The Molecular Dynamic Study of the Stacking Interaction in RNA Dinucleosides, the Dependence of Stacking Ability on RNA Sequence

The stacking interaction of nucleic acid bases is studied by molecular dynamic simulation in the ApA, ApC, CpA and CpC ribodinucleoside monophosphates. The rate of stacking was analyzed during the molecular dynamic simulation using three geometry parameters – the distance between the nitrogen atoms of nucleic acid bases linked to sugar with the glysosidic bond and by two angles that depend on mutual orientation of bases. Intervals for the three geometry descriptors that correspond to the “stacked” conformation of RNA dinucleoside used here were set in accord with the commonly accepted ideal stacked conformation (ideal A-RNA) known from the solid state studies (Nucleic Acid Database). A high stacking ability is observed for the ApA and CpA dinucleosides compared to the stacking calculated for the ApC dinucleoside and CpC ones, CpA ≥ ApA > ApC ~ CpC. The stacking ability further depends on RNA sequence along the 5’→ 3’ backbone, CpA > ApC. Introduction Understanding of the physical and biological function of DNA and RNA molecules essentially relies on their structure and dynamics. One very important feature is the stacking interaction that is one of the major driving forces in folding and stabilization of the three-dimensional structure of nucleic acids (NA). Base stacking phenomena are not yet completely understood. Stacking interaction depends on mutual position of NA bases and its amplification occurs when the NA bases lie in parallel orientation with maximal spatial overlap. Stacking interaction results from the π−π interaction between aromatic bases that is stabilized by London dispersion forces and hydrophobic effect. Stacking in NAs and its sequential dependence were recently studied by Norberg and Nilsson [1995] and the following order of stacking ability in RNA dinucleosides was proposed: purine-purine > purine-pyrimidine and pyrimidine-purine > pyrimidinepyrimidine [Norberg and Nilsson, 1995]. However, the analysis by Norberg was based on very short molecular dynamic (MD) simulation and on calculation of the potential of mean force. The analysis of the stacking interaction and the dependence on the sequence of the ribodinucleoside monophosphates presented in this work is based on the dynamical behavior during relatively long MD simulation. We have used the MD simulation technique that is nowadays commonly used in dynamical studies of biomolecules and that could give detailed information about the interactions in NAs. In the present study we were focused particularly on the dynamical behavior of base stacking in RNA dinucleosides and its sequential dependence with respect to the nonequivalent ends 5’ and 3’ of the NA backbone. Since the present study of dynamic behavior carried out for RNA dinucleoside is also related to some experimental data (NMR and X-ray) the obtained results thus could give us deeper inside into the nature of the stacking problem. Methodology Geometry descriptors of the RNA dinucleosides Dinucleoside monophosphate (N-p-N) schematically shown in Figure 1 represents the simplest molecule for studying sequentially dependent properties in NAs. Backbone the N-p-N residue is generally described by eight torsion angles γ1, δ1, ε1, ζ1, α2, β2, γ2, δ2, two sugar pucker coordinates and two glycosidic torsion angles χ that determine the orientation of nucleic base with respect to the sugar (Figure 1). The same geometry parameters were also used for determination of the conformational families in ribosomal RNA that represents up to now the larges statistical ensemble of the RNA solid state structures [Schneider et al., 2004]. Both 5’and 3’ends of our molecular model are terminated by the hydroxyl group and the NA bases B1 and B2 (Figure 1) were selected to be adenine (A) or cytosine (C). The strength of stacking interaction depends to the large extent on the spatial overlap of NA bases and will therefore differ for the purine or pyrimidine base to base interaction and their mutual combinations. The ApA, ApC, CpA and CpC dinucleosides form closed ensemble with respect to WDS'07 Proceedings of Contributed Papers, Part III, 78–83, 2007. ISBN 978-80-7378-025-8 © MATFYZPRESS