Fidelity discrimination in DNA polymerase beta: differing closing profiles for a mismatched (G:A) versus matched (G:C) base pair

Abstract

Understanding fidelity-the faithful replication or repair of DNA by polymerases-requires tracking of the structural and energetic changes involved, including the elusive transient intermediates, for nucleotide incorporation at the template/primer DNA junction. We report, using path sampling simulations and a reaction network model, strikingly different transition states in DNA polymerase beta's conformational closing for correct dCTP versus incorrect dATP incoming nucleotide opposite a template G. The cascade of transition states leads to differing active-site assembly processes toward the "two-metal-ion catalysis" geometry. We demonstrate that these context-specific pathways imply different selection processes: while active-site assembly occurs more rapidly with the correct nucleotide and leads to primer extension, the enzyme remains open longer, has a more transient closed state, and forms product more slowly when an incorrect nucleotide is present. Our results also suggest that the rate-limiting step in pol beta's conformational closing is not identical to that for overall nucleotide insertion and that the rate-limiting step in the overall nucleotide incorporation process for matched as well as mismatched systems occurs after the closing conformational change.

Figure 1

Overall captured reaction kinetics profile for pol β's closing transition followed by chemical incorporation of dNTP for G:C and G:A systems. The barriers to chemistry (dashed peaks) are derived from experimentally measured k_pol values [3-5,29]. The profiles were constructed by employing reaction coordinate characterizing order parameters (χ₁–χ₅) in conjunction with transition path sampling (Appendix B). The order parameters χ₁–χ₅ serve as reaction coordinates to characterize the transition states TS 1–TS 5 in the matched G:C system, as well as TS 1–TS 4 in the mismatched G:A system. The potential of mean force along each reaction coordinate is computed for each conformational event (Appendix B, Figs. S2 and S3). The relative free energies of the metastable states and the free-energy barrier characterizing each transition state are calculated with BOLAS [18].

Figure 2

Conformational landscapes for the rotation and flipping of Arg258 and Phe272, in the conformational closing pathway of pol β for G:C versus G:A systems.

Figure 3

Evolution of average distances of ligands coordinating the catalytic and nucleotide-binding Mg²⁺ ions along the reaction coordinate for G:C and G:A. Metastable states 1 to 7 evolve the system in the closing pathway. The extent of thumb closing (χ₁ at top), and a crucial distance for the chemical reaction (O3′ of last primer (guanine) residue to P_α of dCTP in bottom plot) are also provided. Coordination and distances are diagrammed on the right: catalytic site ready for the phosphoryl transfer reaction. Circled area represents the QM region

Figure 4

Comparison of reaction kinetics for G:C and G:A systems. The temporal evolution of open (blue band), closed (red band) and product (black band) species are derived based on 100 evolution trajectories from binary (open) complexes. Inset describes the reaction networks according to profiles in Fig. 1. The networks are solved with the stochastic algorithm of Gillespie [19] (Appendix D). The spread in the kinetics (thickness of bands shown) represents the inherent stochasticity of the system and is not due to variations in the values of the individual rate-constants (k_ijs in Table I). The effect of uncertainties in the k_ij values on the time evolution has not been considered here.

Table of Content Graphics