To elucidate how human DNA polymerase β (pol β) discriminates dATP from dCTP when processing 8-oxoguanine (8-oxoG), we analyze a series of dynamics simulations before and after the chemical step with dATP and dCTP opposite an 8-oxoG template started from partially open complexes of pol β. Analyses reveal that the thumb closing of pol β before chemistry is hampered when the incorrect nucleotide dATP is bound opposite 8-oxoG; the unfavorable interaction between active-site residue Tyr^sup 271^ and dATP that causes an anti to syn change in the 8-oxoG (syn):dATP complex explains this slow motion, in contrast to the 8-oxoG (anti):dCTP system. Such differences in conformational pathways before chemistry for mismatched versus matched complexes help explain the preference for correct insertion across 8-oxoG by pol β. Together with reference studies with a nonlesioned G template, we propose that 8-oxoG leads to lower efficiency in β's incorporation of dCTP compared with G by affecting the requisite active-site geometry for the chemical reaction before chemistry. Furthermore, because the active site is far from ready for the chemical reaction after partial closing or even full thumb closing, we suggest that pol β is tightly controlled not only by the chemical step but also by a closely related requirement for subtle active-site rearrangements after thumb movement but before chemistry.

DNA base and sugar lesions caused by cellular reactive oxygen species contribute to mutagenesis and cardnogenesis (1). One of the most prevalent lesions in the genome is 7, 8-dihydro-8-oxoguanine (8-oxoG) (2) (Fig. 1 b), in which C8 is oxidized to a carbonyl group and N7 is transformed to an -NH- group. The 8-oxoG nucleoside can adopt two glycosidic conformations: the ami conformation when paired with a complementary C (Fig. 1 c), and the syn conformation when paired with A to form a Hoogsteen basepair (Fig. 1 a); the latter avoids a clash between the deoxyribose backbone and the 8-oxo group. When it is not repaired, the mismatched 8-oxoG:A basepair can introduce G:C to T:A transversion mutations (i.e., the mispairing of 8-oxoG with A will result in thymine in the next replication cycle). Such mutations contribute significantly to somatic mutations associated with spontaneous cell transformations (3-6).

High- and low-fidelity DNA polymerases, such as human DNA polymerase β and Sulfoiobus solfataricus P2 DNA polymerase IV (Dpo4), often encounter 8-oxoG (7-11) and can bypass it in the replication or repair process. Steady-state kinetic data suggest that pol β prefers dCTP over dATP incorporation by twofold and that dCTP insertion is only threefold less efficient opposite 8-oxoG than opposite a normal dG( 12).

A partial explanation to the accommodation of 8-oxoG: dCTP comes from the ternary crystal structure of the pol β/ substrate complex with 8-oxoG (13): the structure shows that the matched 8-oxoG (anti):dCTP basepair is easily tolerated by pol β due to a 184° flip of the template backbone (O3'-P-O5'-C5'), thereby avoiding the clash between O8 and O5' and O1P of 8-oxoG. The mismatched 8-oxoG (syn):dATP basepair, in contrast, is not as stable as the matched pair in pol β. Crystallography has not captured the syn conformation of 8-oxoG pairing with dATP in pol β but rather captured ami 8-oxoG stacking with dAMP (12).

Besides the crystal structures of pol β 8-oxoG complexes, other DNA polymerases across different families with 8-oxoG at the template position have also been resolved crystallographically, These include RB69 from the B-family (10), T7 DNA polymerase (14,15) and Bacillus fragment (BF) from the A-family (11), and Dpo4 from the Y-family (16). However, none of these wild-type enzyme complexes captured syn 8-oxoG pairing with dATP at the active site. Recently, only by mutating Lys^sup 536^ to Ala has the ternary form of T7 DNA polymerase resolved the 8-oxoG (syn):dATP basepair (15).

Kinetic data for pol β, RB69. Dpo4, wild-type T7, LyS^sup 536^ Ala T7, and BF (10-12,15,16) indicate preference ratios of dCTP over dATP incorporation opposite 8-oxoG to be 2:1.20:1, ~70:1.2:1,1:20, and 1:9, respectively. We thus can infer that the incorporation of 8-oxoG (iyn):dATP mismatch is energetically Jess favorable in pol β, wild-type T7. RB69, BF, and Dpo4, but more favorable in LyS^sup 536^ Ala T7, than the corresponding 8-oxoG (anti):dCTP basepair.

Although the nascent basepair 8-oxoG:dAMP is distorted in the pol β/DNA crystal structure, a modeling of dATP pairing with syn 8-oxoG in the closed pol β revealed no serious steric clashes at the active site (17). Since pol β maintains fidelity in DNA replication through an inducedfit mechanism-the thumb subdomain of pol β closes when the correct nucleotide complementary to the template residue binds (18-22) and opens after the chemical reaction to release the reaction products and translocate DNA-the induced-fit cycle may also contribute to differentiating dATP and dCTP insertions opposite 8-oxoG and dG (12). Specifically, dATP is less efficiently inserted opposite 8-oxoG than dCTP in pol β, though dATP insertion is ~10^sup 5^ more efficient compared with it opposite dG. Furthermore, computational studies of pol β (22-28) have shown that conformational pathways of pol β before and after the chemical reaction play a vital role in determining its fidelity. With the ultimate goal of understanding the fidelity mechanism of pol β in processing 8-oxoG, we investigate here the dynamic process of pol β's conformational pathways before and after the chemical reaction by performing a series of six dynamics simulations starting from intermediate (partially open) structures where the incoming nucleolides dCTP and dATP pair with 8-oxoG in anti or syn orientations, Although insertion rates of dATP and dCTP opposite 8-oxoG in pol β differ slightly, atomic-level simulations can help unravel systematic differences in their conformational pathways to explain biological observations. In fact, our transition path sampling simulations further dissected the conformational and energetics pathways of correct and incorrect nucleotide insertions opposite 8-oxoG in pol β (29). The resolved free energy barriers in that work along with the results here suggest that the different transition states and sequences of conformational events during thumb closing for the two systems could be correlated to different stabilities of the respective closed states and associated insertion efficiencies.