DNA with Parallel Strand Orientation: A Nanometer Distance Study

Article pubs.acs.org/JPCB

DNA with Parallel Strand Orientation: A Nanometer Distance Study with Spin Labels in the Watson−Crick and the Reverse Watson−Crick Double Helix Dorith Wunnicke,∥,† Ping Ding,∥,‡,§ Haozhe Yang,‡,§ Frank Seela,*,‡,§ and Heinz-Jürgen Steinhoff*,† †

Department of Physics and ‡Department of Chemistry, University of Osnabrück, Barbarastrasse 7, 49069 Osnabrück, Germany Laboratory of Bioorganic Chemistry and Chemical Biology, Center for Nanotechnology, Heisenbergstrasse 11, 48149 Münster, Germany

§

S Supporting Information *

ABSTRACT: Parallel-stranded (ps) DNA characterized by its sugar− phosphate backbones pointing in the same direction represents an alternative pairing system to antiparallel-stranded (aps) DNA with the potential to inhibit transcription and translation. 25-mer oligonucleotides were selected containing only dA·dT base pairs to compare spin-labeled nucleobase distances over a range of 10 or 15 base pairs in ps DNA with those in aps DNA. By means of the copper(I)-catalyzed Huisgen−Meldal−Sharpless alkyne−azide cycloaddition, the spin label 4-azido-2,2,6,6-tetramethylpiperidine-1-oxyl was clicked to 7ethynyl-7-deaza-2′-deoxyadenosine or 5-ethynyl-2′-deoxyuridine to yield 25mer oligonucleotides incorporating two spin labels. The interspin distances between spin labeled residues were determined by pulse EPR spectroscopy. The results reveal that in ps DNA these distances are between 5 and 10% longer than in aps DNA when the labeled DNA segment is located near the center of the double helix. The interspin distance in ps DNA becomes shorter compared with aps DNA when one of the spin labels occupies a position near the end of the double helix.

■

INTRODUCTION

2′-Deoxyribonucleic acid (DNA) is the carrier of genetic information in living organisms. This information is stored in form of an antiparallel stranded (aps) duplex DNA stabilized by Watson−Crick base pairing; however, DNA is known to adopt a wide range of conformations. Parallel-stranded (ps) DNA represents a unique DNA structure characterized by its sugar− phosphate backbones pointing in the same direction (Figure 1a). Such a chain orientation also exists in triplexes, Gquadruplexes, and i-motif DNA;1−4 however, in these structures the ps alignment is caused by Hoogsteen or other base pair motifs, which are entirely different from those in aps DNA. The ps DNA duplex structure was originally suggested by Pattabiraman5 and realized later by Jovin,1 Seela,6−8 and others.9−12 DNA with parallel strand orientation is remarkably stable, although less stable than its antiparallel congener,1 due to weaker stacking interactions and the weaker hydrogen bonding of the dA·dT reverse Watson−Crick (Donohue) base pair (Figure 1b). Spectroscopic properties, enzymatic recognition, as well as drug-binding properties of ps DNA differ from that of native DNA with antiparallel chains.13−15 Furthermore, it has been reported that ps DNA is less hydrated than the corresponding aps DNA.16 ps DNA is not a substrate for many cellular nucleases or restriction enzymes.17 Thus, parallel hybridization results in enhanced life time, making this approach interesting for antisense therapy. Antisense oligonu© 2015 American Chemical Society

Figure 1. (a) Motifs of strand orientation. (b) dA·dT base pair in the Watson−Crick mode (motif I) for aps DNA and in the reverse Watson−Crick mode (motif II) for ps DNA.

Special Issue: Wolfgang Lubitz Festschrift Received: March 27, 2015 Revised: June 26, 2015 Published: June 29, 2015 13593

DOI: 10.1021/acs.jpcb.5b02935 J. Phys. Chem. B 2015, 119, 13593−13599

Article

The Journal of Physical Chemistry B

spectrometry as well as enzymatic hydrolysis. Melting curves were measured with a Cary-100 Bio UV−vis spectrophotometer equipped with a Cary thermoelectrical controller. The temperature was measured continuously in the reference cell with a Pt-100 resistor with a heating rate of 1 °C min−1. The thermodynamic data of duplex formation were calculated by the Meltwin 3.0 program. Pulse EPR Experiments. For all pulse EPR experiments, 30−40 μL of the sample solution in 0.1 M NaCl, 10 mM MgCl2 (pH 7.0) with a DNA concentration of 60 μM was used. 10% glycerol was added as cryoprotectant. Pulse EPR measurements were performed at 50 K and X-band frequencies (∼9.4 GHz) using a Bruker Elexsys 580 spectrometer. The four-pulse PELDOR sequence25 was applied, and for orientation-selective experiments the observer frequency was varied to achieve a 40, 65, and 80 MHz frequency offset between observer and pump pulses. All other parameters were kept unmodified with respect to previously specified values.23 For analyzing the experimentally determined time traces the three orientation selective PELDOR traces corresponding to the respective spin-labeled oligonucleotide were subsequently summed up (orientation averaging)26 and analyzed by means of using Tikhonov regularization (DeerAnalysis2008).27 Interspin distance distributions were calculated by fitting the background-corrected summed dipolar evolution function. Interspin Distance Modeling and Calculation. Molecular models for aps and ps DNA with spin-labeled nucleotides were set up using YASARA dynamics.23,28 Parameters for ps DNA were obtained by adjusting the implemented parameters for aps DNA according to published data.11 The radial distances, ρ, of the nitroxide groups with respect to the helix axes were determined for different rotamers of the spin label side chains, yielding values between 1.0 and 1.2 nm for aps DNA and between 1.0 and 1.3 nm for ps DNA. (See Figure 5a.) To calculate the distance between the nitroxide groups of the spin labels as a function of the number, N, of base pairs separating the spin labels, we assume a simple geometrical model with a straight helix. Here the positions of the spins of the two nitroxide groups are represented by two points given in cylindrical coordinates, P1(ρ, ϕ1, z1) and P2(ρ, ϕ2, z2), with radial distance ρ from the z axis, azimuth angle, φi, and height, zi. The distance, r, between the spins is then given by the rise per base pair, Δz, the twist angle per base pair, Δϕ, and N according to

cleotides bind to RNA with antiparallel chain orientation and inhibit protein synthesis. Parallel hybridization expands this application by an alternative pairing system and has the potential of applications in antisense technology and other fields of biotechnology. Theoretical and NMR studies have shown that the d(A·T)based ps DNA forms a helix with a number of properties similar to aps DNA, while others are different.12,18 The orientation of the nucleobases in ps DNA is anti and the sugar conformation is similar to aps DNA. The helix sense of ps DNA is righthanded. MD simulations based on NMR constraints revealed backbone torsional angles similar to those found in aps DNA and similar helix diameters. In ps DNA, the two grooves have almost identical widths of 8 to 9 Å, whereas significantly different widths are found in aps DNA for the minor (6 Å) and major (12 Å) grooves.12 Of high importance are the significant differences found for the helicoid parameters (rise, twist, propeller twist, tip) of ps and aps DNA, consequently leading to different interactions among the stacked base pairs.12 Recently, it was shown by our laboratory that modifications performed on the sugar moiety of ps DNA have a negative impact on the stability.19 In aps DNA, 2′-fluoro substituents lead to a strong duplex stabilization, while the same modification performed on ps DNA leads to destabilization. When all sugar residues were modified, duplexes could not be formed. A similar observation was made on RNA. It was not possible to obtain stable RNA duplexes with parallel strand orientation.19,20 Taken together, the existing knowledge of DNA or RNA structures with parallel chain orientation in solution is unsatisfactory. NMR spectroscopy does not allow the direct measurement of helicoidal parameters.12 Instead, these parameters have to be deduced from the interproton distances of nucleotide nearest neighbors. The accuracy of such measurements has to be very high to determine the overall structure of the DNA molecule. Spectroscopic techniques such as EPR spectroscopy in combination with site-directed spin labeling and Förster resonance energy transfer (FRET) spectroscopy provide direct information about inter- and intramolecular distances on the nanometer scale but require the introduction of a reporter group. The high sensitivity and accuracy of pulse electron double resonance (PELDOR) or double electron−electron resonance (DEER) for nanometer distance measurements on spin-labeled oligonucleotides have been demonstrated.21−24 In the present study, two nitroxide spin labels were covalently linked in a certain distance to nucleobase residues. To this end, the 5-position of 2′-deoxyuridine and the 7-position of 2′deoxy-7-deazaadenosine were selected. These are appropriate modification sites that should not perturb the structure of aps and ps DNA. Here we report on the first nanometer distance study performed on spin-labeled ps DNA, and we correlate the obtained data with those of a corresponding aps DNA using EPR spectroscopy.

r=

2 ⎛ ⎛ (N − 1) ·Δφ ⎞⎞ ⎟⎟ ((N − 1) ·Δz) + ⎜2ρ sin⎜ ⎝ ⎠⎠ 2 ⎝ 2

Distances were calculated for two different sets of helical parameters Δz and Δϕ: The first set was chosen according to the average values found for crystallized aps B-DNA,29 Δz = 0.332 nm, Δϕ = 35.4°, and radial distance ρ = 1 nm; the second set was chosen to fit the present experimental data for ps DNA, Δz = 0.35 nm, Δϕ = 36°, and ρ = 1 and 1.3 nm.

■

■

EXPERIMENTAL METHODS Synthesis and Characterization of Oligonucleotides. The protocols of synthesis and characterization of oligonucleotides and the functionalization of oligonucleotides with 4-azido2,2,6,6-tetramethylpiperidine-1-oxyl (2) using Huisgen−Meldal−Sharpless alkyne−azide cycloaddition (CuAAC) were previously reported.23 The oligonucleotides were purified by reversed-phase HPLC and characterized by MALDI-TOF mass

RESULTS AND DISCUSSION The copper(I)-catalyzed Huisgen−Meldal−Sharpless alkyne− azide cycloaddition (CuAAC), the so-called “click” reaction, was found to be an efficient method for spin labeling of nucleic acids.23 By this means, the (4-azido-2,2,6,6-tetramethylpiperidine-1-oxyl; 4-azido TEMPO) spin label 2 was clicked to 7ethynyl-7-deaza-2′-deoxyadenosine (1) or 5-ethynyl-2′-deoxy13594


Article


Figure 2. Structures of ethynyl-substituted nucleosides and 4-azido TEMPO conjugates.

ization of 6 with 7 yielded a ps duplex, while hybridization of 6 with 8 gave the aps duplex. By the same methodology, oligonucleotide duplexes with different chain orientation containing two spin-labeled 7-deaza-2′-deoxyadenosine residues or two spin labeled 2′-deoxyuridine residues were prepared. Consequently, both modifications are present in the same strand, while the second strand is unmodified (Table 1). This strategy is possible because dT as well as the spinlabeled derivatives show ambiguous base pairingWatson− Crick pairing for aps DNA and reverse Watson−Crick pairing for ps DNA (Figure 1b, motifs I and II). Hence, the chain orientation of DNA is controlled by the selection of a particular sequence motif.9,34 Melting experiments were carried out to evaluate duplex stabilities of spin-labeled and unlabeled duplexes with ps and aps chain orientation. (For Tm values, see Table 2 and

uridine (4) to generate nucleoside conjugates 3 and 5, respectively (cf. Figure 2).23 For the present study, a series of 25-mer oligonucleotides (Table 1) containing only dA and dT residues were synthesized Table 1. Oligonucleotides With and Without Spin Labels Used for the Construction of aps and ps Duplex DNA and Their Molecular Masses Determined by MALDI-TOF Mass Spectrometry.a [M-1]− (Da) oligonucleotide

calc.

found

5′-d(AAAAAAAAAATAATTTTAAATATTT)-3′ (6) 5′-d(TTTTTTTTTTATTAAAATTTATAAA)-3′ (7) 3′-d(TTTTTTTTTTATTAAAATTTATAAA)-5′ (8) 5′-d(AAAAAAAA1ATAATTTT1AATATTT)-3′(9) 5′-d(AAAAAAAAA1TAATTTTA1ATATTT)-3′(10) 5′-d(AAAAA1AAAATAATTTTAA1TATTT)-3′(11) 5′-d(AAAAAAAAAATAA4TTTAAATA4TT)-3′(12) 5′-d(AAAAAAAA3ATAATTTT3AATATTT)-3′(13) 5′-d(AAAAAAAAA3TAATTTTA3ATATTT)-3′(14) 5′-d(AAAAA3AAAATAATTTTAA3TATTT)-3′(15) 5′-d(AAAAAAAAAATAA5TTTAAATA5TT)-3′(16)

7687 7624 7624 7732 7732 7732 7076 8127 8127 8127 8101

7687 7624 7624 7731 7733 7732 7078 8128 8128 8128 8102

Table 2. Tm Values of Antiparallel-Stranded and ParallelStranded DNA and Interspin Distance Values Determined by Pulse EPR Experiments

Determined as [M-1]− in the linear negative mode with an Applied Biosystems Voyager DE PRO spectrometer using 3-hydroxypicolinic acid (3-HPA) as matrix.

a

on solid support employing phosphoramidite chemistry. The sequence pattern was originally reported by Jovin.30 dA·dT rich oligonucleotide duplexes have characteristic local conformations slightly different from DNA containing all four canonical nucleobases;31,32 however, they readily form ps duplexes without introducing further modifications. Oligonucleotides 9−12 incorporating the clickable nucleoside residues 1 and 4 were synthesized; the corresponding phosphoramidites have been previously reported.23,33 In all cases, two 4-azido TEMPO spin labels (2) were simultaneously introduced into the single-stranded oligonucleotides 9−12 by the “click” reaction,23 resulting in three single-stranded oligonucleotides containing spin labels linked to 7-deaza-2′deoxyadenosine (13−15) and one single-stranded oligonucleotide containing spin labels linked to 2′-deoxyuridine (16). The spin-labeled nucleotides are separated by 8 or 13 base pairs (the number of base pairs including the two spin labeled ones resulting in 10 or 15 bp, respectively). The modified oligonucleotides were purified by HPLC and characterized by MALDI-TOF mass spectra as well as by enzymatic hydrolysis (Table 1 and Supporting Information). The unlabeled duplexes with parallel or antiparallel chain orientations were constructed according to Jovin.30 Hybrid-

DNA duplex

Tm (°C)a

number of base pairs

aps DNA 13·8 ps DNA 13·7 aps DNA 14·8 ps DNA 14·7 aps DNA 15·8 ps DNA 15·7 aps DNA 16·8 ps DNA 16·7

52 39 52 39 52 39 53 40

10 10 10 10 15 15 10 10

r (nm)b

w (nm)c

± ± ± ± ± ± ± ±

0.5 0.7 0.5 0.7 0.6 (0.5)d 0.6 0.7

3.01 3.16 3.12 3.30 4.92 5.44 3.03 2.86

0.01 0.01 0.02 0.01 0.05 0.10 0.03 0.02

a

Measured at 260 nm in 0.1 M NaCl, 10 mM MgCl2, and 10% glycerol (pH 7.0) with 5 μM single-strand concentration and 10% excess of the unlabeled complementary oligonucleotide. bError values of interspin distance values, r, were calculated using the validation tool of (DeerAnalysis2008)27 as previously described.26 cError of the distribution width, w (full width at half-maximum), is estimated to be ±10%. dAt the border of the accessible distance range, here for r > 5 nm, distribution widths are underestimated.27

Supporting Information.) Throughout all experiments, aps DNA duplexes exhibit 13−15 °C higher Tm values than the corresponding ps duplexes reflecting a higher stability for the aps DNA duplexes. For both ps as well as aps DNA, the introduction of spin labels decreases the Tm values by 1−3 °C, indicating that the spin labels do not perturb the duplex structure significantly. All melting curves show a monophasic profile demonstrating a one-step transition (see the Supporting Information). The previous results prove that the spin labels are well-accommodated in ps and aps DNA, even though the grooves of ps DNA are different from those of aps DNA.12,18 13595


Article


Figure 3. Spin-labeled dA·dT base pair in the Watson−Crick mode (aps DNA) and the reverse Watson−Crick mode (ps DNA).

label distance is up to 0.4 nm (∼13%) shorter than those of the other ps DNA duplexes of the same base pair distance. Compared with aps DNA, it is shortened by ∼6%. Strikingly, the distance distribution width, w, is not significantly increased for ps duplex 16·7 compared with the other ps DNA duplexes. This can be explained by “breathing” motions with the terminal spin label fluctuating mainly perpendicular to the helix axis. Larger deviations of helicoids parameters at the 3′-end of ps DNA from their average values have also been observed by NMR experiments.12 According to the previously mentioned differences in base sequence, spin label location and orientation the spin label distance data cannot be directly compared with the others. According to Egli et al., in ps DNA the base planes as well as the backbone are positioned perpendicular to the axis,20 and helicoid parameters of base pairs are different.12 Consequently, the adoption of various conformations such as an A-form is impossible. This is supported by the fact that RNA duplexes with ps orientation are extremely unstable. To compare our experimental results with known helicoid parameters the distances between the NO groups of the spin labels were calculated for aps and ps DNA assuming a straight helix conformation (see Experimental Methods). Crystal data for aps DNA yield an average axial rise per base pair of 0.332 nm and an average twist per base pair of 35.4°.29 In addition, the average radial distance between the NO-group and the helix axis has to be considered in the calculations. On the basis of MD simulations23 and molecular modeling of the spin labeled DNA segments, the range of radial distances of the NO group from the helix axis was estimated (Figure 5a). For aps DNA, this distance was found to cover the range between 1 and 1.2 nm, whereas for ps DNA this range extends to 1.3 nm. The comparison of the NO−NO distances shown in Figure 5b reveals that the present experimental distance values for three sets of spin-labeled aps DNA samples (squares) with 4, 10, and 15 base pairs between the spin labels agree with the values calculated with the averaged helicoid parameters of crystallized aps DNA29 (continuous line). This is strong evidence of the

Next, interspin distances of ps and aps DNA were determined using pulse EPR spectroscopy (PELDOR) (cf. Table 2, Figure 4, and Supporting Information). Both aps and ps DNA duplexes incorporating spin labels within a 10 base pair distance show distances around 3 nm, while those within a 15 base pair spacing display distances around 5 nm. The PELDOR spectra reveal a more structured conformation of the double-stranded DNA duplex compared with single-stranded DNA for both aps and ps DNA, as expected from previous studies (see Supporting Information).23,26 The distance distribution widths found for ps DNA are larger by 20 to 30% compared with those for aps DNA, indicating more flexibility for the spin labels in ps DNA. For DNA duplexes incorporating the spin labels at the 7-position of the 7-deaza-2′deoxyadenosine residue, within a 10 and 15 base pair distance, the interspin distance values are increased by about 5−10% for ps DNA compared with aps DNA. Next, the spin labels were linked to the 5-position of the pyrimidine base. Because we want to keep the sequence unaltered, we had to shift the spin-labeled segment from a nearcentral to a near-terminal site when using the spin-labeled dT analogue, in both ps and aps DNA. As a consequence, the more central segments modified with the spin-labeled dA analogue contain five dT and three dA residues, while the more terminal segment modified with the spin-labeled dT analogue contains an equal number of four dA and four dT residues. In addition, the orientation of the spin label linked to the 5-position of the pyrimidine base differs from the one of the spin label linked to the 7-deazapurine moiety, as shown in Figure 3. Because the Tm values of the ps DNA duplexes are significantly lower than those of the aps DNA duplexes and the ends of the helices are “breathing”, the terminal base pairs of the ps duplex 16·7 are expected to be more flexible than in aps DNA. Consequently, the spin-labeled dT residue near the end of the ps helix is probably in a single-stranded state. All of these factors do not influence the spin label distance distribution of aps duplex 16·8 compared with the other aps duplexes of the same base pair distance; however, this is different for ps duplex 16·7. The spin 13596


Article


Figure 5. (a) Left: Schematic representation of spin-labeled aps (top) and ps (bottom) DNA. Parameters for ps DNA were used according to published data.11 The depicted example shows spin labels (red) introduced at positions 9 and 18 in ps and aps DNA using the UCSF Chimera package.35 Right: View along the helix axis of aps DNA; the yellow arrow illustrates the helix axis−NO radial distance, ρ. (b) Experimental and calculated distances between NO groups of the spin labels as a function of the number of base pairs separating the spin labels. Interspin distances calculated with ρ = 1 nm and averaged helicoid parameters (Δz = 0.332 nm, Δϕ = 35.4°) determined from crystallized aps B-DNA29 (black line) agree with the experimental data for aps DNA (squares). The experimental distance value for aps DNA with spin labels separated by four base pairs was taken from Wunnicke et al.26 Distances calculated with Δz = 0.35 nm, Δϕ = 36°, and ρ = 1 and 1.3 nm (lower and upper boundary of the gray area, respectively) fit the experimental distance values for ps DNA (triangles) with the exception of ps DNA 16·7 (circle), where one of the spin labels is close to the helix 3′-end.

Figure 4. Interspin distances determined from PELDOR data for aps and ps DNA. Left: DNA sequences used in the present study with spin-labeled positions (conjugates 3 and 5) highlighted in bold. The number of base pairs between the spin labels (10 or 15) is indicated by bars. Right: Probability density distributions, P(r), of the interspin distances obtained by Tikhonov regularization (DeerAnalysis2008)27 of the PELDOR traces (black, aps DNA; green, ps DNA). The widths of the distributions reflect the degree of structural heterogeneity of the frozen conformers that is related to conformational flexibility under physiological conditions. The values for the interspin distances of maximum probability and for the distribution widths are given in Table 2. (See Experimental Methods and Supporting Information for details).

applicability of PELDOR spectroscopy for accurate distance measurements on spin-labeled DNA. For ps DNA 13·7, 14·7, and 15·7, the experimentally observed distances are increased by between 5 and 10% compared with those of aps DNA (Figure 5b). This cannot be explained by a larger radial distance of the spin label side chain from the helix axis because its value has hardly any influence on the NO−NO distances for 13·7 and 14·7 (see gray area in Figure 5b). The increased distances thus reflect different helicoid parameters for ps DNA. A distance curve calculated with values of 0.35 nm and 36° for the axial rise and twist angle per base pair, respectively, agrees with the experimental data of ps DNA (gray area in Figure 5b). Thus, the axial rise value for the present ps DNA sequences is increased by ∼5% compared with that of aps DNA. Unlike Xray crystallography, NMR does not directly provide values for helicoid parameters, but the values have to be determined from simulated structures based on the NMR data. Such models performed on dodecamer ps DNA duplexes with A:T stretches suggest increased twist angles (38°) compared with aps DNA.12

In contrast with our results, these NMR data indicate smaller average axial rise for ps DNA than that determined for aps DNA; however, axial rise was found to vary over the length of the ps duplex being larger than that for aps DNA at the 3′end.12 Slightly increased sequence-dependent twist angles for A:T-based ps DNA (35.1 ± 0.8°) compared with aps DNA were also observed in a theoretical study, whereas the values for the axial rise were found to be similar ((0.34 ± 0.01) nm).18 The numerical values reported for ps DNA in this theoretical study are in reasonable agreement with our experimental results.

■

CONCLUSIONS Different from NMR measurements on ps DNA, which determine the distance of neighboring nucleobases, the PELDOR EPR technique gives us insight into longer range distances in oligonucleotides. In the present study, pulse EPR spectroscopy is used to study ps DNA. To this end, 25-mer 13597


Article


(2) Frank-Kamenetskii, M. D.; Mirkin, S. M. Triplex DNA Structures. Annu. Rev. Biochem. 1995, 64, 65−95. (3) Burge, S.; Parkinson, G. N.; Hazel, P.; Todd, A. K.; Neidle, S. Quadruplex DNA: Sequence, Topology and Structure. Nucleic Acids Res. 2006, 34, 5402−5415. (4) Guéron, M.; Leroy, J.-L. The i-Motif in Nucleic Acids. Curr. Opin. Struct. Biol. 2000, 10, 326−331. (5) Pattabiraman, N. Can the Double Helix be Parallel? Biopolymers 1986, 25, 1603−1606. (6) Seela, F.; Wei, C. The Base-Pairing Properties of 7-Deaza-2′deoxyisoguanosine and 2′-Deoxyisoguanosine in Oligonucleotide Duplexes with Parallel and Antiparallel Chain Orientation. Helv. Chim. Acta 1999, 82, 726−745. (7) Seela, F.; He, Y.; Wei, C. Parallel-Stranded Oligonucleotide Duplexes Containing 5-Methylisocytosine-Guanine and IsoguanineCytosine Base Pairs. Tetrahedron 1999, 55, 9481−9500. (8) Ming, X.; Ding, P.; Leonard, P.; Budow, S.; Seela, F. ParallelStranded DNA: Enhancing Duplex Stability by the ’G-Clamp’ and a Pyrrolo-dC Derivative. Org. Biomol. Chem. 2012, 10, 1861−1869. (9) Otto, C.; Thomas, G. A.; Rippe, K.; Jovin, T. M.; Peticolas, W. L. The Hydrogen-Bonding Structure in Parallel-Stranded Duplex DNA is Reverse Watson-Crick. Biochemistry 1991, 30, 3062−3069. (10) Yang, X.-L.; Sugiyama, H.; Ikeda, S.; Saito, I.; Wang, A. H.-J. Structural Studies of a Stable Parallel-Stranded DNA Duplex Incorporating Isoguanine:Cytosine and Isocytosine:Guanine Basepairs by Nuclear Magnetic Resonance Spectroscopy. Biophys. J. 1998, 75, 1163−1171. (11) Geinguenaud, F.; Mondragon-Sanchez, J. A.; Liquier, J.; Shchyolkina, A. K.; Klement, R.; Arndt-Jovin, D. J.; Jovin, T. M.; Taillandier, E. Parallel DNA Double Helices Incorporating isoG or m5isoC Bases Studied by FTIR, CD and Molecular Modeling. Spectrochim. Acta, Part A 2005, 61, 579−587. (12) Parvathy, V. R.; Bhaumik, S. R.; Chary, K. V. R.; Govil, G.; Liu, K.; Howard, F. B.; Miles, H. T. NMR Structure of a Parallel-Stranded DNA Duplex at Atomic Resolution. Nucleic Acids Res. 2002, 30, 1500− 1511. (13) Jain, A. K.; Bhattacharya, S. Groove Binding Ligands for the Interaction with Parallel-Stranded ps-Duplex DNA and Triplex DNA. Bioconjugate Chem. 2010, 21, 1389−1403. (14) Li, H.; Peng, X.; Leonard, P.; Seela, F. Binding of Actinomycin C1 (D) and Actinomin to Base-Modified Oligonucleotide Duplexes with Parallel Chain Orientation. Bioorg. Med. Chem. 2006, 14, 4089− 4100. (15) Fritzsche, H.; Akhebat, A.; Taillandier, E.; Rippe, K.; Jovin, T. M. Structure and Drug Interactions of Parallel-Stranded DNA Studied by Infrared Spectroscopy and Fluorescence. Nucleic Acids Res. 1993, 21, 5085−5091. (16) Shchyolkina, A. K.; Borisova, O. F.; Chernov, B. K.; Tchurikov, N. A. Parallel-Stranded DNA with Mixed Sequence. Evidence for Conformational Transition in Solution at Low Water Activity. J. Biomol. Struct. Dyn. 1994, 11, 1237−1249. (17) Rippe, K.; Jovin, T. M. Substrate Properties of 25-nt ParallelStranded Linear DNA Duplexes. Biochemistry 1989, 28, 9542−9549. (18) Cubero, E.; Luque, F. J.; Orozco, M. Theoretical Studies of d(A:T)-Based Parallel-Stranded DNA Duplexes. J. Am. Chem. Soc. 2001, 123, 12018−12025. (19) Ingale, S. A.; Leonard, P.; Tran, Q. N.; Seela, F. Duplex DNA and DNA-RNA Hybrids with Parallel Strand Orientation: 2′-Deoxy-2′fluoroisocytidine, 2′-Deoxy-2′-fluoroisoguanosine, and Canonical Nucleosides with 2′-Fluoro Substituents Cause Unexpected Changes on the Double Helix Stability. J. Org. Chem. 2015, 80, 3124−3138. (20) Pallan, P. S.; Lubini, P.; Bolli, M.; Egli, M. Backbone-Base Inclination as a Fundamental Determinat of Nucleic Acid Self- and Cross-Pairing. Nucleic Acids Res. 2007, 35, 6611−6624. (21) Wunnicke, D.; Strohbach, D.; Weigand, J. E.; Appel, B.; Feresin, E.; Suess, B.; Müller, S.; Steinhoff, H.-J. Ligand-Induced Conformational Capture of a Synthetic Tetracycline Riboswitch Revealed by Pulse EPR. RNA 2011, 17, 182−188.

duplexes built up entirely from dA·dT base pairs were used and constructed as reported by Jovin.30 Two TEMPO azides were introduced in the same oligonucleotide chain by the click reaction. PELDOR experiments indicate that the distance between nitroxide groups of the spin labels in ps DNA is increased by 5−10% over aps DNA when the number of base pairs separating the spin labels is the same. This result is valid for DNA helix turns located near the center of the double helix. When one of the spin labels is positioned close to the end of the helix the situation changes and the distance for ps DNA becomes shorter than for aps DNA. We suppose that this is due to the thermodynamically less stable terminus in ps DNA compared with aps DNA. Thus, even small local changes in the DNA structuremore stable helix stacks near the center and unstable helix stacks near the endcould be detected in ps DNA. Regarding the overall length of aps and ps DNA, our data suggest that ps DNA is slightly longer than aps DNA considering that the spin labels are in the same distance as the nucleobases they are linked to.

■

ASSOCIATED CONTENT

S Supporting Information *

Structures of phosphoramidites, Tm values of antiparallel- and parallel-stranded oligonucleotide duplexes containing ethynylated nucleosides; melting profiles of aps and ps duplexes; enzymatic hydrolysis of modified oligonucleotides; backgroundcorrected summed dipolar evolution data and distance distributions; background-corrected summed dipolar evolution data; and references. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.5b02935.

■

AUTHOR INFORMATION

Corresponding Authors

*H.-J.S.: E-mail: [email protected]. Tel: +49 5419692675. Fax: +49 5419692656. *F.S.: E-mail: [email protected]. Tel: +49 25153406856. Author Contributions ∥

D.W. and P.D. contributed equally to this work.

Funding

The research of the manuscript was supported by a grant of the DFG to H.-J.S., STE640/12. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Dr. S. Budow-Busse for her continuous support throughout the preparation of the manuscript. Financial support by the ChemBiotech, Münster, Germany (to P.D., H.Y., and F.S.) and by the DFG (D.W. and H.-J.S.) is gratefully acknowledged.

■

ABBREVIATIONS ps, parallel-stranded; aps, anti-parallel-stranded; CuAAC, copper(I)-catalyzed Huisgen−Meldal−Sharpless alkyne−azide cycloaddition; PELDOR, pulse electron double resonance

■

REFERENCES

(1) Van de Sande, J. H.; Ramsing, N. B.; Germann, M. W.; Elhorst, W.; Kalisch, B. W.; von Kitzing, E.; Pon, R. T.; Clegg, R. C.; Jovin, T. M. Parallel Stranded DNA. Science 1988, 241, 551−557. 13598


Article

The Journal of Physical Chemistry B (22) Krstić, I.; Hänsel, R.; Romainczyk, O.; Engels, J. W.; Dötsch, V.; Prisner, T. F. Long-Range Distance Measurements on Nucleic Acids in Cells by Pulsed EPR Spectroscopy. Angew. Chem., Int. Ed. 2011, 50, 5070−5074. (23) Ding, P.; Wunnicke, D.; Steinhoff, H.-J.; Seela, F. Site-Directed Spin-Labeling of DNA by the Azide-Alkyne ’Click’ Reaction: Nanometer Distance Measurements on 7-Deaza-2′-deoxyadenosine and 2′-Deoxyuridine Nitroxide Conjugates Spatially Separated or Linked to a ’dA-dT’ Base Pair. Chem. - Eur. J. 2010, 16, 14385−14396. (24) Jakobsen, U.; Shelke, S. A.; Vogel, S.; Sigurdsson, S. T. SiteDirected Spin-Labeling of Nucleic Acids by Click Chemistry: Detection of Abasic Sites in Duplex DNA by EPR Spectroscopy. J. Am. Chem. Soc. 2010, 132, 10424−10428. (25) Pannier, M.; Veit, S.; Godt, A.; Jeschke, G.; Spiess, H. W. DeadTime Free Measurement of Dipole-Dipole Interactions Between Electron Spins. J. Magn. Reson. 2000, 142, 331−340. (26) Wunnicke, D.; Ding, P.; Seela, F.; Steinhoff, H.-J. Site-Directed Spin Labeling of DNA Reveals Mismatch-Induced Nanometer Distance Changes Between Flanking Nucleotides. J. Phys. Chem. B 2012, 116, 4118−4123. (27) Jeschke, G.; Chechik, V.; Ionita, P.; Godt, A.; Zimmermann, H.; Banham, J.; Timmel, C. R.; Hilger, D.; Jung, H. DeerAnalysis2006 − a Comprehensive Software Package for Analyzing Pulsed ELDOR Data. Appl. Magn. Reson. 2006, 30, 473−498. (28) (a) Krieger, E.; Darden, T.; Nabuurs, S. B.; Finkelstein, A.; Vriend, G. Making Optimal Use of Empirical Energy Functions: ForceField Parameterization in Crystal Space. Proteins: Struct., Funct., Genet. 2004, 57, 678−683. (b) Krieger, E.; Koraimann, G.; Vriend, G. Increasing the Precision of Comparative Models with YASARA NOVAa Self-parameterizing Force Field. Proteins: Struct., Funct., Genet. 2002, 47, 393−402. (29) Olson, W. K.; Gorin, A. A.; Lu, X. J.; Hock, L. M.; Zhurkin, V. B. DNA Sequence-dependent Deformability Deduced from Protein− DNA Crystal Complexes. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 11163−11168. (30) Ramsing, N. B.; Rippe, K.; Jovin, T. M. Helix-Coil Transition of Parallel-Stranded DNA. Thermodynamics of Hairpin and Linear Duplex Oligonucleotides. Biochemistry 1989, 28, 9528−9535. (31) Ulyanov, N. B.; Bauer, W. R.; James, T. L. High-resolution NMR Structure of an AT-rich Sequence. J. Biomol. NMR 2002, 22, 265−280. (32) Vorlícková, M.; Kypr, J. Conformational Variability of Poly(dAdT).Poly(dA-dT) and Some Other Deoxyribonucleic Acids Includes a Novel Type of Double Helix. J. Biomol. Struct. Dyn. 1985, 3, 67−83. (33) Graham, D.; Parkinson, J. A.; Brown, T. DNA Duplexes Stabilized by Modified Monomer Residues: Synthesis and Stability. J. Chem. Soc., Perkin Trans. 1 1998, 1131−1138. (34) Ramsing, N. B.; Jovin, T. M. Parallel-Stranded Duplex DNA. Nucleic Acids Res. 1988, 16, 6659−6676. (35) 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.

13599


DNA with Parallel Strand Orientation: A Nanometer Distance Study

Recommend Documents