Continued primer synthesis at stalled replication forks contributes to checkpoint activation

Abstract

Stalled replication forks activate and are stabilized by the ATR (ataxia-telangiectasia mutated and Rad3 related)-mediated checkpoint, but ultimately, they must also recover from the arrest. Although primed single-stranded DNA (ssDNA) is sufficient for checkpoint activation, it is still unknown how this signal is generated at a stalled replication fork. Furthermore, it is not clear how recovery and fork restart occur in higher eukaryotes. Using Xenopus laevis egg extracts, we show that DNA replication continues at a stalled fork through the synthesis and elongation of new primers independent of the checkpoint. This synthesis is dependent on the activity of proliferating cell nuclear antigen, Pol-delta, and Pol-epsilon, and it contributes to the phosphorylation of Chk1. We also used defined DNA structures to show that for a fixed amount of ssDNA, increasing the number of primer-template junctions strongly enhances Chk1 phosphorylation. These results suggest that new primers are synthesized at stalled replication forks by the leading and lagging strand polymerases and that accumulation of these primers may contribute to checkpoint activation.

Figure 1.

Small DNA products accumulate in response to aphidicolin. (A) A 9-kb plasmid was incubated in HSE and NPE containing α-[³²P]dCTP with or without aphidicolin (15 µM Aph). Three samples were taken at the indicated times after NPE addition. DNA was isolated from the first sample, run on a denaturing polyacrylamide gel, and autoradiographed (top). The second was run on a chloroquine agarose gel and stained with SybrGold (middle; U-form refers to unwound DNA). The third was run on an SDS-PAGE gel and immunoblotted for Chk1 phosphorylation with PCNA as a loading control. (B) Lengths and intensities of DNA intermediates in A were calculated using ImageQuant software (n = 3). (C–E) Xenopus sperm chromatin (2,500 nuclei/µl) and α-[³²P]dCTP were added to mock- or Pol-α–depleted LSE. Recombinant Pol-α complex was added where indicated. (C) Depletion was verified by immunoblotting with Orc2 as a loading control. (D) Aliquots taken at the indicated times were analyzed for replication. (E) Mock and depleted extracts were treated with 15 µM aphidicolin, and samples were analyzed for continued synthesis as in A.

Figure 2.

Primer synthesis and elongation continues at a stalled fork. (A and B) A 9-kb plasmid was replicated in HSE and NPE with 15 µM aphidicolin (Aph). Radionucleotides were added at the indicated times after NPE addition (A) or at the start of the reaction (B), and samples were analyzed as in Fig. 1 A. (C) Sperm chromatin (4,000 nuclei/µl) was replicated in HSE and NPE with 15 µM aphidicolin. 30 min after NPE addition, p27^KIP was added, and 5 min later, α-[³²P]dCTP was added. Samples were taken at the indicated times and analyzed for Chk1 phosphorylation and small nascent DNAs as in Fig. 1 A.

Figure 3.

The ATR–9-1-1 pathway is not required for primer synthesis. (A–D) Sperm chromatin (2,000 nuclei/µl) was replicated in mock- or ATRIP-depleted LSE with or without 15 µM aphidicolin (Aph). (A) Samples were immunoblotted to verify depletion. (B–D) Aliquots were taken at the indicated times and analyzed for replication (B), Chk1 phosphorylation (C), and primer synthesis (D) as in Fig 1. (E–G) Sperm chromatin (2,500 nuclei/µl) was replicated in mock- or Rad1-depleted LSE containing recombinant 9-1-1 complex and 15 µM aphidicolin as indicated. (E and G) Aliquots were analyzed for Chk1 phosphorylation (E) and primer synthesis (G) as in Fig. 1 A. (F) In parallel, chromatin was isolated from a third aliquot, and bound proteins were immunoblotted with the indicated antibodies.

Figure 4.

Relationship between nascent DNA synthesis, Pol-α hyperloading, and TopBP1 function. (A) Sperm chromatin (2,500 nuclei/µl) was replicated in LSE containing 30 µM aphidicolin (Aph). At the indicated times, parallel samples were analyzed for Chk1 phosphorylation and chromatin binding as in Fig. 3 (E and F). 5 min before each time point, an aliquot was incubated with α-[³²P]dCTP and analyzed for nascent DNA synthesis as in Fig. 1 A. (B) Experimental schematic for C–F. Sperm chromatin (4,000 nuclei/µl) was replicated in LSE and isolated after 45 min to yield initiated chromatin. Initiated chromatin was replicated in mock- or TopBP1-depleted LSE containing α-[³²P]dCTP and p27^KIP with or without 30 µM aphidicolin. (C) Initiated chromatin was blotted for chromatin-bound proteins. (D) Replication was analyzed in mock- and TopBP1-depleted extracts lacking aphidicolin as in Fig. 1 D. (E) Parallel samples were taken 60 min after the addition of initiated chromatin and analyzed for Chk1 phosphorylation or chromatin binding as in Fig. 3 (E and F). (F) Primer synthesis was analyzed in aphidicolin-treated extracts as in Fig. 1 A, and intensities were calculated using ImageQuant software and normalized relative to the 90-min sample from mock-depleted extract. Error bars represent standard error (n = 8).

Figure 5.

Pol-δ, Pol-ε, and PCNA are required for the accumulation of small nascent DNAs. (A–F) Sperm chromatin (2,500 nuclei/µl) was added to mock-, Pol-δ–, Pol-ε–, or PCNA-depleted LSE with or without 15 µM aphidicolin (n = 3). (A and D) Depletion was verified by immunoblotting. (B and E) Replication was monitored in the absence of aphidicolin as in Fig. 1 D. (C) Synthesis of small nascent DNAs was monitored in aphidicolin-treated extracts and quantitated as in Fig. 1 B. (F) Small nascent DNAs were monitored as in C.

Figure 6.

Pol-δ, Pol-ε, and PCNA are required for optimal Chk1 phosphorylation. (A–C) Untreated or UV-irradiated (1,000 J/m²) sperm chromatin (2,500 nuclei/µl) was added to mock-, Pol-δ–, Pol-ε–, or PCNA-depleted LSE in the presence or absence of 15 µM aphidicolin (Aph). (A) Chk1 phosphorylation was analyzed as in Fig. 1 A. (B) Chk1 phosphorylation at 60 min was normalized to total Chk1 levels and quantitated using FluorChem analysis software. Error bars indicate standard error (n = 3). (C) Samples were taken at 30 min and analyzed for chromatin binding as in Fig. 3 F.

Figure 7.

Primer synthesis contributes strongly to Chk1 phosphorylation. (A) Schematic of purified ssM13 structures. Biotinylated ends are represented by a dot. (B) Purified ssM13 structures were coupled to streptavidin and added at 6 ng/µl to NPE containing 300 µM aphidicolin. Samples were taken after 20 min and blotted for the proteins indicated. Chk1 phosphorylation was quantitated using Photoshop. (C) Purified ssM13 and 6–80-mer structures were added to NPE containing 300 µM aphidicolin (aph). The indicated concentration of the 6–80 mer is shown, and ssM13 was added so the total concentration of DNA was 144 ng/µl. Sperm chromatin (2,500 nuclei/µl) was also replicated in HSE and NPE containing 15 µM aphidicolin. Samples were analyzed at 20 (structures) or 60 min (chromatin) for Chk1 phosphorylation and quantitated as in Fig. 6 B. (B and C) Error bars indicate standard error (n = 3).