Structural and Dynamical Impact of a Universal Fluorescent

Subscriber access provided by READING UNIV

Article

Structural and Dynamical Impact of a Universal Fluorescent Nucleoside Analogue Based on 3-Hydroxychromone Inserted Into a DNA Duplex Loussiné Zargarian, Akli Ben Imeddourene, Krishna Gavvala, Nicolas P. F. Barthes, Benoît Y. Michel, Cyril Assongo Kenfack, Nelly Morellet, Brigitte René, Philippe Fossé, Alain Burger, Yves Mély, and Olivier Mauffret J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.7b08825 • Publication Date (Web): 27 Nov 2017 Downloaded from http://pubs.acs.org on November 30, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Structural and Dynamical Impact of a Universal Fluorescent Nucleoside Analogue Based on 3-Hydroxychromone Inserted Into a DNA Duplex Loussiné Zargarian1, Akli Ben Imeddourene1, Krishna Gavvala2, Nicolas P.F. Barthes3, Benoit Y.Michel3, Cyril A. Kenfack 4, Nelly Morellet5, Brigitte René1, Philippe Fossé1, Alain Burger3, Yves Mély2 & Olivier Mauffret1*. 1

LBPA, Ecole normale supérieure Paris-Saclay, UMR 8113 CNRS, Université Paris-Saclay, 61 Avenue du Pdt

Wilson 92235 Cachan cedex, France. 2

Laboratoire de Biophotonique et Pharmacologie, Faculté de Pharmacie, UMR 7213 CNRS, Université de

Strasbourg, 74 route du Rhin, 67401 Illkirch, France. 3

Institut de Chimie de Nice, UMR 7272, Université Côte d'Azur, CNRS, Parc Valrose, 06108 Nice Cedex 2,

France. 4

Laboratoire d'Optique et Applications, Centre de Physique Atomique Moléculaire et Optique Quantique,

Université de Douala, BP 85580, Douala, Cameroon. 5

Institut de Chimie des Substances Naturelles, CNRS UPR 2301, Université Paris-Sud, Université Paris Saclay,

1 Avenue de la Terrasse, 91190 Gif-sur-Yvette, France.

* To whom correspondence should be addressed. Email: [email protected]

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 30

ABSTRACT Recently, a 3-hydroxychromone based nucleoside 3HCnt has been developed as a highly environment-sensitive nucleoside surrogate to investigate protein-DNA interactions. When it is incorporated in DNA, the probe is up to 50-fold brighter than 2-aminopurine, the reference fluorescent nucleoside. Although the insertion of 3HCnt in DNA was previously shown to not alter the overall DNA structure, the possibility of the probe inducing local effects cannot be ruled out. Hence, a systematic structural and dynamic study is required to unveil the 3HCnt’s limitations and to properly interpret the data obtained with this universal probe. Here, we investigated by NMR a 12-mer duplex, in which a central adenine was replaced by 3HCnt. The chemical shifts variations and NOE contacts revealed that the 3HCnt is well inserted in the DNA double helix with extensive stacking interactions with the neighbor base pairs. These observations are in excellent agreement with the steady-state and time-resolved fluorescence properties indicating that the 3HCnt fluorophore is protected from the solvent and does not exhibit rotational motion. The 3HCnt insertion in DNA is accompanied by the extrusion of the opposite nucleobase from the double helix. Molecular dynamics simulations using NMR-restraints demonstrated that 3HCnt fluorophore exhibits only translational dynamics. Taken together, our data showed an excellent intercalation of 3HCnt in the DNA double helix, which is accompanied by localized perturbations. This confirms 3HCnt as a highly promising tool for nucleic acid labeling and sensing.


2

Page 3 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Environment sensitive fluorescent nucleoside analogues are of major interest for the development of fluorescent based-modified nucleic acid sensors and their applications in biology, biotechnology and diagnostics 1, 2. These labels are highly desirable since they allow a well-defined geometry and location within the nucleic acid as well as the possibility to locally monitor structural changes in nucleic acids. They also open new avenues for the construction of new materials

3-6

. According to their chemical structure, the fluorescent

nucleoside analogues can be classified in two major families: nucleobase mimics possessing base pairing properties and aromatic fluorophores lacking the H-bonding face. Events that affect the microenvironment of these fluorescent analogues can be monitored through changes in fluorescence intensity, fluorescence anisotropy, excited-state lifetime and emission color. The most popular probe 2-aminopurine (2AP), is an isomer of adenine used as an intensiometric probe 2, 7, 8. 2AP absorbs in the red edge of the nucleic acid/protein absorption domain and fluoresces with large Stokes shift but is strongly quenched in DNA due to dynamic contacts with the neighboring bases 9, 10. Additionally the sensitivity of the probe to changes in its environment is limited. Quenching or poor environment sensitivity are common drawbacks of fluorescent nucleobase mimics. These severe limitations have stimulated further researches. In this context, substituting natural bases by aromatic fluorophores lacking the Hbonding face is an attractive approach to make oligodeoxynucleotides (ODN) fluorescent with more red shifted absorptions, high quantum yields and large Stokes shifts, which facilitate selective excitation and easier detection. This type of labeling also offers the opportunity to develop advanced emissive nucleosides based on new mechanisms of response. By combining various fluorophores, Kool and collaborators have created a large panel of new emissive nucleosides covering a wide range of colors and useful as label for biological applications 1113

. We designed 3HCnt, a nucleoside with a 2-thienyl-3-hydroxychromone that was shown to

act as a universal substitute for natural bases (Schemes 1A&B) 14, 15. The 3HCnt fluorophore exhibits dual emission namely from the normal (N*) and tautomer (T*) as the result of an excited-state intramolecular proton transfer (ESIPT, Supplementary Scheme 1). Two-color dyes are very attractive because the emission intensities of the two bands are highly sensitive to polarity and hydration, thus providing facile and straightforward quantification for sensing through ratiometric detection (color change). When incorporated into DNA, 3HCnt is up to 50-fold brighter than 2AP. The overall conformation of the duplex appeared unaffected as evidenced by CD spectroscopy and only slight alterations have been quantified with thermal


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 30

denaturation methods. 3HCnt has been used to monitor the local conformational changes of a single-stranded oligonucleotide (ODN) upon binding of the HIV-1 nucleocapsid protein

14

.

The fluorescent nucleoside was also successfully used to monitor the flipping of a neighbor methylcytosine, as a result of the binding of ubiquitin-like containing PHD and RING finger domains 1 (UHRF1), a protein involved in the replication of the DNA methylation pattern 16. Finally, in a comparison study with commercially available fluorescent nucleobase analogues, 3HCnt probe demonstrated the highest sensitivity and gave the most detailed information about the conformational changes of DNA induced by Endonuclease VIII binding and processing

17

. Thus, 3-HCnt appears as a powerful tool for investigating protein-DNA

interactions. All our data are consistent with the notion that the hydrophobic 3HCnt chromophore intercalates into DNA as a result of its hydrophobic character and large area of contact with the surrounding bases (Scheme 1B). Nevertheless, in the absence of structural information in DNA, predicting its precise positioning is challenging since 3HCnt cannot form Watson-Crick base pairs. Several obvious questions can be raised. What is the impact of 3HCnt on the local structure and dynamics of DNA? In duplexes, when a natural base pair is opposite to 3HCnt (Scheme 1C), does it adopt a twist or zipper mode, 3 or does 3HCnt push the opposite natural base in an extra-helical position (e.g. flipped mode)?18 Getting a precise picture by NMR of the structural and dynamical impacts of 3HCnt insertion inside a duplex is of major importance to interpret the data resulting from the interaction of proteins with 3HCnt-labeled DNAs. This kind of studies is also of major value to propose improvements of the molecule's properties. In the present study, we have investigated the structural and dynamic properties of a dodecamer duplex labeled in a central position by the 3HCnt fluorescent nucleoside substituting an adenine residue (Scheme 1). The obtained structure was compared to that of the non-labeled duplex.


4

Page 5 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Scheme 1. Working hypotheses (A) Chemical structure of 3HCnt, the fluorescent nucleoside analogue bearing the 2-thienyl-3-hydroxychrome fluorophore. Absorption and dual emission features and proton used for NMR assignments of 3NCnt are indicated. (B) Comparison of the 3HCnt dye with the DNA base pairs. (C) Possible different modes of 3HCnt intercalation. (D) Sequences of the unmodified (ODN1a) and modified (ODN1b) duplexes studied. S7 indicates the 3HCnt fluorescent nucleotide.

Experimental Section Synthesis Unmodified single-stranded oligodeoxynucleotides (ODNs) (5'-CCGCTTAAACGC-3’ and 5'-GCGTTTAAGCGG-3’) were synthesized and HPLC-purified by Microsynth AG (Switzerland). The labeled oligodeoxynucleotide 5'-CCGCTTSAACGC-3’(S7= 3HCnt) was produced as described previously 14.

Absorption, denaturation, fluorescence and CD studies Absorption spectroscopy Extinction coefficients at 260 nm were used to determine the concentration of single-stranded sequences.

Extinction

5'-CCGCTTAAACGC-3’

coefficients and

for

the

non-labeled

5'-GCGTTTAAGCGG-3’

were

sequences

ODNs

108,400 M–1cm–1

and

116,000 M–1cm–1, respectively. The extinction coefficient for the labeled single strand ODN 5'-CCGCTTSAACGC-3’ was 106,400 M–1cm–1. This value was corrected for the 10,000 M-1cm–1 extinction coefficient of S at 260 nm. The modified and unmodified ODNs


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 30

sequences were annealed with their complementary ODN sequence to form a duplex at a concentration of 2 µM in 10 mM cacodylate, pH 7.0, 100 mM NaCl, 1 mM EDTA. Absorption spectra were recorded on a Cary 400 spectrophotometer (Varian). The pathlength of the cell was 1 cm.

Circular dichroism measurements Circular dichroism (CD) spectra were recorded on a 810 CD spectrophotometer (Jasco) at 20°C at a duplex concentration of 2 µM solution in 10 mM cacodylate pH 7.0, 100 mM NaCl, 1 mM EDTA). The pathlength of the cell was 1 cm.

Denaturation studies Melting curves were recorded by following the temperature-dependence of the absorbance changes at 260 nm of the sample (2 µM concentration of each strand) in 10 mM cacodylate pH 7.0, 100 mM NaCl, 1 mM EDTA). Absorption spectra were recorded in a Peltier thermostated cell holder on a Cary 4 spectrophotometer (Varian) with a pathlength of cell of 1 cm. The temperature range for denaturation measurement was 5– 80 °C. Speed of heating was 0.3 °C/min. The melting curves were converted into a plot of α versus temperature, where α represents the fraction of single-strands in the duplex state. The melting temperatures were extracted by fitting the first derivative of these curves by a Gaussian.

Steady-state Fluorescence spectroscopy Fluorescence spectra were recorded at 20°C on a FluoroMax 4.0 (Jobin Yvon, Horiba). All the samples were prepared in 10 mM cacodylate pH 7.0, 100 mM NaCl, 1 mM EDTA). The ODN concentration was adjusted to an absorbance of 0.05 at the absorption maximum of 3HCnt. Spectra were corrected for the buffer background signal, lamp fluctuations, and detector spectral sensitivity. The quantum yield (QY) of the labeled duplexes was determined by using quinine sulfate (QY=0.546 in 0.5 M H2SO4) 19 as a reference. Time-resolved fluorescence measurements Time-resolved fluorescence and anisotropy measurements were performed with the timecorrelated single-photon counting technique (TCSPC)20. Excitation at 315 nm was provided by a pulse-picked frequency tripled Ti-sapphire laser (Tsunami, Spectra Physics), pumped by a Millenia X laser (Spectra Physics). Emission was collected through a polarizer set at magic angle and an 8 nm band-pass monochromator (Jobin-Yvon H10) set at emission maximum.


6

Page 7 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Single photon events were detected with a microchannel plate Hamamatsu R3809U photomultiplier coupled to a Philips 6954 pulse preamplifier and recorded on a multichannel analyzer (Ortec 7100). The instrumental response function was recorded with a polished aluminum reflector, and its full width at half-maximum was 40 ps. Fluorescence intensity decays [I(t)] were analyzed as a sum of exponentials: I(t) = Σαiexp(-t/τi), where τi values are the fluorescence lifetimes and αi values are the associated amplitudes such that Σαi= 1. The mean lifetime was calculated according to the relationship 〈 τ〈 = Σαiτi. For time-resolved fluorescence anisotropy decay measurements, the polarized fluorescence decays for the parallel [I(t)] and perpendicular [I⊥(t)] emission polarizations were first collected at the emission maxima of the probe. The anisotropy decay function r(t) was constructed from these I(t) and I⊥(t) decays using the following equation,

where G is the correction factor for the sensitivity of detector to the polarized emission detection. The anisotropy decay fitting was done using the following functional form,

where r0 is the fundamental anisotropy that describes the inherent depolarization of the fluorophore and air is the pre-exponential term attributed to the fraction of the ith rotational relaxation time, i.e., τir.

NMR experiments The modified and unmodified ODNs (Scheme 1) were annealed with the complementary ODN to form a duplex at a concentration of 1.5 mM in 500 µL (5 mm tube) for the unmodified and 0.8 mM in 180 µL (3 mm tube) for the modified oligonucleotide. The buffer was composed of phosphate sodium (for a total ionic strength of 0.1 M), pH 6.5 and 0.2 mM EDTA. The samples were exchanged and lyophilized with D2O three times for the observation of non-exchangeable protons. For the observation of exchangeable protons, the samples were dissolved in 500 µL (unmodified ODN) or 180 µL (modified ODN) 9:1 H2O/D2O mixture. NOESY, COSY-DQF, TOCSY and 1H-13C HSQC experiments in D2O for the unmodified sample were collected at 20 and 25 °C with a Bruker Avance 500 MHz


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 30

spectrometer. The same experiments with the modified sample were collected on an Avance 950 MHz spectrometer equipped with a cryoprobe unit (Gif sur Yvette, Regional NMR Center) at 5, 10 and 20 °C. NOESY mixing times of 100, 200 and 300 ms were collected. The NMR spectra for the exchangeable protons were recorded from 5 to 70 °C with 3-5°C increments. The NOESY experiments in H2O/D2O mixture were collected at 10 °C and 20 °C and mixing times of 100, 200 and300 ms. Water suppression was achieved with a gradient Watergate pulse sequence 21. Chemical shifts were referenced to an internal standard of TSP (Trimethylsilylpropanoic acid).

Data processing and restraints The data were processed using Topspin (Bruker Biospin) and NMR-Pipe 22. The spectra were analyzed using the Sparky software (Goddard, R.D & Kneller, University of California, San Francisco). The NOESY data at several mixing times were used to assign "qualitative" and conservative distance restraints between two protons using the following categories: "strong" (2.5-4.5 Å), "medium" (3-4.5 Å), "weak" (4.5-6.5 Å), and "exclusive" (6-12/15 Å) restraints. Only interresidues distances were taken into account and only the ten residues located in the center of the oligonucleotide were considered for restraints, i.e. residues 5-9 and 16-21. The restraints have been applied as distance restraints under the form of a well with a square bottom with parabolic sides out to a defined distance (option iald=0 in sander AMBER9). A value of 20 kcal/mol/Å2 was used as the force constant.

Restrained Molecular dynamics simulations MD simulations were performed with the "Parmbsc0 εζ OLI" force field 9 and 14 programs

23

using the AMBER

. Parmbsc0 εζ OLI simulations were performed at constant temperature

24

(300K) and pressure (1bar) using the Berendsen algorithm 25. The integration time-step was 2 fs and covalent bonds involving hydrogen were constrained using SHAKE

26

. Long-range

electrostatic interactions were treated using the particle mesh Ewald (PME) approach 27. Nonbonded interactions were treated with a 9 Å direct space cut-off. The center-of-mass motion was removed every 10 ps. The dodecamer was built in standard B-DNA conformation using Nucleic Acid Builder 28 and the seventh residue in strand 1 was substituted with the fluorescent nucleotide 3HCnt using UCSF Chimera

29

. The probe has been constructed using TRIPOS software. Quantum

mechanical calculations were performed using the Gaussian software package. Geometries were optimized using the Gaussian default optimization criteria. The 3HF/6-31G* level of


8

Page 9 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


theory was employed. Topology files of the aromatic part of the fluorophore and the glycosidic linkage were generated using antechamber. Nevertheless, originally Parmbsc0 εζ OLI

force-field parameters were used for the deoxyribose fragment of the S7 residue and its 3'

and 5' phosphodiester backbone bonds. The modified dodecamer was neutralized using 22 Na+ ions in explicit TIP3P water molecules 30. The primary boxes of water were truncated octahedrons with solvent extending 10 Å around the DNA. The water molecules and counterions were energy minimized and equilibrated at 100 K around the constrained DNA residues (and dye) for 100 ps in the NVT (mole number and volume fixed, constant temperature; canonic ensemble). The system was then heated from 100 to 300 K in 10 ps by 5K increments with harmonic positional restraints of 25.0 kcal/mol/Å2 on the 24 residues. The molecular dynamics simulations were continued in NPT (isobar, isothermal) conditions, without notable change in volume. The positional restraints were gradually removed over 250 ps and followed by the production phase. During this last phase, the NMR restraints are introduced using harmonic restraints and force constants of 20 kcal/mol/Å2. These latter restraints are applied exclusively on the residues 5-9 and 16-21 and only interresidues distances, in limited number, have been restrained. No others restraints were imposed, notably no hydrogen bonds for any base pair was imposed to prevent the base pair opening. The production phase is 20 ns long and MD snapshots were saved every 1ps. Control simulations were computed in the same conditions as the previous ones but omitting the NMR restraints. The resulting structure obtained after 10 ns of production was then used as an input structure for a new run of calculation using the NMR restraints. The interest of this procedure is to use another starting point for the simulations and to check the convergence of the calculations. As described in the text, when the NMR restraints are not used, the dye is totally looped out in solution and the T18 residue stacked inside the double-helix after 10 ns of production in such conditions. The molecular dynamics were analyzed using the CPPTRAJ 31 module of AMBER that allow determination of the RMSD (Root Mean Square Deviations), RMSF (Root Mean Square Fluctuations, see below) as well as calculations of averaged structures. When the RMSD were calculated for averaged structures; the averaged structures were obtained from stable 1000 ps portions of the trajectory taken from the indicated regions in Supplementary Figure S6. In addition, the averaged structures were energy-minimized using positional restraints. The best energy structures were extracted from the dynamics output files using a combination of Perl


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 30

and Python (www.python.org) scripts. Most of the data were extracted and plotted using Python and R scripts (R foundation). The Root Mean Square Fluctuations (RMSF) permit the measure of the atomic average mobility during molecular dynamics simulations. The atomic fluctuations were computed with the program CPPTRAJ from AmberTools 14. In this study, the average mass-weighted fluctuation of heavy atoms is calculated for each residue. In a first step, each frame of the trajectory was superimposed with the first frame via a structural alignment of heavy atoms. We calculated the RMSF for the full nucleotide, its backbone and its base according to the following

equation,

where

protons

were

excluded

from

the

calculations;

Molecules were examined and plotted for figures using Pymol.

Results CD and Tm First, we checked the effect of 3HCnt on the stability and secondary structure of the duplex by monitoring the temperature-induced absorbance changes at 260 nm and recording the CD spectra of labeled and non-labeled duplexes. Thermal denaturation studies showed that the substitution of A by 3HCnt at position 7 weakly impacted the duplex stability (Tm = 48.9 compared to 51.3°C). The CD spectra of the labeled and unlabeled duplexes confirmed the general B-helix conformation of the duplex (Supplementary Figure S1).

UV-visible and fluorescence characteristics The UV absorbance and fluorescence properties of 3HCnt in the duplex were characterized (Supplementary Table S1). In buffer at pH 7.0, the labeled ODN shows an absorption band centered at 382 nm and two well-resolved emission bands centered at 437 and 541 nm, respectively (Supplementary Figure S2). The quantum yield is high (40%). The short and long-wavelength emission bands originate from the emission of the excited states of the normal (N*) and tautomer (T*) forms, respectively. The emission spectrum of the labeled


10

Page 11 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


duplex with a largely dominant T* band was similar to the spectra of the free probe in poorly polar solvents

14

. Moreover, its high quantum yield (40%) is in line with an intercalation

between the flanking base pairs, which favors the flat and more emissive conformation of the 3HC fluorophore. Taken together, our data are very close to those previously reported for 3HCnt-labeled duplexes with T or A flanking nucleobases and suggest that the 3HCnt fluorophore is in a poorly hydrated environment likely being intercalated into DNA 14.

NMR Assignments of the non-exchangeable DNA protons The evaluation of the structural and dynamic impacts of the probe insertion in a DNA doublehelix needs the precise comparison of the chemical shifts between the labeled duplex (ODN1b) and the unmodified duplex (ODN1a) (see Scheme 1). The unmodified duplex displayed a well-resolved spectrum that could be easily assigned even at a moderate field (500 MHz) (Supplementary Figure S3). The modified duplex yielded also a well-resolved spectrum but was complicated by the presence of minor forms for several residues. Specific line broadening that were temperature dependent were observed for these residues, strongly suggesting conformational exchange phenomena. At 20°C, the presence of the minor forms was minimal and most of the analysis and extraction of data have been performed at this temperature. The increased complexity and the lower concentration of the modified duplex led us to record spectra at a very high field (950 MHz) for optimal signal to noise ratio and to bypass the difficulties associated to spectral overlap. The sequential assignment was performed using standard protocols with 2D 1H-1H NMR supplemented by 1H-13C experiments. The assignments and chemical shift measurements were performed at 20°C. The H6/H8-H2'/H2"/CH3 region with the assignments are shown in Figure 1 for the two strands of the modified duplex. The fluorescent nucleoside insertion leads to a break in the sequential nOes that are mainly associated with the H10/H11 3HCnt (S7) protons, which appear extremely broadened and barely detectable. In contrast, it is possible to observe the H2'/H2" S7 to H8A8, but not the H1' S7-H8A8 nOe suggesting a local deviation from the standard B-DNA conformation or effects associated to the resonance broadenings (Figure 1A). In the complementary strand, a break in the H2'/H2"-H6/H8 internucleotide sequential nOes occurs at the T17-T18 and T18-A19 junctions while an H1'T18-H8A19 nOe is observed (Supplementary Figure S4). These data strongly suggest a partial extrusion of the T18 nucleobase and its associated deoxyribose sugar (Scheme 2). The chemical shifts associated to this particular conformation are commented in the following section.


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 30

Figure 1: Expanded plot of a 1H 950 MHz NOESY spectrum in D2O buffer for the labeled duplex showing the sequential nOes from the aromatic H6/H8 to H2'/H2"/CH3 protons. The spectrum was recorded at 20°C with a 300 ms mixing time. (A) Connectivities are plotted for the strand bearing the S7 (=3HCnt) residue. The intra-nucleotide cross-peaks are indicated by the name of the residue. Several cross-peaks involving resonances of the S7 residue are indicated by the symbols S7#. The details are: S7#1: H8A8-H2"S7; S7#2: H8A8-H2'S7; S7#3: CH3T17-H5S7; S7#4: CH3T17-H7S7; S7#5:H2'T17-H7S7; S7#6: H2'/H2"S7-H10S7; S7#7: CH3T17-H6S7; S7#8: H2'T17-H6S7; S7#9: CH3T18-H6S7; S7#10: H2"T17-H6S7. (B) Connectivities for the complementary strand. The two cross-peaks labeled T18 with a star are the H1'-H2-/H2" T18 intraresidues cross-peaks


12

Page 13 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Scheme 2. Representation of the S7 and opposite T18 nucleotide, and their neighboring nucleotides. The front and the back show the major and minor grooves, respectively. The most important nOe contacts are indicated in red arrows. Unobserved correlations described in the text are indicated in dashed blue arrows. Non-exchangeable chemical shift perturbations The differences in chemical shifts between the unmodified and modified duplexes are shown (for each strand) for the aromatic protons (Figure 2A, C) and the sugar protons (Figure 2B, D), respectively. These data show that the main perturbations are localized in the immediate vicinity of S7 including the two base pairs located in 5' and 3' from the insertion site, no effect being detected in the more distant base pairs. The very large positive shifts observed for the base (0.55 ppm) and sugar protons (1.13 ppm) of T18 indicates a decreased shielding suggesting an extrusion of the base and the sugar of this residue from the double helix in agreement with the nOe pattern observed and commented above. The high-field negative shifts observed for the T17 aromatic (0.35 ppm) and sugar protons (0.4 ppm) suggest that this residue undergoes stacking interactions different from those occurring in standard B-DNA, probably as a consequence of its stacking with the 3HCnt ring. It also appears that the T6 deoxyribose protons, but not the aromatic ones, display anomalously positive low field shifts (0.44 ppm). Note that globally the effects are larger for the residues at the 3' side of S7, as compared to the 5' side. The observed chemical shifts provide information on the conformation of the modified duplex and are in favor of an insertion of the probe between the neighboring base pairs while the T18 residue is excluded from the double helix.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 30

Figure 2: Chemical shift differences at 20°C between the modified duplex and the unmodified duplex (∆δ=δmodified-δunmodified). (A) Aromatic protons of the modified strand; (B) sugar protons of the modified strand; (C) aromatic protons of the complementary strand; (D) sugar protons of the complementary strand. Positive values are indicative of decreased shielding and negative values related to increased shielding. NMR assignments of exchangeable DNA protons The imino and amino protons of the modified and non-modified duplexes have been assigned by recording NMR spectra at 10 °C. The NOESY spectra permitted the observation of the main imino-imino and imino to amino cross-peaks in the far downfield region of the 1H spectrum. The 1d imino proton of the unmodified (bottom) and modified (top) duplexes are shown in Figure 3A and the chemical shift differences of the imino protons between the modified and unmodified duplexes at 10 °C are shown in Figure 3B. Two imino protons (T17 and T18) are missing in the spectrum of the modified duplex while two others T6 (0.27 ppm) and T16 (0.4 ppm) are upfield-shifted relative to the spectrum of the unmodified duplex. In contrast, the T5 imino proton and the five G imino protons do not display any significant change indicating that the corresponding base pairs are weakly or not affected by S7. Globally, the observed upfield-shifts as well as the different shifts from non-exchangeable protons do support the intercalation of the probe inside the duplex. However, the fact that two imino protons (T17H3 and T18H3) are missing do not exclude the possibility of an intercalation accompanied with a significant disruption of the base pair A8-17 close to the intercalation site. The disappearance of the T18 proton is expected since this residue is opposite to the S7 nucleotide that cannot pair with a natural nucleobase. The disappearance of the imino proton of the A8-T17 base pair indicates that S7 strongly modifies the structure or dynamics of this


14

Page 15 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


base pair. Additionally, the 3' neighbor A9-T16 base pair has its imino proton strongly broadened indicating that its dynamics is likely impacted. The imino proton chemical shift differences (Figure 3B) confirm that the perturbations are localized in the vicinity of the fluorescent nucleobase, while more distant nucleotides are minimally affected. Interestingly, the effect of S7 on the neighbor base pairs is clearly asymmetrical leading to a larger perturbation at the 3' side as compared to the 5' side, in agreement with the results from the analysis of the non-exchangeable protons described above. The T5 and T6 imino protons are only slightly modified in terms of line broadening (despite that T6 is adjacent to the S7 probe). The T6 ribose chemical shifts are however modified suggesting a close proximity and a direct stacking between the deoxyribose of T6 and the aromatic part of 3HCnt.

Figure 3: Imino protons spectra and their associated chemical shifts (A) Downfield region of the 1H NMR spectrum of the modified (top) and unmodified (bottom) duplex showing the imino proton resonances at 10°C and 500 MHz. Assignments were performed using 2D experiments and are indicated for each resonance by the name of the residue bearing the imino proton. (B) Chemical shift differences at 10°C of the imino protons between the modified duplex and the unmodified duplex (∆δ=δmodified-δunmodified). The stars indicate the imino protons that are not observed. 2-Thienyl-3-hydroxychromone proton assignments It is essential to characterize the different protons of the S7 nucleoside to determine precisely the orientation of the probe inside the double helix. However, most of these protons are


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 30

strongly broadened, probably due to the probe dynamics that occur at various timescales (discussed below). This hindered the assignment task. However, the spin system constituted by the H5, H6, H7 and H8 protons in the S7 3HCnt group (see Scheme 1) was detected and assigned thanks to the observation of new peaks, relative to the unmodified duplex, in the NOESY, TOSCY and COSY spectra. The obtained assignments were in agreement with the downfield shift of H5 proton and the stronger coupling between H6-H5 and H8-H7 protons as compared to H6-H7; these latter features being already observed for chromones

32

. The line

broadening effects led to difficulties and ambiguities in the assignment task; with an ambiguity in the respective assignments of H8 and H7 protons of 3HCnt. Calculations were therefore performed with two different hypotheses: (i) with H7 (6.7 ppm) and H8 (6.33 ppm) and (ii) the reverse assignment. Calculations conducted with the hypothesis (i) give much more satisfying low-energy structures. Hypothesis (i) was also in better agreement with the known chemical shifts for the 3HCnt

32

. Assignments of the 2-thienyl-3-hydroxychromone

group allowed identification of the critical nOes contacts between its protons and those of DNA (Scheme 2). Several patterns of nOes gave important information on the duplex structure. There are nOes between S7-H5 and T17-CH3 (peak S7 #3 in Figure 1A) while no nOe is observed between S7-H5 and the H2' and H2" protons of T17. In contrast, the close S7-H6 proton displays nOe contacts both to the CH3 and H2'/H2" protons of T17 (S7 #7, S7 #8, S7 #10). Interestingly, nOe contacts are observed in the minor groove of DNA between S7-H8 and A8-H2, and S7-H8 and A19-H2.

Observation of conformational exchange effects Several resonances of the modified duplex are considerably broadened relative to other ones indicating that specific residues of the duplex are affected by conformational exchange. As shown in Figure 4, it is quite clear that A19 and A8 H8 resonances (8.2-8.3 ppm) at 10 °C are broader than the resonances of their immediate neighbors in the sequence, i.e. A9 (8.05 ppm) and A20 (7.93 ppm) H8. In addition, A19 and A8 H8 major resonances displayed exchange cross-peaks with minor forms both in NOESY and TOCSY spectra. The positions of these minor forms are indicated in Figure 4 for A19, A8 and T17 aromatic resonances. The minor conformers display maximal intensities at 10 °C. The affected residues are T17, A8 and A19, these residues are adjacent to the probe insertion site. In contrast, T18 H6 resonance is not significantly broadened.


16

Page 17 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 4: 1D 1H spectrum of the modified duplex recorded at 950 MHz and 10 °C showing the aromatic region. The main peaks of the region are labeled with the name of the residue for H6 or H8 protons. Additionally, H2 and H1' protons are indicated with the proton and residue names. The T17 H6 as well as the A8 and A19 H8 resonances are characterized by the presence of two forms in conformational exchange at 10 °C, the major peak is indicated by the name of the residue (A8, A19, T17) and the minor peak is indicated by the name of the residue and a star. Because all the protons of the probe are significantly broadened at a similar extent (see the S7 H6 signal in Figure 4), S7 probably triggers significant dynamics in the duplex core with a timescale in the milli-microsecond range. Indeed, the motions in these timescales are known to affect resonance linewidths

33-35

and complicate the assignment work as well as the

structure determination.

Structure determination and analysis of the dynamics To determine the structure of the modified duplex, we performed restrained molecular dynamics simulations with AMBER using a recently updated nucleic acid force field and an explicit representation of the solvent and the counter-ions. Dynamics calculations have been performed using 53 distance-type restraints corresponding exclusively to inter-residue restraints for residues 5-9 and 16-20, including restraints between the fluorescent probe and neighbor DNA residues. However, no restraints were imposed on the hydrogen bonds and the phosphodiester backbone angles. The restraints deduced from the nOes effects are of qualitative type (strong, medium, weak). Our strategy was to impose no restraints on residues distant by more than two base pairs from S7, because the chemical shifts and nOes patterns associated with these residues were similar to those of the unmodified


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 30

duplex. This strategy was made possible thanks to the progresses in the quality of the force fields used for nucleic acids

36-38

. However, the observed complex dynamics, with probable

equilibrium between several conformers, prevented us to obtain very high-resolution structures as it is the case in several recent works dealing with modified DNA 39-41 and led us to rather use restrained molecular dynamics to propose models reflecting the experimental data. The low number of restraints used as well as their qualitative features allowed us to observe true dynamics on several residues during the 20 ns production phase. The molecular dynamics were performed with and without the experimental restraints. Calculations performed with the experimental restraints and with two totally different starting structures (see Materials and Methods), provided structures in which the 3-hydroxychromone was intercalated inside the double helix and well-stacked with the neighboring base pairs. This insertion led to the complete extrusion of T18 from the double helix. The dynamics and conformational variability during the 20 ns of the restrained dynamics have been analyzed using the RMSDs, as described in the Materials and Methods section. The RMSDs data indicate that the 10 inner residues (9 nucleotides and the probe, Supplementary Figure S5C) are characterized by dynamic fluctuations of larger amplitude as compared to the outer residues (Supplementary Figure S5B). To characterize the residues that are the most involved in these fluctuations, we analyzed the RMSD at the base pair level. Interestingly, the S7-T18 pair exhibits the largest amplitude variations (Supplementary Figure S5G). Moreover, the A8-T17 base pair located downstream of the S7-T18 pair, is more sensitive to the probe fluctuations (See Supplementary Figure S5H (A8-T17)) than the T6-A19 base pair located downstream (Supplementary Figure S5F (T6-A19)), consistent with the imino proton data (see above). Analysis at the single residue level indicates that the S7 residue displays very large fluctuations with transitions between different conformations that are occupied during several nanoseconds (Figure 5C). Only the 3' residues (A8 and A9) display large fluctuations coupled with those of S7, while no significant fluctuations are observed for the 5' residues (T5 and T6).


18

Page 19 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5: Local rmsd (Å) calculated for different subsets of base pairs and single residues indicated under each graph during the 20 ns of the production phase performed using the NMR restraints. The RMSD is calculated for: residues 2-11 and 14-23 (i.e. the non-terminal residues) (A), the S7-T18 base pair (B) and different residues taken individually (C to I). The reference molecule used for all the calculations is the molecule 2000 from the production phase. The results obtained with this latter snapshot display the best visualization of the differences existing between the various phases of the dynamics. To get further information on the different conformational states and their dynamics, we identified the ten structures of lowest total energy. Supplementary Figure S6 displays the local RMSD values for the S7 residue, and the positions of the ten lower energy conformers are indicated along the dynamics run. The RMSD differences between these ten different structures, computed for the ten central base pairs of the dodecamer, are presented in Supplementary Table S2 (the RMSD values have been calculated for all atoms and the 10 central residues). Four representative structures characterized by large RMSD variations at the level of the S7 residue are displayed in Supplementary Figure S7. Interestingly, the four conformers differ mainly by the translation of the thiophenyl and hydroxychromone moieties in a plane approximately perpendicular to the helix axis (Supplementary Figure S7, B and C). Therefore, the hydroxychromone dynamics appears to be confined in a direction perpendicular to the helix axis. A representative view of the central part of the duplex is shown in Figure 6A. The structure is an averaged and minimized structure taken from the region of minimal energy, the point 7605 in Supplementary Figure S6. A strong stacking of S7 and a complete extrusion of T18 residue characterize the structure. The stacking of the S7 probe involves the surrounding A8-T17 and


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 30

T6-A19 base pairs (Figure 6, B and C). It is interesting to note that the 2-thienyl part strongly overlaps with the A8 residue located at the 3'-side of the probe, while a weak stacking occurs between T6 and this part of the probe.

Figure 6: Different views of the central part of an average structure for 1000 ps taken from the minimal energy region of the modified duplex (7605, see Supplementary Figure S6). (A) Only the six central residues including the S7 probe of the modified duplex are shown. Atoms are shown with a classical color code. (B) Expanded view of the stacking pattern of the A8T17 base pair (in black) with S7 (red) and T18 (green, blue, red) residues for the same structure. (C) Same as (B) but with T6-A19 base pair (grey). Investigation of S7 dynamics by time-resolved fluorescence spectroscopy To independently assess the dynamics of S7 in the duplex, we also investigated its excited state decay by picosecond time-resolved fluorescence measurements (Figure 7). Fluorescence transients of this environment-sensitive probe collected at the emission maximum of its T* band exhibit a ~95 ps lifetime growth component followed by a longer decay lifetime of 7.8 ns (Figure 7A). The fast growth component is typical for 3HCnt derivatives and can be attributed to the ESIPT process. Fluorescence decays were also collected over the whole range of its emission spectrum to determine whether there exists single or multiple emissive species. Interestingly, we found a single exponential decay with a long-lived lifetime of ~7.8 ns all over the T* band. This lifetime component is much higher than that observed for the free nucleoside in any tested solvent

42

, likely as the result of the restriction of the relative


20

Page 21 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


motions of the aromatic moieties that favors the highly emissive flat conformation of the probe

43, 44

. This hypothesis is fully consistent with the NMR data showing that the flat

conformation of the probe is favored in all conformations, as a consequence of its stacking with the adjacent base pairs (Figure 6).

Figure 7: Time-resolved fluorescence spectroscopy measurements. Fluorescence intensity (A) and anisotropy (B) decays of the modified duplex collected at T* band. Excitation wavelength is 315 nm. Next, we collected the time-resolved fluorescence anisotropy decay of the modified duplex to monitor the rotational dynamics of S7. As shown in Figure 7B, the anisotropy decay exhibits a mono exponential decay with a rotational correlation time value of 4.7 ns. This single correlation time component indicates that only the overall tumbling of the duplex can be monitored by the probe. Besides, this rotational correlation time is in good agreement with the theoretical value of the rotational time (6.4 ns) calculated by assuming the 12 base pairs DNA as a cylindrical rod of 4 nm in length and 2 nm in width 45. Moreover, as no local motion has been detected, this result confirms that the fluorophore is well stacked between the base pairs.

Discussion The fluorescent 3HCnt based-nucleoside used in this study is a highly prospective probe as it possesses an exceptional sensitivity to its environment and is up to 50-fold brighter than 2aminopurine, the standard fluorescent nucleoside analogue. These favorable properties led us to use it for investigating protein-nucleic acid interactions

14, 16, 46

. However, proper

interpretation of these data requires a careful analysis of the impact of the probe on the structure and dynamics of the labeled oligonucleotides. Globally our NMR and fluorescence data demonstrate a deep insertion of the 2-thienyl-3Hhydroxychromone moiety inside the double helix where it is stabilized by stacking with the surrounding base pairs (Figure 6). This conclusion is fully supported by the steady-state and time-resolved fluorescence data, showing that the restricted environment provided by these


21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 30

stacking interactions favor the highly emissive flat conformation of the probe. Interestingly, the two adjacent base pairs (T6-A19 and A8-T17) appear to be correctly paired in the restrained molecular dynamics calculations, although the T17 imino proton is not observable and that of T16 is significantly broadened in the NMR spectra. In contrast, T6 the immediate neighbor of the probe is only slightly affected, except for its chemical shift. The strongest consequence of S7 intercalation in the duplex is the ejection of the T18 nucleobase from the double helix. The duplex remains in the B-type and presents rather weak distortions, as illustrated by the chemical shifts of exchangeable and non-exchangeable protons. The duplex appears to be not particularly bent but a significant minor groove widening is observed at the probe site (not shown). Therefore, the effect of probe appears rather local, affecting only the opposite base and the two adjacent base pairs exhibit some modifications. These local changes appear quite acceptable, if we consider that nucleoside analogues very close to natural nucleobases, such as 2-aminopurine led to significant change in DNA dynamics and structure 47. The utility and limits of 3HCnt have been illustrated in a recent paper where the base flipping properties of the set and ring (SRA) domain of UHRF1 were explored by incorporating 3HCnt at several positions close to the methylated cytosine (mC)

16

. At two

positions, 3HCnt was found to fully preserve the binding properties of SRA and to be highly sensitive to the SRA-induced mC flipping. In contrast, when 3HCnt was opposite mC or flanking mC at 5’ position, the binding properties and base flipping properties of SRA were altered, in full line with the structural and dynamic consequences of 3HCnt insertion evidenced in this study. Thus, the present NMR data will be instrumental to properly design and interpret experiments using 3HCnt. The proposed structures (Figure 6 and Supplementary Figure S7) help to interpret the chemical shift perturbations of the aromatic and sugar protons induced by the probe. Indeed, the large effects observed for the aromatic and sugar protons of T18 could be explained by the base and sugar extrusion of this residue. The T17 large high-field variation can be rationalized by its stacking with the S7 hydroxychromone ring. The aromatic resonances of T6 display nearly no chemical shift variation indicating that the stacking of T6 nucleobase with S7 is similar to that with A7 in the unmodified duplex. Concerning the sugar protons, the large variations observed for T6 and T17 are consistent with their proximity with the 2-thienyl and 3-hydroxychromone parts of S7, respectively. Taken together, the chemical shift variations are well consistent with the obtained structures. These structures also support the limited decrease (2.4°C) of the melting temperature of the modified duplex. Indeed, the strong stacking interaction between the aromatic ring of S7, which has a size and shape similar to a


22

Page 23 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


base pair, and the adjacent base pairs likely limits the destabilizing effect of T18 ejection towards the solvent. The observed stacking of the probe in the duplex is also well consistent with the absorbance and fluorescence data

14

, suggesting that the probe is shielded in the

duplex with a reduced access to solvent. Besides the structural features, an interesting aspect of the present study is the variety of dynamic effects induced by the probe. Indeed, two different dynamic processes occurring in very different time scales, were observed. One kind of dynamic processes was related to the selective line broadening of protons of several residues (A8, A19 and T17) and to the appearance of additional signals in NMR spectra; such processes being characteristic of slow conformational exchanges in the microsecond to millisecond range

34

. These exchanges are

restricted to the residues in the immediate vicinity of the probe (A19, A8, T17), as evidenced by the differences in line broadening in Figure 4. A large broadening can be also observed for most of the proton resonances of the dye. The exact nature of the conformational exchanges generating these effects is not known and needs additional studies. Similar observations have been made for modified bases

48

and DNA structures with pyrene substituting a base

the case of universal base analogues

49

40

or in

or intercalating nucleotide analogue (INA)

41

.

Interestingly, these examples display common features with 3HCnt, as they lack WatsonCrick bonding face but are well accommodated inside the double helix thanks to strong stacking properties with the neighbor bases. They exhibit at the same time significant but localized conformational dynamics leading to resonance broadenings. The second class of dynamics was observed during the 20 ns of the restrained dynamics simulation. Despite the restraints, significant motions were observed during the molecular dynamics simulations, similar to those reported on dodecamers bearing a residue substituted by a coumarin 50. These motions are thought to correspond to a translation of the S7 aromatic moieties in a plane approximately perpendicular to the helix axis (Supplementary Figure S7B and C). This translational motion is consistent with the time-resolved anisotropy data, because being perpendicular to the helix axis and intercalated between the neighbor base pairs, the S7 fluorophore cannot rotate explaining the absence of short rotational correlation time in the anisotropy decay. In addition, the A8 and A9 residues at the 3' side display correlated motions with S7 dynamics while the residues located at the 5' side are not affected. The larger coupling between S7 and the 3' base pair as compared to the 5' base pair could be related to the stronger stacking interaction between S7 and the A8-T17 base pair as compared to the A19-T6 base pair (Figure 7b). However, the observation of these dynamics at the nanosecond timescale is indirect and not explicitly related to experimental data. As this dynamical


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 30

behavior is observed in spite of the restraints on the inner residues, this underlines the important intrinsic dynamic properties of the S7 residue. One important point is that these dynamics impact significantly the downstream base pair that appears strongly destabilized according to the proton data. Because the base pair lifetime (several milliseconds) and dye dynamics (several nanoseconds) are in very different timescales, any correlation between them is difficult to establish. Interestingly, similar molecular dynamics simulations with AMBER and explicit solvent and counter-ions, but without NMR restraints have been performed on dodecamers with a base pair replaced by a coumarin opposite to an abasic site 50. During the simulation, the coumarin aromatic moiety was observed to shift towards the abasic site in a direction perpendicular to the helix axis and with a timescale similar to that observed with 3HCnt. Such translational motions are also thought to explain the fast dynamics observed between pyrenes possessing a large unpolarized hydrophobic surface and natural bases

40

. The 3HCnt probe possesses an

unpolarized six membered ring part that could slide on the surfaces of neighbor bases more easily than natural DNA bases that possess well-localized bond dipoles.

Conclusions This work demonstrates the perfect intercalation and stacking of the 3HCnt fluorophore with the surrounding bases in a DNA helix. This strong intercalation readily explains its favorable fluorescence properties and notably its high quantum yield. In addition, we give a detailed description of the structural and dynamic impact of the 3HCnt insertion. The main drawback associated with the insertion of 3HCnt is the base flipping of the opposite nucleobase. This local disruption of the DNA structure in the vicinity of the probe may perturb the proper recognition by target proteins, and must therefore be properly taken into account for correctly interpreting protein-nucleic acid interaction data, and for properly designing the sequence of the labeled oligonucleotides for such experiments

16

. A likely solution to avoid this flipping

effect is to replace the opposite natural base by an abasic site. Though abasic sites could introduce local changes in the structure and dynamics of duplexes

51

, they are probably the

preferred option to optimally use 3HCnt as a fluorescent mimic of a base pair in DNA duplexes. Molecular dynamics simulation further revealed that the 3HCnt fluorophore underwent translational dynamics in a plane approximately perpendicular to the helix axis. These translational dynamics destabilize to some extent the two downstream base pairs. The effect on the other base pairs is marginal, explaining the limited impact of 3HCnt insertion on the


24

Page 25 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


melting temperature of the duplex. Thus, the design of sequences labeled with the 3HCnt should integrate all this information for properly monitoring protein-DNA interactions. Taken together, our data confirm 3HCnt as a promising fluorescent nucleoside analogue.

SUPPORTING INFORMATION DESCRIPTION The Supplementary data are composed of 2 Tables and 7 figures. Spectroscopic (CD, fluorescence) data of the modified duplex. Additional NOESY spectra of the modified duplex. Local RMSD calculated for different subsets of residues from the restrained molecular dynamics simulation. Superimposed views of the main averaged best energy structures.

SUPPORTING INFORMATION The supporting information is available free of charge on the ACS publication web site Spectroscopic data of the labeled duplex; RMSD of the ten best-energy structures; Schematic representation of the ESIPT reaction; CD spectra of labeled and non-labeled duplexes; absorption and emission spectra of labeled duplex; Expanded plot of 500 MHz NOESY spectrum of H2'/H2"-H6/H8 region of non-labeled duplex; Expanded plot of 950 MHz NOESY of H6/H8-H1' region of the labeled duplex; local RMSD of different subset of the labeled duplex during the restrained- molecular dynamics; local RMSD of the S7 residue during the 20 ns of the restrained molecular dynamics simulation; superimposed views of four averaged structures.

ACKNOWLEDGEMENTS This work was supported by CNRS. Funding for Open access charge: Centre National de la Recherche Scientifique (CNRS LBPA UMR 8113). Financial support from the TGIR-RMNTHC Fr3050 CNRS for conducting the research is gratefully acknowledged. This work was supported by Agence Nationale de la Recherche (ANR) (Fluometadn ANR-12-BS08-0003-02 and Pico2, ANR-15-CE11-0006-03) and the Fondation pour la Recherche Médicale (FRM) (DCM20111223038). We thank Dr. Brigitte Hartmann and Romain Retureau for help and advice in the conduction of molecular dynamics calculations with AMBER.

References


25


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1. 2.

3.

4. 5. 6.

7.

8.

9.

10.

11. 12. 13.

14.

15.

16.

17.

Page 26 of 30

Wilhelmsson, L. M. Fluorescent Nucleic Acid Base Analogues. Quart. Rev. Biophys. 2010, 43, 159-183. Sinkeldam, R. W.; Greco, N. J.; Tor, Y. Fluorescent Analogs of Biomolecular Building Blocks: Design, Properties, and Applications. Chem. Rev. 2010, 110, 2579-2619. Wojciechowski, F.; Leumann, C. J. Alternative DNA Base-Pairs: From Efforts to Expand the Genetic Code to Potential Material Applications. Chem. Soc. Rev. 2011, 40, 5669-5679. Malinovskii, V. L.; Wenger, D.; Haner, R. Nucleic Acid-Guided Assembly of Aromatic Chromophores. Chem. Soc. Rev. 2010, 39, 410-422. Teo, Y. N.; Kool, E. T. DNA-Multichromophore Systems. Chem. Rev. 2012, 112, 4221-4245. Ensslen, P.; Wagenknecht, H. A. One-Dimensional Multichromophor Arrays Based on DNA: From Self-Assembly to Light-Harvesting. Acc. Chem. Res. 2015, 48, 27242733. Ward, D. C.; Reich, E.; Stryer, L. Fluorescence Studies of Nucleotides and Polynucleotides .I. Formycin 2-Aminopurine Riboside 2,6-Diaminopurine Riboside and Their Derivatives. J. Biol. Chem. 1969, 244, 1228-1237. Rachofsky, E. L.; Osman, R.; Ross, J. B. Probing Structure and Dynamics of DNA with 2-Aminopurine: Effects of Local Environment on Fluorescence. Biochemistry 2001, 40, 946-956. Jones, A. C.; Neely, R. K. 2-Aminopurine as a Fluorescent Probe of DNA Conformation and the DNA-Enzyme Interface. Quart. Rev. Biopys. 2015, 48, 244279. Voltz, K.; Leonard, J.; Touceda, P. T.; Conyard, J.; Chaker, Z.; Dejaegere, A.; Godet, J.; Mely, Y.; Haacke, S.; Stote, R. H. Quantitative Sampling of Conformational Heterogeneity of a DNA Hairpin Using Molecular Dynamics Simulations and Ultrafast Fluorescence Spectroscopy. Nucleic Acids Res. 2016, 44, 3408-3419. Teo, Y. N.; Wilson, J. N.; Kool, E. T. Polyfluorophore Labels on DNA: Dramatic Sequence Dependence of Quenching. Chemistry 2009, 15, 11551-11558. Dai, N.; Kool, E. T. Fluorescent DNA-Based Enzyme Sensors. Chem. Soc. Rev. 2011, 40, 5756-5770. Singh, V.; Wang, S.; Kool, E. T. Genetically Encoded Multispectral Labeling of Proteins with Polyfluorophores on a DNA Backbone. J. Am. Chem. Soc. 2013, 135, 6184-6191. Dziuba, D.; Postupalenko, V. Y.; Spadafora, M.; Klymchenko, A. S.; Guerineau, V.; Mely, Y.; Benhida, R.; Burger, A. A Universal Nucleoside with Strong Two-Band Switchable Fluorescence and Sensitivity to the Environment for Investigating DNA Interactions. J. Am. Chem. Soc. 2012, 134, 10209-10213. Spadafora, M.; Postupalenko, V. Y.; Shvadchak, V. V.; Klymchenko, A. S.; Mely, Y.; Burger, A.; Benhida, R. Efficient Synthesis of Ratiometric Fluorescent Nucleosides Featuring 3-Hydroxychromone Nucleobases. Tetrahedron 2009, 65, 7809-7816. Kilin, V.; Gavvala, K.; Barthes, N. P.; Michel, B. Y.; Shin, D.; Boudier, C.; Mauffret, O.; Yashchuk, V.; Mousli, M.; Ruff, M.; Granger, F.; Eiler, S.; Bronner, C.; Tor, Y.; Burger, A.; Mely, Y. Dynamics of Methylated Cytosine Flipping by Uhrf1. J. Am. Chem. Soc. 2017, 139, 2520-2528. Kuznetsova, A. A.; Kuznetsov, N. A.; Vorobjev, Y. N.; Barthes, N. P.; Michel, B. Y.; Burger, A.; Fedorova, O. S. New Environment-Sensitive Multichannel DNA


26

Page 27 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


18.

19.

20. 21.

22.

23.

24.

25.

26. 27. 28. 29.

30. 31.

32. 33. 34. 35. 36.

Fluorescent Label for Investigation of the Protein-DNA Interactions. Plos One 2014, 9, e:100007. Beuck, C.; Singh, I.; Bhattacharya, A.; Heckler, W.; Parmar, V. S.; Seitz, O.; Weinhold, E. Polycyclic Aromatic DNA-Base Surrogates: High-Affinity Binding to an Adenine-Specific Base-Flipping DNA Methyltransferase. Ang. Chem. Int. Ed. 2003, 42, 3958-3960. Melhuish, W. H. Quantum Efficiencies of Fluorescence of Organic Substances Effect of Solvent and Concentration of Fluorescent Solute. J. Phys. Chem. 1961, 65, 229-235. Lakowicz, J. R. 2006. Principles of Fluorescence Spectroscopy. Piotto, M.; Saudek, V.; Sklenar, V. Gradient-Tailored Excitation for SingleQuantum Nmr-Spectroscopy of Aqueous-Solutions. J. Biomol. NMR 1992, 2, 661665. Delaglio, F.; Grzesiek, S.; Vuister, G. W.; Zhu, G.; Pfeifer, J.; Bax, A. Nmrpipe - a Multidimensional Spectral Processing System Based on Unix Pipes. J. Biomol. NMR 1995, 6, 277-293. Zgarbova, M.; Luque, F. J.; Sponer, J.; Cheatham, T. E.; Otyepka, M.; Jurecka, P. Toward Improved Description of DNA Backbone: Revisiting Epsilon and Zeta Torsion Force Field Parameters. J. Chem. Theory Comput. 2013, 9, 2339-2354. Case, D. A.; Cheatham, T. E., 3rd; Darden, T.; Gohlke, H.; Luo, R.; Merz, K. M., Jr.; Onufriev, A.; Simmerling, C.; Wang, B.; Woods, R. J. The Amber Biomolecular Simulation Programs. J. Comput. Chem. 2005, 26, 1668-1688. Berendsen, H. J. C.; Postma, J. P. M.; van Gunsteren, W. F.; DiNola, A.; Haak, J. R. Molecular Dynamics with Coupling to an External Bath. J. Chem. Phys. 1984, 81, 3684-3690. van Gunsteren, W. F.; Berendsen, H. J. C. Algorithms for Macromolecular Dynamics and Constraint Dynamics. Mol. Phys. 1977, 34, 1311-1327. Darden, T.; York, D.; Pedersen, L. Particle Mesh Ewald: An N·Log(N) Method for Ewald Sums in Large Systems. J. Chem. Phys. 1993, 98, 10089-10092. Macke, T. C., D.A. 1998. Modeling Unsuual Nucleic Acid Structures in Molecular Modeling of Nucleic Acids, Washington, DC. Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.; Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. Ucsf Chimera - a Visualization System for Exploratory Research and Analysis. J. Comput. Chem. 2004, 25, 1605-1612. Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D. Comparison of Simple Potential Functions for Simulating Liquid Water. J. Chem. Phys. 1983, 79, 926-935. Roe, D. R.; Cheatham, T. E. Ptraj and Cpptraj: Software for Processing and Analysis of Molecular Dynamics Trajectory Data. J. Chem. Theory Comput. 2013, 9, 30843095. Ellis, G. P. 1977. Chromenes, Chromanones and Chromones Wiley. Palmer, A. G. Nmr Characterization of the Dynamics of Biomacromolecules. Chem. Rev. 2004, 104, 3623-3640. Palmer, A. G., 3rd. Chemical Exchange in Biomacromolecules: Past, Present, and Future. J. Magn. Res. 2014, 241, 3-17. Mittermaier, A. K.; Kay, L. E. Observing Biological Dynamics at Atomic Resolution Using Nmr. Trends Biochem. Sci. 2009, 34, 601-611. Ben Imeddourene, A.; Xu, X. Q.; Zargarian, L.; Oguey, C.; Foloppe, N.; Mauffret, O.; Hartmann, B. The Intrinsic Mechanics of B-DNA in Solution Characterized by Nmr. Nucleic Acids Res. 2016, 44, 3432-3447.


27


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

37.

38.

39.

40.

41.

42.

43.

44.

45. 46.

47.

48.

49.

50.

Page 28 of 30

Pasi, M.; Maddocks, J. H.; Beveridge, D.; Bishop, T. C.; Case, D. A.; Cheatham, T. C.; Dans, P. D.; Jayaram, B.; Lankas, F.; Laughton, C.; Mitchell, J.; Osman, R.; Orozco, M.; Perez, A.; Petkeviciute, D.; Spackova, N.; Sponer, J.; Zakrzewska, K.; Lavery, R. Mu Abc: A Systematic Microsecond Molecular Dynamics Study of Tetranucleotide Sequence Effects in B-DNA. Nucleic Acids Res. 2014, 42, 12272-12283. Ben Imeddourene, A.; Elbahnsi, A.; Gueroult, M.; Oguey, C.; Foloppe, N.; Hartmann, B. Simulations Meet Experiment to Reveal New Insights into DNA Intrinsic Mechanics. PLoS Comput. Biol. 2015, 11, e:1004631. Shanmugam, G.; Kozekov, I. D.; Guengerich, F. P.; Rizzo, C. J.; Stone, M. P. Structure of the 1,N-2-Etheno-2 '-Deoxyguanosine Lesion in the 3 '-G(Epsilon Dg)T-5 ' Sequence Opposite a One-Base Deletion. Biochemistry 2010, 49, 2615-2626. Smirnov, S.; Matray, T. J.; Kool, E. T.; de los Santos, C. Integrity of Duplex Structures without Hydrogen Bonding: DNA with Pyrene Paired at Abasic Sites. Nucleic Acids Res. 2002, 30, 5561-5569. Nielsen, C. B.; Petersen, M.; Pedersen, E. B.; Hansen, P. E.; Christensen, U. B. Nmr Structure Determination of a Modified DNA Oligonucleotide Containing a New Intercalating Nucleic Acid. Bioconjugate Chem. 2004, 15, 260-269. Dziuba, D.; Karpenko, I. A.; Barthes, N. P.; Michel, B. Y.; Klymchenko, A. S.; Benhida, R.; Demchenko, A. P.; Mely, Y.; Burger, A. Rational Design of a Solvatochromic Fluorescent Uracil Analogue with a Dual-Band Ratiometric Response Based on 3-Hydroxychromone. Chemistry 2014, 20, 1998-2009. Sholokh, M.; Zamotaiev, O. M.; Das, R.; Postupalenko, V. Y.; Richert, L.; Dujardin, D.; Zaporozhets, O. A.; Pivovarenko, V. G.; Klymchenko, A. S.; Mely, Y. Fluorescent Amino Acid Undergoing Excited State Intramolecular Proton Transfer for SiteSpecific Probing and Imaging of Peptide Interactions. J. Phys. Chem. B 2015, 119, 2585-2595. Gavvala, K.; Barthes, N. P. F.; Bonhomme, D.; Dabert-Gay, A. S.; Debayle, D.; Michel, B. Y.; Burger, A.; Mely, Y. A Turn-on Dual Emissive Nucleobase Sensitive to Mismatches and Duplex Conformational Changes. RSC Adv. 2016, 6, 8714287146. Ortega, A.; de la Torre, J. G. Hydrodynamic Properties of Rodlike and Disklike Particles in Dilute Solution. J. Chem. Phys. 2003, 119, 9914-9919. Kuznetsova, A. A.; Kuznetsov, N. A.; Vorobjev, Y. N.; Barthes, N. P. F.; Michel, B. Y.; Burger, A.; Fedorova, O. S. New Environment-Sensitive Multichannel DNA Fluorescent Label for Investigation of the Protein-DNA Interactions. Plos One 2014, 9. Dallmann, A.; Dehmel, L.; Peters, T.; Mugge, C.; Griesinger, C.; Tuma, J.; Ernsting, N. P. 2-Aminopurine Incorporation Perturbs the Dynamics and Structure of DNA. Angew. Chem. Int. Ed. 2010, 49, 5989-5992. Huang, H.; Wang, H.; Lloyd, R. S.; Rizzo, C. J.; Stone, M. P. Conformational Interconversion of the Trans-4-Hydroxynonenal-Derived (6s,8r,11s) 1,N(2)Deoxyguanosine Adduct When Mismatched with Deoxyadenosine in DNA. Chem. Res. Toxicol. 2009, 22, 187-200. Gallego, J.; Loakes, D. Solution Structure and Dynamics of DNA Duplexes Containing the Universal Base Analogues 5-Nitroindole and 5-Nitroindole 3Carboxamide. Nucleic Acids Res. 2007, 35, 2904-2912. Furse, K. E.; Corcelli, S. A. Effects of an Unnatural Base Pair Replacement on the Structure and Dynamics of DNA and Neighboring Water and Ions. J. Phys. Chem. B 2010, 114, 9934-9945.


28

Page 29 of 30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


51.

Furse, K. E.; Corcelli, S. A. Dynamical Signature of Abasic Damage in DNA. J. Am. Chem. Soc. 2011, 133, 720-723.


29


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 30

TOC Graphic


30

Structural and Dynamical Impact of a Universal Fluorescent

Recommend Documents