| Literature DB >> 26119247 |
Nohad Gresh1,2, Judit E Sponer3,4, Mike Devereux5, Konstantinos Gkionis3, Benoit de Courcy1,2, Jean-Philip Piquemal1,2, Jiri Sponer3,4.
Abstract
Until now, atomistic simulations of DNA and RNA and their complexes have been executed using well calibrated but conceptually simple pair-additive empirical potentials (force fields). Although such simulations provided many valuable results, it is well established that simple force fields also introduce errors into the description, underlying the need for development of alternative anisotropic, polarizable molecular mechanics (APMM) potentials. One of the most abundant forces in all kinds of nucleic acids topologies is base stacking. Intra- and interstrand stacking is assumed to be the most essential factor affecting local conformational variations of B-DNA. However, stacking also contributes to formation of all kinds of noncanonical nucleic acids structures, such as quadruplexes or folded RNAs. The present study focuses on 14 stacked cytosine (Cyt) dimers and the doubly H-bonded dimer. We evaluate the extent to which an APMM procedure, SIBFA, could account quantitatively for the results of high-level quantum chemistry (QC) on the total interaction energies, and the individual energy contributions and their nonisotropic behaviors. Good agreements are found at both uncorrelated HF and correlated DFT and CCSD(T) levels. Resorting in SIBFA to distributed QC multipoles and to an explicit representation of the lone pairs is essential to respectively account for the anisotropies of the Coulomb and of the exchange-repulsion QC contributions.Entities:
Mesh:
Substances:
Year: 2015 PMID: 26119247 DOI: 10.1021/acs.jpcb.5b01695
Source DB: PubMed Journal: J Phys Chem B ISSN: 1520-5207 Impact factor: 2.991