Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (142K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Lin, Z.
Right arrow Articles by de los Santos, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lin, Z.
Right arrow Articles by de los Santos, C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 2385-2391  

Solution structure of duplex DNA containing an extrahelical abasic site analog determined by NMR spectroscopy and molecular dynamics
Introduction
Materials And Methods
   Synthesis and purification
   NMR methods
   Computational methods
Results
   NMR spectra of non-exchangeable protons
   Exchangeable protons
   Phosphorous spectra
   Three-dimensional structures
Discussion
   NMR spectra
   The tetrahydrofuran residue
   AP-duplex structure
   Biological implications
Acknowledgements
References

Solution structure of duplex DNA containing an extrahelical abasic site analog determined by NMR spectroscopy and molecular dynamics

Solution structure of duplex DNA containing an extrahelical abasic site analog determined by NMR spectroscopy and molecular dynamics

Zhen Lin, Kwong-Ngai Hung, Arthur P. Grollman, Carlos de los Santos*

Department of Pharmacological Sciences, State University of New York at Stony Brook, Stony Brook, NY 11794-8651, USA

Received January 20, 1998; Revised and Accepted April 2, 1998

ABSTRACT

Translesional DNA synthesis past abasic sites proceeds with the preferential incorporation of dAMP opposite the lesion and, depending on the sequence context, one or two base deletions. High-resolution NMR spectroscopy and molecular dynamics simulations were used to determine the three-dimensional structure of a DNA heteroduplex containing a synthetic abasic site (tetrahydrofuran) residue positioned in a sequence that promotes one base deletions. Analysis of NMR spectra indicates that the stem region of the duplex adopts a right-handed helical structure and the glycosidic torsion angle is in anti orientation for all residues. NOE interactions establish Watson-Crick alignments for all canonical base pairs of the duplex. Measurement of distance interactions at the lesion site shows the abasic residue excluded from the helix. Restrained molecular dynamics simulations generated three-dimensional models in excellent agreement with the spectroscopic data. These structures show a regular duplex region and a slight bend at the lesion site. The tetrahydrofuran residue extrudes from the helix and is highly flexible. The model reported here, in conjunction with a previous study performed on abasic sites, explains the structural bias of one-base deletion mutations.

INTRODUCTION

The loss of DNA bases by the spontaneous hydrolysis of N-glycosylic bonds results in the formation of abasic sites (apurinic/apyrimidinic sites, AP sites). This process, which in mammalian cells may produce up to 10 000 AP sites per day (1), leaves primarily an inter-converting mixture of 2[prime]-deoxy-[alpha]-d-ribose, 2[prime]-deoxy-[beta]-d-ribose and a small percentage of the open form of the sugar. Base modifications that favor the hydrolytic process, and the repair of damaged bases by DNA glycosylases, accelerate the formation of AP sites (2). Naturally occurring abasic sites are somewhat unstable, because the aldehydic (open) form is subject to [beta]-elimination and strand scission (3). This limitation can be overcome by using 3-hydroxy-2-(hydroxymethyl)-tetrahydrofuran (F), a closed-form analog of AP sites (4,5). DNA containing this modified tetrahydrofuran group is efficiently cleaved by endonuclease IV, exonuclease III and ApeI, the major AP endonucleases involved in DNA repair in Escherichia coli and mammalian cells, and serves as a template for AMV reverse transcriptase as well as prokaryotic and eukaryotic DNA polymerases (5-10).

The mutagenic properties of natural and synthetic AP sites have been investigated using randomly or site-specifically modified substrates. In vitro studies using bacterial DNA polymerases revealed the preferential incorporation of dAMP opposite AP sites (5,6,11). In E.coli, translesional synthesis past abasic sites is facilitated by induction of the SOS response, (12) and proceeds mainly by incorporation of dAMP opposite the lesions and a small percentage of deletion mutations (13,14). DNA sequences containing an abasic site flanked by a thymidine residue at the 5[prime] side have been identified as a `hot spot' for [Delta]-1 mutations (15). In contrast, a broader mutational spectrum is observed in eukaryotic cells. In simian kidney cells, replication of double-stranded vectors containing a single AP site leads to the preferential incorporation of dAMP (16). In the same cellular system, replication of a single-stranded vector carrying a unique abasic site showed random incorporation of bases opposite the lesion (17). In human lymphoblastoid cells, deletions and frameshift mutations accounted for 55% of the mutations and, among single base substitutions, incorporation of dGMP was prevalent (18). On the other hand, preferential incorporation of dCMP opposite a unique natural AP site has been reported in Saccharomyces cerevisiae (19).

Three-dimensional structures of DNA duplexes containing natural and synthetic abasic sites have been solved by NMR spectroscopy. In these studies, the duplex retained a B-form geometry and the AP site induced only local perturbations, limited to the lesion site and flanking base pairs (20-27). In duplexes containing purine (dG or dA) residues opposite the lesion, both the abasic site and opposing purine are stacked inside the helix (20,22-25,27). However, when pyrimidine (dC or T) bases are opposite the AP site, the duplex structure shows increased conformational flexibility. Depending on the temperature, the abasic sugar residue, and sometimes the opposing pyrimidine base, are extrahelical (21,26). Less is known about the structure of DNA duplexes containing abasic site lesions in the context of frameshift mutations. Proton NMR spectroscopy and molecular mechanics have been used to determine the structure of a (GC)-rich oligodeoxynucleotide duplex containing an extrahelical abasic site residue (28). In the refined models, the abasic sugar adopted two conformations outside of the helix, associated with a different pucker on the 3[prime]-flanking sugar. In the present study, we describe the solution structure of a d(G-A-A-C-T-F-G-C-T-A-C)·d(G-T-A-G-C-A-G-T-T-C) heteroduplex (referred to throughout this paper as the AP-duplex), where F is a 3-hydroxy-2-(hydroxymethyl)-tetrahydrofuran residue (Fig. 1), as determined by high resolution NMR spectroscopy and restrained molecular dynamics simulations. The numbering scheme for the duplex sequence employed in this paper is shown in Figure 1.


Figure 1. Structure of the abasic site analog (a); AP-duplex sequence and numbering scheme (b).


MATERIALS AND METHODS

Synthesis and purification

The synthesis of a modified tetrahydrofuran residue and incorporation into DNA sequences by phosphoramidite chemistry procedures has been described (5). Purification of products by reverse phase HPLC and isolation of the sodium salt follows methods previously reported (29). NMR samples consisted of 3.2 × 10-3 mol of AP-duplex dissolved in 0.6 ml of 10 mM phosphate buffer, pH 6.9, containing 50 mM NaCl and 1 mM EDTA in either 99.96% D2O or 90% H2O-10% D2O.

NMR methods

NMR spectra were recorded on Varian spectrometers operating at 600 and 500 MHz. Proton spectra were referenced relative to the (2,2,3,3-d4) sodium 3-trimethylsilyl-propionate signal at 0 p.p.m., and phosphorous spectra relative to TMP. Phase-sensitive (30) NOESY (50, 90, 140, 200, 270 and 300 ms mixing times), COSY, DQF-COSY and TOCSY (120 ms isotropic mixing time) spectra in D2O buffer were collected at 25°C with a repetition delay of 1.5 s, during which the residual water signal was suppressed by saturation. A NOESY (220 ms mixing time) spectrum in H2O buffer was recorded at 5°C using a jump-return reading pulse (31). Time domain data sets consisted of 2048 by 256 complex data points in the t2 and t1 dimensions, respectively. A [1H-31P]-correlation spectrum was recorded at 25°C using indirect detection mode (32) with 2048 by 128 complex data points in t2 and t1 dimensions, respectively. NMR data were processed using Felix95 (Biosym) running on Silicon Graphics computers. Time domain data were multiplied by shifted sinebell window functions prior to Fourier transformation. No baseline correction was applied to the transformed spectra.

Computational methods

Molecular dynamics (MD) simulations were performed on Silicon Graphics and Sun computers using X-PLOR 3.1 (33). Visualization of the structures was done using Insight95 (Biosym) and Midas Plus (UCSF, Computer Graphics Laboratory) and the structural parameters obtained with Curves (34,35).

Simulations were run `in vacuum'using an all atom force field derived from CHARMM (36). Partial atomic charges on phosphate groups were not reduced, resulting in deoxynucleotide residues with a net charge of -1. Cross-peak volumes measured from all NOESY spectra collected in D2O buffer were input to the program MARDIGRAS (UCSF) which calculates interproton distances, taking into consideration spin diffusion effects (37). A single set of distance bounds was obtained by adding "0.4 Å to the averaged value determined by MARDIGRAS. A total of 300 experimental distances were enforced during the simulation using an empirical square-well energy potential function with a force constant of 30 kcal/(mol Å2). Watson-Crick alignments were enforced by distance restraints on all canonical base pairs of the duplex. Backbone dihedral angles outside the lesion site were restrained by an empirical square well potential function with a width encompassing a range for both A- and B-form DNA. The SHAKE algorithm (38) was employed to maintain the length of all covalent bonds involving hydrogen atoms. At the lesion site, the [epsis] (C4[prime]-C3[prime]-O3[prime]-P) torsion angles of T5 and F6 residues were determined based on their experimental JH3[prime]-P coupling constant, using the Karplus relation J = 15.3 cos2 [phiv] - 6.1 cos [phiv] + 1.6 (39) and [epsis] = (-[phiv] - 120°). Of the four possible solutions, a cis conformational range ([epsis] = 330-30°) was considered forbidden. Subsequently, potential energy functions with a double minimum between 100-115° and 210-270° were enforced during the simulation. Similarly, based on the large JH5[prime](5[prime][prime])-P coupling constant value, the [beta] (P-O5[prime]-C5[prime]-C4[prime]) torsion angle of F6 was constrained within the gauche+ (30-90°) or gauche- (270-330°) conformational ranges by using a double-minimum potential energy function.

Sugar puckers and glycosidic torsion angles were not restrained during the simulations.

Five different initial models containing a tetrahydrofuran residue were built by replacing the aromatic ring of adenosine by a proton. Two of the initial models were 10mer duplexes, derived from B-form DNA, having an additional tetrahydrofuran residue in the major or minor groove of the helix. The other three models were derived from 11mer duplexes by deleting a T on one strand and modifying a dA residue, either complementary to the deleted T or staggered towards the 3[prime] or 5[prime] direction. The resulting structures were termed F-major, F-minor, F-opposite, F-3[prime]-staggered and F-5[prime]-staggered. Initial models were energy-minimized by the conjugate gradient method prior to MD. Simulations consisted of an equilibration step during which the temperature was increased from 100 to 500 K in 4 ps and the scale of the NOE distance restraint potential function was gradually augmented. Once the equilibration temperature was obtained, the simulation continued at 500 K for 20 ps, followed by cooling to 300 K in 10 ps, and 105 ps of restrained dynamics at constant temperature. Seven MD simulations were performed, starting from each initial model, using a different set of initial velocities. Coordinates of the last 5 ps of the simulation were averaged and subjected to 1000 steps of energy minimization, yielding distance-refined structures. Converging structures were defined as those having rms deviations in the helical region of the duplex smaller than 0.5 Å with respect to the average structure.

RESULTS

NMR spectra of non-exchangeable protons

The 1D proton NMR spectrum of the AP-duplex recorded in D2O buffer, pH 6.9, 25°C is characterized by the presence of sharp signals and the absence of minor resonances (Supplementary Material, Fig. 1). Assignment of the non-exchangeable base and sugar protons followed the analysis of NOESY, COSY and TOCSY spectra using standard procedures (40,41). Figure 2 shows distance interactions detected on a NOESY (300 ms mixing time) spectrum between the base (purine H8/pyrimidine H6) and the sugar H1[prime] proton regions. Indicative of a right-handed helix, each base proton of the duplex exhibited NOE peaks to its own and 5[prime]-flanking H1[prime] sugar protons. The presence of the tetrahydrofuran residue did not impede the observation of sequential distance interactions at the lesion site. On the lesion-containing strand, NOE peaks were observed between G7(H8) and T5(H1[prime], H3[prime]) (Fig. 2, peaks A and E, respectively) as well as T5(H6, H2[prime], H2[prime][prime]) protons (data not shown). On the complementary strand, normal distance interactions connected A17(H8) and C16(H1[prime], H2[prime], H2[prime][prime]) protons. In addition, inter-strand conectivities were detected between A17(H2) and the H1[prime] sugar protons of T5 and G7 (Fig. 2, peaks B and D, respectively).


Figure 2. Contour plots of a portion of a NOESY (300 ms mixing time) spectrum recorded in D2O buffer, pH 6.9, 25°C, depicting distance interactions around the lesion, between the base (7.1-8.3 p.p.m.) and the H1[prime]/H3[prime] (5.5-6.3 p.p.m.) proton regions. Labeled peaks are assigned as follows: C4, C4(H6)-C4(H1[prime]); T5, T5(H6)-T5(H1[prime]); G7, G7(H8)-G7(H1[prime]); C8, C8(H6)-C8(H1[prime]); G15, G15(H8)-G15(H1[prime]); C16, C16(H6)-C16(H1[prime]); A17, A17(H8)-A17(H1[prime]); G18, G18(H8)-G18(H1[prime]); A, G7(H8)-T5(H1[prime]); B, A17(H2)-T5(H1[prime]); C, A17(H2)-A17(H1[prime]); D, A17(H2)-G18(H1[prime]); E, G7(H8)-T5(H3[prime]). Asterisks indicate the cytosine H5-H6 cross-peaks.

Assignment of the tetrahydrofuran protons started with the identification of COSY interactions between F6(H1[prime]/H1[prime][prime]) and F6(H2[prime]/H2[prime][prime]) protons, which occurred in a region of the spectrum devoid of other signals (Supplementary Material, Fig. 2). Subsequent analysis of DQF-COSY, TOCSY (120 ms isotropic mixing time) and NOESY (100 ms mixing time) spectra completed their assignment. None of the tetrahydrofuran protons showed distance interactions to the base or H1[prime], H2[prime], H2[prime][prime] and H3[prime] sugar protons of flanking T5 and G7 residues. The chemical shift of the non-exchangeable protons measured at 25°C is listed in Table 1.

Table 1. Proton and phosphorous chemical shiftsa
  H8/H6b H2/H5/CH3b H1[prime]b H2[prime]b H2[prime][prime]b H3[prime]b H4[prime]b T(N3H)/G(N1H)c,d C(N4H)/A(H2)c 31Pb
G1 7.81 - 5.50 2.43 2.65 4.80 4.16 12.62 - -
A2 8.17 7.46 5.91 2.74 2.88 5.04 4.39 - 7.36 -5.01
A3 8.11 7.71 6.12 2.62 2.82 5.01 4.45 - 7.65 -4.89
C4 7.20 5.09 5.77 1.83 2.39 4.68 4.19 - 7.74/6.60 -4.88
T5 7.30 1.51 6.07 2.43 2.43 4.91 4.21 13.24 - -4.90
F6 - - 3.99 2.25 2.12 4.63 4.13 - - -4.00
G7 7.58 - 5.87 2.44 2.63 4.68 4.36 12.58 - -4.44
C8 7.47 5.21 5.81 2.09 2.44 4.70 4.19 - 8.18/6.68 -5.02
T9 7.42 1.59 5.70 2.10 2.45 4.85 4.12 13.67 - -5.13
A10 8.25 7.57 6.23 2.64 2.83 4.99 4.38 - 7.48 -4.86
C11 7.26 5.18 6.01 2.11 2.12 4.27 3.99 - nd -4.72
G12 7.90 - 5.94 2.60 2.76 4.79 4.20 12.64 - -
T13 7.36 1.38 5.70 2.13 2.45 4.88 4.19 13.67 - -4.78
A14 8.17 7.39 6.05 2.70 2.88 5.03 4.39 - 7.30 -4.72
G15 7.61 - 5.73 2.46 2.63 4.91 4.37 12.88 - -4.77
C16 7.17 5.12 5.61 1.86 2.30 4.72 4.14 - 8.15/6.43 -5.00
A17 8.01 7.46 5.94 2.61 2.82 4.97 4.33 - 7.39 -4.91
G18 7.63 - 5.85 2.50 2.70 4.91 4.37 12.74 - -4.77
T19 7.18 1.29 5.97 2.53 2.53 4.78 4.19 13.89 - -4.93
T20 7.39 1.60 6.14 2.51 2.51 4.86 4.13 14.04 - -4.88
C21 7.51 5.55 6.22 2.27 2.27 4.55 3.98 - nd -4.61
aValues in p.p.m. relative to TSP (1H) or TMP (31P) at 0 p.p.m.
bMeasured at 25°C in D2O phosphate buffer (10 mM), pH 6.9, 50 mM NaCl.
cMeasured at 5°C in 10% D2O phosphate buffer (10 mM), pH 6.9, 50 mM NaCl.
dAssignment of G1(N1H) and G12(N1H) imino protons could be reversed.
nd, not detected.

Exchangeable protons

The 1D spectrum of the exchangeable protons recorded in 10% D2O buffer, pH 6.8, 5°C showed seven partially resolved signals, between 12.2 and 14.2 p.p.m., which account for the 10 imino protons of the AP-duplex (

Supplementary Material, Fig. 3). Their assignment followed the analysis of a NOESY (200 ms mixing time) spectrum shown in Figure 3. Strong NOE interactions between thymidine imino and AH2 protons, and guanine imino and cytidine amino protons identified A·T and G·C base pairs of the duplex (Fig. 3A, peaks A-E and F/F[prime]-H/H[prime], respectively).


Figure 3. Contour plots of portions of a NOESY (150 ms mixing time) spectrum recorded in 10% D2O buffer, pH 6.8, 5°C. The spectra display distance connectivities between (A) imino and base/amino proton regions and (B) the symmetrical imino proton region. Labeled peaks are assigned as follows: (A) A, T20(N3H)-A2(H2); B, T19(N3H)-A3(H2); C, T13(N3H)-A10(H2); D, T9(N3H)-A14(H2); E, T5(N3H)-A17(H2); F, G18(N1H)-C4(N4H)hb; F[prime], G18(N1H)-C4(N4H)nhb; G, G15(N1H)-C8(N4H)hb; G[prime], G15(N1H)-C8(N4H)nhb; H, G7(N1H)-C16(N4H)hb; H[prime], G7(N1H)-C16(N4H)nhb; I, T20(N3H)-A3(H2); J, T19(N3H)-A2(H2); K, G15(N1H)-A14(H2); L, G18(N1H)-A17(H2); M, G18(N1H)-A3(H2) and N, G7(N1H)-A17(H2). (B) A, T20(N3H)-T19(N3H); B, T19(N3H)-G18(N1H); C, T9(N3H)-G15(N1H); D, T5(N3H)-G18(N1H); E, G15(N1H)-G7(N1H) and F, T5(N3H)-G7(N1H). (N4H)hb and (N4H)nhb refer to the hydrogen-bonded and non hydrogen-bonded cytidine amino protons, respectively.

We observed NOE peaks between T20(N3H)-A3(H2), T19(N3H)-A2(H2), G15(N1H)-A14(H2), G18(N1H)-A17(H2) and G18(N1H)-A3(H2) protons located on consecutive base pairs of the duplex (Fig. 3A, peaks I-M, respectively). In addition, a weak NOE interaction connects G7(N1H) and A17(H2) protons, which are located on base pairs flanking the tetrahydrofuran residue (Fig. 3A, peak N). Further evidence of base-pair stacking are the NOE peaks observed among the imino protons of the duplex (Fig. 3B, peaks A-F). The chemical shifts of the exchangeable protons measured at 5°C are listed in Table 1.

Phosphorous spectra

The 31P-NMR 1D spectrum of the AP-duplex at 25°C spans ~1.1 p.p.m., with two phosphates resonating slightly down-field from the normal region observed on DNA duplexes. The specific assignment of all phosphorus signals of the duplex followed the analysis of the 1H-31P correlation spectrum recorded in D2O buffer, pH 6.8, 25°C, shown in Figure 4. Each phosphorous exhibited COSY cross-peaks to the H3[prime] sugar proton of the 3[prime]-flanking nucleotide. Thus, the two most down-field signals were assigned to T5-p-F6 and F6-p-G7 flanking the tetrahydrofuran residue. Phosphorous chemical shifts measured at 25°C are listed in Table 1.


Figure 4. Contour plot of an inverse-detected 1H-31P correlation spectrum recorded in D2O buffer, pH 6.8, 25°C. Phosphorous signals flanking the tetrahydrofuran are identified in the figure. Labeled peaks are assigned as follows: A, T5(H3[prime])-T5-p-F6; B, F6(H3[prime])-F6-p-G7; C, F6(H5[prime],H5[prime][prime])-T5-p-F6. JH3[prime]-P values of residues at the lesion site are, T5 = 6.7 Hz and F6 = 8.6 Hz.

The [epsis] torsion angles of the T5 and F6 residues were determined from the experimental JH3[prime]-P coupling constant values (Fig. 4, peaks A and B). In both cases, [epsis] values were within the anti+(+ac)/anti-(-ac) ranges. Determination of JH5[prime]-P and JH5[prime][prime]-P coupling constants is generally impaired by the severe overlap of the H5[prime] and H5[prime][prime] proton signals and the lack of stereospecific assignments. Due to small JH5[prime]-P and JH5[prime][prime]-P values, weak H5[prime]-P and H5[prime][prime]-P cross-peaks are usually observed in DNA duplexes (Fig. 4). This characteristic is the result of a preferred [beta]-trans conformation, which implies small values for both coupling constants simultaneously. On the AP-duplex, a strong COSY cross-peak was detected between the T5-p-F6 phosphorous and the overlapping F(H5[prime], H5[prime][prime]) protons (Fig. 4, peak C), excluding a trans conformation for the [beta] torsion angle of the F residue. Since F(H5[prime]) and F(H5[prime][prime]) proton signals are isochronous, we could not differentiate between [beta]-gauche- (JH5[prime]-P > JH5[prime][prime]-P) and [beta]-gauche+ (JH5[prime][prime]-P > JH5[prime]-P ) conformations.

Table 2. Rms-deviations for converging structuresa
Initial modelsb AP duplexc Helical regiond Furan residuee
F-minor 0.61 0.43 1.41
F-minor 0.68 0.47 1.42
F-opposite 0.62 0.42 1.48
F-opposite 0.55 0.41 1.45
F-5[prime]-staggered 0.54 0.40 1.45
F-5[prime]-staggered 0.56 0.48 1.51
F-3[prime]-staggered 0.54 0.46 1.39
F-3[prime]-staggered 0.59 0.41 1.47
aValues in Angstroms relative to an average structure calculated from 35 distance-refined models.
bNone of the refinements that started with the F residue on the major groove converged within 0.5 Å of the average. For each of the other initial structures, two refined models fulfilled the selection criteria.cOverall heteroduplex.
d10mer helical region, only.
eAbasic residue, only.

Three-dimensional structures

MD simulations were performed with an average of 17 experimental restraints on unmodified nucleotides of the sequence, but only four on the tetrahydrofuran residue. Five refined structures were immediately rejected, because the final values of [epsis] and [beta] torsion angles at the lesion site were outside the experimental bounds. The remaining structures exhibited rmsd in the atom position from 0.60 to 1.60 Å, with respect to their average, no violations of experimental dihedral restraints and no NOE-violation larger than 0.30 Å. Analysis of the structures showed that rmsd were unevenly distributed throughout the duplex. The helical part (excluding the tetrahydrofuran) was well defined, while the lesion site was quite flexible, having average rmsd of 0.60 and 1.40 Å, respectively. Thus, structures having rmsd [le]0.5 Å for the helical part (total of eight) were selected as converging models and used to describe the AP-duplex structure. Distribution of rmsd in the heavy atom positions is given in Table 2.

An overlap view of the converging structures, as seen from the major groove of the helix, is displayed in Figure 5. The structures belong to the B-form DNA family with sugar puckers on the C2[prime]-endo/C1[prime]-exo range and anti glycosidic torsion angles. The presence of the abasic residue induces a small bend on the helix in the direction of the major groove. However, Watson-Crick base-pair alignments remain intact throughout the helical region of the duplex. On the lesion-containing strand, the sugar-phosphate backbone is perturbed at the lesion site by the presence of the tetrahydrofuran residue, while the backbone of the partner strand is quite regular. Figure 6 shows a close view of the central 4 base pair segment of the duplex, as seen from the minor groove of the helix. Base pairs flanking the lesion are properly aligned and almost undisturbed. Only T5·A17, 5[prime]-flanking the tetrahydrofuran residue, showed increased propeller twist and is slightly buckled. In contrast, normal values are seen for G7·C16. The helix is under twisted between the T5·A17 and G7·C16 base pairs, but this perturbation is immediately compensated for by a proportional over twist at the G7·C16-C8·G15 step.


Figure 5. Overlapping view of the three-dimensional models obtained after restrained molecular dynamics as seen from the major groove. The picture shows the convergence of the T5·A17 (red) and G7·C16 (yellow) base pairs of the duplex contrasting with the conformational freedom exhibited by the tetrahydrofuran residue (blue).


Figure 6. Stereo view of the central d(C4-T5-F6-G7-C8)[bull]d(G15-C16-A17-G18) segment as seen from the minor groove side. T5·A17 (red) and G7·C16 (yellow) exhibit proper Watson-Crick alignment while F6 (blue) is very flexible.

The abasic lesion is extruded from the helix and, as suggested by the values of rmsd, its position is flexible. In some models, the tetrahydrofuran residue is inclined toward one of the grooves of the helix, while in others remains equidistant. Figure 7 depicts the position of the tetrahydrofuran residue, as seen from the helical axis looking in the direction of the sugar-phosphate backbone. The flexibility of the lesion is the result of the conformational freedom of the backbone angles. Of the torsion angles restrained during the simulation, only [epsis] showed some propensity for the +ac range while [beta] exhibited no preference between the gauche+ and gauche- ranges. Among the angles not restrained during the simulation, only [delta] (C5[prime]-C4[prime]-C3[prime]-O3[prime]) showed some conformational preference for the ±sc range. In contrast, the pucker of the tetrahydrofuran residue was well defined, falling in the C2[prime]-exo range (amplitude 12-34°) in the converging structures.


Figure 7. Conformational freedom of the tetrahydrofuran residue (blue) in the converging structures. The picture shows a view from the helical axis of the duplex looking in the direction of the sugar-phosphate backbone. The F6 phosphate group (top) and the O3[prime] atoms (bottom) are colored in red.


DISCUSSION

NMR spectra

The fact that we have detected sharp proton and phosphorous signals, together with their assignment to one set of NMR resonances, suggests that the AP-duplex adopts a single averaged structure in solution. The sequential distance connectivities observed in the NOESY spectra between the base (purine H8/pyrimidine H6) and the H1[prime], H2[prime], H2[prime][prime], H3[prime] sugar protons of the 3[prime]-attached nucleotide establishes a right-handed helical structure. Furthermore, weak intensities characterized the intra-residue base to H1[prime] NOE cross-peaks (Fig. 2), specifically at short mixing times, suggesting long distances among these protons of the AP-duplex. This observation is consistent with residues having glycosidic torsion angles on the anti conformation (40,41). Consistent with this orientation, Watson-Crick alignments were established by distance interactions detected in the NOESY spectrum recorded in 10% D2O buffer, between T(N3H)-A(H2) and G(N1H)-C(N4H1)/C(N4H2) protons (Fig. 3A, peaks A-E and F/F[prime]--H/H[prime], respectively).The JH1[prime]-H2[prime] and JH1[prime]-H2[prime][prime] cross-peak patterns detected in the phase-sensitive COSY spectrum together with the absence of strong NOE interactions between the base (purine H8/pyrimidine H6) and its own H3[prime] sugar proton, determined that sugar puckers are in the south conformation (42). Distance connectivities among the imino protons, and between imino and A(H2) protons of adjacent base pairs of the sequence (Fig. 3), indicated favorable base-pair stacking throughout the AP-duplex (41). The perturbation of the DNA backbone induced by the presence of a tetrahydrofuran residue is somewhat limited. The phosphorous spectra shows that only T5-p-F6 and F6-p-G7 resonate outside the region usually observed on B-form DNA duplexes (-4.5 to -5.0 p.p.m.) (Fig. 4). Thus, distortions of the sugar-phosphate backbone are limited to the lesion site on the modified strand.

Table 3.
Structure calculations
   Deviations from idealized geometrya
      Rmsd of bond lengths (Å) 0.005 ± 0.0001
      Rmsd of bond angles (°) 3.07 ± 0.03
      Rmsd of improper angles (°) 0.24 ± 0.02
      van der Waals energy (kcal/mol) -312 ± 3
      Watson-Crick energy (kcal/mol)b -95 ± 3
      NOE energy (kcal/mol Å2) 29 ± 4
      Rmsd of NOE violations (Å) 0.057 ± 0.002
Structural parametersa,c
         Helical bend (°) 12 ± 5
   Base-pair parameters
         T5·A17 propeller (°) -23 ± 4
         T5·A17 buckle (°) -5 ± 2
         G7·C16 propeller (°) 5 ± 3
         G7·C16 buckle (°) 2 ± 2
   Inter base-pair parameters
         T5·A17-G7·C16 roll (°) 18 ± 5
         C4·G18-T5·A17 twist (°) 31 ± 3
         T5·A17-G7·C16 twist (°) 26 ± 3
         G7·C16-C8·G15 twist (°) 44 ± 2
aAverage values measured for converging structures.
bSum of electrostatic and NOE terms.
cAs defined by Curves (35). Canonical values of the base-pair parameters in B-form DNA are propeller, 3.7°; buckle, 0°; roll, 0°; twist, 36°.

The tetrahydrofuran residue

Several spectroscopic characteristics define structural features of the AP-duplex at the lesion site. NOE interactions encountered between G7(H8) and the H1[prime], H2[prime], H2[prime][prime] and H3[prime] sugar protons of T5 establish distances shorter than 4.5 Å between these protons of the duplex. Furthermore, in several regions of the NOESY spectra not hindered by signal overlap, we have failed to observe distance connectivities between tetrahydrofuran protons and the 3[prime]- and 5[prime]-flanking residues. This fact cannot be ascribed to broadening of the tetrahydrofuran signals due to molecular motion, since they were clearly identified in the phase-sensitive COSY, COSY45, DQF-COSY and TOCSY spectra without any apparent increase in the line width. Thus, we interpret the lack of distance interactions as confining the F residue to conformations outside the helix.

AP-duplex structure

Distance-restrained molecular dynamics was able to drive quite different initial models to structures having rmsd smaller than 0.7 Å with respect to their average. In evaluating the precision of the three-dimensional models, we must consider the disparity in the number of experimental restraints enforced during the simulation for different regions of the AP-duplex. The helical region of the duplex, with an average of 17 restraints/residue, showed excellent convergence, while the tetrahydrofuran, with only four experimental constraints, was highly flexible (Table 2).

The duplex structure is a fairly regular helix where the F residue has been forced out (Fig. 5). Stacking of the T5·A17 and G7·C16 base pairs is almost unaffected by the presence of the lesion. The tetrahydrofuran H1[prime][prime] proton points toward the bulk solvent suggesting that, in natural abasic sites, the OH group at this position could be readily stabilized by hydrogen bonding interactions. Our three-dimensional model resembles the structure reported by Cuniasse et al., where an extrahelical F residue is flanked by G·C base pairs (28). Perhaps the most notable difference between the structures is the small bend in the helical axis observed here (Table 3). It is possible that the reason for this subtle difference is the presence of a less stable A·T base pair 5[prime]-flanking the lesion.

As mentioned above, the limited number of experimental restraints impeded the precise determination of the tetrahydrofuran conformations in the AP-duplex structure. There was no apparent spectroscopic reason, other than distances >5 Å, that could account for the absence of distance interactions between protons of the abasic and flanking residues. Molecular dynamics simulations showed that a large number of conformations are accessible to the F residue, while maintaining long distances with the flanking bases. The lack of conformational preference exhibited by the backbone dihedral angles of the F residue makes possible this structural freedom. If we consider that the energy barrier associated with variations of backbone angles is low, the conclusion emerging from our studies is that, in solution, the AP-duplex is a stable and regular B-form helix with the extrahelical tetrahydrofuran residue oscillating between the DNA grooves.

Biological implications

The aim of our study was to address the question of a structural bias for [Delta]-1 frameshift mutations induced by abasic sites. The three-dimensional model reported here is entirely consistent with the mechanism proposed for frameshift mutations in unmodified and chemically modified DNA duplexes (43-45). Following incorporation of dAMP opposite the abasic site during DNA synthesis, a slipped intermediate could form with the lesion looped-out of the helix. Only in the context of a 5[prime]-flanking thymidine residue can such a structure be stabilized by Watson-Crick hydrogen bonds, thereby facilitating translesional synthesis and resulting in the observed `hot spot' for [Delta]-1 mutations induced by AP sites. The three-dimensional model presented here is essentially identical to the structures reported by Cuniasse et al., where an F residue was flanked by G·C base pairs (28). However, in vitro and in vivo mutagenic studies performed to date have failed to observe frameshift mutations in such a sequence context. Therefore, the lack of [Delta]-1 mutations, in a 5[prime]-dG lesion-flanking sequence context, is the result of the pronounced preference of DNA polymerases to incorporate dAMP opposite the AP sites, and not the consequence of the three-dimensional structure of the mutagenic intermediate.

ACKNOWLEDGEMENTS

We thank Mr Robert Rieger and Ms Cecilia Torres for preparing and purifying modified oligodeoxynucleotides. Atomic coordinates of the NMR refined structures have been deposited in the Brookhaven Data Bank. This research is supported by NIH Grant CA47995. The NMR facility at SUNY Stony Brook is supported by NSF Grants CHE8911350 and CHE9413510 and NIH Grant 1S10RR554701.

REFERENCES

1. Lindahl, T. (1982) Annu. Rev. Biochem. 51, 61-87.

2. Loeb, L. A. and Preston, B. D. (1986) Annu. Rev. Genet. 20, 201-230.

3. Weiss, B. and Grossman, L. (1978) Adv. Enzymol. Relat. Areas Mol. Biol. 60, 1-34.

4. Millican, T. A., Mock, G. A., Chauncey, M. A., Patel, T. P., Eaton, M. A., Gunning, J., Cutbush, S. D., Niedele, S. and Mann, J. (1984) Nucleic Acids Res. 12, 7435-7453. MEDLINE Abstract

5. Takeshita, M., Chang, C.-N., Johnson, F., Will, S. and Grollman, A. P. (1987) J. Biol. Chem. 262, 10171-10179.

6. Randall, S. K., Eritja, R., Kaplan, B., Petruska, J. and Goodman, M. F. (1987) J. Biol. Chem. 262, 6864-6870.

7. Takeuchi, M., Lillis, R., Demple, B. and Takeshita, M. (1994) J. Biol. Chem. 269,21907-21914.

8. Paz-Elizur, T., Takeshita, M., Goodman, M., O'Donnell, M. and Livneh, Z. (1996) J. Biol. Chem. 271, 24662-24669.

9. Paz-Elizur, T., Takeshita, M. and Livneh, Z. (1997) Biochemistry 36, 1766-1773. MEDLINE Abstract

10. Wilson, D. M., Takeshita, M., Grollman, A. P. and Demple, B. (1995) J. Biol. Chem. 270,16002-16007.

11. Sagher, D. and Strauss, B. (1985) Nucleic Acids Res. 13, 4285-4298. MEDLINE Abstract

12. Schaaper, R. M. and Loeb, L. A. (1981) Proc. Natl. Acad. Sci. USA 78, 1773-1777.

13. Kunkel, T. A. (1984) Proc. Natl. Acad. Sci. USA 81, 1494-1498.

14. Lawrence, C. W., Borden, A., Banerjee, S. K. and LeClerc, J. E. (1990) Nucleic Acids Res. 18, 2153-2157. MEDLINE Abstract

15. Shibutani, S., Takeshita, M. and Grollman A. P. (1997) J. Biol. Chem. 272, 13916-13922.

16. Takeshita, M. and Eisenberg, W. (1994) Nucleic Acids Res. 22, 1897-1902. MEDLINE Abstract

17. Cabral Neto, J. B., Cabral, R. E., Margot, A., Le Page, F., Sarasin, A. and Gentil, A. (1994) J. Mol. Biol. 240, 416-420.

18. Klinedinst, D. K. and Drinkwater, N. R. (1992) Mol. Carcinog. 6, 32-42.

19. Gibbs, P. E. and Lawrence, C. W. (1995) J. Mol. Biol. 251,229-236.

20. Cuniasse, Ph., Sowers, L. C., Eritja, R., Kaplan, B., Goodman, N. F., Cognet, J. A. H., LeBret, M., Guschlbauer, W. and Fazakerly, G. V. (1987) Nucleic Acids Res. 15, 8003-8022. MEDLINE Abstract

21. Cuniasse, Ph., Fazakerly, G. V., Guschlbauer, W., Kaplan, B. and Sowers, L. C. (1990) J. Mol. Biol. 213, 303-314.

22. Kalnick, M. W., Chang, C. N., Grollman, A. P. and Patel, D. J. (1988) Biochemistry 27, 924-931.

23. Kalnick, M. W., Chang, C. N., Johnson, F., Grollman, A. P. and Patel, D. J. (1989) Biochemistry 28, 3373-3383.

24. Withka, J. M., Wilde, J. A., Bolton, P. H., Mazumder, A. and Gerlt, J. A. (1991) Biochemistry 30, 9931-9940. MEDLINE Abstract

25. Goljer, I., Kumar, S. and Bolton P. H. (1995) J. Biol. Chem. 270, 22980-22987.

26. Singh, M. P., Hill, G. C., Péoc'h, D., Rayner, B., Imbach, J.-L. and Lown, W. (1994) Biochemistry 33, 10271-10285. MEDLINE Abstract

27. Coppel, Y., Berthet, N., Coulombeau, C., Coulombeau, C., Garcia, J. and Lhomme, J. (1997) Biochemistry 36, 4817-4830. MEDLINE Abstract

28. Cuniasse, Ph., Sowers, L. C., Eritja, R., Kaplan, B., Goodman, N. F., Cognet, J. A. H., LeBret, M., Guschlbauer, W. and Fazakerly, G. V. (1989) Biochemistry 28, 2018-2026. MEDLINE Abstract

29. Korobka, A., Cullinan, D., Cosman, M., Grollman, A. P., Patel, D. J., Eisenberg, M. and de los Santos, C. (1996) Biochemistry 35, 13310-13318. MEDLINE Abstract

30. States, D. J., Habekorn, R. A. and Ruben, D. J. (1982) J. Magn. Res. 48, 286-292.

31. Plateau, P. and Gueron, M. (1982) J. Am. Chem. Soc. 104, 7310-7311.

32. Sklenar, V., Mishayiro, H., Zon, G., Miles, H. T. and Bax, A. (1986) FEBS Lett. 208, 94-98. MEDLINE Abstract

33. Brunger, A. (1993) X-PLOR Version 3.1 A System for X-Ray Crystallography and NMR. Yale University Press, New Haven, CT.

34. Lavery, R. and Sklenar, H. (1988) J. Biomol. Struct. Dyn. 6, 63-91.

35. Lavery, R. and Sklenar, H. (1989) J. Biomol. Struct. Dyn. 6, 655-667.

36. Brooks, B., Bruccoleri, R., Olafson, B., States, D., Swaminathan, S. and Karplus, M. (1983) J. Comp. Chem. 4, 187-217.

37. Borgias, B. A. and James, T. L. (1990) J. Magn. Reson. 87, 475-487.

38. Ryckaert, J. P., Ciccoti, G. and Berendsen, H. J. C. (1977) J. Comput. Phys. 23, 327-341.

39. Lankhorst, P. P., Haasnoot, C. A. G., Erkelens, C. and Altona, C. (1984) J. Biomol. Struct. Dyn. 1, 1387-1405.

40. Hare, D. R., Wemmer, D. E., Chou, S. H., Drobny, G. and Reid, B. R. (1983) J. Mol. Biol. 171, 319-336.

41. Wüthrich, K. (1986) In NMR of Proteins and Nucleic Acids. John Wiley, New York, p.208.

42. Majumdar, A. and Hosur, R. V. (1992) Prog. NMR Spectrosc. 24, 109-158.

43. Kunkel, T. A. (1990) Biochemistry 29, 8003-8011. MEDLINE Abstract

44. Schaaper, R. M., Koffel-Schwartz, N. and Fuchs, R. P. P. (1990) Carcinogenesis 11 1087-1095. MEDLINE Abstract

45. Shibutani, S. and Grollman, A. P. (1993) J. Biol. Chem. 268, 11703-11710.

*To whom correspondence should be addressed. Tel: +1 516 444 3649; Fax: +1 516 444 3218; Email: cds@pharm.sunysb.edu

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 13 May 1998
Copyright©Oxford University Press, 1998.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
A. Bibillo, D. Lener, A. Tewari, and S. F. J. Le Grice
Interaction of the Ty3 Reverse Transcriptase Thumb Subdomain with Template-Primer
J. Biol. Chem., August 26, 2005; 280(34): 30282 - 30290.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
V. Simonelli, L. Narciso, E. Dogliotti, and P. Fortini
Base excision repair intermediates are mutagenic in mammalian cells
Nucleic Acids Res., August 2, 2005; 33(14): 4404 - 4411.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
C. Maufrais, G. V. Fazakerley, J. Cadet, and Y. Boulard
Structural study of DNA duplex containing an N-(2-deoxy-{beta}-D-erythro-pentofuranosyl) formamide frameshift by NMR and restrained molecular dynamics
Nucleic Acids Res., October 15, 2003; 31(20): 5930 - 5940.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (142K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (13)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Lin, Z.
Right arrow Articles by de los Santos, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lin, Z.
Right arrow Articles by de los Santos, C.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?