N.m.r. determination of the solution conformation and dynamics of the A.G mismatch in the d(CGCAAATTGGCG)2 dodecamer

Dynamic basis for dA•dGTP and dA•d8OGTP misincorporation via Hoogsteen base pairs

Replicative errors contribute to the genetic diversity needed for evolution but in high frequency can lead to genomic instability. Here, we show that DNA dynamics determine the frequency of misincorporating the A•G mismatch, and altered dynamics explain the high frequency of 8-oxoguanine (8OG) A•8OG misincorporation. NMR measurements revealed that Aanti•Ganti (population (pop.) of >91%) transiently forms sparsely populated and short-lived Aanti+•Gsyn (pop. of ~2% and kex = kforward + kreverse of ~137 s-1) and Asyn•Ganti (pop. of ~6% and kex of ~2,200 s-1) Hoogsteen conformations. 8OG redistributed the ensemble, rendering Aanti•8OGsyn the dominant state. A kinetic model in which Aanti+•Gsyn is misincorporated quantitatively predicted the dA•dGTP misincorporation kinetics by human polymerase β, the pH dependence of misincorporation and the impact of the 8OG lesion. Thus, 8OG increases replicative errors relative to G because oxidation of guanine redistributes the ensemble in favor of the mutagenic Aanti•8OGsyn Hoogsteen state, which exists transiently and in low abundance in the A•G mismatch.

Influence of pH and Mg(ii) on the catalytic core domain 5 of a bacterial group II intron

RNA molecules fold into complex structures that allow them to perform specific functions. To compensate the relative lack of diversity of functional groups within nucleotides, metal ions work as crucial co-factors. In addition, shifted pK_as are observed in RNA, enabling acid-base reactions at ambient pH. The central catalytic domain 5 (D5) hairpin of the Azotobacter vinelandii group II intron undergoes both metal ion binding and pH dependence, presumably playing an important functional role in the ribozyme's reaction. By NMR spectroscopy we have here characterized the metal ion binding sites and affinities for the hairpin's internal G-A mismatch, bulge, and pentaloop. The influence of Mg(ii) and pH on the local conformation of the catalytically crucial region is also explored by fluorescence spectroscopy.

Photophysical Characterization of Enhanced 6-Methylisoxanthopterin Fluorescence in Duplex DNA

The structure and dynamic motions of bases in DNA duplexes and other constructs are important for understanding mechanisms of selectivity and recognition of DNA-binding proteins. The fluorescent guanine analogue, 6-methylisoxanthopterin 6-MI, is well suited to this purpose as it exhibits an unexpected 3- to 4-fold increase in relative quantum yield upon duplex formation when incorporated into the following sequences: ATFAA, AAFTA, or ATFTA (where F represents 6-MI). To better understand some of the factors leading to the 6-MI fluorescence increase upon duplex formation, we characterized the effect of local sequence and structural perturbations on 6-MI photophysics through temperature melts, quantum yield measurements, fluorescence quenching assays, and fluorescence lifetime measurements. By examining 21 sequences we have determined that the duplex-enhanced fluorescence (DEF) depends on the composition of bases adjacent to 6-MI and the presence of adenines at locations n ± 2 from the probe. Investigation of duplex stability and local solvent accessibility measurements support a model in which the DEF arises from a constrained geometry of 6-MI in the duplex, which remains H-bonded to cytosine, stacked with adjacent bases and inaccessible to quenchers. Perturbation of DNA structure through the introduction of an unpaired base 3' to 6-MI or a mismatched basepair increases 6-MI dynamic motion leading to fluorescence quenching and a reduction in quantum yield. Molecular dynamics simulations suggest the enhanced fluorescence results from a greater degree of twist at the X-F step relative to the quenched duplexes examined. These results point to a model where adenine residues located at n ± 2 from 6-MI induce a structural geometry with greater twist in the duplex that hinders local motion reducing dynamic quenching and producing an increase in 6-MI fluorescence.

Protonation-Dependent Base Flipping at Neutral pH in the Catalytic Triad of a Self-Splicing Bacterial Group II Intron

NMR spectroscopy has revealed pH-dependent structural changes in the highly conserved catalytic domain 5 of a bacterial group II intron. Two adenines with pK(a) values close to neutral pH were identified in the catalytic triad and the bulge. Protonation of the adenine opposite to the catalytic triad is stabilized within a G(syn)-AH(+) (anti) base pair. The pH-dependent anti-to-syn flipping of this G in the catalytic triad modulates the known interaction with the linker region between domains 2 and 3 (J23) and simultaneously the binding of the catalytic Mg(2+) ion to its backbone. Hence, this here identified shifted pK(a) value controls the conformational change between the two steps of splicing.

TEMPO-derived spin labels linked to the nucleobases adenine and cytosine for probing local structural perturbations in DNA by EPR spectroscopy

Three 2´-deoxynucleosides containing semi-flexible spin labels, namely (T)A, (U)A and (U)C, were prepared and incorporated into deoxyoligonucleotides using the phosphoramidite method. All three nucleosides contain 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) connected to the exocyclic amino group; (T)A directly and (U)A as well as (U)C through a urea linkage. (T)A and (U)C showed a minor destabilization of a DNA duplex, as registered by a small decrease in the melting temperature, while (U)A destabilized the duplex by more than 10 °C. Circular dichroism (CD) measurements indicated that all three labels were accommodated in B-DNA duplex. The mobility of the spin label (T)A varied with different base-pairing partners in duplex DNA, with the (T)A•T pair being the least mobile. Furthermore, (T)A showed decreased mobility under acidic conditions for the sequences (T)A•C and (T)A•G, to the extent that the EPR spectrum of the latter became nearly superimposable to that of (T)A•T. The reduced mobility of the (T)A•C and (T)A•G mismatches at pH 5 is consistent with the formation of (T)AH(+)•C and (T)AH(+)•G, in which protonation of N1 of A allows the formation of an additional hydrogen bond to N3 of C and N7 of G, respectively, with G in a syn-conformation. The urea-based spin labels (U)A and (U)C were more mobile than (T)A, but still showed a minor variation in their EPR spectra when paired with A, G, C or T in a DNA duplex. (U)A and (U)C had similar mobility order for the different base pairs, with the lowest mobility when paired with C and the highest when paired with T.

Mispair-specific recruitment of the Mlh1-Pms1 complex identifies repair substrates of the Saccharomyces cerevisiae Msh2-Msh3 complex

DNA mismatch repair is initiated by either the Msh2-Msh6 or the Msh2-Msh3 mispair recognition heterodimer. Here we optimized the expression and purification of Saccharomyces cerevisiae Msh2-Msh3 and performed a comparative study of Msh2-Msh3 and Msh2-Msh6 for mispair binding, sliding clamp formation, and Mlh1-Pms1 recruitment. Msh2-Msh3 formed sliding clamps and recruited Mlh1-Pms1 on +1, +2, +3, and +4 insertion/deletions and CC, AA, and possibly GG mispairs, whereas Msh2-Msh6 formed mispair-dependent sliding clamps and recruited Mlh1-Pms1 on 7 of the 8 possible base:base mispairs, the +1 insertion/deletion mispair, and to a low level on the +2 but not the +3 or +4 insertion/deletion mispairs and not on the CC mispair. The mispair specificity of sliding clamp formation and Mlh1-Pms1 recruitment but not mispair binding alone correlated best with genetic data on the mispair specificity of Msh2-Msh3- and Msh2-Msh6-dependent mismatch repair in vivo. Analysis of an Msh2-Msh6/Msh3 chimeric protein and mutant Msh2-Msh3 complexes showed that the nucleotide binding domain and communicating regions but not the mispair binding domain of Msh2-Msh3 are responsible for the extremely rapid dissociation of Msh2-Msh3 sliding clamps from DNA relative to that seen for Msh2-Msh6, and that amino acid residues predicted to stabilize Msh2-Msh3 interactions with bent, strand-separated mispair-containing DNA are more critical for the recognition of small +1 insertion/deletions than larger +4 insertion/deletions.

Probing nucleobase mismatch variations by electrochemical techniques: exploring the effects of position and nature of the single-nucleotide mismatch

Electrochemical impedance spectroscopy (EIS) has been used as an ultrasensitive tool for label-free detection of single-nucleotide mismatches in double-stranded DNA (ds-DNA) films. In this study, we have explored the effects of the position and of the type of single-nucleotide mismatch in ds-DNA on gold surfaces and were able to distinguish mismatch positions and mismatch pairs. The single-nucleotide mismatches A-C, A-A and A-G were introduced at three positions within the sequence in bottom, middle and top positions of ds-DNA, the films were studied by EIS, and the impedance results were interpreted with the help of equivalent circuits. The DeltaR(ct), the difference in charge transfer resistance before and after the addition of Zn(2+), was used to distinguish single-nucleotide mismatch within the DNA sequences. Importantly, the mismatch pair is easily distinguishable at the middle position. A purine-pyrimidine mismatch can be distinguished from purine-purine mismatch by its lower DeltaR(ct) value. In addition, all ds-DNA films were studied by scanning electrochemical microscopy in the absence and presence of Zn(2+), allowing us to distinguish a range of mismatched films from matched ds-DNA film.

Time-resolved fluorescent properties of 8-vinyl-deoxyadenosine and 2-amino-deoxyribosylpurine exhibit different sensitivity to their opposite base in duplexes

8-Vinyl-deoxyadenosine (8VA) has been recently introduced as a fluorescent analogue of adenosine that is less perturbing and less quenched than the well-established 2-amino-deoxyribosylpurine (2AP) probe when inserted in oligonucleotides. To further validate 8VA as a fluorescent substitute of A, we compared the ability of 8VA and 2AP in sequences of the type d(CGT TTT XNX TTT TGC) (with N=8VA or 2AP and X=T and C) to discriminate the nature of the opposite base (Y) in duplexes. For both probes, systematic variations in the amplitudes of the short- and long-lived lifetimes of the fluorescence intensity decays as well as in the amplitude of the fast rotational correlation time of the fluorescence anisotropy decays were observed as a function of the nature of Y. From these parameters, we inferred a stability order 8VA-T > 8VA-G > 8VA-A > 8VA-C, similar to the stability order with the native A base, but different from the stability order with 2AP. Using a combination of molecular mechanics and ab initio calculations, we found that the time-resolved parameters of 8VA, but not the 2AP ones, correlate well with the geometry and the strength of the A-Y base-pairing interaction. This may be rationalized by the smaller structural and electronic perturbations induced by the vinyl group in position 8 as compared to the amino group at position 2. As a consequence, substitution of A by 8VA in a base pair was found to only minimally modify the structure and interaction energy of the base pair. Thus, 8VA can be used as a fluorescent substitute of the natural A, to straightforwardly discriminate the nature of the opposite base. This may find interesting applications notably in the elucidation of the mechanisms and dynamics of the DNA mismatch repair system.

Formation of purine-purine mispairs by Sulfolobus solfataricus DNA polymerase IV

DNA damage that stalls replicative polymerases can be bypassed with the Y-family polymerases. These polymerases have more open active sites that can accommodate modified nucleotides. The lack of protein-DNA interactions that select for Watson-Crick base pairs correlate with the lowered fidelity of replication. Interstrand hydrogen bonds appear to play a larger role in dNTP selectivity. The mechanism by which purine-purine mispairs are formed and extended was examined with Solfolobus solfataricus DNA polymerase IV, a member of the RAD30A subfamily of the Y-family polymerases, as is pol eta. The structures of the purine-purine mispairs were examined by comparing the kinetics of mispair formation with adenine versus 1-deaza- and 7-deazaadenine and guanine versus 7-deazaguanine at four positions in the DNA, the incoming dNTP, the template base, and both positions of the terminal base pair. The time course of insertion of a single dNTP was examined with a polymerase concentration of 50 nM and a DNA concentration of 25 nM with various concentrations of dNTP. The time courses were fitted to a first-order equation, and the first-order rate constants were plotted against the dNTP concentration to produce k pol and K d (dNTP) values. A decrease in k pol/ K d (dNTP) associated with the deazapurine substitution would indicate that the position is involved in a crucial hydrogen bond. During correct base pair formation, the adenine to 1-deazaadenine substitution in both the incoming dNTP and template base resulted in a >1000-fold decrease in k pol/ K d (dNTP), indicating that interstrand hydrogen bonds are important in correcting base pair formation. During formation of purine-purine mispairs, the k pol/ K d (dNTP) values for the insertion of dATP and dGTP opposite 7-deazaadenine and 7-deazaguanine were decreased >10-fold with respect to those of the unmodified nucleotides. In addition, the rate of incorporation of 1-deaza-dATP opposite guanine was decreased 5-fold. These results suggest that during mispair formation the newly forming base pair is in a Hoogsteen geometry with the incoming dNTP in the anti conformation and the template base in the syn conformation. These results indicate that Dpo4 holds the incoming dNTP in the normal anti conformation while allowing the template nucleotide to change conformations to allow reaction to occur. This result may be functionally relevant in the replication of damaged DNA in that the polymerase may allow the template to adopt multiple configurations.

Secondary structure and stability of the selenocysteine insertion sequences (SECIS) for human thioredoxin reductase and glutathione peroxidase

We have used high resolution NMR and thermodynamics to characterize the secondary structure and stability of the selenocysteine insertion sequences (SECIS) of human glutathione peroxidase (58 nt) and thioredoxin reductase (51 nt). These sequences are members of the two classes of SECIS recently identified with two distinct structures capable of directing selenocysteine incorporation into proteins in eukaryotes. UV melting experiments showed a single cooperative and reversible transition for each RNA, which indicates the presence of stable secondary structures. Despite their large size, the RNAs gave well resolved NMR spectra for the exchangeable protons. Using NOESY, the imino protons as well as the cytosine amino protons of all of the Watson-Crick base pairs were assigned. In addition, a number of non-canonical base pairs including the wobble G.U pairs were identified. The interbase-pair NOEs allowed definition of the hydrogen-bonded structure of the oligonucleotides, providing an experimental model of the secondary structure of these elements. The derived secondary structures are consistent with several features of the predicted models, but with some important differences, especially regarding the conserved sequence motifs.

Solution structure of Cobalt(III)hexammine complexed to the GAAA tetraloop, and metal-ion binding to G.A mismatches

The solution structure of a 22 nt RNA hairpin and its complex with Co(NH(3))(6)(3+) bound to the GAAA tetraloop has been determined by NMR spectroscopy. Co(NH(3))(6)(3+) has a similar geometry to Mg(H(2)O)(6)(2+) and can be used as a probe for binding sites of completely solvated magnesium ions. The hairpin contains tandem G.A mismatches, similar to the P5abc region of a group I intron, and is closed by a GAAA tetraloop. The tandem G.A mismatches are imino hydrogen bonded in contrast with the sheared G.A mismatches found in a different context in the crystal structure of the P4-P6 domain. Chemical shift changes of the imino protons upon titration of the RNA hairpin with Mg(2+) and with Co(NH(3))(6)(3+) were used to identify ion-binding sites. Paramagnetic resonance broadening upon titration with Mn(2+) was also used. The titration curves gave dissociation binding constants for the magnesium ions in the millimolar range, similar to the binding in the major groove of RNA at tandem G.U base-pairs. Although the largest chemical shift change occurred at an imino proton of one of the G.A base-pairs, no nuclear Overhauser enhancement cross-peaks between the cobalt ligand and neighboring RNA protons were seen, presumably due to the high mobility of the Co(NH(3))(6)(3+) at this site. Nuclear Overhauser enhancement cross-peaks between Co(NH(3))(6)(3+) and the GAAA tetraloop were observed, which allowed the determination of the structure of the tetraloop binding site. The Co(NH(3))(6)(3+) is bound in the major groove of the GAAA tetraloop with hydrogen bonds to guanine base N7 and to phosphate oxygen atoms of the tetraloop.

AT selectivity and DNA minor groove binding: modelling, NMR and structural studies of the interactions of propamidine and pentamidine with d(CGCGAATTCGCG)2

A molecular modelling strategy has been developed to identify potential binding sites for bis(amidine) ligands in the minor groove of duplex DNA. Calculations of interaction energy for propamidine and pentamidine with d(CGCGAAT TCGCG)2 show that this duplex contains two symmetrically equivalent binding sites of identical affinity, each displaced by 0.3-0.4 bp from the centre of the AT segment. The ligands occupy groove sites spanning approximately 4 and 4-5 bp, respectively with asymmetric binding to the 5'-AATT sequence. The DNA-bis(amidine) interactions have been examined by high-resolution 1H-NMR. The patterns of induced changes in DNA proton chemical shift and the DNA-ligand NOEs confirm that both agents bind in the AT minor groove in a non-centrosymmetric fashion. Detailed structures were determined for each complex using a NOE-restrained simulated annealing procedure, showing that the B-type DNA conformation is not significantly altered upon complexation with either ligand. The free DNA duplex has previously been shown to be extensively hydrated in the minor groove [Kubinec, M.G. and Wemmer, D.E. (1992) J. Am, Chem. Soc. 114, 8739-8740 Liepinsh, E. Otting, G. and Wüthrich, K. (1992) Nucleic Acids Res. 20. 6549-6553]. We detect hydration water close to the A(H2) protons in the presence of propamidine, which may stabilise certain waters against exchange. This conclusion supports recent crystallographic analyses, suggesting that such ligands may use water molecules to bridge between amidinium protons and host DNA bases Details of the ligand interactions with AT-tract DNA duplexes can now be compared for the subsequences 5'-AAT, 5'-AATT and 5'-AAATTT.

DNA structural forms

In the years that have passed since the publication of Wolfram Saenger's classic book on nucleic acid structure (Saenger, 1984), a considerable amount of new data has been accumulated on the range of conformations which can be adopted by DNA. Many unusual species have joined the DNA zoo, including new varieties of two, three and four stranded helices. Much has been learnt about intrinsic DNA curvature, dynamics and conformational transitions and many types of damaged or deformed DNA have been investigated. In this article, we will try to summarise this progress, pointing out the scope of the various experimental techniques used to study DNA structure, and, where possible, trying to discern the rules which govern the behaviour of this subtle macromolecule. The article is divided into six major sections which begin with a general discussion of DNA structure and then present successively, B-DNA, DNA deformations, A-DNA, Z-DNA and DNARNA hybrids. An extensive set of references is included and should serve the reader who wishes to delve into greater detai.

The thermodynamics of DNA structures that contain lesions or guanine tetrads

It is becoming increasingly apparent that energetic as well as structural information is required to develop a complete appreciation of the critical interrelationships between structure, energetics, and biological function. Motivated by this recognition, we have reviewed in this article the current state of the thermodynamic databases associated with lesion-containing DNA duplexes and DNA quadruplexes, while highlighting important considerations concerning the methods used to obtain the requisite data.

Interaction of minor-groove-binding diamidine ligands with an asymmetric DNA duplex. NMR and molecular modelling studies

The aromatic diamidines berenil and propamidine bind reversibly to A+T-rich sites in the minor groove of B-form DNA duplexes. Based on extensive solution and crystallographic information we have designed a non-self-complementary double-stranded DNA sequence, d(GCAATGAGCG).d(CGCTCATTGC), that should contain a near-ideal binding site for berenil and a poorer site for the larger propamidine molecule, viz. d(AAT).d(ATT). 1H-NMR studies show that both ligands bind with 1:1 stoichiometry to the embedded 5'-AAT site, and induce numerous shifts of NMR resonances of DNA protons located in the minor groove. In addition, interactions with each strand can be distinguished by NOE spectroscopy due to the inherent asymmetry of the DNA. Detailed modelling based on experimental data show that no significant distortion of the B-DNA duplex is induced by either ligand. Sufficient NOE data were obtained to determine the position of the bound ligands in each complex. These conclusions are in agreement with predictions from molecular modelling calculations that provide a microscopic energy profile for interaction with the minor groove tract. Such calculations reveal an unexpected heterogeneous 5'-ATGA binding site that includes a spanned guanosine. This secondary binding site accounts for the extensive chemical shift perturbation induced by these ligands. The structures of the free DNA and the reversible complexes formed with each ligand molecule have been refined using an NOE-restrained isothermal annealing procedure. These structures confirm that the introduced ligands effect minimal perturbation of the helix, with binding to the 5'-AAT base sequence largely determined by specific non-bonded interactions.

Determination of fast dynamics of nucleic acids by NMR

Double-stranded oligonucleotides of < 10 base pairs are adequately described as an isotropic rotor, using the correlation time for the cytosine H6-H5 vector. For longer fragments, the cylindrical model should be used for detailed analysis of NOEs. The appropriate correlation times can be calculated using the formulae of Tirado and Garcia de la Torre or derived from measurements of the cross-relaxation rate constants for cytosine (or uridine) H6-H5. Order parameters describing the degree of motion of different vectors on the subnanosecond time scale vary substantially, with typical values of S2 > 0.8 for base vectors and 0.5-0.8 for intrasugar and base-sugar vectors. Order parameters for terminal nucleotides are typically significantly smaller than for internal nucleotides, which may also mean that their conformation will be less well determined in the formalism of a unique structure. The CSA relaxation rates of the phosphodiesters appear to be insensitive to internal motions and may, therefore, provide the most accurate estimate of the overall tumbling time in nucleic acid fragments. Using a combination of relaxation data for different nuclei and different spectrometer frequencies may be expected to yield detailed information about fast motions in nucleic acid fragments.

Properties of multiple G.A mismatches in stable oligonucleotide duplexes

The solution structure of the deoxydecanucleotide [sequence: see text] has been determined by NMR methods. This duplex, which contains six G.A mismatches and four Watson-Crick base pairs, is thermodynamically more stable than a decamer where T.A base pairs are substituted for the G.A mismatches, and is less stable than the duplex that contains G.C base pairs. Circular-dichroism spectroscopy indicates an overall B-like conformation for the decamer, but stronger than usual base stacking. 1H-NMR spectroscopy revealed that the N1H groups of the mismatched guanine residues are not hydrogen bonded, and 31P-NMR showed the presence of BII phosphate conformations for the GpA steps. Detailed analysis of the NMR data showed that all nucleotides have anti glycosidic torsion angles and S type sugar puckers. The G.A mismatches pair in the amino form as originally proposed by Li et al. [Li, Y., Zon, G. & Wilson, W. D. (1991) Proc. Natl Acad. Sci. USA 88, 26-30], which results in extensive base-base stacking between the tandem G.A base pairs and their nearest neighbours. The terminal G.A base pairs are less stable than the central base pairs and show evidence of an equilibrium between two conformations, one involving BII phosphate.

Thermodynamic stability and solution conformation of tandem G.A mismatches in RNA and RNA.DNA hybrid duplexes

G.A mismatches form a variety of hydrogen-bonded structures in DNA, most of which destabilise the duplex. Tandem G.A mismatches in the context YGAR (Y = pyrimidine, R = purine), however, form base pairs using the amino group of the guanine residue [Li, Y., Zon. G. & Wilson, W.D. (1991) Proc. Natl Acad. Sci. USA 88, 26-30], which permits extensive base-base stacking, leading to a slight stabilisation of the helix [Ebel, S., Lane, A. N. & Brown, T. (1992) Biochemistry 31, 12083-12086]. We have measured the thermodynamic stability of several RNA and RNA.DNA hybrid duplexes containing tandem G.A mismatches. The RNA duplexes are intrinsically much more stable than the corresponding DNA duplexes and the mutations are destabilising in all cases. NOE and coupling-constant data show that all of the sugars are in the C3'-endo range of conformations, and glycosidic torsion angles are in the range -160 degrees to -180 degrees in [sequence: see text]. Both sequential NOE intensities and circular-dichroism measurements indicate that the global conformation of the mismatched RNA is A-like. The N1H group of the mismatched guanine residue is not involved in hydrogen bonding with the adenine residue, indicating the presence of the amino-pairing scheme. Determination of the structure using 'loose' NMR-derived constraints shows that the potential energies of the imino-paired and amino-paired forms are similar, but substantially higher than energy-minimised RNA. Using tighter constraints derived from more extensive analysis of one-dimensional and two-dimensional NOE data showed that the amino-paired structure agrees with the constraint data better than the imino-paired structure, and also accounts for unusual chemical shifts and the lack of hydrogen bonding of the guanine N1H group. Resulting molecular models show that the amino-paired mismatches are not as extensively stacked on the neighbouring part of the duplex as in the B-DNA analogues, largely accounting for the lower thermodynamic stability in the RNA duplexes.

Nuclear magnetic resonance measurements of slow conformational dynamics in macromolecules

The combined use of rotating-frame relaxation methods, temperature-dependent measurements of line shapes and magnetization transfer experiments allows in favorable cases the examination in some detail of exchange processes that occur on the millisecond time scale. It is possible to determine not only the rate constants, but also the activation parameters and chemical shifts even for events that are in fast exchange on the chemical shift time scale. Such measurements complement the information obtainable from heteronuclear relaxation methods that probe mainly the fast librational motions in macromolecules and may provide information important for functional studies of biological macromolecules.

Solution structure of a conserved DNA sequence from the HIV-1 genome: restrained molecular dynamics simulation with distance and torsion angle restraints derived from two-dimensional NMR spectra

The three-dimensional solution structure of a trisdecamer DNA duplex sequence, d(AGCTTGCCTTGAG).d(CTCAAGGCAAGCT), from a conserved region of HIV-1 genome's long terminal repeat, has been investigated using NMR spectroscopy and restrained molecular dynamics calculations. Interproton distances derived from two-dimensional nuclear Overhauser enhancement (2D NOE) experiments, using the iterative complete relaxation matrix algorithm MARDIGRAS, and torsion angles for sugar rings, estimated from stimulated fitting of double-quantum-filtered correlation (2QF-COSY) spectra, were obtained [Mujeeb, A., Kerwin, S. M., Egan, W. M., Kenyon, G. L., & James, T. L. (1992) Biochemistry 31, 9325-38]. These structural restraints have now been employed as the basis for structure refinement using restrained molecular dynamics (rMD) to search conformational space for structures consistent with the experimental restraints. Specifically, upper and lower bounds on the restraints were incorporated into the AMBER (version 4.0) total potential energy function of the system, the bounds being used to define the width of a flat-well penalty term in the AMBER force field. Confidence in the time-averaged structure obtained is engendered by convergence to essentially the same structure (root-mean-square deviation approximately 0.9 A) when two quite different DNA models, A-DNA and B-DNA (RMSD approximately 6.5 A), were employed as starting structures and when various initial trajectories were used for the rMD runs. The derived structure is further supported by the total energy calculated, the restraint violation energy, the restraint deviations, and the fit with experimental data. For the latter, the sixth-root residual index indicated a good fit of the determined structure with experimental 2D NOE spectral intensities (R1x < 0.07), and the RMS difference between vicinal proton coupling constants calculated for the derived structure and experimental coupling constants were also in reasonable agreement (JRMS = 0.9 Hz). While the structure of the trisdecamer is basically in the B-DNA family, some structural parameters manifest interesting local variations. The helix parameters indicate that, compared with classical B-DNA, the structure is longitudinally more compressed. Local structural variations at the two TG steps in particular together create bending into the major groove of the duplex. Comparison of the two-CTTG-tetrads in the duplex reveals that they have similar structures, with the TT moieties being almost identical; however, the -CTTG-pur sequence has a larger roll and slide for the -TG- step than for the -CTTG-pyr sequence, in accord with published X-ray crystallographic conclusions.

DNA methylation by N-methyl-N-nitrosourea: methylation pattern changes in single- and double-stranded DNA, and in DNA with mismatched or bulged guanines

The detection of abnormal DNA base pairing arrangements and conformations is chemically probed in synthetic 32P-end-labeled deoxyribonucleotide oligomers using N-methyl-N-nitrosourea (MNU) and 2,12,-dimethyl-3,7,11,17-tetraazabicyclo-[11.3.1]heptadeca-1 -[17],2,11,13,15 pentaene-Ni (II) (Ni-complex) with KHSO5. The DNA targets studied are single-stranded (s-s) DNA, double-stranded (d-s) DNA, d-s DNA with G-G, G-A and G-T mismatches, d-s DNA with a single bulged G and d-s DNA with two bulged G's. The effect of the non-Watson--Crick structures on the formation of N7-methylguanine (N7-MeG) by MNU and the oxidation of G by Ni-complex is reported along with the Tm's and circular dichroism spectra of the different duplex oligomers. The results for MNU and Ni-complex show that the qualitative and quantitative character of the cleavage patterns at a G3 run change with the nature of the abnormal base pairing motif. Based on the DNA substrates studied, the results indicate that a combination of reagents which report electronic and steric perturbations can be a useful approach to monitor DNA mismatches and bulges.

The activating transcription factor region within the E2A promoter exists in a novel conformation

The ATF (activating transcription factor) binding site within the E2A promoter region of adenovirus is shown to exist in a novel conformation in vitro via nuclear magnetic resonance methods. This novel conformation may be important to the protein DNA recognition process. This conformation has characteristics of both A- and B-form DNA. From circular dichroism and through-space-based NMR experiments, it is clear that the overall helical structure is B-like. However, the 1H-1H coupling constant information indicates that most of the sugar puckers of the individual nucleotides are in the C3'-endo/C4'-exo range which is more characteristic of A-form DNA. The sugar conformation can also be described as a mixture of two states, C3'-endo and C2'-endo, where many of the sugars exist mainly in the C3'-endo state. These data show that the conformation of the sugar puckers does not determine the nature of the overall base stacking on the DNA. Helical parameters were calculated from NOE build-up curves for half of the dinucleotide pairs. Severe spectral overlap on the nuclear Overhauser based spectra prevented determination of the helical parameters for all of the dinucleotide base-pairs. Energy minimization and molecular dynamic simulation methods using the sugar pucker and glycosidic torsion angles determined from the NMR data as constraints were carried out in order to demonstrate that such a conformation was energetically favorable.

Conformational properties of the -35 region of the trp promoter in solution: comparison of the wild-type sequence with an AT transversion

The majority of the 1H NMR resonances of the protons in a tetradecamer containing the -35 region of the trp promoter d(GCTGTTGACAATTA): d(TAATTGTCAACAGC) and in the TA transversion have been assigned. The conformational properties of the nucleotides have been determined and compared in the two duplexes. Analysis of spin-spin coupling and NOEs shows that all sugar puckers are in the south domain (i.e. near C2' endo) and the glycosidic torsion angles are anti (chi approximately 110 degrees). The NMR data are consistent with the duplex being in the B family of conformations. Significant differences in chemical shifts between the two molecules were observed only for nearest neighbours to the transversion site, suggesting the absence of long range conformational effects. This was confirmed by the similarity of coupling constants and NOEs. Other properties are also not greatly affected at positions more than two base pairs from the mutation site. These results are consistent with the hypothesis that unconstrained oligonucleotides are highly flexible, and can readily accommodate significant perturbations of the local structure, such as a transversion.

Structural studies of the 5'-phenazinium-tethered matched and G-A-mismatched DNA duplexes by NMR spectroscopy

The mechanism through which modified oligo-DNA analogues act as antisense repressors at the transcriptional and translational level of gene expression is based on the information content in the nucleotide sequence which is determined by the specific base pairing. The efficiency of such action is largely determined by the stability of the duplex formed between the oligonucleotide reagent and the target sequence and also by the mismatched base pairing, such as G-A, that occurs during replication or recombination. We herein report that the phenazinium (Pzn)-tethered matched duplex p(d(TGTTTGGC)):(Pzn)-p(d(CCAAACA)) (III) (Tm = 50 degrees C) has a much larger stability than the parent matched duplex p(d(TGTTTGGC)):p(d(CCAAACA)) (I) (Tm = 30 degrees C). On the other hand, the Pzn-tethered G-A-mismatched duplex p(d(TGTTTGGC)):(Pzn)-p(d(ACAAACA)) (IV) (Tm = 34 degrees C) is only slightly more stable than its parent mismatched duplex p(d(TGTTTGGC)):p(d(ACAAACA)) (Tm = 25 degrees C). A detailed 500 MHz NMR study and constrained MD refinements of NMR-derived structures have been undertaken for the DNA duplexes (I), (II), (III) and (IV) in order to understand the structural basis of stabilization of Pzn-tethered matched DNA duplex (delta Tm = 20 degrees C) compared to mismatched duplex (delta Tm = 9 degrees C). Assignment of the 1H-NMR (500 MHz) spectra of the duplexes has been carried out by 2D NOESY, HOHAHA and DQF-COSY experiments. The torsion angles have been extracted from the J-coupling constants obtained by simulation of most of the DQF-COSY cross-peaks using program SMART. The solution structure of the duplexes were assessed by an iterative hybride relaxation matrix method (MORASS) combined with NOESY distances and torsion angles restrained molecular dynamics (MD) using program Amber 4.0. The standard Amber 4.0 force-field parameters were used for the oligonucleotide in conjunction with the new parameters for Pzn residue which was obtained by full geometry optimization using ab initio program (3-21G basis set). It has been shown that mismatched G-A bases are in the anti-anti conformation. The mismatched 7G-1A form stable base pairs through inter-strand hydrogen bonds (N7(A)...HN2(G) (1.92 A) with a subtended angle of 176 degrees and N3(G)...HN6(A) (2.01 A) with a subtended angle of 153 degrees (the 'amino-type' hydrogen bond)) and a propeller twist of 36 degrees for 7G-1A residues.(ABSTRACT TRUNCATED AT 400 WORDS)

NMR and molecular modeling studies of the interaction of berenil and pentamidine with d(CGCAAATTTGCG)2

The interaction of two anti-trypanosomal agents, berenil and pentamidine, with the A+T-rich dodecamer d(CGCAAATTTGCG)2 has been examined by high-resolution 1H-NMR, optical spectroscopy, and molecular modeling. Proton assignments for the free DNA and each DNA-ligand complex were obtained using nuclear Overhauser enhancement spectroscopy and total correlation spectroscopy. Complexation induces large changes in chemical shift for protons in the DNA minor groove for the A5-T9 segment, and intermolecular NOEs reveal contacts between the DNA bases and each ligand. The asymmetric binding site for berenil indicated by the NMR data suggests that at least two overlapping sites are involved. Rapid exchange between symmetrically-equivalent binding sites, via dissociative rearrangement, is consistent with retention of twofold degeneracy for both the ligand and the DNA host. Calculations of binding energy confirm that this DNA duplex contains overlapping sites of similar binding affinity. In contrast, the larger pentamidine molecule occupies a site that spans four or five bp, with asymmetric binding to the minor-groove 5'-ATTT sequence. The B-type conformation of the DNA is not altered substantially by either ligand.

Crystal structure of an oligonucleotide duplex containing G.G base pairs: influence of mispairing on DNA backbone conformation

The structure of the G.G mispaired dodecanucleotide d(CGCGAATTGGCG)2 has been solved by x-ray crystallography and refined to an R factor of 18.8% at 2.2 A resolution for 3513 reflections. The dodecamer crystallizes as a B-type DNA double helix. It contains two G(anti).G(syn) base pairs--i.e., G-4/G-16(anti).G-21/G-9(syn). The Hoogsteen base pairing involves atoms O-6 and N-7 of the guanine in the syn conformation with atoms N-1 and N-2 of the anti-paired purine. One G.G base pair has a bifurcated hydrogen bond between G-4(N-1)...G-21(N-7) and G-4(N-1)...G-21(O-6). There is little overall structural distortion of the double helix induced as a consequence of the mispairing. The helical width is significantly increased by comparison with the structure of the native duplex, and the minor groove width in the 5'-AATT region is decreased. The G.G base pairing induces high-BII phosphate conformations at residues G-9 and T-20 in addition to more normal BII conformations at G-10 and G-22. It is suggested that these backbone aberrations provide signals for the facile repairability of G.G mispairs in DNA.

Very stable mismatch duplexes: structural and thermodynamic studies on tandem G.A mismatches in DNA

We have used ultraviolet melting techniques to compare the stability of several DNA duplexes containing tandem G.A mismatches to similar duplexes containing tandem A.G, I.A, and T.A base pairs. We have found that tandem G.A mismatches in 5'-Y-G-A-R-3' duplexes are more stable than their I.A counterparts and that they are sometimes more stable than tandem 5'-Y-T-A-R-3' sequences. This is not the case for tandem G.A mismatches in other base stacking environments, and it suggests that tandem G.A mismatches in 5'-Y-G-A-R-3' sequences have a unique configuration. In contrast to tandem 5'-G-A-3' mismatches, tandem 5'-A-G-3' mismatches were found to be unstable in all cases examined.

Solution conformation of a deoxynucleotide containing tandem G.A mismatched base pairs and 3'-overhanging ends in d(GTGAACTT)2

We have used 31P and 1H NMR spectroscopy and circular dichroism to define the solution conformation of d(GTGAACTT)2 which contains tandem G.A mismatched base pairs and 3'-overhanging TT ends. Measurements of coupling constants and NOE intensities show that the sugar puckers of the nucleotides are predominantly in the south domain (i.e., near C2'-endo) and that the glycosidic torsion angles are anti. The sequential NOE intensities indicate the presence of a right-handed helix. Analysis of the 31P and 1H NMR spectra of the duplex shows that the tandem mismatch forms a block in which there are unusual backbone torsion angles (i.e., in the BII state), within an otherwise B-like structure. The chemical shift of the N1H of the mismatched guanosine and NOEs between the mismatched base pairs and their nearest neighbors are inconsistent with the imino pairing present in single A.G mismatches or in the X-ray structure of a tandem mismatch [Privé, G. G., et al. (1987) Science 238, 498-503] but the data are consistent with the amino pairing found by Li et al. (1991) [Li, Y., et al. (1991) Proc. Natl. Acad. Sci. U.S.A. 88, 26-30]. The strong base-base stacking both within the tandem G.A block and between the G.A mismatches and their other nearest neighbors offsets the intrinsic destabilizing effects of the mismatch. Further, the 3'-TT overhangs stack onto the ends of the helix and stabilize the duplex against fraying, which accounts for the observed increase in the melting temperature compared with the flush-ended duplex.

NMR study of parallel-stranded tetraplex formation by the hexadeoxynucleotide d(TG4T)

Multistranded DNA structures based upon guanine association have been proposed to be important in the structure of chromosome telomeres and in immunoglobulin class switching. Nucleic acids containing runs of guanine bases form a number of structures in vitro, including fold-back structures (Fig. 1a) and parallel-stranded quadruplex structures in DNA and RNA. The features of fold-back structures have now been determined at high-resolution. The different structures are probably based on a tetrad of hydrogen-bonded guanine bases (Fig. 1b), with buffer conditions and sequence effects mediating isomerization between the different forms. Here we use NMR spectroscopy to investigate the solution structure of the complex formed by the hexadeoxynucleotide d(TG4T) in the presence of sodium ions. We have observed the formation of a parallel-stranded quadruplex containing hydrogen-bonded tetrads of guanine. The parallel-stranded form differs significantly from the fold-back form, with individual nucleotide conformations being closer to those of B-form DNA.

The conformational variability of an adenosine.inosine base-pair in a synthetic DNA dodecamer

A crystal structure analysis of the synthetic deoxydodecamer d(CGCAAATTIGCG) which contains two adenosine.inosine (A.I) mispairs has revealed that, in this sequence, the A.I base-pairs adopt a A(anti).I(syn) configuration. The refinement converged at R = 0.158 for 2004 reflections with F greater than or equal to 2 sigma(F) in the range 7.0-2.5A for a model consisting of the DNA duplex and 71 water molecules. A notable feature of the structure is the presence of an almost complete spine of hydration spanning the minor groove of the whole of the (AAATTI)2 core region of the duplex. pH-dependent ultraviolet melting studies have suggested that the base-pair observed in the crystal structure is, in fact, a protonated AH+ (anti).I(syn) species and that the A.I base-pairs in the sequence studied display the same conformational variability as A.G mispairs in the sequence d(CGCAAATTGGCG). The AH+(anti).I(syn) base-pair predominates below pH 6.5 and an A(anti).I(anti) mispair is the major species present between pH 6.5 and 8.0. The protonated base-pairs are held together by two hydrogen bonds one between N6(A) and O6(I) and the other between N1(A) and N7(I). This second hydrogen bond is a direct result of the protonation of the N1 of adenosine. The ultraviolet melting studies indicate that the A(anti).I(anti) base-pair is more stable than the A(anti).G(anti) base-pair but that the AH+(anti).I(syn) base pair is less stable than its AH+(anti).G(syn) analogue. Possible reasons for this observation are discussed.

Conformational properties of the G.G mismatch in d(CGCGAATTGGCG)2 determined by NMR

The conformational properties of the DNA duplex d(CGCGAATTGGCG)2, which contains two noncomplementary G.G base pairs, have been examined in aqueous solution by 1H and 31P NMR as a function of temperature. The G.G mismatch is highly destabilizing, with a Tm value 35 K below that observed for the native EcoRI dodecamer. The dodecamer appears symmetric in the NMR spectra and exists largely as an average B-type DNA conformation. However, the 1H and 31P NMR spectra give evidence of considerable conformational heterogeneity at the mismatched nucleotides and their nearest neighbors, which increases with increasing temperature. There is no evidence for a significant population of the syn purine conformation. The imino protons of the mispaired bases G4 and G9 are degenerate, resonate at high field, and exchange readily with solvent. These results indicate that the mispaired bases are only weakly hydrogen-bonded and are only partially stacked into the helix. On raising the temperature, the duplex shows increasing exchange between two or more conformations originating from the mismatch sites. However, these additional conformations maintain their Watson-Crick hydrogen bonding. The increase in chemical exchange is consistent with a quasimelting process for which the G.G sites provide local nuclei. Extensive modeling studies by dynamic annealing have confirmed that the G(anti).G(anti) conformation is favored and that the mispairs are poorly stacked within the helix. The results explain both the poor thermal stability and low hypochromicity of this duplex.