Reaction of hydroxyl radicals with azacytosines: a pulse radiolysis and theoretical study.

Pulse radiolysis and density functional theory (DFT) calculations at B3LYP/6-31+G(d,p) level have been carried out to probe the reaction of the water-derived hydroxyl radicals (*OH) with 5-azacytosine (5Ac) and 5-azacytidine (5Acyd) at near neutral and basic pH. A low percentage of nitrogen-centered oxidizing radicals, and a high percentage of non-oxidizing carbon-centered radicals were identified based on the reaction of transient intermediates with 2,2'-azinobis(3-ethylbenzothiazoline-6-sulfonate), ABTS2-. Theoretical calculations suggests that the N3 atom in 5Ac is the most reactive center as it is the main contributor of HOMO, whereas C5 atom is the prime donor for the HOMO of cytosine (Cyt) where the major addition site is C5. The order of stability of the adduct species were found to be C6-OH_5Ac*>C4-OH_5Ac*>N3-OH_5Ac*>N5-OH_5Ac* both in the gaseous and solution phase (using the PCM model) respectively due to the additions of *OH at C6, C4, N3, and N5 atoms. These additions occur in direct manner, without the intervention of any precursor complex formation. The possibility of a 1,2-hydrogen shift from the C6 to N5 in the nitrogen-centered C6-OH_5Ac* radical is considered in order to account for the experimental observation of the high yield of non-oxidizing radicals, and found that such a conversion requires activation energy of about 32 kcal/mol, and hence this possibility is ruled out. The hydrogen abstraction reactions were assumed to occur from precursor complexes (hydrogen bonded complexes represented as S1, S2, S3, and S4) resulted from the electrostatic interactions of the lone pairs on the N3, N5, and O8 atoms with the incoming *OH radical. It was found that the conversion of these precursor complexes to their respective transition states has ample barrier heights, and it persists even when the effect of solvent is considered. It was also found that the formation of precursor complexes itself is highly endergonic in solution phase. Hence, the abstraction reactions will not occur in the present case. Finally, the time dependent density functional theory (TDDFT) calculations predicted an absorption maximum of 292 nm for the N3-OH_5Ac* adduct, which is close to the experimentally observed spectral maxima at 290 nm. Hence, it is assumed that the addition to the most reactive center N3, which results the N3-OH_5Ac* radical, occurs via a kinetically driven process.