Single-molecule measurements of synthesis by DNA polymerase with base-pair resolution

Abstract

The catalytic mechanism of DNA polymerases involves multiple steps that precede and follow the transfer of a nucleotide to the 3'-hydroxyl of the growing DNA chain. Here we report a single-molecule approach to monitor the movement of E. coli DNA polymerase I (Klenow fragment) on a DNA template during DNA synthesis with single base-pair resolution. As each nucleotide is incorporated, the single-molecule Förster resonance energy transfer intensity drops in discrete steps to values consistent with single-nucleotide incorporations. Purines and pyrimidines are incorporated with comparable rates. A mismatched primer/template junction exhibits dynamics consistent with the primer moving into the exonuclease domain, which was used to determine the fraction of primer-termini bound to the exonuclease and polymerase sites. Most interestingly, we observe a structural change after the incorporation of a correctly paired nucleotide, consistent with transient movement of the polymerase past the preinsertion site or a conformational change in the polymerase. This may represent a previously unobserved step in the mechanism of DNA synthesis that could be part of the proofreading process.

Fig. 1.

Mechanism for DNA synthesis. (A) Kinetic mechanism for DNA synthesis by KF (4). (B) Models for DNA-polymerase dynamics are shown schematically. Fingers, palm, and thumb subdomains located in the polymerase domain are labeled F, P, and T, respectively. Exonuclease domain is labeled E. Cy3 and Cy5 fluorophores are located on the template and polymerase, respectively. Catalytic active site in polymerase (pink) and exonuclease (orange) domains are located ≈30 Å apart. Putative transient intermediate site is shown in yellow and is positioned at a location consistent with the proposed mechanism and observed FRET efficiencies. (Left diagram) Initial binding of the polymerase forms an open complex with the primer/template (15-mer/28-mer) located at position n (FRET ≈0.68). (Second from left diagram) Nucleotide incorporation leads to translocation past the next preinsertion site, placing the nascent base pair in the transient intermediate site (FRET ≈0.47). (Third from left diagram) Polymerase then returns to the preinsertion site (FRET ≈0.62) where sampling for the next correct nucleotide occurs. (Right panel) Positioning a 16-mer/28-mer primer/template having a terminal G–A mismatch results in the primer terminus moving to the exonuclease site (FRET ≈0.26).

Fig. 2.

Dwell-time analysis of FRET time trajectories. A Cy3-labeled 28-mer template annealed to a 15-mer primer was immobilized on a glass slide and the fluorescence intensity time trajectories from immobilized single molecules were measured as described previously (15, 32). (A) Cy5-labeled KF was introduced and the Cy3 (blue) and Cy5 (red) fluorescence intensities were detected. (B) smFRET time trajectory for the single molecule shown in (A). Each peak represents a binding event to the immobilized primer/template. (C) Histograms showing times between binding events (left) and duration of binding events (right) were fit to a single-exponential decay to determine the association rate (k₁) and dissociation rate (k_-1) for KF binding. The dissociation constant (K_D, 20.3 ± 5.5 nM) was determined from the ratio of the association rate (k₁) to dissociation rate (k_-1).

Fig. 3.

Histograms of FRET efficiencies from single-molecule traces using primers of different lengths. Traces of FRET efficiencies for Cy5-labeled KF binding to Cy3-linked 28-mer templates annealed to primers of different lengths were used to generate binding histograms. The primer lengths used and numbers of molecules analyzed were as follows: length = 15, n = 100; length = 16, n = 100; length = 17, n = 100; length = 18, n = 79. Distributions were fit to asymmetric Gaussians as described in Materials and Methods.

Fig. 4.

DNA synthesis on a Cy3-labeled 15-mer/28-mer primer/template in the presence of dTTP, dATP, and dGTP. (A) FRET 15-mer primer/template duplex. The positions of the primer/template terminus with increasing primer length are indicated. (B) Single nucleotide (dTTP) incorporation trajectory (n = 46). FRET first increases to ≈0.68 (polymerase bound to the 15-mer/28-mer), then drops to ≈0.4 (arrow) before rising to ≈0.62 (polymerase bound to a 16-mer/28-mer). FRET finally drops to zero, indicating polymerase dissociation or acceptor photobleaching. (C) Two-nucleotide (dTTP and dATP) incorporation trajectory (n = 15). smFRET initially drops from ≈0.68 to ≈0.47 (arrow) before rising to ≈0.62 (polymerase bound to a 16-mer/28-mer). The second incorporation shows FRET decrease from ≈0.62 to ≈0.23 (arrow) before rising to ≈0.47 (polymerase bound to 17-mer/28-mer). (D) Three-nucleotide (dTTP, dATP, dGTP) incorporation trajectory (n = 7). FRET drops sequentially from ≈0.68 to ≈0.62, then to ≈0.47, and finally to ≈0.23 (polymerase bound to a 18-mer/28-mer). The first two incorporations showed a decrease in FRET immediately after each incorporation before returning to the expected value (arrows). The third incorporation shows a FRET drop to 0 before returning to ≈0.23, which may correspond to the polymerase translocation, dissociation, or acceptor blinking. (E) Dwell-time distribution for the combined incorporation (n = 97). Dashed lines correspond to the estimated 15, 16, 17, and 18-mer FRET values for each trace. Note that these values may changes slightly for each particular single molecule complex within the width of the FRET histograms shown in Fig. 3.

Fig. 5.

FRET time-trajectory of KF binding to the 16-mer/28-mer G–A terminally mismatched primer/template. (A) KF binding to the mismatched primer/template produces three distinct FRET efficiencies, the highest corresponding to the expected FRET when bound at the polymerase active site, an intermediate FRET that matches binding to the transient site observed during synthesis, and a new FRET state that is assigned to binding to the exonuclease site. (B) FRET histogram for KF binding to a mismatched primer/template.