Multianalytical Study of the Binding between a ... - ACS Publications

Subscriber access provided by University of Otago Library

Article

Multi-analytical study of the binding between a small chiral molecule and a DNA aptamer: evidences for asymmetric steric effect upon 3’ vs. 5’-end sequence modification Lylian Challier, Rebeca Miranda-Castro, Bertrand Barbe, Claire Fave, Benoit Limoges, Eric Peyrin, Corinne Ravelet, Emmanuelle Fiore, Pierre Labbé, Liliane Coche-Guerente, Eric Ennifar, Guillaume Bec, Philippe Dumas, Francois Mavre, and Vincent Noel Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b04046 • Publication Date (Web): 07 Nov 2016 Downloaded from http://pubs.acs.org on November 7, 2016

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.

Analytical Chemistry 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 10

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

Analytical Chemistry

Multi-analytical study of the binding between a small chiral molecule and a DNA aptamer: evidences for asymmetric steric effect upon 3’ vs. 5’-end sequence modification Lylian Challier,a‡ Rebeca Miranda-Castro,a‡ Bertrand Barbe,a Claire Fave,b Benoît Limoges,b Eric Peyrin,c Corinne Ravelet,c Emmanuelle Fiore,c Pierre Labbé,d Liliane Coche-Guérente,d Eric Ennifar,e Guillaume Bec,e Philippe Dumas,e François Mavré,b* Vincent Noël a* a

ITODYS, UMR 7086 CNRS, Univ. Paris Diderot, Sorbonne Paris Cité, 15 rue Jean-Antoine de Baïf, F-75205 Paris Cedex 13, France. b Laboratoire d’Electrochimie Moléculaire, UMR 7591 CNRS, Univ. Paris Diderot, Sorbonne Paris Cité, 15 rue JeanAntoine de Baïf, F-75205 Paris Cedex 13, France. c Département de Pharmacochimie Moléculaire, UMR 5063 CNRS, Université Grenoble Alpes, 470 rue de la chimie, 38400 Saint-Martin d’Hères, France. d Département de Chimie Moléculaire, CNRS, UMR 5250, Université Grenoble Alpes, FR 2607, 570 rue de la chimie, B.P. 53, 38400 Grenoble, France. e “Architecture et Réactivité de l’ARN”, Biophysique et Biologie Structurale, Institut de Biologie Moléculaire et Cellulaire du CNRS, Université de Strasbourg, 15, rue René Descartes, 67084 Strasbourg, France. KEYWORDS : Enantioselective binding, aptamer, microdroplet electrochemistry, kinITC, quartz microbalance, fluorescence polarization, induced-fit.

ABSTRACT: Nucleic acid aptamers are involved in a broad field of applications ranging from therapeutics to analytics. Deciphering the binding mechanisms between aptamers and small ligands is therefore crucial to improve and optimize existing applications and to develop new ones. Particularly interesting is the enantiospecific binding mechanism involving small molecules with nonprestructured aptamers. One archetypal example is the chiral binding between L-tyrosinamide and its 49-mer aptamer for which neither structural nor mechanistic information is available. In the present work, we have taken advantage of a multiple analytical characterization strategy (i.e., using electroanalytical techniques such as kinetic rotating droplet electrochemistry, fluorescence polarization, isothermal titration calorimetry, and quartz crystal microbalance) for interpreting the nature of binding process. Screening of the binding thermodynamics and kinetics with a wide range of aptamer sequences revealed the lack of symmetry between the two ends of the 23-mer minimal binding sequence, showing an unprecedented influence of the 5’ aptamer modification on the bimolecular binding rate constant kon, and no significant effect on the dissociation rate constant koff. The results we have obtained lead us to conclude that the enantiospecific binding reaction occurs through an induced-fit mechanism, wherein the ligand promotes a primary nucleation binding step near the 5’-end of the aptamer followed by a directional folding of the aptamer around its target from 5’-end to 3’-end. Functionalization of the 5'-end position by a chemical label, a polydA tail, a protein, or a surface influences the kinetic/thermodynamic constants up to two orders of magnitude in the extreme case of a surface immobilized aptamer while significantly weaker effect is observed for a 3'-end modification. The reason is that steric hindrance must be overcome to nucleate the binding complex in the presence of a modification near the nucleation site.

small molecules) in different areas ranging from therapeutics8,9 to diagnosis10 and sensing.11,12 However, in spite of the large number of aptamers selected against a wide range of molecules, only a few efforts have been made to characterize the binding mechanism between aptamers and their target, such studies being performed only for a restricted number of popular molecules of interest.6 Still, mechanistic characterization of the binding process through the determination of structural, thermodynamic and kinetic properties are obviously a major prerequisite in order to rationally design further applications.

INTRODUCTION Chemical and structural diversity of single-stranded oligonucleotides (DNA or RNA) are at the origin of a wide functional variety. A spectacular example is the ability to bind molecular targets with high specificity and affinity. These binding properties may naturally occur1-4 or be revealed for synthetic oligonucleotides, i.e. aptamers, by in vitro selection processes such as SELEX.5-7 The recent progress in these selection methods allows now to fulfill the increasing demand in selective receptors for targets of interest (cells, viruses, proteins and

1

ACS Paragon Plus Environment


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

For instance, many of the aptamer-based analytical detection strategies involve either the labeling of the aptamer oligonucleotide (i.e., by a molecular probe, a protein, a nanoparticle, etc.) or the anchoring of the 3’ or 5’ extremity of the aptamer to a surface (aptasensor).9 Such labeling requires at least key information concerning the binding mechanism in order to optimize its impact on the recognition process. Single-stranded oligonucleotides often possess readily ascertainable structures that may predefine a target binding site, and most of the aptamers studied so far adopt such a prestructuration due either to high content in canonical base pairing (theophylline,13 cocaine,14 ATP15) or to the formation of a Gquadruplex structure (thrombin).16 For those systems, the correlation between NMR and molecular dynamics has proven its efficiency for the aptamer/target complex mapping.15-17 Conversely, in the critical case for which there is no trivial pre-structuration of the aptamer, no predefined binding site and no accessible structural information on the final aptamer/target complex, elucidation of the binding mechanism is even more difficult to achieve. One archetypal example is the binding between L-tyrosinamide (L-Tym) and its 49-mer aptamer18 for which neither structural nor mechanistic information is available. Recognition between this aptamer and its target is highly enantioselective (D-Tym is not recognized by the aptamer) and was thus exploited in the development of aptamer-based enantiomeric excess measurements.19-21 It is therefore a relevant system for the better understanding of enantioselective recognition process. Mechanism elucidation requires the cross-correlation of multiple analytical techniques (titration and/or kinetic) for a wide range of binding conditions (ionic force, buffer composition, temperature) as well as binding partners (target analogs, mutated or truncated aptamer sequences). Current techniques used for such a purpose are optical-based techniques (fluorescence,4,22 SPR23) but also non optical methods such as isothermal titration calorimetry24 or capillary electrophoresis.25 Bearing in mind the cost and/or time required to perform such experiments, rapid and quantitative methods able to operate with small sample volumes are highly desirable. Here we propose to take advantage of a previously reported homogeneous electrochemical method that was developed in the framework of an aptamer-based enantiomeric excess measurement using L-tyrosinamide as a model system.20 On one hand, the measurement principle relies on the difference in diffusion rate of the bound and free forms of the small molecule target, which makes this strategy intrinsically versatile since no conformational change is required for an efficient transduction. On the other hand, electrochemistry holds competitive advantages toward previously cited methods, particularly in terms of sensitivity, reproducibility, and ease of quantification of small sample volumes, but also in terms of rapidity and simplicity of measurement. The principle was also extended to the measurement of binding kinetics with the help of our previously developed kinetic rotating droplet electrochemical method26 which, compared to the traditional mixing-based kinetics methods, has the advantages to be affordable, simple, easy to implement and able to operate with very low amount of reagents. The objective of the work is thus to decipher the mechanistic features of the L-tyrosinamide (L-Tym) binding to its aptamer as an illustrative example. Here, the accurate determination of dissociation and kinetic constants for truncated or mutated

Page 2 of 10

sequences of the anti-L-Tym aptamer as well as for protein- or surface-functionalized sequences is carrying out in order to identify the factors affecting the binding mechanism. The present study should thus pave the way for optimization in the development of aptamer-based analytical strategies.

EXPERIMENTAL SECTION Chemical and biochemicals. 2-Amino-1,3-propanediol hydrochloride (Aldrich), 2-amino-1,3-propanediol (Aldrich), NaCl (Fluka), MgCl2 (Sigma), L- and D -tyrosinamide (Aldrich) were used without further purification. Synthesis of 2((ferrocenylmethyl)amino)-3-(4-hydroxyphenyl)propanamide (L-Tym-Fc) was previously described.20 The oligonucleotides were synthesized and HPLC-purified by Eurogentec (Angers, France) and their concentration was controlled by UV-Vis spectroscopy using a Nanodrop 2000 spectrophotometer (ThermoFisher). All the used sequences are reported in SI. Electrochemical apparatus. Electrochemical measurements were performed with an AUTOLAB PGSTAT 100 (Metrohm) controlled by computer, and the data were acquired using GPES 4.9007 software (EcoChemie, The Netherlands). The electrochemical cell is made of 3 screen-printed electrodes on a flat PET substrate. Working electrode (0.017 cm2) and counter electrode (≈ 0.05 cm2) were screen-printed carbon electrodes (ink: Electrodag PF-470 A) deposited on the top of an underlying Ag layer (Electrodag 418 SS). Pseudo reference is directly an Ag layer.26 Detailed description of the experimental rotating droplet setup26 as well as the electrodes conditioning protocol are supplied in SI. Fluorescence Polarization. The binding buffer consisted of 10 mM Tris–HCl, pH 7.5, 50 mM NaCl, 10 mM MgCl2. The aptamer solutions were prepared in water and stored at -20 °C. The working aptamer solutions were obtained by adequate dilution of the stock solution in 1.25 × concentrated binding buffer. Prior to the first utilization, the working solutions were heated at 80 ° C for 5 min and left to stand at RT for 30 min. The Tym solutions were prepared in water. All solutions were filtered prior to use through 0.45 µm membranes. To construct the titration curves, the aptamer and analyte solutions were mixed into the individual wells (final volume 100 µL) at RT. Blank wells of the microplate received 100 µL of the binding buffer. The final concentration of the aptamer probe was 10 nM. All experiments were done in triplicate. The microplate was immediately placed into the microplate reader for the measurement. Fluorescence anisotropy readings were taken on a Tecan Infinite F500 microplate reader (Mannedorf, Switzerland) using black, 96-well Greiner Bio-One microplates (ref: 675086, Courtaboeuf, France). Excitation was set at 485 ± 20 nm and emission was collected with 535 ± 25 nm bandpass filters. The anisotropy (r, see Eq. 1) was calculated by the instrument software, as classically reported.19 =

(1)

where Iw and Ivh are the vertically and horizontally polarized components of the emission after excitation by vertically polarized light. The instrumental correction factor G was determined from standard solutions according to the manufacturer's instructions. Dissociation constant (Kd) was obtained by fitting the experimental anisotropy change as a function of L-Tym concentration by equation 2.

2


Page 3 of 10

6

(A) 1.0

a

(B)

(C)

b

80

0

Ip/Ip

d e

40

4

0.8

c

60

Kd / µM

100

I (nA)

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


0.6

2

20 0.4

0

0 0.4

0.5

0.6

0.7

0.8

0

5

10

E (V vs. Ag/AgCl)

15

20

25

20

25

[Aptn] (µM)

30

35

40

45

50

Aptn

Figure 1. (A) Voltammograms (v = 0.05 V s-1) of 5 mM Tris buffer containing 50 mM NaCl, 5 mM MgCl2, 1 µM [Os(bpy)3]2+, 5 µM L-Tym and (a) 0 µM, (b) 2 µM, (c) 5 µM, (d) 10 µM (e) 20 µM Apt23. (B) Normalized catalytic peak current as a function of aptamer concentration for aptamer of different length: (■) Apt49, (●) Apt33, () Apt31 (●) Apt28 and () Apt23, and () 49-mer scramble oligonucleotide. Error bars: standard deviations from 3 measurements. Plain lines: fits of eq 2 to the experimental data. (C) Variation of Kd as a function of the aptamer length (●) truncated from the 5’-end, (■) Apt5’-26A-23 and (■) Apt23-26A-3’

=

Os(bpy)32+, 5 µM L-Tym, and increasing amounts of 49-mer aptamer (Apt49, see Scheme 1). The peak current change as a function of aptamer concentration is used to determine the dissociation constant (Kd) of the binding reaction by fitting eq S5 to the experimental data (Figure 1B). A Kd value of 4.9 ± 0.5 µM is recovered. This methodology gives reproducible Kd values in a short time and with low reagent consumption (the titration curve was performed with a single droplet of ca. 50 µL solution). It is therefore a method of choice for a screening of different aptamer sequences as well as various experimental conditions in order to understand the main binding process features. We first aimed at determining the minimal sequence able to bind the target.28-30 To that end, the original 49-mer sequence (Apt49)18 was sequentially truncated from both the 3’-end and 5’-end and Kd values systematically determined for each truncated sequence with our electrochemical method. Loss of only one base at the 3’-terminal of Apt49 (i.e., cytosine C1) was enough to completely lose the binding properties towards LTym (no assessable Kd).

(2)

where r is the recorded anisotropy, rf is the anisotropy of the free strand and C the L-Tym concentration. Quartz Crystal Microbalance with Dissipation monitoring (QCM-D). QCM-D measurements were performed using QSense, E1-E4 instrument (Biolin Scientific –Sweden) equipped with one (or four) flow chamber(s) and polished ATcut piezoelectric quartz crystals covered 100 nm gold (QSX 301). Besides measurements of bound mass, which is provided from changes in the resonance frequency, f, of the piezoelectric sensor resonator, the QCM-D technique also provides structural information of biomolecular films via changes in the damping, D, of the crystal. f and D were measured at the fundamental resonance frequency (5 MHz) as well as at the 3rd, 5th, 7th, 9th, 11th and 13th. Experiments were conducted in a continuous flow of buffer at 50 µL/min by using a peristaltic pump (ISM935C, Ismatec, Zurich, Switzerland). The temperature of the E1-E4 QCM-D platform and all solutions were stabilized (via a Thermomixer (Ependorf)) to ensure stable operations at 24°C. All buffers were previously degassed in order to avoid bubble formation in the QCM-D measuring chamber. Thermal titration calorimetry. ITC was performed with an ITC200 (Microcal-Malvern) operated in the multiple injection mode. Data were acquired with high-gain mode and mixing at 1000 rpm. Injection of the ligand was performed at 0.5 µl s-1.

Scheme 1. Aptamer sequences

RESULTS This is in contrast to the results obtained with the stepwise sequence length reduction from the 5'-end which did not cause any significant change of the affinity binding until the truncation has reached the thymine T38 (Figure 1C). Upon further 5’-truncation of the aptamer, however, a progressive significant increase in the affinity binding was observed until reaching asymptotically a minimal value of Kd = 0.8 ± 0.2 µM when the aptamer is cut up to the T23 base (Figure 1C). A further deletion of the T23 base finally leads to a complete loss of the affinity binding to L-Tym. The bases C1 and T23 are therefore essential for the binding to L-Tym. The final truncated sequence of 23-bases length (Apt23) appears thus as the minimal

Determination of the minimal binding sequence by electrochemical titration. As recalled in the introduction, the principle of the electrochemical titration relies on the difference in diffusion rate of the bound and free forms of a small molecule target. In the case of an electroactive target, the current intensity is used to determine the relative amount of bound and free target during a titration experiment, from which the dissociation constant can be extracted. The electrochemical detection of L-Tym is made possible thanks to redox-mediated catalytic oxidation of its phenol function as described previously.20,27 Figure 1A shows the cyclic voltammograms (CVs) obtained at a screen-printed electrode immersed in a solution of 1 µM

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 10

limitation of the FP detection when applied to short DNA sequences.

binding sequence (MBS). Furthermore, identification of the MBS allows to interpret previous work published by Zhou et al. into which the anti-L-Tym aptamer was hybridized near its 5’ terminus to a fluorophore-labeled complementary strand of 12 mer length.31 The complementary strand was displaced exclusively in the presence of the cognate target according to a competitive exchange mechanism. Such competition is explained by the fact that the hybridized aptamer sequence involves 6 of the MBS bases (T18 to T23), including the necessary T23. The remaining 6 bases of the 12mer, as shown here, are not involved in the L-Tym binding process and will readily dissociate at the working temperature. Such mechanism is very similar to a toehold-mediated strand displacement.32 The resulting improved Kd is almost one order of magnitude lower than the starting Kd of the 49-mer aptamer, a behavior that seems to be related to some steric hindrance effect or unproductive intra-molecular interactions arising from the 5’tail of the aptamer.28-30 Bases from 23 to 49 have hence a detrimental effect on the affinity binding, an effect that is clearly length dependent. Whether this negative effect is also base-dependent or not was further investigated. The MBS was elongated by a poly-dA tail containing 26 adenosines at the 5'-end (Apt5'-26A-23) to finally reach a total size of 49 bases. The 5' side elongation leads to a Kd of 5.0 ± 0.3 µM. This value is close to the one of Apt49 (i.e. 4.9 µM), allowing us to conclude that the loss of affinity by the extra 26 bases is not sequence specific but sequence-length dependent. Interestingly, position of the polyA tail (either on 3’- or 5’-end) has been found to have an influence on the binding affinity. Indeed, when the 3'-end is elongated by the same 26A tail (Apt23-26A-3'), a Kd value of 1.1 ± 0.2 µM, almost identical to the Apt23 (0.8 µM), is determined. We can conclude that only modification of the MBS 5'-end would generate a loss in the binding strength of the aptamer for L-Tym. Fluorescence Polarization. The fact that MBS is a short 23mer length DNA sequence without stable preexisting secondary structure suggests a binding process involving a complete dynamic aptamer folding and structuration around its target. A previous study has shown that labeling the Apt49 with a fluorescent probe (i.e., fluorescein or Texas red) allows for monitoring the specific binding event through FP and that the labeling position (5’ vs. 3’) were not equivalent in the magnitude of the FP signal change.19 It was notably observed that the fluorescence anisotropy change upon binding was greater when the fluorescent probe is attached at the 3’-end. This is in line with our present findings which show that the 3’-end of the aptamer (and C1) is directly involved in the binding process conversely to the 5’-end. FP titration curves of the Apt23 and Apt49 sequences, both labeled with a fluorescein probe in 3’ position, were established as a function of increasing L-Tym concentration (Figure S1). From nonlinear regression fits of the titration curves, Kd values of 1.63 ± 0.07 µM and 0.44 ± 0.14 µM were obtained for the Apt49 and Apt23, respectively, values consistent with higher affinity for the Apt23. It is worthwhile to note the significantly weaker fluorescence anisotropy amplitude in the case of Apt23 compared with Apt49, a behavior which can result from the different sizes of the aptamer. This lower FP amplitude leads thus to a lower detection sensitivity of the binding process, which can cause less accurate determination of Kd. Such a behavior also underlines a

Table 1. Variation with the aptamer sequence of thermodynamic parameters from ITC Seq. Apt49 Apt23 Apt23

T (°C) 15 15 20

Kd (µM) 0.63 ± 0.06 0.19 ± 0.02 0.32 ± 0.03

∆H (kcal.mol-1) −16.1 ± 0.2 −22.2 ± 0.3 −27.1 ± 0.3

T∆ ∆S (kcal mol-1) −7.6 ± 0.1 −12.6 ± 0.15 −17.7 ± 0.2

Isothermal Titration Calorimetry (ITC). ITC titration curves were performed with the Apt49 and Apt23 sequences and increasing concentrations of L-Tym. The data were processed with the AFFINImeter software which provides the user with classical thermodynamic results, but also with kinetic information in terms of association and dissociation rate constants thanks to the implementation of a simplified, and yet efficient, version of kinITC24.33 ITC data were collected at two aptamer concentrations (i.e., 15 and 25 µM) and two temperatures (15 and 20 °C) with increasing concentration of L-Tym (see Table 1 and Figures S2 to S4). ITC titrations were performed with the same buffer, to work under the same conditions as the electrochemical experiments. Data processing with the AFFINImeter software yielded a Kd value of 0.32 ± 0.03 µM at 20 °C for the Apt23, which is in good agreement with the values of 0.8 µM and 0.44 µM determined by electrochemistry and fluorescence anisotropy, respectively. ITC results show a large enthalpy variation of the binding reaction in the range 16-27 kcal mol-1 (Table 1) accompanied by an important negative ∆S that is the same order of magnitude as previously reported for Apta49.34 Results relative to Apt23 show an additional enthalpic contribution (-6.1 kcal mol-1), which is mostly compensated by an entropic decrease (-5 kcal mol-1). This is consistent with general expectations of enthalpy-entropy compensation for ligand binding to a macromolecule.35 Influence on the binding interaction of a bulky label attached at the 3’ and 5’ extremities of the 23-mer aptamer. To further evaluate the effect of a bulky substituent on the binding properties of Apt23 to L-Tym, we have established the electrochemical titration curves of 3’- and 5’-biotinylated Apt23 specifically anchored to the binding site of an avidin (forming thus aptamer-biotin/avidin complexes wherein the aptamer extremity is affixed to a bulky protein). In order to avoid interferences from the tyrosine residues of avidin that could take part on the electrochemical signal of the oxidation of the L-Tym phenol moiety, the electrochemical titration curves were performed by monitoring the current response decrease of a ferrocenyl-labeled L-Tym (Fc-L-Tym) as a function of increasing concentrations of the aptamer-biotin/avidin complexes (see SI for experimental details), a detection strategy that is similar to the one we have previously described.20,27 The binding affinity of Apt23 for Fc-L-Tym was found slightly lower (Kd = 1.2 ± 0.3 µM) than the one determined with the non-labeled target (Kd = 0.8 ± 0.2 µM). Similar values of Kd were also determined for the binding of 5’- and 3’-biotinylated Apt23 to L-Tym showing that the biotin residue with its linker only marginally affects the aptamer binding properties to its target, independently of the labeling position on the 5’- or 3’end (Figure S5). In the presence of an excess of avidin during

4


the titration experiment of the 3’-end biotinylated Apt23/avidin complex (Apt3’avidin in Table 2), a 5-fold increase of Kd (i.e. 6.4 ± 0.9 µM) was determined compared to the same system without avidin, a result which can either arise from steric hindrance in the L-Tym/aptamer complex formation or also possibly from non-specific competitive interactions between the aptamer and the avidin. Nevertheless, this remains a small change in comparison with the increase of Kd value determined with the 5’-end biotinylated Apt23/avidin complex (i.e., 31 ± 10 µM, see Apt5’avidin in Table 2). Kd is indeed one order of magnitude higher than in the absence of avidin. This loss of affinity is comparable to the loss of affinity obtained for the MBS aptamer extended at the 5’-end by a polydA tail. Such a result clearly indicates that the recognition process is largely impacted by the presence of an avidin on the 5'-end of the aptamer, which, likewise to the polydA tail, tends to sterically affect the aptamer binding to L-Tym. The effect of surface immobilization. To complete our study, we have studied the binding of L-Tym to an aptamerimmobilized gold surface, a widely used format in analytical strategies (aptasensor). Here, a quartz crystal microbalance with dissipation monitoring (QCM-D) was used to estimate the heterogeneous affinity binding constant of L-Tym towards gold surfaces functionalized by different alkylthiol-terminated Apt23 (the latter were prepared according to a standard protocol in such a way to generate a mixed self-assembled monolayer based on 6-mercaptohexanol as a diluent and on the desired hexanethiol-terminated aptamer).36 The studied sequences are reported in Table 2.

30 7 25

3’ modification

5’ modification HS-C6H12-

-C6H12-SH -T6-C6H12-SH -T6-C6H12-SH

5

15

4

10

3

2

-1

-3

2

5

1 0 20

25

30

35 40 Aptan

45

50

Figure 2. Variation of (●) kon and (■) koff values for Apt n-mer (with n varying from 23 to 49); and kon values for (●) Apt23-26A-3’ and (▲) Apt5’-2326A-3.

higher than the dissociation constant determined in homogeneous solution, the better affinity for the 3’-end over the 5’end anchored aptamer again underlines an anisotropic contribution of the aptamer to the binding process. These results highlight the importance of paying attention to immobilization strategies when dealing with a heterogeneous binding. Note also that these results show that removing the aptamer away from the surface with extra bases is not necessarily sufficient for recovering a similar affinity binding than in homogeneous solution (as it was previously suspected by Chang et al23). Characterization of the binding reaction kinetics by rotating droplet electrochemistry. To gain a more dynamical picture of the binding mechanism as well as to investigate the influence of Mg2+ on the recognition mechanism, binding kinetics were next investigated at room temperature (ca. 20°C). For such purpose, we took advantage of a kinetic rotating electrochemical method that we have recently developed for monitoring the reaction progress kinetics of a binding reaction, and in particular the specific binding of L-Tym to its 49-mer aptamer.26 Briefly, the methodology is based on the fast injection and mixing of a reactant solution (1–10 µL of the aptamer) in a reaction droplet (50 µL of the target) rapidly rotated over the surface of a nonmoving working electrode and on the recording of the ensuing transient faradaic current decrease associated to the specific binding of the aptamer to the target. The simple bimolecular reaction scheme given by Eq. 3 was at first considered.

Table 2. Kd of the different Apt23 modified sequences extract from QCMD measurements.

Aptamer Apt 1 Apt 2 Apt 3 Apt 4

6

20

-1

-1

10 kon (L.mol .s )

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


10 koff (s )

Page 5 of 10

(µM) nd nd 80 ± 3b 3.4 ±0.6b

When the aptamer was grafted to the gold surface directly via a hexanethiol anchoring group coupled either to the 3’-end or 5’-end of Apt23 (i.e., Apt1 and Apt2), no QCM frequency change could be observed during successive injections of LTym (Figure S8). One possible explanation is that the coimmobilized 6-mercaptohexanol in the mixed layer hinders the availability of the first bases positioned closest to the surface (i.e., C1 for Apt1 or T23 for Apt2, two essential bases for the binding of L-Tym). In the case of Apt3 and Apt4 (which both contain a T6 spacing sequence between the aptamer and the anchoring alkylthiol group), increasing concentrations of LTym leads to a significant change of both QCM frequency and energy dissipation (Figure S7a). As a control, addition of DTym showed no variation of the QCM signal (data not shown). The steady-state QCM frequency change recorded at equilibrium for each injected L-Tym concentration can thus be used to estimate the dissociation constant based on a classical Langmuir isotherm (Figure S7b). For the gold surface functionalized with Apt4 (i.e. anchored through the 3’-end), an apparent surface dissociation constant = 3.4 ± 0.6 µM was determined, while for the Apt3 (i.e., anchored through the 5’-end) a 20-fold higher value was inferred (i.e., 80 ± 3 µM). Though these values are about two orders of magnitude

L-Tym

+ D-Apt49

kon → ← koff

L-Tym/D-Apt49

(3)

From the global fit (see SI for details) of the transient current responses recorded at different aptamer concentrations, values of kon= (2.2 ± 0.1) × 104 M-1 s-1 and koff = 0.02 ± 0.01 s-1 were obtained for the Apt23 (see SI for details).26 From kinITC-ETC experiments, kon and koff values of (5.5 ± 0.7) × 104 M-1 s-1 and 0.018 ± 0.002 s-1 were determined, respectively. From the koff/kon ratio, a Kd of 0.7 µM can be calculated which is similar to the one obtained by cyclic voltammetry under nonhydrodynamic conditions. Applying the same kinetic rotating electrochemical method to Apt49 (data not shown) lead to kon and koff values of (5.6 ± 0.1) × 103 M-1 s-1 and 0.02 ± 0.01 s-1, respectively. Kinetic rate constants were also systematically determined for the different truncated aptamers that were used to plot the

5


25

12 10 8 6 4 2 0

-1 -2

20 15 10 0

10 2+

20

[Mg ] (mM)

30

0

10

20

Page 6 of 10

decreased. This behavior can be rationalized in the framework of a square-scheme mechanism (Scheme S2), with the assumptions that (i) only one Mg2+ can bind per aptamer and (ii) the reaction with Mg2+ is fast regarding to the ligand binding kinetics, so fast that the Mg2+ binding to the aptamer or aptamer/target complex is always at equilibrium. Kinetics constants for the magnesium-free binding, , and , , are 9000 mol-1 L s-1 and 0.11 s-1, respectively. For the magnesium-aided binding reaction, , and , , are 30000 M-1 s-1 and 0.02 s-1, respectively. Finally, the magnesium equilibrium constants for !"",#$ the target-free (,!"",#$ ) and target-bound (,

) aptamer are 3 mM and 0.16 mM, respectively, allowing to conclude that the apparent affinity of the Mg2+ is increased in the presence of L-Tym. The Mg2+ concentration is therefore an important parameter on which one can play with to finely tune the binding strength of the aptamer to the target, with a rather good predictive capacity.

koff,app (10 s )

-1 -1 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

kon,app ( 10 M s )


30

2+

[Mg ] (mM)

Figure 3. kon,app (left) and koff,app (right) values for Apt23 at different Mg2+ concentrations (0, 0.25, 1, 10, 20, and 30 mM) The ionic strength was kept constant by adjusting the NaCl concentration. The plain curves represent the fitting according to eqs S13 to S15.

graph in Figure 1C, including the aptamers Apt23-26A-3' and Apt5'-26A-23. Figure S10 shows the time-course current change for four different truncated aptamers. From fits of these experimental data to the general expression describing the kinetics of a reversible binding reaction, it was possible to extract the kinetic rates for each of the truncated aptamers as reported in Figure 2. The results show that only kon is significantly affected by the change in the aptamer sequence since the stepwise truncation of the aptamer from Apt49 to Apt23 causes a fourfold increase of kon (0.56 × 104 to 2.2 × 104 M-1 s-1) while koff remains nearly constant (ca. 0.02 s-1). Note that the same trend was obtained from kinITC-ETC analysis, showing a three-fold increase of kon (1.6 × 104 to 4.7 × 104 M-1 s-1) when going from Apt49 to Apt23, but no significant change of koff (ca. 0.01 s-1) (see Figures S2 to S4). The variations of kon/koff agree with the observed change of Kd in Figure 1C. Moreover, the absolute values of kon and koff were found to be independent on the nature of the extra 26-mer sequence protruding from position 23 of the aptamer. This is supported by the fact that a 26-mer polydA tail extension on the 5’-end of Apt23 (Apt5'-26A-23) presents similar kon and koff rate constants than Apt49. On the contrary, the same kon values obtained for both the MBS and MBS elongated by a poly(A26) tail at its 3’-end demonstrated that aptamer elongation from 3’-end does not impact the binding kinetics. On account of these results, it is legitimate to assume that the binding process is largely controlled by the association rate which tends to substantially increase as the 5’-end of the Apt49 is progressively truncated up to obtain the Apt23. Lastly, we decided to better evaluate the impact of the divalent cation Mg2+ on the binding kinetics of the recognition. This cation has been previously considered as essential in the aptamer binding process, acting both on the stabilization of the aptamer secondary structure and on its conformational switching dynamics upon binding.3,4,37-39 For example, Mg2+ is required for the binding of theophylline to its aptamer, showing a Kd of ca. 300 nM in the presence of 10 mM Mg2+. This aptamer still binds theophylline in the absence of divalent metals ions, but with a 104-fold lower affinity.14,40 The same trend was observed for adenosine.41 In the case of a nonprestructured aptamer, it is interesting to understand the influence of such a divalent cation on the thermodynamic and kinetic binding properties. For that purpose, the rate constants were determined at various Mg2+ concentrations. As shown in Figure 3, Apt23 exhibits an asymptotic increase of kon with the Mg2+ concentration, while koff is in contrast asymptotically

DISCUSSION As depicted in Scheme 2, two main binding mechanisms can be proposed, a priori, for the recognition process between the small L-Tym molecule and its aptamer (we neglect here the role of Mg2+). To a certain extent, these two limiting binding mechanism can be identified to the well-known induced-fit and conformational search models extensively used to describe folding of natural riboswitches.42-44 In the first one (Scheme 2A), the aptamer folds in the right conformation before the target binds and then takes part in a reversible interaction with the as pre-structured aptamer. In the second one (Scheme 2B), binding of the target occurs at first at some position on the non-structured aptamer which then promotes an ensuing induced-fit folding of the aptamer around the target. Here, the minimal binding sequence was determined to be a 23-nucleotides long sequence that results from the truncation of 26 bases on the 5’-end of the initial 49-mer sequence. This relatively small size aptamer has a low probability to generate stable pre-structured conformations at room temperature and under our buffer condition. Hence, the conformational capture-type mechanism is a priori unlikely compared to an induced-fit process wherein the contact between the target and the aptamer induces a complete folding and restructuration of the nucleic acid around the target. It is also reasonable to consider that every process in which the aptamer is undergoing a huge organization, i.e. a transition from a disordered state to a more organized one, will start by a localized single step that can be termed as a nucleation. Interestingly, our results in Figure 1 showed that a protruding oligonucleotide tail on the 5’-end of the MBS Apt23 is to some extent detrimental to the binding process while addition of extra bases on the 3’-end of Apt23 did not change the aptamer binding affinity and kinetics. This contrasting behavior reveals a lack of symmetry on the role and implication of the two extremities of the MBS aptamer in the binding process. It also suggests the occurrence of a nonspecific steric hindrance and/or occlusion at the 5’-end of Apt23 when supplementary bases are added to the 5’-extremity (a nonspecificity supported by the fact that the steric effect is independent on the nature of the protruding sequence). Therefore, the high degree of freedom required at the 5’-extremity Scheme 2. (A) Conformational search vs. (B) induced-fit process

6


Page 7 of 10

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

Analytical Chemistry while a 5'-modifications decrease the binding strength by up to two orders of magnitude. On account of these results, it was next interesting to see how steric hindrance can influence the binding kinetics. Kinetic investigations conducted on aptamer sequence with different lengths revealed an association rate constant kon that significantly increased with the truncation from 49 to 23 bases (i.e., increasing from 5600 to 22000 M-1 s-1, respectively) while the dissociation rate constant koff remains approximately constant. Most of the DNA or RNA aptamers witch have been investigated to date show kon values ranging from 104 to 105 M-1 s-1, whereas koff are within 10-3 to 10-1 s-1.23,25,40,45-46 The values reported here for the anti-L-Tym aptamer lies in the low range, which is probably the consequence of the high energy cost required for subsequent aptamer folding around its target. Interestingly, the variation in kon almost entirely accounts for the difference in the affinity binding with length (Figure 2). This conclusion is in contrast to the results reported with the highly pre-structured anti-cocaine aptamer.25 This anti-cocaine aptamer has a stable secondary/tertiary structure composed of three loops with a cross-shaped junction that clearly defines a target binding pocket. Mutations of the sequence near the binding pocket were shown to alter its stability and shape and therefore to lead in cocaine affinity binding changes. Kinetic analysis of the cocaine/aptamer binding has clearly showed that only koff is impacted by these mutations while kon remains constant. Hence, in the case of a highly pre-structured binding pocket, affinity is predominantly controlled by the target position and adjustment within the pocket than by the entrance rate of the target in the pocket. In our case, this is contrarily the association kinetic constant, kon, that is principally affected by the aptamer modification (here extra protruding bases at the 5’-end) while the dissociation rate constant remains the same, suggesting that the folded state is poorly affected by those extra bases. The sensitivity of kon to the 5’-end modification can be partly understood when considering that there exists here no preconfigured binding pocket and that an organization of the strand, starting by a nucleation process, is required. This nucleation process necessitates the L-Tym/ aptamer system to overcome a given activation barrier that strongly depends on steric hindrance effects, as for instance the presence of extra non-participating bases in the vicinity of the nucleation site that would increase the entropic cost for organization. Also interesting is the steeper transition of the kon value with extrabases than the one observed for affinity constant (compare Figures 1 and 2). The steep transition somehow defines a threshold elongation of 10mer for which it affects the binding through steric constraints. Again, this effect is dissymmetric since strand elongation at the 3’-end does not affect the recognition kinetics (Figure 2). We also examinated the influence of the presence or absence of Mg2+ on the kinetics. L-Tym is recognized in the absence of Mg2+, suggesting that the aptamer can fold properly around its target without divalent cations. The kinetics have been here analyzed in the framework of a four-state model involving the equilibrium binding of aptamer with two partners, either first with Mg2+ followed by L-Tym of first L-Tym followed by Mg2+. From this reaction scheme, kinetics of the elementary binding steps as well as the Mg2+ thermodynamics have been derived, a detailed kinetic analysis that is rarely done. On one hand, increasing amounts of divalent cations permit to speed

of Apt23 for an optimal binding reaction is the consequence of a binding reaction which initiates at the 5’-end of Apt23 sincesteric hindrance is anticipated to be of greater importance near the extremity where the nucleating site is located. Nucleation of the L-Tym/aptamer complex happens near the 5’-end and might involve directly T23. The organization of the aptamer around the target is then probably completed from this nucleus to the other-end of the aptamer in the 5’-to-3’ direction, since C1 is equally essential for the recognition. To further prove that steric hindrance occurs and that there is a dissymmetric role of the two strand ends, we investigated the effect of a bulky substituent attached at the aptamer extremities. For such a purpose, we have determined the affinity binding constants of L-Tym toward the 5’- and 3’-biotinylated Apt23 sequence specifically bounded to avidin. The introduction of a biotin residue at the 5’- and 3’-end of Apt23 was found to have no significant effect on the affinity binding of Apt23 to L-Tym (Table 2), while in the presence of an avidin bound to the aptamer extremity, an homothetic shift the two dissociation constants to higher values was obtained (Table 2). This shift indicates that the bulky protein contributes to additional steric effects in the binding process but without removing the dissymmetric effect between the 5’- and 3’-end of the aptamer in the binding mechanism. We also investigated the influence of a surface immobilization of the aptamer on its binding properties. A T6 spacer between the aptamer and the alkylthiol anchoring group (required for the grafting on the QCM gold surface) was found essential to preserve the aptamer binding capacity. Besides, when the aptamer is immobilized through the 3’-end, the apparent affinity constant is improved by a factor > 20 compared with the one immobilized through the 5’-end. Surfaces represent an extreme case of steric hindrance and the simple addition of few extra nucleotides (here 6 thymines) to distance the aptamer from the surface is therefore not enough to recover an affinity binding similar to the one obtained in homogenous solution. However, it is worth noting that for a MBS removed far from the surface (dT6 spacing), the determined value of the affinity constant are close to the one found for bulkysubstituent-modified aptamer, i.e. avidin-modified, independently of the modified extremity (see Table 2). As a consequence further spacing from the surface would not necessarily leads to a further gain in affinity. We can therefore consider a similar effect of both avidin and surface attachment through a dT6 spacer in term of steric hindrance, a process that limits the binding affinity of the L-Tym to the MBS. Overall, modifications of the 3'-end by a chemical label, a polydA tail, a protein, or a surface does not significantly affect the affinity,

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

up the binding process (from 9000 to 30000 M-1 s-1). This therefore reveals that Mg2+ promotes the oligonucleotide conformation needed to nucleate the complex. On the other hand, variation of dissociation rate constant with Mg2+ concentration (from 0.11 to 0.02 s-1) highlights the stabilizing role of the divalent cation. All of the above results are in line with a binding mechanism where there is no preconfigured binding site and for which a nucleation step followed by a folding step are taking place. The nucleation step, in the case of the Apt23, would occur near the 5’-end and is promoted by the presence of Mg2+. Then the aptamer would fold in a preferential pathway around the target and this folding step is directional from the 5’-end to the 3’end. An interesting question that rises at that point is the influence of the chiral nature of our target in the folding process. It has already been shown that this recognition is highly enantioselective.19-21 From a general point of view, such recognition could either be due to the chiral nature of a binding pocket or to the chiral nature of the oligonucleotide folding around the target. We may imagine, in our case, since there is no predefined binding pocket that the directional folding around the chiral target may be at the origin of the enantioselective recognition. We previously showed that the three main points of interactions of the L-Tym are the amide, the amine and the phenolic group.27,47 Those are likely the three interaction points required for enantioselective recognition as proposed by the model of enantioselective binding from Easson and Steadman.48 On one hand, the two firsts are essential in the recognition since their alteration lead to the complete loss of binding ability. On the other hand, considering the fact that Lphenylalaninamide can be recognized19 and the fact that the phenolic –OH group of the L-Tym can be functionalized without drastic consequence on the binding,27 it can be stated to some extent that the interaction of the aptamer with the -OH group is not essential to the binding. This suggests that this OH group does not participate in the activating process of the binding that is the nucleation process. We may therefore hypothesize that nucleation only involves both the amide and amine functional groups and that the subsequent folding occurs in a specific and directional way, from 5’- to 3’-, around the phenyl function. If the nucleation involves only two functional groups of the L-Tym, this 2-points interaction is not sufficient to selectively favor the recognition of one enatiomeric form. In that case, this is the organization around the phenolic group that would induce the enantioselection. Otherwise the nucleation itself would be the selective step. Further investigations are required to fully answer this issue and progress will be reported in due course.

Page 8 of 10

interesting in the case of oligonucleotides harvested from natural biological systems. Besides, this is a method of choice since the signal change does not depend on conformational change of the aptamer during binding or on a detailed structural knowledge of the complex. Therefore, this method can be easily extended to other ligand-bioreceptor binding reactions as soon as the molecular weight of the two partners is significantly different. Our approach allows discussing on two major issues: mechanism elucidation and practical information about appropriate aptamer implementation (labeling, immobilization,…) for applications. From a mechanistic point of view, the enantiospecific binding reaction occurs through an induced-fit mechanism. The minimal binding sequence of the anti-L-Tym aptamer was easily determined. Analysis of the binding constants for a range of functionalized minimal binding sequences has highlighted a lack of symmetry in the role of the two strand-ends. Functionalization at the 3'-end by a chemical label, a polydA tail, a protein, or a surface did not significantly affect the binding affinity and this position should therefore be preferred to immobilize the sequence on a surface for biosensors development. On contrary, any modification of the 5'-end position influences the kinetic/thermodynamic constants up to two orders of magnitude in the extreme case of a surface immobilized aptamer. The reason is that, steric hindrance must be overcome to nucleate the binding complex in the presence of a modification near the nucleation site. The process leading to the formation of the aptamer/L-Tym complex is in conclusion somehow directional from the 5’- to the 3’-end. This trend was confirmed from the determination of the association rates, which were shown to be preponderant in the binding affinity. The dissymmetry in the binding process between the two opposite extremities is important in the development of optimal transduction strategies and applications. Therefore, it is clear that any modification for labeling or surface immobilization has to be preferentially performed through a favorite extremity (here 3’-end) in order to avoid affinity or kinetic losses.

ASSOCIATED CONTENT Supporting Information. Synthesis protocols, full aptamer sequences, data analysis for thermodynamic and kinetic constant extraction, additional figure for ITC, FP, QCM-D and electrochemistry.

AUTHOR INFORMATION Corresponding Author [email protected], [email protected].

CONCLUSION Determining the binding thermodynamic and kinetic characteristics between aptamers and small ligands for a wide range of conditions is important for understanding their recognition mechanisms as well as for developing practical applications. The simple electrochemical method proposed here allows the screening of different aptamer sequences in a very short time, with good relevance compared with more demanding methods as fluorescence polarization and ITC. The electrochemical technique offers the possibility to work with small volumes, reducing the cost of biological material. This is particularly

Note ‡

These authors contribute equally to this research.

ACKNOWLEDGMENT This work was supported by Agence Nationale pour la Recherche (ANR ESCTASE project).

8


Page 9 of 10

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

Analytical Chemistry (25) Bao, J.; Krylova, S.; Reinstein, O.; Johnson, P.; Krylov, S. Anal. Chem., 2011, 83, 8387 (26) Challier, L.; Miranda-Castro R.; Marchal, D.; Noël, V.; Mavré, F.; Limoges, B. J Am Chem Soc. 2013, 135, 14215. (27) Moreau, J.; Challier, L.; Lalaoui, N.; Mavré, F.; Noël, V.; Limoges, B.; Schöllhorn, B.; Fave, C. Chem. - Eur. J 2014, 20, 2953. (28) Kubik, M. F.; Stephens, A. W.; Schneider, D.; Marlar, R. A.; Tasset, D. Nucleic Acids Res. 1994, 22, 2619. (29) Kaur, H.; Yung, L.-Y. L. PLoS ONE 2012, 7, e31196. (30) Le, T. T.; Chumphukam, O.; Cass, A. E. G. RSC Advances 2014, 4, 47227. (31) Zhu, Z., Schmidt, T; Mahrous, M.; Guieu, V.; Perrier, S.; Ravelet, C.; Peyrin, E. Anal Chim Acta. 2011, 707, 191. (32) Jung, C.; Ellington, A.D. Acc. Chem. Res. 2014, 47, 1825. (33) Dumas, P.; Ennifar, E.; Da Veiga, C.; Bec, G.; Palau, W.; Di Primo, C.; Piñeiro, A.; Sabiin, J.; Eva Muñoz, E.; Javier Rial, J. Extending ITC to kinetics with kinITC in Methods in Enzymology Vol 567 (2016) Chap. 7. Ed. Andrew Feig. (34) Po-Hsun, L.; Shih-Lun, Y.; Ming-Shen, L.; Yung, C.; Selva Roselins, L.; Akon, H.; Wen-Yih, C. J. Phys. Chem. B 2008, 112, 6665. (35) Olsson, T. S.; Ladbury, J. E.; Pitt, W. R.; Williams M. A. Protein Sci. 2011, 20, 1607. (36) Xiao, Y.; Lai, R.; Plaxco, K. Nat. Prot. 2007, 2, 2875. (37) Buck, J.; Noeske, J.; Wohnert, J.; Schwalbe, H. Nucleic Acids Res. 2010, 38, 4143. (38) Buck, J.; Wacker, A.; Warkentin, E.; Wohnert, J.; WirmerBartoschek, J.; Schwalbe, H. Nucleic Acids Res. 2011, 39, 9768. (39) Gebhardt, K.; Shokraei, A.; Babaie, E.; Lindqvist, B. H. Biochemistry 2000, 39, 7255. (40) Jucker, F. M.; Phillips, R. M.; McCallum, S. A.; Pardi, A. Biochemistry 2003, 42, 2560. (41) André, C.; Xicluna A., Guillaume, Y.-C. Electrophoresis 2005, 26, 3247. (42) Da Costa, J. B.; Andreiev, A. I.; Dieckmann, T. Biochemistry 2013, 52: 6575. (43) Förster, U.; Weigand, J.E.; Trojanowski, P.; Suess, B.; Watchtveilt J. Nucleic Acid Res., 2011, 22, 1. (44) Furtig, J.; Noeske, B.; Wbhnert, J.; Schwalbe Ens, Schwalbe, H. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 15699-15704. (45) Wickiser, J. K.; Cheah, M. T; Breaker, R. R; Crothers, D. M. Biochemistry 2005, 44, 13404. (46) Gilbert, S.D.; Stoddard, C. D.; Wise, S. J.; Batey, R. T., J. Mol. Biol. 2006, 359, 754. (47) Souard, F.; Perrier, S; Noel, V; Fave, C.; Fiore, E.; Peyrin, E.; Garcia, J.; Vanhaverbeke, C. C. Chem. - Eur. J 2015, 21, 15740. (48) Easson, L.H.; Stedman, E. Bio-chem. J. 1933, 1257

REFERENCES (1) Zhang, J.; Lau, M. W.; Ferre-D’Amar, A. R. Biochemistry 2010, 49, 9123. (2) Serganov, A.; Patel, D. J. Curr. Opin. in Struct. Biol. 2012, 22, 279. (3) Reining, A.; Nozinovic, S.; Schlepckow, K.; Buhr, F.; Furtig, B.; Schwalbe, H. Nature 2013, 499, 355. (4) Holmstrom, E. D.; Polaski, J. T.; Batey, R. T.; Nesbitt, D. J. J. Am. Chem. Soc. 2014, 136, 16832. (5) Klussmann, S. (2006) (Editor), The Aptamer Handbook: Functional Oligonucleotides and Their Applications, Wiley. (6) Mascini, M.; Palchetti, I.; Tombelli, S. Angew. Chem. Int. Ed. 2012, 51, 1316. (7) McKeague, M.; DeRosa, M. C. J. Nucleic Acids 2012, 2012, 748-768 (8) Keefe, A. D.; Pai, S.; Ellington, A. Nat. Rev. Drug. Discov. 2010, 9, 537. (9) Fang X., Tan W. Acc. of Chem. Res. 2010, 43, 48. (10) Cho, E. J.; Lee, J.-W.; Ellington, A. D. Annual Rev. of Anal. Chem. 2009, 2, 241. (11) Hayat, A.; Marty, J. L.; Front in Chem. 2014, 2, 41. (12) Zhou, W.; Huang, P-J. J.; Ding, J.; Liu, J., Analyst 2014, 139, 2627. (13) Stojanovic, M. N.; de Prada, P.; Landry, D. W. J. Am. Chem. Soc., 2001, 123, 4928. (14) Zimmermann, G. R.; Jenison, R. D.; Wick, C. L.; Simorre, JP.; Pardi, A. Nature Struct. Biol. 1997, 4, 644. (15) Huizenga, D. E.; Szostak, J. W. Biochemistry 1995, 34, 656. (16) Macaya, R. F.; Schultze, P.; Smith, F. W.; Roe, J. A.; Feigon, J. Proc. Natl. Acad. Sci. U. S. A. 1993, 90, 3745. (17) Buck, J.; Fuortig, B.; Noeske, J.; Wohnert, J.; Schwalbe, H. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 15699. (18) Vianini, E.; Palumbo, M.; Gatto, B. Bioorg. Med. Chem. 2001, 9, 2543. (19) Ruta, J.; Perrier, S.; Ravelet, C.; Fize, J.; Peyrin, E. Anal Chem. 2009, 81, 7468. (20) Challier, L.; Mavré, F.; Moreau, J.; Fave, C.; Schollorm, B.; Marchal, D.; Peyrin, E.; Noel, V.; Limoges, B. Anal. Chem. 2012, 84, 5415. (21) Feagin, T. A.; Olsen, D. P. V.; Headman, Z. C.; Heemstra, J. M. J. Am. Chem. Soc. 2015, 137, 4198. (22) Xiao, H.; Edwards, T. E.; Ferré-D’Amaré A. R. Chem. Biol. 2008, 15, 1125. (23) Chang, A. L.; McKeague, M.; Liang, J. C.; Smolke C. D. Anal. Chem. 2014, 86, 3273. (24) Burnouf, D.; Ennifar, E.; Guedich, S.; Puffer, B.; Hoffmann, G.; Bec, G.; Disdier, F.; Baltzinger, M.; Dumas, P. J. Am. Chem. Soc. 2012, 134, 559-565

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 10

Insert Table of Contents artwork here


10

Multianalytical Study of the Binding between a ... - ACS Publications

Recommend Documents