Eukaryotic origin-dependent DNA replication in vitro reveals sequential action of DDK and S-CDK kinases

Abstract

Proper eukaryotic DNA replication requires temporal separation of helicase loading from helicase activation and replisome assembly. Using an in vitro assay for eukaryotic origin-dependent replication initiation, we investigated the control of these events. After helicase loading, we found that the Dbf4-dependent Cdc7 kinase (DDK) but not S phase cyclin-dependent kinase (S-CDK) is required for the initial origin recruitment of Sld3 and the Cdc45 helicase-activating protein. Likewise, in vivo, DDK drives early-firing-origin recruitment of Cdc45 before activation of S-CDK. After S-CDK activation, a second helicase-activating protein (GINS) and the remainder of the replisome are recruited to the origin. Finally, recruitment of lagging but not leading strand DNA polymerases depends on Mcm10 and DNA unwinding. Our studies identify distinct roles for DDK and S-CDK during helicase activation and support a model in which the leading strand DNA polymerase is recruited prior to origin DNA unwinding and RNA primer synthesis.

Figure 1. An Assay for Replisome Assembly in vitro

(A) Schematic of replisome assembly assay. ARS1 origin DNA was treated with three sequential incubations: step 1, Mcm2-7 loading in G1 extract supplemented with Cdc6; step 2, DDK phosphorylation of Mcm2-7; step 3, replisome assembly in S-phase extract. (B) Protein, substrate, and extract requirements for the replisome assembly assay. Replisome assembly assays were performed with or without DDK-inactivated yRH182 S-phase extract (yRH182-S, lanes 1-4) or with alpha-factor- (G1), hydroxyurea- (HU) or nocodazol (Noc)-arrested extracts made from yRH182 expressing active DDK and overexpressing Cdc45, Dpb11, Sld2 and Sld3 (lanes 5-10). Unless indicated, WT ARS1 DNA and Cdc6 were used in all reactions. ARS1-A^-B2^- is an ARS1 mutant lacking ORC binding sites. Additional DDK (125 ng) was added to the 2^nd extract in lanes 8-10. Changes in ORC DNA association were likely due to release of ORC after pre-RC formation (lanes 5 and 8, Tsakraklides and Bell, 2010) and ORC rebinding after Cdc45, GINS and Mcm10 recruitment.

Figure 2. Interdependent Recruitment of Replisome Proteins

Depletion of replisome proteins reveals interdependent origin DNA association. S-phase extracts were depleted for the indicated protein prior to replisome assembly assays. Associated proteins were analyzed by immunoblot. Depleted protein and extracts were: Sld3, yRH208-S; Cdc45, yRH182-S; Sld2, yRH207-S; Dpb11, yRH209-S; GINS, yRH223-S; Mcm10, yRH183-S. For each panel, Cdc6 was omitted from reaction 1, S-phase extracts were depleted for the indicated protein in reactions 3-4 and the corresponding purified protein (see Fig. S1) was added in reaction 4. Note: purified Cdc45-FLAG and MBP-Mcm10 lack HA and myc tags, respectively. GINS was detected with a polyclonal antibody in lanes 25-28.

Figure 3. Long DNA Templates Supports Polymerase Loading and Replication Initiation

(A) Circular templates show increased DNA Pol α association. Replisome assembly assays were performed with 1 kb linear ARS1 DNA or pARS1/WT plasmid. Lines and ovals below images indicate biotinylated linear and circular templates, respectively. (B) Analysis of replication products. Replication assays were performed using yRH182-S extract on pARS1/WT template. Left, native gel of DNA products, ethidium bromide stain. The location of relaxed plasmid is indicated. Center, autoradiogram of the native gel. Right, autoradiogram of replication products analyzed by alkaline gel electrophoresis. The presence of Cdc6 during Mcm2-7 loading is indicated. (C) Protein, template, and nucleotide requirements of the replication assay. Replication assays were performed with yRH182-S extract. Reaction lacking Cdc6 during helicase loading are indicated. Immunoblot (upper panels) and alkaline gel analysis (lower panels) of proteins and replication products are shown. Templates used: lanes 1-4, pARS1/WT (5.6 kb); lanes 5-7, pUC19-ARS1 (3.7 kb); lanes 8 and 9, circular pARS1/Nco-Nco (7.6 kb); lanes 10 and 11, linear pARS1/Nco-Nco (7.6 kb). Lanes 7, 9 and 11 use A-B2- derivatives of the indicated DNA. ATPγS reactions replaced ATP and the ATP regenerating system with 1 mM ATPγS in step three of the assay. + aphid, 100 μg/ml aphidicolin in step three. (D) Time course of Mcm10 recruitment and replication product accumulation. Replication assays using yRH182-S extract and pARS1/WT were analyzed by immunoblot of origin-associated proteins (upper panels) and nucleotide incorporation (lower panel). (E-F) Relative contributions of overexpressed Cdc45 (C45), Dpb11 (D11), Sld2 (S2) and Sld3 (S3). Replication assays using pARS1/WT plasmid template and yRH182-S (lane 1) or yRH191-S (no overexpressed proteins, lanes 2-9) extracts were supplemented with the indicated purified replication proteins. Relative level of replication products (lower panel) was quantified and plotted in (F). (G) Heavy-light analysis of replication products. Replication reactions were performed in the presence of 500 μM BrdUTP in place of dTTP. Replication products were fractionated by CsCl gradient and detected by scintillation counter (black line). Heavy-heavy and light-light controls are shown (gray line). The CsCl density (mg/ml) of the highest point in each peak is indicated.

Figure 4. DDK and S-CDK Are Required For Distinct Stages of Origin Activation

(A-B) S-CDK and DDK are required for replisome assembly. Replisome assembly assays using extracts yRH166-S (A) or yRH229-S (B) were analyzed as described in Fig. 1. DDK kinase activity was eliminated by omitting DDK from reaction step two. S-CDK activity was blocked by the addition of GST-Sic1 to reaction step three. DDK with CDK, DDK was omitted from reaction step two and added to reaction step three in which S-CDK is also active. CDK →DDK, reaction step two was eliminated. After step three, purified GST-Sic1 and DDK were added sequentially and incubation was continued for 20 min. (C) S-CDK and DDK are required for DNA replication. Replication assays using yRH182-S and pARS1/WT plasmid template. Mcm4-Pi immunoblot was probed with the Mcm4-phospho-S82-D83 phosphospecific antibody that recognizes a DDK target site in Mcm4 (Randell et al., 2010). (D) Sld3 binding to ARS305 in G1 requires DDK. Either wild-type CDC7 or congenic cdc7-4 strains including myc-tagged Sld3 were arrested in nocodazol and released into 25°C or 32°C media containing α-factor (Fig. S3) and analyzed by ChIP using anti-Mcm2-7 or anti-myc antibodies. Samples were analyzed by PCR using primers recognizing ARS305 and two non-origin sequences (ARS305+17kb and ARS306+6kb). (E) Cdc45 binding to early origins in G1 requires DDK. Either wild-type CDC7 or congenic cdc7-4 strains including myc-tagged Cdc45 were arrested in media containing α-factor at 25°C (Fig. S3) and analyzed by ChIP-Chip using anti-myc antibodies. The average log₂ ratio of IP to input signal from two experiments are plotted for chromosome III (wild-type, orange; cdc7-4, blue). Three early origins (ARS305, ARS306 and ARS307) and one late origin (ARS316) are indicated.

Figure 5. Mcm10 is Required for the Recruitment of Pol α and Pol δ to Origin DNA

(A) Effect of Mcm10 depletion on DNA polymerase origin association. Replication assays were performed with pARS1/WT plasmid template and extract yRH183-S (lanes1-3), yRH185-S (lanes 4-6), or yRH187-S (lanes 7-9). As indicated, extracts were depleted of Mcm10 and supplemented with MBP-Mcm10. (B) Relative levels of DNA polymerase association. Two (Pol α and Pol ε) or three (Pol δ) iterations of the experiment in A were quantified and plotted. Polymerase recruitment in the undepleted extract was set to 1. Error bars = standard deviation from the mean.

Figure 6. ATP Hydrolysis is Required for RPA Loading and for the Loading of a Subset of Polymerases

(A) ATP hydrolysis requirement for RPA loading. Replisome assembly assays were performed using pARS1/WT plasmid templates and extract yRH184-S. Where indicated, ATPγS was added after DDK phosphorylation of Mcm2-7 in place of ATP and the ATP regenerating system in reaction step three. (B) Cdc45 was required for RPA loading and DNA replication. Replication assays were performed using pARS1/WT templates and extract yRH188-S. As indicated, extract was depleted for Cdc45 and supplemented with purified Cdc45-3HA/3Flag. (C) ATP hydrolysis requirement for DNA polymerase recruitment. Replisome assembly reactions were performed using pARS1/WT plasmid templates and extracts yRH188-S (lanes 1-4), yRH184-S (lanes 5-7), yRH186-S (lanes 8-10), or yRH182-S (lanes 11-13). ATP: reactions were performed under standard conditions. +Sic1: GST-Sic1 added to reaction step three. ATPγS+AS-CDK: During reaction step three, ATP and the ATP regenerating system were replaced with 1 mM ATPγS, analog-specific CDK (CDK-AS) and 0.5 mM 6-benzyl-ATP.

Figure 7. A Model for the Events Leading to Replication Initiation

The Mcm2-7 helicase is loaded to origin DNA in an inactive form during late M/G1 phase of the cell cycle to form the pre-RC. II. Upon entry into S phase, DDK targets Mcm2-7 for phosphorylation, allowing recruitment of Sld3 and Cdc45. III. S-CDK activation and phosphorylation of Sld2 and Sld3 trigger the recruitment of Sld2, Dpb11, GINS, and Pol ε, and the subsequent recruitment of Mcm10. The formation of the Cdc45-Mcm2-7-GINS complex activates the helicase, triggering melting of origin DNA. IV. Pol α and Pol δ are loaded on the unwound DNA in a Mcm10-dependent process to complete replisome assembly.