Nature of Nucleic Acid−Base Stacking: Nonempirical ab Initio and Empirical Potential Characterization of 10 Stacked Base Dimers. Comparison of Stacked and H-Bonded Base Pairs

Abstract

Ab initio (MP2/6-31G*(0.25)) interaction energies were calculated for almost 240 geometries of 10 stacked nucleic acid−base pairs: A···A, C···C, G···G, U···U, A···C, G···A, A···U, G···C, C···U, and G···U; in some cases uracil was replaced by thymine. The most stable stacked pair is the G···G dimer (−11.3 kcal/mol), and the least stable is the uracil dimer (−6.5 kcal/mol). The interaction energies of H-bonded base pairs range from −25.8 kcal/mol (G···C) to −10.6 kcal/mol (T···T). The stability of stacked pairs originates in the electron correlation, while all the H-bonded pairs are dominated by the HF energy. The mutual orientation of the stacked bases is, however, primarily determined by the HF interaction energy. The ab initio base stacking energies are well reproduced by the empirical potential calculations, provided the atomic charges are derived by the same method as used in the ab initio calculations. Some contributions previously postulated to significantly influence base stacking (induction interactions, π−π interactions) were not found. Base stacking was also investigated in six B-DNA and two Z-DNA base pair steps; their geometries were taken from the oligonucleotide crystal data. The many-body correction was estimated at the HF/MINI-1 level. The sequence-dependent variations of the total base pair step stacking energies range from −9.9 to −14.7 kcal/mol. The range of the calculated many-body corrections to the stacking energy is 2 kcal/mol. The ab initio calculations exclude the consideration that the unusual conformational properties of the CpA(TpG) steps might be associated with attractive induction interactions of the exocyclic groups of DNA bases and the aromatic rings of bases.

PM6 quantum chemical study of the H-bonded and stacked associates of the adenine and thymine DNA bases: The nature of base stacking

“…As the exact geometry of the A/T stacks in water is not known [25], we followed the procedure of Ref. [30] to generate initial configurations for the stacked associates. We considered several configurations, starting from undisplaced stacked dimers (with the axis connecting the centres of mass of the two bases perpendicular to the base planes).…”

Comparison of Intrinsic Stacking Energies of Ten Unique Dinucleotide Steps in A-RNA and B-DNA Duplexes. Can We Determine Correct Order of Stability by Quantum-Chemical Calculations?

“…Calculations performed with the second-order Møller-Plesset method using medium-sized AO basis set containing diffuse polarization functions provided meaningful results for base pairing, 30 stacking, 31 for interactions of DNA bases and base pairs with metal cations 32,33 as well as for interactions of DNA base pairs with drugs. 34,35 With the advent of fast and reliable quantum chemical methods, such as RI-DFT-D (resolution of identity density functional theory augmented with dispersion term), 36,37 the size of the system can be considerably increased, representing thus a major progress in ab initio description of nucleic acid systems.…”

“…The d-polarization functions are capable of filling the gap between the interacting monomers, which significantly improves the description of base stacking interactions compared with the standard 6-31G* basis set. The MP2/6-31G*(0.25) method has been the most widely used ab initio technique to study aromatic stacking in the past literature 31,35,50,51,60,85,100,101,[107][108][109][110][111][112][113][114][115][116][117][118][119][120] and has only recently been replaced by CBS(T). The MP2/6-31G*(0.25) method is entirely sufficient to capture the nature of base stacking, and therefore all conclusions reported with this method remain qualitatively valid.…”

Balance of Attraction and Repulsion in Nucleic-Acid Base Stacking: CCSD(T)/Complete-Basis-Set-Limit Calculations on Uracil Dimer and a Comparison with the Force-Field Description

“…We have shown that AMBER, the leading molecular-mechanics force field for nucleic acids, provides a surprisingly good description of base stacking. 35 However, this does not mean that the force field provides an exact description of stacking. In fact, the MM force fields, albeit often providing a very insightful description of nucleic-acid structure and dynamics, are known to have limitations.…”