Multiple and cooperative binding of fluorescence light-up probe thioflavin T with human telomere DNA G-quadruplex.

TLDR: The binding mechanism of ThT with several variants of the human telomeric sequence in the presence of monovalent cations is investigated and results are crucial for the sensible design and interpretation of G-quadruplex detection assays using fluorescent ligands in general and ThT in particular.

Abstract: Thioflavin T (ThT), a typical probe for protein fibrils, also binds human telomeric G-quadruplexes with a fluorescent light-up signal change and high specificity against DNA duplexes. Cell penetration and low cytotoxicity of fibril probes having been widely established, modifying ThT and other fibril probes is an attractive means of generating new G-quadruplex ligands. Thus, elucidating the binding mechanism is important for the design of new drugs and fluorescent probes based on ThT. Here, we investigated the binding mechanism of ThT with several variants of the human telomeric sequence in the presence of monovalent cations. Fluorescence titrations and electrospray ionization mass spectrometry (ESI-MS) analyses demonstrated that each G-quadruplex unit cooperatively binds to several ThT molecules. ThT brightly fluoresces when a single ligand is bound to the G-quadruplex and is quenched as ligand binding stoichiometry increases. Both the light-up signal and the dissociation constants are exquisitely sensitive to the base sequence and to the G-quadruplex structure. These results are crucial for the sensible design and interpretation of G-quadruplex detection assays using fluorescent ligands in general, and ThT in particular.

Figures

Figure 3. Titration of 5 µM 22AG with ThT. (A) Fluorescence titration curve in 100 mM KCl (red) or NH4OAc (black). (B and C) ESI-MS spectra of 22AG with (B) 0 µM ThT and (C) 30 µM ThT, showing up to four ligands bound per 22AG molecule. (D) ESI-MS titration curve with equilibrium concentrations of each form deduced from the relative mass spectral peak areas. The lines correspond to fitting using sequential macroscopic binding events (KDn values are summarized in Table 2). Error bars in (A) and (D) represent standard error of triplicate measurements.

Figure 4. ESI-MS spectra of 5 µM dsDNA in the absence (A) and presence (B) of 30 µM ThT, and of 5 µM 45G in the absence (C) and presence (D) of 30 µM ThT. All experiments were carried out in 100 mM NH4OAc solution at 22°C (room temperature).

Figure 2. (A) Fluorescence intensity of ThT at 485 nm with various concentrations of 22AG (black circles), 21G (red circles), 22GT (blue circles), 46AG (red diamonds), 45G (black diamonds), ssDNA (black triangles), dsDNA (red triangles), or triplex DNA (green triangles) in 100 mM KCl buffer. (B) Fluorescence intensity at 485 nm of ThT with various concentrations of 22AG in buffers containing 100 mM KCl (black), CH3COONH4 (red), NaCl (blue), or LiCl (green). All measurements were carried out with 0.85 μM ThT at 25°C. Excitation wavelength is 450 nm.

Table 1. Dissociation constants (KD) at 25°C and Hill constants (n) of ThT-G-quadruplex binding. These parameters were evaluated from fluorescence titration (ex = 450 nm, em = 485 nm) of 0.85 µM ThT by various telomeric DNA sequences in buffers with different cations. All cation concentrations are 100 mM.

Table 2. Macroscopic, consecutive dissociation constants [KD,n (µM)] determined from ESI-MS in 100 mM NH4OAc solution at 22°C.

Figure 1. (A) Solution structure of human telomere DNA (4). (B) Chemical structure of a Gquartet. M+ indicates a monovalent cation coordinating guanine O6. (C) Chemical structure of ThT.

Full text

HAL Id: inserm-01388017
https://www.hal.inserm.fr/inserm-01388017
Submitted on 26 Oct 2016
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of sci-
entic research documents, whether they are pub-
lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diusion de documents
scientiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Multiple and cooperative binding of uorescence
light-up probe Thioavin T with human telomere DNA
G-quadruplex
Valérie Gabelica, Ryuichi Maeda, Takeshi Fujimoto, Hidenobu Yaku, Takashi
Murashima, Naoki Sugimoto, Daisuke Miyoshi
To cite this version:
Valérie Gabelica, Ryuichi Maeda, Takeshi Fujimoto, Hidenobu Yaku, Takashi Murashima, et al.. Mul-
tiple and cooperative binding of uorescence light-up probe Thioavin T with human telomere DNA G-
quadruplex. Biochemistry, American Chemical Society, 2013, 52, pp.5620 - 5628. �10.1021/bi4006072�.
�inserm-01388017�

This is an author postprint version of an article published in:
Biochemistry(2013)52,5620‐5628.http://dx.doi.org/10.1021/bi4006072
Diffusion on institutional repository has been authorized by the Editor on October 24, 2016.
Multiple and cooperative binding of fluorescence
light-up probe Thioflavin T with human telomere
DNA G-quadruplex
Valérie Gabelica
‡¶,
*, Ryuichi Maeda
#
, Takeshi Fujimoto
#
, Hidenobu Yaku
#,||,§
, Takashi Murashima
#,||
,
Naoki Sugimoto
#,||
, Daisuke Miyoshi
#,||,
*
‡
Physical Chemistry and Mass Spectrometry Laboratory, Department of Chemistry, University of
Liège, B-4000 Liège, Belgium,
#
Faculty of Frontiers of Innovative Research in Science and
Technology (FIRST), Konan University, 7-1-20 Minatojima-minamimachi, Chuo-ku, Kobe 650-0047,
Japan,
||
Frontier Institute for Biomolecular Engineering Research (FIBER), Konan University, 7-1-20
Minatojima-minamimachi, Chuo-ku, Kobe 650-0047, Japan,
§
Advanced Technology Research
Laboratories, Panasonic Corporation, 3-4 Hikaridai, Seika-cho, Soraku-gun, Kyoto 619-0237, Japan
¶
Present addresses: (1) Univ. Bordeaux, IECB - ARNA laboratory, F-33600 Pessac, France, (2)
INSERM, U869 - ARNA laboratory, F-33000 Bordeaux, France)

2
Abbreviations and Textual Footnotes:
†
This work is supported by Grants-in-Aid for Scientific Research from the Ministry of Education,
Culture, Sports, Science and Technology, Japan (MEXT) [24350089 and 24107525 to D.M.], the
Scientific Research on Innovative Areas ‘‘Nanomedicine Molecular Science’’ (No. 2306), the
‘‘Strategic Research Foundation at Private Universities’’ (2009-2014) (to N.S.), the Kurata Grants, the
Kurata Memorial Hitachi Science and Technology Foundation (to D.M.), the Fonds de la Recherche
Scientifique-FNRS [FRFC 2.4528.11 to V.G.], and the Japan Society for the Promotion of Science
[research fellowship to T.F.].
* To whom correspondence should be addressed. Tel: +81 78 303 1426; Fax: +81 78 303 1495; Email:
miyoshi@center.konan-u.ac.jp. Correspondence can also be addressed to: v.gabelica@iecb.u-bordeaux.fr
Abbreviations.
DNA, deoxyribonucleic acid; ThT, Thioflavin T; HPLC, high performance liquid
chromatography; ESI-MS, electrospray ionization mass spectrometry; CD, circular dichroism; G-rich
sequence, guanine-rich sequence.

3
ABSTRACT: Thioflavin T (ThT), a typical probe for protein fibrils, also binds human telomeric G-
quadruplexes with a fluorescent light-up signal change and high specificity against DNA duplexes. Cell
penetration and low cytotoxicity of fibril probes having been widely established, modifying ThT and
other fibril probes is an attractive means of generating new G-quadruplex ligands. Thus, elucidating the
binding mechanism is important for the design of new drugs and fluorescent probes based on ThT. Here,
we investigated the binding mechanism of ThT with several variants of the human telomeric sequence
in the presence of monovalent cations. Fluorescence titrations and electrospray ionization mass
spectrometry (ESI-MS) analyses demonstrated that each G-quadruplex unit cooperatively binds to
several ThT molecules. ThT brightly fluoresces when a single ligand is bound to the G-quadruplex and
is quenched as ligand binding stoichiometry increases. Both the light-up signal and the dissociation
constants are exquisitely sensitive to the base sequence and to the G-quadruplex structure. These results
are crucial for the sensible design and interpretation of G-quadruplex detection assays using fluorescent
ligands in general, and ThT in particular.

4
Some of the most frequently encountered repetitive DNA sequences throughout the genomes of most
organisms are guanine (G)-rich sequences and their complementary cytosine (C)-rich strands (1). G-rich
sequences can adopt a stable four-stranded structure known as the G-quadruplex (Figure 1A) (2-4). G-
quadruplexes can be formed with four Hoogsteen-paired coplanar guanines called a G-quartet (Figure
1B) (2). The best known G-rich (G-quadruplex forming) sequence is found at the ends of the
chromosomes, the telomeres (5,6). Moreover, bioinformatic studies have identified more than 300,000
G-rich sequences in the human genome (7,8). Interestingly, G-rich sequences are enriched in the
untranslated regions of proto-oncogenes, indicating that G-quadruplexes are critical participants in the
gene regulation system (9,10). However, the actual roles and functional mechanisms of G-quadruplexes
remain unclear.
G-quadruplex ligands developed to investigate the functional roles of G-quadruplexes should have
both high affinity for G-quadruplexes and also high specificity against DNA duplexes (11-14), because
the most abundant DNA structure in a living cell is the canonical duplex. Moreover, a bright light-up
signal upon G-quadruplex binding is ideal for the detection and imaging of G-quadruplexes (15). Thus,
various G-quadruplex probes, which can transduce the binding event into a detectable signal, have been
developed (16-21). However, the use of fluorescent probes for the detection of G-quadruplexes is still
insufficient due to difficulties in designing fluorescent ligands that combine high specificity for DNA
with significant changes in emission quantum yield upon binding. Moreover, such ligands should be
able to penetrate cells and have low toxicity for cellular and in vivo imaging.
Here, we report on Thioflavin T, or “ThT” (4-(3,6-dimethyl-1,3-benzothiazol-3-ium-2-yl)-N,N-
dimethylaniline) (Figure 1C), a new G-quadruplex ligand possessing very unique fluorescent properties.
Mohanty and coworkers recently reported that ThT, which is a typical probe for protein fibrils, binds to
the human telomeric G-quadruplex sequence 22AG (dAGGG(TTAGGG)
3
) with a fluorescent light-up
signal change and very low background, and almost no fluorescent response with DNA duplexes (22).
Various ThT-DNA G-quadruplex binding modes have been discussed, but the ligand binding

Frequently asked questions

Q1. What are the contributions in "Multiple and cooperative binding of fluorescence light-up probe thioflavin t with human telomere dna g-quadruplex" ?

Gabelica et al. this paper combined three independent probing techniques ( ESI-MS, fluorescence, and CD ) to decipher the Thioflavin T binding mechanism to human telomeric DNA G-quadruplexes. 

Q2. How many ligands can stack onto a single G-quartet?

Since ThT is smaller than daunomycin (which has five six-membered rings), it is possible that up to three ThT molecules stack onto a single G-quartet. 

Q3. What is the effect of ThT on the CD spectrum of 22AG?

It was reported previously that ThT induces changes in the CD spectra of 22AG in the presence of 50 mM KCl, lowering the peak at 270 nm, and thereby indicating a transition towards more “antiparallel” strand arrangements (22). 

Q4. What can be inferred from the macroscopic dissociation constants?

Cooperative binding of ThT. Positive cooperativity can be inferred by examining the values of the macroscopic dissociation constants and considering statistics. 

Q5. How many ligands can stack onto one G-quadruplex?

In the case of the isoquinoline alkaloid berberine and its derivatives, up to 6 ligand molecules bind to one G-quadruplex with some binding modes (52). 

Q6. What is the role of the 5'-A in the light-up fluorescence response?

The presence of the 5’-A, or of a full G-quadruplex unit in the neighborhood of ligand binding, could further restrict the torsional motions of the bond between the rings and be responsible for further enhancement of the fluorescence response with those sequences. 

Q7. At what concentration is the CD spectrum of 22AG affected?

In the presence of NH4OAc, the 1:1 complex is already present at 10 µM ThT, but at that ThT concentration the CD spectrum of 22AG is not affected. 

Q8. What is the effect of the stacking of aromatic amino acids on the fluorescence response?

the presence of a 5’-A, or of a full G-quadruplex unit and a TTA linker, increases not only the fluorescence response as discussed above but also the binding affinity of ThT. 

Q9. What was the titration of the oligonucleotide?

Before measurement, the samples were heated at 90°C for 10 min, gently cooled at 0.5°C min-1, and incubated at 25°C for 1 h.DNA oligonucleotides were titrated with 0.85 µM ThT (ThT by DNA fluorescence titration). 

Q10. What is the way to determine the ligand binding mode?

(3) Finally, spectroscopic techniques sensitive to the DNAconformation (e.g., CD spectroscopy for G-quadruplexes) also provide important complementary information on the ligand binding mode, and the conformation of the complexes can be probed only when their abundance is predominant over that of the free DNA. 

Q11. What is the affinity of ThT to a G-quadruplex?

The affinity is independent of the thermal stability of the G-quadruplex (K+ > Na+ > NH4+ >> Li+), and these results suggest that ThT preferentially binds to pre-formed (3+1) Gquadruplex folds. 

Q12. What is the effect of stacking of aromatic amino acids on ThT fluorescence?

In addition, the stacking of aromatic amino acids has also been shown to enhance ThT fluorescence, even though ThT may not be constricted to a planar conformation (48). 

Q13. What is the reason for the fluorescence enhancement?

This strongly suggests that preferential ThT binding to 22AG compared to duplex DNA is the major reason for the selectivity of the fluorescence detection. 

Q14. What is the likely binding site for ThT in human telomeric DNA?

A likely binding site for ThT molecules in human telomeric DNA is around the 5'-A (A1); the neighboring G-quartet plane of G1, G11, G15, and G21; and the neighboring loop composing T12, T13, and A14 (highlighted in Figure 1A). 

Q15. What is the effect of fluorescence on the ligand binding?

In the case of ThT binding to human telomeric G-quadruplexes, fluorescence showed how the response of ThT changes with the total number of ligands bound to the G-quadruplexes. 

Q16. What is the probability of a site being populated first?

If the sites are independent (no cooperativity) but non-equivalent, the most affine binding sites are more likely to be populated first, so the increase of macroscopic KD with n is even larger. 

Q17. How many ThT bindings can be seen on a single G-quadruplex?

ESI-MS titration experiments of ThT on 45G (which can be seen as one 22GT unit connected to a 22AG unit by a thymine linker) showed a larger number of ThT bindings, up to seven.