Hydroxyl radical reactions with adenine: reactant complexes, transition states, and product complexes

In order to address problems such as aging, cell death, and cancer, it is important to understand the mechanisms behind reactions causing DNA damage. One specific reaction implicated in DNA oxidative damage is hydroxyl free-radical attack on adenine (A) and other nucleic acid bases. The adenine reaction has been studied experimentally, but there are few theoretical results. In the present study, adenine dehydrogenation at various sites, and the potential-energy surfaces for these reactions, are investigated theoretically. Four reactant complexes [A···OH]* have been found, with binding energies relative to A+OH* of 32.8, 11.4, 10.7, and 10.1 kcal mol(-1). These four reactant complexes lead to six transition states, which in turn lie +4.3, -5.4, (-3.7 and +0.8), and (-2.3 and +0.8) kcal mol(-1) below A+OH*, respectively. Thus the lowest lying [A···OH]* complex faces the highest local barrier to formation of the product (A-H)*+H(2)O. Between the transition states and the products lie six product complexes. Adopting the same order as the reactant complexes, the product complexes [(A-H)···H(2)O]* lie at -10.9, -22.4, (-24.2 and -18.7), and (-20.5 and -17.5) kcal mol(-1), respectively, again relative to separated A+OH*. All six A+OH* → (A-H)*+H(2)O pathways are exothermic, by -0.3, -14.7, (-17.4 and -7.8), and (-13.7 and -7.8) kcal mol(-1), respectively. The transition state for dehydrogenation at N(6) lies at the lowest energy (-5.4 kcal mol(-1) relative to A+OH*), and thus reaction is likely to occur at this site. This theoretical prediction dovetails with the observed high reactivity of OH radicals with the NH(2) group of aromatic amines. However, the high barrier (37.1 kcal mol(-1)) for reaction at the C(8) site makes C(8) dehydrogenation unlikely. This last result is consistent with experimental observation of the imidazole ring opening upon OH radical addition to C(8). In addition, TD-DFT computed electronic transitions of the N(6) product around 420 nm confirm that this is the most likely site for hydrogen abstraction by hydroxyl radical.