Altogether eight keto and enol tautomers of guanine were studied theoretically in the gas phase, in a microhydrated environment (1 and 2 water molecules) and in bulk water. The structures of isolated, as well as mono- and dihydrated tautomers were determined by means of the RI-MP2 method using the extended TZVPP (5s3p2d1f/3s2p1d) basis set. The relative energies of isolated tautomers included the correction to higher correlation energy terms evaluated at the CCSD(T)/aug-cc-pVDZ level. The relative enthalpies at 0 K and relative free energies at 298 K were based on the above-mentioned relative energies and zero-point vibration energies, temperature-dependent enthalpy terms and entropies evaluated at the MP2/6-31G level. The keto form having hydrogen atom at N7 is the global minimum while the canonical form having hydrogen atom at N9 represents the first local minimum at all theoretical levels in vacuo and in the presence of 1 and 2 water molecules. All three unusual rare tautomers having hydrogens at N3 and N7, at N3 and N9, and also at N9 and N7 are systematically considerably less stable and can be hardly detected in the gas phase. The theoretical predictions fully agree with existing theoretical as well as experimental results. The effect of bulk solvent on the relative stability of guanine tautomers was studied by self-consistent reaction field and molecular dynamics free energy calculations using the thermodynamic integration method. Bulk solvent, surprisingly, strongly favored these three rare tautomers over all remaining low-energy tautomers and probably only these forms can exist in water phase. The global minimum (tautomer with hydrogens at N3 and N7) is by 13 kcal/mol more stable than the canonical form (3rd local minimum). Addition of one or two water molecules does not change the relative stability order of isolated guanine tautomers but the respective trend clearly supports the surprising stabilization of three rare forms.

Altogether, 14 amino and imino tautomers of adenine were studied theoretically in the gas phase, in a microhydrated environment (one and two water molecules), and in bulk water environment using the thermodynamic integration method (MD-TI), conductor-like polarizable continuum model (C-PCM, COSMO), and a hybrid model (C-PCM + one to three explicit water molecules). The structures and relative energies of various tautomers were determined at the RI-MP2 level using the TZVPP basis set. The relative enthalpies at 0 K and relative free energies at 298 K were based on relative energies and zero-point vibration energies, temperature-dependent enthalpy terms, and entropies evaluated at the MP2/6-31G** level. The effect of bulk solvent on the relative stability of adenine tautomers was studied by molecular dynamics free energy calculations using the thermodynamic integration method and self-consistent reaction field. The dipole moment of the canonical form is rather small (2.8 D) but three rare imino tautomers have very large dipole moments (more than 10 D). The canonical form is the global minimum at all theoretical levels in the gas phase, in a microhydrated environment, and in the bulk water. Two unusual rare amino tautomers having hydrogens at N3 and N7, respectively, are less stable in the gas phase by more than 7 kcal/mol and represent the first and the second local minimum. Microhydration, as well as bulk water, stabilizes these unusual tautomers, and the energy gap between them and the canonical form is reduced, but the canonical tautomer remains the global minimum in all three phases. Relative free energies (T = 298 K) of these two unusual tautomers in the bulk water evaluated by molecular dynamics free energy calculations are 2.5 and 2.8 kcal/mol, which supports their coexistence in this phase. The C-PCM results agree well with the MD-TI data, and the agreement became close when considering not only the bare tautomers but their complexes with several water molecules representing first solvation shell. Other tautomers are considerably less stable (by 12−45 kcal/mol), and neither a microhydrated environment nor bulk water can change this unfavorable tautomeric equilibrium. The theoretical data predicting the coexistence of the canonical form and the N3 and the N7 tautomers in bulk water nicely agreed with experimental data obtained from NMR measurements of the adenine tautomers in DMSO (Laxer, A.; Major, D. T.; Gottlieb, H. E.; Fischer, B. J. Org. Chem. 2001, 66, 5463.)

Altogether 13 keto and enol tautomers of uracil and 13 keto and enol tautomers of thymine were studied theoretically in the gas phase, in a microhydrated environment (1 and 2 water molecules) and in a water environment. Bulk water was described using the thermodynamic integration method, Conductor-like polarizable continuum model (C-PCM, COSMO) and hybrid model (C-PCM + 1-2 explicit water molecules). The structures of various tautomers were determined at the RI-MP2 level using the TZVPP basis set while relative energies were determined at the CCSD(T) level. The relative free energies at 298 K were based on the relative energies mentioned above and zero-point vibration energies, and temperature dependent enthalpy terms and entropies evaluated at the MP2/6-31G** level. The effect of bulk solvent on the relative stability of uracil and thymine tautomers was studied using molecular dynamics free energy calculations by means of the thermodynamic integration method and self-consistent reaction field. Despite the completely different nature of these methods they provide comparable solvation free energies. Besides theoretical investigation, experimental detection of uracil and thymine tautomers was performed by means of steady-state fluorescence. We conclude that it is impossible to utilize the method used by Suwaiyan and Morsy (M. A. Morsy, A. M. Al-Somali and A. Suwaiyan, J. Phys. Chem. B, 1999, 103(50), 11205) for tautomer detection, even if a very sensitive fluorimeter is used. Theoretical relative energies and free energies for isolated uracil and thymine tautomers support the existence of the canonical form only. The microhydrated environment and bulk solvent stabilize enol forms more than the canonical keto one, but gas phase destabilization of these enol forms is too high. Population of rare enol forms of uracil and thymine in bulk water will thus be very low and canonical structure will also be dominant in this phase.

Correlated ab initio as well as semiempirical quantum chemical calculations and molecular dynamics simulations were used to study the intercalation of cationic ethidium, cationic 5-ethyl-6-phenylphenanthridinium and uncharged 3,8-diamino-6-phenylphenanthridine to DNA. The stabilization energy of the cationic intercalators is considerably larger than that of the uncharged one. The dominant energy contribution with all intercalators is represented by dispersion energy. In the case of the cationic intercalators, the electrostatic and charge-transfer terms are also important. The DeltaG of ethidium intercalation to DNA was estimated at -4.5 kcal mol(-1) and this value agrees well with the experimental result. Of six contributions to the final free energy, the interaction energy value is crucial. The intercalation process is governed by the non-covalent stacking (including charge-transfer) interaction while the hydrogen bonding between the ethidium amino groups and the DNA backbone is less important. This is confirmed by the evaluation of the interaction energy as well as by the calculation of the free energy change. The intercalation affects the macroscopic properties of DNA in terms of its flexibility. This explains the easier entry of another intercalator molecule in the vicinity of an existing intercalation site.

The relative stability of all possible 5-bromouracil tautomers was studied theoretically in a gas phase, in a microhydrated environment (with one water molecule), and in bulk water. Tautomer structures were determined by gradient optimization at the correlated ab initio quantum chemical level with an extended basis set of atomic orbitals. The role of water was examined by using a self-consistent reaction field method. The relative stabilization and free energies in the gas phase, the microhydrated environment, and the bulk water clearly support the preference of the canonical keto form of 5-bromouracil in all mentioned environments. An increased abundance of enol tautomers when passing from uracil to 5-bromouracil is not supported by our calculations. Thus, the tautomeric model of the mutagenic activity of 5-bromouracil proposed previously [Hu et al. Biochemistry (2004) 43, 6361] can be refuted. The validity of other mutagenic models was also discussed, and finally a new mechanism for explaining the mutagenic activity of halogenuracils based on their different behaviors in triplet excited states was suggested.

The hydration Gibbs energies of adenine, cytosine, guanine, thymine, uracil and isoguanine are determined by the molecular dynamics-thermodynamic integration method (MD-TI) and using continuous COSMO and hybrid models. The role of solvents in the COSMO model is described by permittivity and by combining the permittivity and specific hydration of single water molecules placed on the energetically most probable position in the hybrid model. The hybrid model describes hydration similarly to the COSMO model; both the continuous methods are in good agreement with the MD-TI method. Differences are small and the use of both models can be recommended. Keywords: COSMO; Hybrid model; Hydration; Gibbs energy; Nucleobases tautomers; Purines; Pyrimidines; Thermodynamic integration; Molecular dynamics; Tautomerism.Contemporary computational chemistry has become very powerful in investigating the properties of molecules in the gas phase. Unfortunately, the gas phase is not biologically relevant. The study of chemical systems in the condensed phase, which is much closer to the natural environment, is far more difficult. This paper compares several theoretical procedures for determining of hydration Gibbs energies (∆G HYD ) and, as a result of this comparison, we recommend the most appropriate solution. Both accuracy and time dependence of the methods is taken into account.One of the "most standard" methods -molecular dynamics (MD) -deals with a large number of explicit water molecules, but averaging a large number of configurations (sampling of the phase space) is very timedemanding. Moreover, only the empirical potential (force field) is reliable at the moment. Another explicit approach relies on a limited amount (1-10) of solvent molecules fully described by the most accurate quantum

No abstract