The eukaryotic leading and lagging strand DNA polymerases are loaded onto primer-ends via separate mechanisms but have comparable processivity in the presence of PCNA

Abstract

Saccharomyces cerevisiae DNA polymerase delta (Pol delta) and DNA polymerase epsilon (Pol epsilon) are replicative DNA polymerases at the replication fork. Both enzymes are stimulated by PCNA, although to different levels. To understand why and to explore the interaction with PCNA, we compared Pol delta and Pol epsilon in physical interactions with PCNA and nucleic acids (with or without RPA), and in functional assays measuring activity and processivity. Using surface plasmon resonance technique, we show that Pol epsilon has a high affinity for DNA, but a low affinity for PCNA. In contrast, Pol delta has a low affinity for DNA and a high affinity for PCNA. The true processivity of Pol delta and Pol epsilon was measured for the first time in the presence of RPA, PCNA and RFC on single-stranded DNA. Remarkably, in the presence of PCNA, the processivity of Pol delta and Pol epsilon on RPA-coated DNA is comparable. Finally, more PCNA molecules were found on the template after it was replicated by Pol epsilon when compared to Pol delta. We conclude that Pol epsilon and Pol delta exhibit comparable processivity, but are loaded on the primer-end via different mechanisms.

Figure 1.

Comparison of the replication efficiency of Pol ε and Pol δ when stimulated by PCNA. Polymerase replication was assessed over time on single-primed circular ssDNA templates, (A) M13mp18 DNA and (B) pBluescript II SK(+). PCNA and RFC were included in lanes 1–5 and 7–11. No PCNA or RFC was added in lanes 6 and 12. Products of the reactions were separated by electrophoresis on a 1% alkaline agarose gel.

Figure 2.

RPA influences the PCNA-dependent stimulation of Pol ε but not Pol δ. RPA was added at varying concentrations. Reactions in lanes 1 and 2 and 5 and 6 contained 10 pmol RPA, and 2 pmol RPA was added in lanes 3 and 4 and 7 and 8. Reactions with Pol ε (lanes 1–4) were incubated for 15 min; reactions with Pol δ (lanes 5–8) were incubated for 4 min. Products of the reactions were separated by electrophoresis on a 1% alkaline agarose gel.

Figure 3.

Processivity of Pol ε and Pol δ. The replication products from a single reaction mix were separated on three different gels: (A) 8% polyacrylamide gel separating products in the range of 0–300 nt, (B) 8% polyacrylamide gel separating products in the range of 177–600 nt and (C) a 2% alkaline agarose gel separating products in the range of 100–3000 nt. All gels were dried on a 3MM Whatmann paper, except the 2% alkaline agarose where the section with the end-labeled 50-mer primer was blotted onto a DE81 paper (as indicated on the left side). Reaction times are indicated above each lane. An end-labeled oligonucleotide was annealed to Bluescript II SK(+) ssDNA (lane 1) and 80 fmol primer–template was added in all reactions. RPA (2.7 pmol) was added to the reactions in lanes 2–5 and 8–11. RPA (21 pmol) was added to reactions in lanes 6 and 7 and 12 and 13. Pol ε (2.4 fmol) was added to lanes 2–7 and Pol δ (2.4 fmol) was added to lanes 8–13. RFC and PCNA were added in lanes 4–7 and 10–13. Lane 14 of gel (C) contains 100 fmol of Pol δ to demonstrate where the full-length template is migrating. A sequencing ladder with the identical template was used as a molecular weight marker to the right of gel (A) and (B). A 100 bp molecular weight marker was used on the 2% alkaline agarose gel and stained with ethidium bromide. The migration of each band is indicated to the right of gel (C).

Figure 4.

Measurement of the affinity between DNA and Pol δ or Pol ε. DNA affinity was analyzed using surface plasmon resonance. Approximately 175 RU of primed and dsDNA templates and 95 RU of ssDNA template were immobilized, reflecting equal molar amounts of substrate. Pol δ or Pol ε was injected at a concentration of 20 and 10 nM. Each injection was repeated twice. After each injection, 0.5 M sodium chloride was injected for 5 s, followed by a buffer blank injection. The top row illustrates the interactions between Pol ε and primed DNA, double-stranded DNA, and single-stranded DNA. The middle row illustrates the interactions between Pol δ and primed DNA, double-stranded DNA and single-stranded DNA. The bottom row illustrates the interactions between Pol ε and primed DNA, double-stranded DNA and single-stranded DNA when the DNA was saturated with RPA.

Figure 5.

Interaction between Pol ε and PCNA in solution. Pol δ or Pol ε, at concentrations of 11.5, 23 and 46 nM, were injected onto PCNA immobilized on the surface of the dextran chip (CM5) (∼500 RU). Each injection was repeated twice, and after each injection, 0.5 M sodium chloride was injected for 5 s, followed by a buffer blank injection.

Figure 6.

Loading of PCNA on a single-primed circular template. ³²P-labeled PCNA was added to a holoenzyme assay with Pol ε, Pol δ or in the absence of polymerase. The reaction with single-primed pBluescript II SK(+) was carried out for 7 min at 30°C, stopped and the PCNA molecules loaded onto the circular template were separated from free PCNA molecules over a BiogelA-5 m column. (A) Elution profiles of reactions where Pol ε or Pol δ was added or the polymerase was omitted. (B) The amount of PCNA loaded onto 0.5 pmol of primed template.

Figure 7.

Assessment of the loading of PCNA as the rate-limiting step in a holoenzyme assay with Pol ε. A series of PCNA-dependent replication assays were carried out, with each reaction having the amount of PCNA indicated on the X-axis. Each series had a constant amount of RFC as follows: filled circle, no RFC added; open circle, 10 fmol RFC; open square, 30 fmol RFC; filled square, 35 fmol RFC; open triangle, 114 fmol RFC and filled triangle, 350 fmol RFC.