC-L1-C2 and PDC-L2-C2).
40
PDC-L1-C2
PDC-L2-C2
10
5
30
20
10
0
22AG
hras-1
22CTA
c-kit87up
0
0
5
10
15
0
5
10
15
PDC-L3-C2
PDC-L2-C3
10
60
40
20
0
Discussion
5
Photophysical properties: Our photophysical studies
demonstrate that the fluorescence of PDC-fluorophore
conjugates is efficiently quenched when the probes are free in
solution, with fluorescence quantum yields from 3.5-fold (PDC-
L1-C2) to 400-fold (PDC-L2-C1) lower with respect to the
corresponding model compounds FL-L. The redox potentials
and the calculated changes of free energy (∆GET,PDC-L) indicate
that PDC moiety serves as an efficient electron acceptor for
photo-excited fluorophores, in accordance with our model of
oxidative photoinduced electron transfer. However, it appears
that the observed differences between fluorophores with respect
to PET efficiency are mostly governed by their vertical excitation
energies E00: thus, the blue-emitting fluorophores (C1 and PY)
demonstrate more negative values of ∆GET,PDC-L (cf. Table 2) and
undergo much more efficient quenching upon conjugation to
PDC than the green-emitting coumarins C2 and C3. Finally, the
analysis of molecular orbital energies from DFT calculations also
confirms the plausibility of the PET process.
0
0
5
10
15
0
5
10
15
c(G4-DNA) / µM
c(G4-DNA) / µM
Figure 8. Binding isotherms from spectrofluorimetric titrations, presented as a
relative increase of the integral fluorescence of the probes (F/F0) in the
presence of various G4-DNA: 22AG (black), hras-1 (red), 22CTA (blue) or
c-kit87up (magenta). The lines represent fitting to the independent-site model
using the parameters given in Table 3. Conditions: c(probe) = 5 µM in K-100
buffer, excitation wavelength: 410 nm, emission range: 420–700 nm; error
bars represent the standard deviation from three independent titrations.
Table 3. Binding constants (Ka / 106 M−1) of PDC-coumarin conjugates with
selected G4-DNA substrates, determined from fluorimetric titrations.[a]
Our results also indicate that in aqueous buffer solutions
most PDC-fluorophore conjugates exist predominantly in an
intramolecular “stacked” conformation, characterized by a strong
hypochromic effect observed in the absorption spectra, as well
as lower fluorescence quantum yields than in an organic solvent
(acetonitrile). This is confirmed by the results of molecular
dynamics simulations indicating the prevalence of the stacked
conformation in aqueous medium (81% of MD snapshots).
However, in the case of PDC-L1-C2, the photophysical data
allow us to infer the formation of intramolecular H-aggregates
(most likely, H-dimers) at concentrations higher than 5 µM. We
suggest that this particular behavior is due to a short linker in
PDC-L1-C2, which precludes efficient intramolecular stacking of
PDC and coumarin moieties in aqueous medium and, instead,
favors the formation of H-type dimers via intermolecular stacking
of coumarin chromophores. Indeed, formation of H-aggregates
G4-DNA
Probe
22AG
hras-1
22CTA
c-kit87up
PDC-L1-C2
PDC-L2-C2
PDC-L3-C2
PDC-L2-C3
0.42 (2:1)
0.82 (2:1)
> 10 (1:1)[b]
4.5 (1:1)
0.76 (1:1)
2.4 (1:1)
4.3 (1:1)
2.5 (1:1)
0.64 (1:1)
0.78 (1:1)
4.3 (1:1)
0.12 (1:1)
0.20 (1:1)
0.14 (1:1)
0.23 (1:1)
0.33 (1:1)
[a] The stoichiometry (probe:G4-DNA) used for the fitting of titration isotherms
to the independent-site model is given in parentheses. [b] The binding
constant could not be accurately determined due to stoichiometry-limited
binding.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
of coumarin derivatives was occasionally observed in aqueous
medium.[76,77]
process and the structure of the resulting ligand–G4 complex
are still not clear.[70,71] Comparing with PDC-L, the effect of the
conjugates on this conformational change is somewhat less
pronounced. In combination with the fact that thermal
stabilization of antiparallel G4-DNA structures induced by PDC–
fluorophore conjugates is lower compared with PDC-L, this
observation suggests that the fluorophore unit reduces the
affinity of the PDC moiety to antiparallel G-quadruplex folds and
therefore the driving force of the aforementioned structural
transformation. In contrast to the structures derived from human
Indeed, we may suggest that, in this case, the intramolecular
“stacked” conformation is unfavorable due to a short linker
between the PDC and coumarin moieties (three sp3 carbon
atoms). This is also in line with the higher fluorescence quantum
yield of this compound in aqueous buffer (φ = 0.05%), with
respect to longer-linker analogues PDC-L2-C2 and PDC-L3-C2
(cf. Table 1).
Altogether, the PDC-fluorophore system behaves quite
similarly to the related viologen-pyrene[52] and viologen-1,8-
naphtahlimide[51] conjugates. This is reasonable taking into
account the fact that the one-electron acceptor potential of
telomeric sequence,
a non-telomeric hybrid G-quadruplex,
bcl2Mid (a hybrid-2 fold), is not subjected to a conformation
change in the presence of ligands.
PDC-L (−0.80 V) is close to that of methyl viologen (MV2+, Ered
=
−0.69 V vs. SCE), a potent electron acceptor and fluorescence
Fluorimetric response of the probes: While the affinities of all
PDC-fluorophore conjugates to each G4-DNA structure are
comparable, their fluorimetric response is much more hetero-
geneous. Specifically, two probes (PDC-L2-C2 and PDC-L3-C2)
demonstrate significant fluorescence enhancement (up to 180-
fold) upon binding to several antiparallel and hybrid G4-DNA,
two probes (PDC-L1-C2 and PDC-L2-C3) demonstrate
moderate “light-up” effect (notably, without systematic
discrimination between different G4-DNA in the case of PDC-L1-
C2), and two probes (PDC-L2-C1 and PDC-L2-PY) do not
demonstrate any fluorescence increase upon interaction with
DNA substrates of our panel. Remarkably, the fluorimetric
response of the first group of probes is most pronounced with
antiparallel G4-DNA hras-1 and 22CTA, but not in the case of
other antiparallel structures, namely c-kit*[78] and TBA featuring
only two G-quartets (cf. Figure 7). Poor detection of TBA with
G4-selective “light-up” probes has often been documented but is
still poorly understood;[16,27,28] taking into account a similar
behavior of c-kit*, a future systematic study of other two-quartet
G4 structures may shed light on this phenomenon. Furthermore,
quencher.[50]
Binding to G4-DNA and conformations changes of
substrates: The results of fluorescence melting experiments
indicate that all PDC-fluorophore conjugates behave as good
G4-DNA ligands, even though their affinity is systematically
reduced by the attachment of coumarin fluorophores, as
compared with the “parent” ligands PDC and PDC-L. At the
same time, the effect of pyrene moiety on G4-DNA binding of
the conjugates is less detrimental. Taking into account that
neither coumarin nor pyrene derivatives show significant
interaction with G4-DNA substrates per se (as demonstrated by
the results obtained with the corresponding model compounds),
the observed drop in affinity is likely due to the energetic penalty
arising (i) from destacking of the fluorophore prior to interaction
of PDC moiety with G4-DNA, in the case that DNA binding
requires unfolding of the conjugate, or (ii) steric clashes with the
substrate (in the case that “stacked” conformation of the probe is
maintained in the complex with G4-DNA).
In contrast to the results of thermal denaturation experiments, high fluorescence response is observed with three variants of
fluorimetric titrations demonstrated significant differences in
binding constants of ligands to various G4-DNA. Specifically, the
lowest Ka values were systematically observed with a parallel
G4-DNA (c-kit87up), while the values obtained with hybrid
(22AG) and two anti-parallel substrates were of the same order
of magnitude (Table 3). This apparent discrepancy is likely due
to the fact that the fluorimetric titrations reveal exclusively the
formation of fluorescent complexes, neglecting the formation of
probe–DNA complexes with (almost) unchanged fluorescence
but a strong capacity to contribute to thermal stabilization (for
example, binding of the probe in a “stacked” conformation). This
is particularly prominent in the case of parallel G4-DNA which
are strongly stabilized by PDC-fluorophore conjugates, yet it is
possible that only the fraction of the probe binds in a “light-up
state”, resulting in low apparent Ka values.
Remarkably, all PDC-fluorophore conjugates are able to
induce conformational changes of hybrid G4-DNA structures
derived from human telomeric sequence (22AG, 25TAG,
24TTG), namely a conversion to an antiparallel structure
evidenced by CD spectroscopy. This behavior has already been
documented for PDC and other ligands of bisquinolin(ium) series
(360A-Br, PhenDC3, pyridostatin), although the details of this
human telomeric sequence (22AG, 25TAG and 24TTG) known
to adopt one or several hybrid conformations in K+ conditions.
Taking into account the ligand-induced conformational change of
these substrates to an antiparallel structure, we may postulate
that the fluorimetric response is due to formation of a complex of
the probes with antiparallel G4-DNA structure (Scheme 2).
Conversely, a hybrid G4-DNA substrate unable of conformation
change (bcl2Mid) does not give a strong fluorimetric response
with our probes.
In the absence of structural data, we cannot elucidate with
certainty why the conformational change (“unstacking”) of the
probe, required for the “light-up effect”, takes place only with
antiparallel G4-DNA substrates. However, we may propose that
mutual interactions of the nucleobases belonging to lateral loops
and governing the antiparallel G4 folds preclude binding of the
conjugates in the “stacked” conformation, and therefore induce
their “unfolding” leading to reduction of the intramolecular PET
and a “light-up” effect (Scheme 2). To the best of our knowledge,
there are no high-resolution structural models on ligands bound
to antiparallel G4-DNA structures; however, binding between the
external G-quartet and the bases of lateral or diagonal loops has
been proposed for PDC and related ligands, in order to explain
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
the ligand-induced structural rearrangements of telomeric
quadruplexes.[70] Notably, the model presented on Scheme 2
does not exclude simultaneous binding of further probe
molecules to G4 substrate in the “stacked” conformation, which
do not contribute to fluorescence enhancement, but stabilize G4-
DNA (as evidenced by melting experiments). Therefore, the
fluorescence response of probe molecules bound to opposite
sides of a G-quadruplex may differ to a large extent.
after a conformational change of the probe upon binding to G4-
DNA. This hypothesis is supported by the fact that quantum
yields of these two probes in acetonitrile is not higher than in
aqueous buffer, in contrast to 7-aminocoumarin derivatives (cf.
Table 1). (ii) Binding to G4-DNA suppresses the intramolecular
PET, but gives rise to an intermolecular PET reaction of
fluorophores with DNA bases (preferably guanine). This
hypothesis is supported by large values of ∆GET,dG (−19.4 and
−8.9 kcal mol−1 for C1-L and PY-L, respectively), in contrast to
less favorable oxidation of guanine by amino-substituted
coumarins (∆GET,dG = −3.8 and 0.4 kcal mol−1 for C2-L and C3-L,
respectively). Finally, a combination of these two processes may
take place, hindering the restoration of the fluorescence.
antiparallel G4
Conclusions
FL
+
In this work, we synthesized and systematically studied six novel
PET
conjugates of
fluorophores, namely coumarin and pyrene derivatives. Through
photophysical and electrochemical experiments we
a G4-DNA ligand PDC with electron-rich
hybrid G4
demonstrated that, in all cases, the intrinsic fluorescence of
these probes is efficiently quenched due to an oxidative PET
process, with PDC moiety serving as electron acceptor. All
conjugates behave as good ligands for various G4-DNA
structures, with a slightly reduced affinity with respect to the
“parent” PDC ligand. At the same time, their capacity to serve as
G4-selective fluorescent probes varies to a large extent, being
defined both by the nature of the fluorophore and the linker.
Thus, the derivatives of pyrene (PY) and 7-methoxycoumarin
(C1) do not show any “light-up effect” and should not be used in
combination with PDC moiety. A derivative of Coumarin 343
(C3) displays a moderate efficiency, while the best results could
be obtained with 7-diethylaminocoumarin (C2) fluorophore. Also,
the performance of the probes appears to be limited in the case
of short linker (L1), which does not allow efficient switching of
the probe conformations. In summary, two of six conjugates
(PDC-L2-C2 and PDC-L3-C2) represent interesting fluorescent
probes, which selectively detect antiparallel G4-DNA substrates
as well as hybrid structures (formed by the variants of human
telomeric sequence) which may be converted into an antiparallel
isoform. In combinations with the data on other related
conjugates,[42,43] these results define the molecular determinants
for effective design of fluorescent probes, which may be
applicable beyond the field of G4-DNA.
parallel G4
Scheme 2. Proposed model of fluorimetric response of PDC-coumarin
conjugates towards various G4-DNA structures. The dashed lines depict the
base–base interactions defining the antiparallel G4 topology.
A comparison of the G4-DNA response in the series PDC-
Ln-C2 demonstrates the beneficial effect of a longer linker on
the fluorimetric response of the probe. Thus, in the case of PDC-
L1-C2, the less efficient “light-up” effect, comparing with two
other derivatives discussed above, is likely due to combination
of two factors: first, the intramolecular stacking of the probe is
inefficient and leads to a higher fluorescence in the unbound
state (cf. Table 1). Second, in the “unstacked” conformation
obtained upon binding to antiparallel G4-DNA, the distance
between the PDC and fluorophore moieties of the probe is still
shorter as compared to the other analogues, which results in a
less efficient suppression of the intramolecular PET process.
Thus, short linkers (less than five sp3 atoms) should be avoided
in design of this type of fluorescent probes.
The last case is represented by the derivatives PDC-L2-C1
and PDC-L2-PY which do not show any “light-up” effect. Since
the affinity of these probes to G4-DNA is at least as good as the
ones of other conjugates (or even higher, in the case of PDC-
L2-PY), as evidenced by melting experiments, the observed lack
of fluorimetric response is likely due to the electronic properties
of the corresponding fluorophores C1 and PY. Thus, two
possibilities may be considered: (i) G4-DNA binding does not
suppress the intramolecular PET process in these probes.
Indeed, these two probes demonstrate most negative values of
∆GET,PDC-L (−30.5 and −34.3 kcal mol−1, respectively; cf. Table 2)
and we may assume that this process is not suppressed even
Experimental Section
Spectrophotometric measurements: Stock solution of all compounds
were prepared in DMSO at a concentration of 2 mM. Absorption spectra
were recorded with a Hitachi U2900 double-beam spectrophotometer in
quartz cells with a pathway of 1 cm, using scan speed of 200 nm min−1
.
Fluorescence emission spectra were recorded with a HORIBA Jobin-
Yvon Fluoromax-3 spectrofluorimeter. Fluorescence quantum yields were
measured by the comparative method with Coumarin 153 in EtOH (φ =
38%) or quinine sulfate in 0.1 M H2SO4 (φ = 59%) as standards.[79] In this
method, four to five solutions of each compound, with 0.02 < A < 0.15,
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
were used to determine the slopes of plots of integral fluorescence
emission intensity vs. absorbance at a given excitation wavelength, and
the quantum yields were calculated using eq. (5),[80]
CD spectroscopy: Circular dichroism spectra were recorded with a
J-710 spectropolarimeter (JASCO) equipped with a Peltier temperature
controller, in quartz cells with a path length of 5 mm. Pre-folded (as
described above) G4-DNA samples were diluted with K-100 buffer to a
final concentration of 5 µM and mixed with ligands (final concentration:
10 µM) or a DMSO control (0.5% v/v). After incubation for 1 h in the dark
at room temperature, CD spectra were recorded at 20 °C from 210 to 330
nm using the following parameters: data pitch, 1 nm; bandwidth, 2 nm;
response, 2 s; scan speed, 50 nm min−1; number of scans, six.
2
Gradx
nx
φx = φst
,
(5)
(Grad )(n )
st
st
where Grad are the slopes, n are solvent refraction indices, and the
subscripts “x” and “st” denote the sample and the quantum yield standard,
respectively.
Thermal denaturation experiments: Fluorimetric melting experiments
were carried out with G4-forming sequences (Supporting Information,
Table S2), double-labeled with 6-carboxyfluorescein (5′, 6-FAM) and 5-
carboxytetramethylrhodamine (3′, TAMRA). For analysis, 0.2 µM of pre-
folded G4-DNA were mixed with 1 µM of ligand (or DMSO control), in
absence or in presence of competitor DNA (self-complementary duplex
ds26, 3 or 10 µM) in a total volume of 25 µL of a corresponding buffer
(21G, 25TAG, 24TTG, ckit87up: K-100, 22CTA and hras-1: K-10, c-myc:
K-1). Thermal denaturation runs were performed with a 7900HT Fast
Real-Time qPCR instrument (Applied Biosystems) using a single heating
ramp of 0.5 °C min−1 and monitoring the fluorescence intensity (F) in the
FAM channel. Tm values were determined through peak analysis of first-
order derivative plots (dF/dT).
Cyclic voltammetry: Electrochemical experiments were performed in a
conventional three-electrode configuration (CH Instruments, CHI 440A).
The working electrode, namely glassy carbon (GC, diameter: 3 mm), was
used for the characterization of dyes in solution. A saturated silver
electrode (Ag/AgCl) and a stainless steel grid were used as reference
and counter-electrode respectively. After polishing, the working electrode
was immersed in the degassed dye solution and cyclic voltammetry
technique was used to investigate the electroactivity of the molecule.
Molecular dynamics: The starting structure of PDC-L2-C2, drawn in a
stretched-out conformation, was optimized with the Minimization module
of the SCHRÖDINGER suite. The structure was minimized using the
OPLS_2005 force field[64] and the Polak–Ribière Conjugate Gradient
method for a maximum of 2500 iterations to a gradient convergence
threshold of 0.05 kJ mol−1 Å−1. The Molecular dynamics calculation was
performed using DESMOND, version 3.6, a part of the SCHRÖDINGER suite.
A cuboid box of 6568 water molecules surrounding the compound was
built in the TIP3P model, and two chloride ions were added to neutralize
the system charges. The system was equilibrated using the default
protocol available in the Molecular Dynamics module. Then, a 12 ns
simulation was performed in the NPT ensemble at 300 K and 1013 hPa,
using the Nosé–Hoover thermostat and the Martyna–Tobias–Klein
barostat. The RESPA integrator with a time step of 2 fs was utilized. The
periodic boundary conditions were applied to the system, with the far
interactions computed by using the particle mesh Ewald summation and
a cutoff radius of 9 Å for the Coulomb interactions. The trajectories were
analyzed with VMD.[81]
Single-wavelength fluorescence measurements: Measurements at a
fixed DNA concentration were performed in 96-well black, flat-bottom
microplates (Corning 3650) filled with samples (200 µL per well)
containing the probe (5 µM) in the absence or in the presence of DNA
samples (oligonucleotides: final strand concentration of 10 µM, ct DNA:
final phosphate concentration of 230 µM) in K-100 buffer. After mixing for
5 min, the microplates was incubated for 1 h in the dark at room
temperature, and fluorescence intensity was measured with a microplate
reader (BMG FLUOstar Omega) using 50 flashes per well and a suitable
combination of excitation and emission filters with a band-pass of 10 nm
(cf. Figure 7 caption). The data, presented as relative fluorescence
enhancement (RFE), were corrected to the signal of “blank” wells
containing only buffer solution using eq. 6:
F(dye+DNA) – F0
RFE =
,
(6)
F(dye) – F0
Quantum chemical calculations: DFT calculations were performed with
Gaussian 09 Rev. D.01.[82] The structure of the PDC-L2-C2 dication from
where F(dye+DNA) is emission intensity of the dye in the presence of a
DNA sample, F(dye) is the average emission intensity of the wells
containing the dye but no DNA, and F0 is the average emission intensity
of the wells containing only the buffer solution.
the final snapshot from MD simulation was fully optimized on
a
B97D/6-31G*/PCM level of theory (bulk solvent: water). The obtained
stationary point was checked for the absence of imaginary frequencies.
Nucleic acids and buffer solutions: The following buffers were used in
this work: K-100 (10 mM LiAsMe2O2, 100 mM KCl, pH 7.2), K-10 (10 mM
LiAsMe2O2, 10 mM KCl, 90 mM LiCl, pH 7.2), K-1 (10 mM LiAsMe2O2,
1 mM KCl, 99 mM LiCl, pH 7.2). Oligonucleotides (sequences: see
Supporting Information, Table S2), purified by RP-HPLC and lyophilized,
were purchased from Eurogentec. Samples were dissolved in a K-100
buffer to a strand concentration of 100 µM, and the double-stranded
structure (duplex 3) was prepared by mixing equal volumes of solutions
of the corresponding single strands. All samples were then heated to
95 °C for 5 min and left to cool to ambient temperature, to assure
hybridization of double-stranded structures and folding of quadruplexes.
Calf thymus DNA solution (10 mg mL−1, Invitrogen) was diluted in K-100
buffer to a final phosphate concentration of 2.3 mM (as checked by
photometry, using the extinction coefficient of 6412 cm−1 M−1 at 260 nm);
this concentration corresponds to the average phosphate concentration
in oligonucleotide samples.
Fluorimetric titrations and calculation of binding constants:
Fluorimetric titrations were performed and analyzed as described
elsewhere,[27] using solutions of the probes in K-100 buffer (c = 5 µM). To
avoid dilution effects, the titrant solutions of G4-DNA (100 µM in K-100
buffer) also contained the same concentration of the probe (c = 5 µM).
Job plots were obtained by means of MacCarthy titration method using a
total concentration (probe + G4-DNA) of 20 µM.[83]
Acknowledgements
The authors thank Dr. Abhijit Saha, Ms. Maëva Gène and Ms.
Sandia Adypagavane for help with synthesis and spectroscopic
experiments. The CNRS and the French Ministry of Higher
Education, Research and Innovation (MESRI) are gratefully
acknowledged for financial support (PhD scholarship to O.R.),
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
[32] S. Wu, L. Wang, N. Zhang, Y. Liu, W. Zheng, A. Chang, F. Wang, S. Li,
D. Shangguan, Chem. Eur. J. 2016, 22, 6037–6047.
[33] B. Jin, X. Zhang, W. Zheng, X. Liu, C. Qi, F. Wang, D. Shangguan,
Anal. Chem. 2014, 86, 943–952.
and Université Paris Sud for the access to Cloud@VD platform
for scientific computing.
Keywords: G-quadruplex • fluorescent probes • electron
transfer • coumarin • pyrene
[34] L. Liu, Y. Shao, J. Peng, C. Huang, H. Liu, L. Zhang, Anal. Chem. 2014,
86, 1622–1631.
[35] Y. Kataoka, H. Fujita, Y. Kasahara, T. Yoshihara, S. Tobita, M.
Kuwahara, Anal. Chem. 2014, 86, 12078–12084.
[1]
[2]
[3]
[4]
H. J. Lipps, D. Rhodes, Trends Cell Biol. 2009, 19, 414–22.
D. Rhodes, H. J. Lipps, Nucleic Acids Res. 2015, 43, 8627–8637.
J. L. Huppert, Chem. Soc. Rev. 2008, 37, 1375–84.
V. S. Chambers, G. Marsico, J. M. Boutell, M. Di Antonio, G. P. Smith,
S. Balasubramanian, Nat. Biotechnol. 2015, 33, 877–881.
R. Hänsel-Hertsch, D. Beraldi, S. V Lensing, G. Marsico, K. Zyner, A.
Parry, M. Di Antonio, J. Pike, H. Kimura, M. Narita, et al., Nat. Genet.
2016, 48, 1267–1272.
[36] S.-B. Chen, W.-B. Wu, M.-H. Hu, T. Ou, L. Gu, J.-H. Tan, Z.-S. Huang,
Chem. Commun. 2014, 50, 12173–12176.
[37] A. K. Jain, V. V. Reddy, A. Paul, M. K., S. Bhattacharya, Biochemistry
2009, 48, 10693–10704.
[38] M.-H. Hu, S.-B. Chen, R.-J. Guo, T.-M. Ou, Z.-S. Huang, J.-H. Tan,
Analyst 2015, 140, 4616–4625.
[5]
[39] A. Banerjee, P. Majumder, S. Sanyal, J. Singh, K. Jana, C. Das, D.
Dasgupta, FEBS Open Bio 2014, 4, 251–259.
[6]
[7]
[8]
[9]
J. L. Huppert, S. Balasubramanian, Nucleic Acids Res. 2007, 35, 406–
413.
[40] A. Laguerre, L. Stefan, M. Larrouy, D. Genest, J. Novotna, M. Pirrotta,
D. Monchaud, J. Am. Chem. Soc. 2014, 136, 12406–12414.
[41] J. Zhou, B. T. Roembke, G. Paragi, A. Laguerre, H. O. Sintim, C.
Fonseca Guerra, D. Monchaud, Sci. Rep. 2016, 6, 33888.
[42] M. Zuffo, F. Doria, V. Spalluto, S. Ladame, M. Freccero, Chem. Eur. J.
2015, 21, 17596–17600.
J. L. Huppert, S. Balasubramanian, Nucleic Acids Res. 2005, 33, 2908–
2916.
H. M. Wong, O. Stegle, S. Rodgers, J. L. Huppert, J. Nucleic Acids
2010, 564946.
A. Bedrat, L. Lacroix, J.-L. Mergny, Nucleic Acids Res. 2016, 44, 1746–
1759.
[43] M. Zuffo, S. Ladame, F. Doria, M. Freccero, Sens. Actuators, B 2017,
245, 780–788.
[10] A. B. Sahakyan, V. S. Chambers, G. Marsico, T. Santner, M. Di Antonio,
S. Balasubramanian, Sci. Rep. 2017, 7, 14535.
[44] C. Granotier, G. Pennarun, L. Riou, F. Hoffschir, L. R. Gauthier, A. De
Cian, D. Gomez, E. Mandine, J. F. Riou, J. L. Mergny, et al., Nucleic
Acids Res. 2005, 33, 4182–4190.
[11] V. T. Mukundan, A. T. Phan, J. Am. Chem. Soc. 2013, 135, 5017–5028.
[12] A. Varizhuk, D. Ischenko, V. Tsvetkov, R. Novikov, N. Kulemin, D.
Kaluzhny, M. Vlasenok, V. Naumov, I. Smirnov, G. Pozmogova,
Biochimie 2017, 135, 54–62.
[45] G. Pennarun, C. Granotier, L. R. Gauthier, D. Gomez, F. Hoffschir, E.
Mandine, J.-F. Riou, J.-L. Mergny, P. Mailliet, F. D. Boussin, Oncogene
2005, 24, 2917–2928.
[13] X. Li, K. Zheng, J. Zhang, H. Liu, Y. He, B. Yuan, Y. Hao, Z. Tan, Proc.
Natl. Acad. Sci. 2015, 112, 14581–14586.
[46] D. Verga, F. Hamon, F. Poyer, S. Bombard, M.-P. Teulade-Fichou,
Angew. Chem. Int. Ed. 2014, 53, 994–998.
[14] B. Heddi, N. Martín-Pintado, Z. Serimbetov, T. M. A. Kari, A. T. Phan,
Nucleic Acids Res. 2016, 44, 910–916.
[47] D. McBrayer, S. Kerwin, Molecules 2015, 20, 16446–16465.
[48] J. F. Betzer, F. Nuter, M. Chtchigrovsky, F. Hamon, G. Kellermann, S.
Ali, M. A. Calméjane, S. Roque, J. Poupon, T. Cresteil, et al., Bioconjug.
Chem. 2016, 27, 1456–1470.
[15] C. K. Kwok, C. J. Merrick, Trends Biotechnol. 2017, 35, 997–1013.
[16] A. Renaud de la Faverie, A. Guédin, A. Bedrat, L. A. Yatsunyk, J.-L.
Mergny, Nucleic Acids Res. 2014, 42, e65.
[49] P. Yang, A. De Cian, M.-P. Teulade-Fichou, J.-L. Mergny, D. Monchaud,
Angew. Chem. Int. Ed. 2009, 48, 2188–2191.
[17] M. Kim, A. Kreig, C. Y. Lee, H. T. Rube, J. Calvert, J. S. Song, S.
Myong, Nucleic Acids Res. 2016, 44, 4807–4817.
[50] G. Jones, S. F. Griffin, C. Y. Choi, W. R. Bergmark, J. Org. Chem. 1984,
49, 2705–2708.
[18] R. del Villar-Guerra, J. O. Trent, J. B. Chaires, Angew. Chem. Int. Ed.
2017, DOI 10.1002/anie.201709184.
[51] T. P. Le, J. E. Rogers, L. A. Kelly, J. Phys. Chem. A 2000, 104, 6778–
6785.
[19] E. Largy, A. Granzhan, F. Hamon, D. Verga, M.-P. Teulade-Fichou,
Top. Curr. Chem. 2012, 330, 111–177.
[52] M. Hariharan, J. Joseph, D. Ramaiah, J. Phys. Chem. B 2006, 110,
24678–24686.
[20] B. R. Vummidi, J. Alzeer, N. W. Luedtke, ChemBioChem 2013, 14,
540–558.
[53] D. T. Browne, J. Eisinger, N. J. Leonard, J. Am. Chem. Soc. 1968, 90,
7302–7323.
[21] D.-L. Ma, Z. Zhang, M. Wang, L. Lu, H.-J. Zhong, C.-H. Leung, Chem.
Biol. 2015, 22, 812–828.
[54] G. Wenska, S. Paszyc, Can. J. Chem. 1984, 62, 2006–2010.
[55] J. F. Constant, P. Laugaa, B. P. Roques, J. Lhomme, Biochemistry
1988, 27, 3997–4003.
[22] Y. J. Lu, Q. Deng, J. Q. Hou, D. P. Hu, Z. Y. Wang, K. Zhang, L. G.
Luyt, W. L. Wong, C. F. Chow, ACS Chem. Biol. 2016, 11, 1019–1029.
[23] J.-W. Yan, W.-J. Ye, S.-B. Chen, W.-B. Wu, J.-Q. Hou, T.-M. Ou, J.-H.
Tan, D. Li, L.-Q. Gu, Z.-S. Huang, Anal. Chem. 2012, 84, 6288–92.
[24] X. Chen, J. Wang, G. Jiang, G. Zu, M. Liu, L. Zhou, R. Pei, RSC Adv.
2016, 6, 70117–70123.
[56] G. Jones, W. R. Jackson, C. Y. Choi, W. R. Bergmark, J. Phys. Chem.
1985, 89, 294–300.
[57] X. Liu, J. M. Cole, K. S. Low, J. Phys. Chem. C 2013, 117, 14731–
14741.
[25] H. Ihmels, L. Thomas, Org. Biomol. Chem. 2013, 11, 480–487.
[26] C.-C. Chang, J.-Y. Wu, C.-W. Chien, W.-S. Wu, H. Liu, C.-C. Kang, L.-J.
Yu, T.-C. Chang, Anal. Chem. 2003, 75, 6177–6183.
[58] R. S. Moog, D. D. Kim, J. J. Oberle, S. G. Ostrowski, J. Phys. Chem. A
2004, 108, 9294–9301.
[59] C. Reichardt, Chem. Rev. 1994, 94, 2319–2358.
[60] L. F. M. de Azevedo, M. G. Trevisan, J. S. Garcia, A. M. S. Lucho, J.
Braz. Chem. Soc. 2014, 25, 469–477.
[27] X. Xie, B. Choi, E. Largy, R. Guillot, A. Granzhan, M.-P. Teulade-
Fichou, Chem. Eur. J. 2013, 19, 1214–1226.
[28] X. Xie, A. Renvoisé, A. Granzhan, M.-P. Teulade-Fichou, New J. Chem.
2015, 39, 5931–5935.
[61] D. Rehm, A. Weller, Isr. J. Chem. 1970, 8, 259–271.
[62] C. A. M. Seidel, A. Schulz, M. H. M. Sauer, J. Phys. Chem. 1996, 100,
5541–5553.
[29] H. Lai, H. Lai, Y. Xiao, S. Yan, F. Tian, C. Zhong, Y. Liu, X. Weng, X.
Zhou, Analyst 2014, 139, 1834–1838.
[63] E. Zahavy, M. A. Fox, J. Phys. Chem. B 1999, 103, 9321–9327.
[64] J. L. Banks, H. S. Beard, Y. Cao, A. E. Cho, W. Damm, R. Farid, A. K.
Felts, T. A. Halgren, D. T. Mainz, J. R. Maple, et al., J. Comput. Chem.
2005, 26, 1752–1780.
[30] J. Yan, Y. Tian, J. Tan, Z. Huang, Analyst 2015, 140, 7146–7149.
[31] M.-Q. Wang, W.-X. Zhu, Z.-Z. Song, S. Li, Y.-Z. Zhang, Bioorg. Med.
Chem. Lett. 2015, 25, 5672–5676.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
[65] V. Berl, I. Huc, R. G. Khoury, M. J. Krische, J. M. Lehn, Nature 2000,
407, 720–723.
[66] A. De Cian, E. DeLemos, J.-L. Mergny, M.-P. Teulade-Fichou, D.
Monchaud, J. Am. Chem. Soc. 2007, 129, 1856–1857.
[67] W. J. Chung, B. Heddi, F. Hamon, M.-P. Teulade-Fichou, A. T. Phan,
Angew. Chem. Int. Ed. 2014, 53, 999–1002.
[68] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104.
[69] M. R. Geraskina, A. S. Dutton, M. J. Juetten, S. A. Wood, A. H. Winter,
Angew. Chem. Int. Ed. 2017, 56, 9435–9439.
[70] A. Marchand, A. Granzhan, K. Iida, Y. Tsushima, Y. Ma, K. Nagasawa,
M. P. Teulade-Fichou, V. Gabelica, J. Am. Chem. Soc. 2015, 137, 750–
756.
[71] M. Bončina, Č. Podlipnik, I. Piantanida, J. Eilmes, M.-P. Teulade-
Fichou, G. Vesnaver, J. Lah, Nucleic Acids Res. 2015, 43, 10376–
10386.
[72] A. De Cian, L. Guittat, M. Kaiser, B. Saccà, S. Amrane, A. Bourdoncle,
P. Alberti, M.-P. Teulade-Fichou, L. Lacroix, J.-L. Mergny, Methods
2007, 42, 183–195.
[73] D. Renčiuk, J. Zhou, L. Beaurepaire, A. Guédin, A. Bourdoncle, J.-L.
Mergny, Methods 2012, 57, 122–128.
[74] A. De Rache, J.-L. Mergny, Biochimie 2015, 115, 194–202.
[75] F. H. Stootman, D. M. Fisher, A. Rodger, J. R. Aldrich-Wright, Analyst
2006, 131, 1145.
[76] A. Alekseev, J. Peltonen, V. Savransky, Thin Solid Films 1994, 247,
226–229.
[77] P. Verma, H. Pal, J. Phys. Chem. A 2012, 116, 4473–4484.
[78] E.-A. Raiber, R. Kranaster, E. Lam, M. Nikan, S. Balasubramanian,
Nucleic Acids Res. 2012, 40, 1499–1508.
[79] A. M. Brouwer, Pure Appl. Chem. 2011, 83, 2213–2228.
[80] A Guide to Recording Fluorescence Quantum Yields, HORIBA Jobin
Yvon, 2014.
[81] W. Humphrey, A. Dalke, K. Schulten, J. Mol. Graph. 1996, 14, 33–38.
[82] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A.
Petersson, et al., Gaussian 09, Revision D.01, Gaussian, Inc.,
Wallingford, CT, 2009.
[83] P. MacCarthy, Anal. Chem. 1978, 50, 2165.
This article is protected by copyright. All rights reserved.

10.1002/chem.201801701
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Xiao Xie, Oksana Reznichenko, Ludovic
Chaput, Pascal Martin, Marie-Paule
Teulade-Fichou, Anton Granzhan*
or
+
Page No. – Page No.
Topology-Selective Fluorescent
“Light-Up” Probes for G-Quadruplex
DNA Based on Photoinduced
Electron Transfer
Binding to G-quadruplex DNA reduces the photoinduced electron transfer in
some of fluorescent probes containing a fluorophore (coumarin or pyrene) linked to
an electron-deficient G4-DNA ligand (PDC). The resulting fluorescence “light-up”
effect can be used for selective detection of antiparallel G4 structures formed by
variants of a human telomeric sequence.
This article is protected by copyright. All rights reserved.