Minimum energy conformations of DNA dimeric subunits: Potential energy calculations for dGpdC, dApdA, dCpdC, dGpdG, and dTpdT

Dinucleotide TpT and its 2'-O-Me analogue possess different backbone conformations and flexibilities but similar stacked geometries

UV irradiation at 254 nm of 2'-O,5-dimethyluridylyl(3'-5')-2'-O,5-dimethyluridine (1a) and of natural thymidylyl(3'-5')thymidine (1b) generates the same photoproducts (CPD and (6-4)PP; responsible for cell death and skin cancer). The ratios of quantum yields of photoproducts obtained from 1a (determined herein) to that from 1b are in a proportion close to the approximately threefold increase of stacked dinucleotides for 1a compared with those of 1b (from previous circular dichroism results). 1a and 1b however are endowed with different predominant sugar conformations, C3'-endo (1a) and C2'-endo (1b). The present investigation of the stacked conformation of these molecules, by unrestrained state-of-the-art molecular simulation in explicit solvent and salt, resolves this apparent paradox and suggests the following main conclusions. Stacked dinucleotides 1a and 1b adopt the main characteristic features of a single-stranded A and B form, respectively, where the relative positions of the backbone and the bases are very different. Unexpectedly, the geometry of the stacking of two thymine bases, within each dinucleotide, is very similar and is in excellent agreement with photochemical and circular dichroism results. Analyses of molecular dynamics trajectories with conformational adiabatic mapping show that 1a and 1b explore two different regions of conformational space and possess very different flexibilities. Therefore, even though their base stacking is very similar, these molecules possess different geometrical, mechanical, and dynamical properties that may account for the discrepancy observed between increased stacking and increased photoproduct formations. The computed average stacked conformations of 1a and 1b are well-defined and could serve as starting models to investigate photochemical reactions with quantum dynamics simulations.

Solvent influence on base stacking

In this paper we present a detailed analysis of the base-stacking phenomenon in different solvents, using nanosecond molecular dynamics simulations. The investigation focuses on deoxyribo- and ribodinucleoside monophosphates in aqueous and organic solutions. Organic solvents with a low dielectric constant, such as chloroform, and solvents with intermediate dielectric constants, such as dimethyl sulfoxide and methanol, were analyzed. This was also done for water, which is highly polar and has a high dielectric constant. Structural parameters such as the sugar puckering and the base-versus-base orientations, as well as the energetics of the solute-solvent interactions, were examined in the different solvents. The obtained data demonstrate that base stacking is favored in the high dielectric aqueous solution, followed by methanol and dimethyl sulfoxide with intermediate dielectric constants, and chloroform, with a low dielectric constant.

Influence of electrostatic interactions on the sugar-phosphate backbone conformation in DNA

The influence of sugar ring flexibility in DNA on the mechanism of the B <===> A conformational transition is studied. The dipole moment of the deoxyribose as a function of its puckered states is calculated by the quantum-mechanical method using the MINDO/3 approximation. The interaction of the sugar dipole with the neighbor molecular groups in polynucleotide chains is estimated. The sugar dipole interaction with phosphate groups and counterions is shown to be strong and capable to deform the pseudorotation potential of deoxyribose. The effective pseudorotation potential of sugar ring in the B and A helices is obtained. The results are used to explain the behavior of Raman bands in the region of sugar-phosphate vibrations. The mechanism of the effect of electrostatic forces on the sugar-phosphate backbone conformation, which is essential for the B <===> A and other structure transitions, is offered.

A Monte Carlo method for generating structures of short single-stranded DNA sequences

A Monte Carlo method has been developed for generating the conformations of short single-stranded DNAs from arbitrary starting states. The chain conformers are constructed from energetically favorable arrangements of the constituent mononucleotides. Minimum energy states of individual dinucleotide monophosphate molecules are identified using a torsion angle minimizer. The glycosyl and acyclic backbone torsions of the dimers are allowed to vary, while the sugar rings are held fixed in one of the two preferred puckered forms. A total of 108 conformationally distinct states per dimer are considered in this first stage of minimization. The torsion angles within 5 kcal/mole of the global minimum in the resulting optimized states are then allowed to vary by +/- 10 degrees in an effort to estimate the breadth of the different local minima. The energies of a total of 2187 (3(7)) angle combinations are examined per local conformational minimum. Finally, the energies of all dinucleotide conformers are scaled so that the populations of differently puckered sugar rings in the theoretical sample match those found in nmr solution studies. This last step is necessitated by limitations in the theoretical methods to predict DNA sugar puckering accurately. The conformer populations of the individual acyclic torsion angles in the composite dimer ensembles are found to be in good agreement with the distributions of backbone conformations deduced from nmr coupling constants and the frequencies of glycosyl conformations in x-ray crystal structures, suggesting that the low energy states are reasonable. The low energy dimer forms (consisting of 150-325 conformational states per dimer step) are next used as variables in a Monte Carlo algorithm, which generates the conformations of single-stranded d(CXnG) chains, where X = A, T and n = 3, 4, 5. The oligonucleotides are built sequentially from the 5' end of the chain using random numbers to select the conformations of overlapping dimer units. The simulations are very fast, involving a total of 10(6) conformations per chain sequence. The potential errors in the buildup procedure are minimized by taking advantage of known rotational interdependences in the sugar-phosphate backbone. The distributions of oligonucleotide conformations are examined in terms of the magnitudes, positions, and orientations of the end-to-end vectors of the chains. The differences in overall flexibility and extension of the oligomers are discussed in terms of the conformations of the constituent dinucleotide steps, while the general methodology is discussed and compared with other nucleic acid model building techniques.

Block-units method for conformational calculations of large nucleic acid chains. I. Block-units approximation of atomic structure and conformational energy of polynucleotides

A new block-units method for rapid conformational calculation of large nucleic acid fragments has been developed. Atomic structure of polynucleotides has been approximated by the block-units structure. Each monomer of the polynucleotide consists of three one-center block units: phosphate, ribose, and nucleic base. Full conformational energy of the polynucleotide is separated into two parts. The first part is the energy of the short-range interactions between adjacent block units and is calculated on the basis of the potential energy surface model of the dinucleotide fragment pXp (where X = A, G, T, U, C). The second part is the energy of the middle- and long-range interactions between separated block units, and is calculated as a sum of the effective interaction energies between centers of the block units. The present block-units method is in agreement within the range of +/- 0.5 kcal/mol with the method of the atom-atom potentials, but the former is 30-100-fold faster. The block-units method is recommended for screening of the probable conformations of the large polynucleotide systems.

Base sequence effects in double helical DNA. I. Potential energy estimates of local base morphology

A series of potential energy calculations have been carried out to estimate base sequence dependent structural differences in B-DNA. Attention has been focused on the simplest dimeric fragments that can be used to build long chains, computing the energy as a function of the orientation and displacement of the 16 possible base pair combinations within the double helix. Calculations have been performed, for simplicity, on free base pairs rather than complete nucleotide units. Conformational preferences and relative flexibilities are reported for various combinations of the roll, tilt, twist, lateral displacement, and propeller twist of individual residues. The predictions are compared with relevant experimental measures of conformation and flexibility, where available. The energy surfaces are found to fit into two distinct categories, some dimer duplexes preferring to bend in a symmetric fashion and others in a skewed manner. The effects of common chemical substitutions (uracil for thymine, 5-methyl cytosine for cytosine, and hypoxanthine for guanine) on the preferred arrangements of neighboring residues are also examined, and the interactions of the sugar-phosphate backbone are included in selected cases. As a first approximation, long range interactions between more distant neighbors, which may affect the local chain configuration, are ignored. A rotational isomeric state scheme is developed to describe the average configurations of individual dimers and is used to develop a static picture of overall double helical structure. The ability of the energetic scheme to account for documented examples of intrinsic B-DNA curvature is presented, and some new predictions of sequence directed chain bending are offered.

Conformational studies of nucleic acids: IV. The conformational energetics of oligonucleotides: d(ApApApA) and ApApApA

Utilizing a new method for modeling furanose pseudorotation (D. A. Pearlman and S.-H. Kim, J. Biomol. Struct. Dyn. 3, 85 (1985)) and the empirical multiple correlations between nucleic acid torsion angles we derived in the previous report (D. A. Pearlman and S.-H. Kim, previous paper in this issue), we have made an energetic examination of the entire conformational spaces available to two nucleic acid oligonucleotides: d(ApApApA) and ApApApA. The energies are calculated using a semi-empirical potential function. From the resulting body of data, energy contour map pairs (one for the DNA molecule, one for the RNA structure) have been created for each of the 21 possible torsion angle pairs in a nucleotide repeating unit. Of the 21 pairs, 15 have not been reported previously. The contour plots are different from those made earlier in that for each point in a particular angle-angle plot, the remaining five variable torsion angles are rotated to the values which give a minimum energy at this point. The contour maps are overall quite consistent with the experimental distribution of oligonucleotide data. A number of these maps are of particular interest: delta (C5'-C4'-C3'-O3')-chi (O4'-C1'-N9-C4), where the energetic basis for an approximately linear delta-chi correlation can be seen: zeta (C3'-O3'-P-O5')-delta, in which the experimentally observed linear correlation between zeta and delta in DNA(220 degrees less than zeta less than 280 degrees) is clearly predicted; zeta-epsilon (C4'-C3'-O3'-P), which shows that epsilon increases with decreasing zeta less than 260 degrees; alpha (O3'-P-O5'-C5')-gamma (O5'-C5'-C4'-C3') where a clear linear correlation between these angles is also apparent, consistent with experiment; and several others. For the DNA molecule studied here, the sugar torsion delta is predicted to be the most flexible, while for the RNA molecule, the greatest amount of flexibility is expected to reside in alpha and gamma. Both the DNA and RNA molecules are predicted to be highly polymorphic. Complete energy minimization has been performed on each of the minima found in the energy searches and the results further support this prediction. Possible pathways for B-form to A-form DNA interconversion suggested by the results of this study are discussed. The results of these calculations support use of the new sugar modeling technique and torsion angle correlations in future conformational studies of nucleic acids.

Vibronic exciton interactions. Resolution and interpretation of the temperature-dependent circular dichroism and absorption spectra of ApA and of dApdA

The circular dichroism and absorption spectra of the stacked and unstacked forms of ApA and dApdA were derived. The unstacked spectra are not identical with the corresponding free nucleoside spectrum. The stacked spectra can be satisfactorily described in terms of the vibronic degenerate exciton theory which suggests that the non-degenerate interactions have a less important influence on the observed circular dichroic spectra than hitherto assumed. From the experimental spectra the magnitudes of the exciton coupling were found to correspond to very fast transfer rates of energy between the adenine moieties. The absorption spectra of the stacked species are consistent with average angles of near 60 degrees and 20 degrees in ApA and dApdA, respectively, between the degenerate transition moments of the adenine moieties.

Conformation of the deoxydinucleoside monophosphate dCpdG modified at carbon 8 of guanine with 2-(acetylamino)fluorene

Minimized conformational potential energy calculations were performed for dCpdG modified with the carcinogen 2-(acetylamino)fluorene (AAF). The major adduct, linked via a covalent bond between guanine C-8 and N-2 of AAF, was investigated. The 12 variable torsion angles and both deoxyribose puckers were independent flexible parameters in the energy minimizations. Three categories of low-energy conformers were calculated in which the guanine was syn and nearly perpendicular to the plane of the fluorene: (1) forms in which fluorene is stacked with cytidine (included among these is the global minimum energy conformation); (2) conformers which preserve guanine-cytidine stacking while placing the fluorene in a base-pair obstructing position; (3) conformers which maintain guanine-cytidine stacking and place the fluorene at the helix exterior, without interfering with base pairing. The Z form is important in this group. In addition, a low-energy conformation with guanine anti, but still nearly perpendicular to fluorene, was computed. Molecular models were constructed for the most important conformations incorporated into larger polymers. These indicated that the fluorene-cytidine stacked forms induce a severe kink in the B helix. Conformers with guanine-cytidine stacking and AAF in a base-pair obstructing position place the AAF at the B-type helix interior with little distortion in the helix direction. Conformers with the guanine-cytidine stack in which AAF does not affect base pairing place the fluorene at the Z or alternate helix exterior. It is suggested that base sequence, extent of modification, and external conditions such as salt concentration determine which of a number of possible conformational effects is actually induced by AAF. The variety of observed experimental results with AAF-modified DNA may reflect there various conformational possibilities.

Force field conformational analysis of aminofluorene and acetylaminofluorene substituted deoxyguanosine

The hepatocarcinogen N-hydroxy-2-acetylaminofluorene forms two C8-substituted deoxyguanosine adducts in vivo. The conformation of these adducts, as well as 2'-deoxyguanosine and 8-amino-2'-deoxyguanosine has been studied with Allinger's force field. Using the glycoside rotation as a reaction coordinate, multidimensional potential energy surface were determined by relaxing all internal degrees of freedom. The calculations indicate the 2'-deoxyguanosine should exist as a mixture of syn and anti forms, that the syn form is slightly favored for 8-amino-2'-deoxyguanosine, that N-(deoxyguanosin-8-yl)-2-acetylaminofluorene (dG-C8-AAF) will only be found in the syn conformation and that, although the syn form is also more stable for N-(deoxyguanosin-8-yl)-2-aminofluorene (dG-C8-AF), it will have a substantially greater proportion of the anti-conformer than is found with dG-C8-AAF. The results of the force field calculations are discussed in relation to the effects these adducts may have on DNA structure.