De novo assembly of genuine replication forks on an immobilized circular plasmid in Xenopus egg extracts

Abstract

We describe an improved model of DNA replication in Xenopus egg extracts, in which a circular plasmid immobilized on paramagnetic beads is used as a template. DNA synthesis occurred on either circular or linear plasmids coupled to the beads, but only DNA synthesis on the circular plasmid was inhibited by geminin and a CDK inhibitor, p21. DNA synthesis on the circular plasmid occurred after a time lag, during which nuclear formation was probably occurring. Although pre-replicative complexes (pre-RCs) were formed soon after mixing plasmids with egg extracts, binding of CDC45, RPA, Pol alpha, delta and epsilon, and PCNA to the circular plasmid was delayed, but still correlated with DNA synthesis. Moreover, p21 inhibited binding of these replication fork proteins to the circular plasmid. Therefore, the circular plasmid, but not the linear plasmid, assembles bona fide replication forks in egg extracts. We conclude that this improved replication system will be useful for studying the mechanism of formation of replication forks in eukaryotic DNA replication.

Figure 1

DNA synthesis on plasmid immobilized on paramagnetic beads in Xenopus egg extracts. (A) A circular form of plasmid pBluescript SK used for coupling to the beads (lane 1) or recovered from the immobilized beads (lane 2) was separated by agarose gel electrophoresis and stained with ethidium bromide. (B) A linear form (lanes 1–4) or circular form (lanes 5–8) of pG5λ6.6 (11 kb) immobilized on the beads was incubated in LSS in the presence or absence of 2 μg/ml geminin for the indicated time periods. (C) A circular form of pEXλ6.6 (12 kb) (lanes 1 and 2) or pKS-EX (5 kb) (lanes 3 and 4) immobilized on the beads was incubated in LSS for 4 h in the presence or absence of 13 μg/ml GST-p21. The ³²P-labeled replication products were separated by agarose gel electrophoresis and detected by autoradiography. Note that the major 23 kb product seen in lanes 1–4 in (B) migrated slightly faster than a nicked circular form of the plasmid. I, II and III denote fully supercoiled circular, nicked circular and linear forms of plasmid, respectively.

Figure 2

A time course analysis of DNA synthesis on circular plasmid beads in Xenopus egg extracts. A circular form of pG5λ6.6 immobilized on the beads was incubated in LSS for the indicated time periods. The reaction in lane 8 was performed using the extracts supplemented with 2 μg/ml geminin. The autoradiogram of the replication products (A) and the amounts of DNA synthesis relative to a 6 h incubation (B) are shown. I, II, and III in (A) denote fully supercoiled circular, nicked circular and linear forms of plasmid, respectively.

Figure 3

Analysis of the proteins bound to plasmid beads during incubation in Xenopus egg extracts. (A) pBluescript-coupled beads were incubated for 30 min in the mock-depleted (lane 1), the ORC-depleted extracts (lanes 2) or the ORC-depleted extracts supplemented with PEG-B fraction (lane 3). (B) pG5λ6.6-coupled beads or the beads alone (no DNA) were incubated in LSS for the indicated time periods in the absence or presence of 13 μg/ml GST-p21. After incubation, the proteins bound to the beads were analyzed by western blotting with the appropriate antibodies as indicated.

Figure 4

Requirement of nuclear formation for the binding of the replication fork proteins to plasmid beads. (A) pG5λ6.6-coupled beads were incubated in LSS prepared by the normal procedure (lanes 2–6) or partially depleted of nuclear membranes (lanes 8–12) for the indicated time periods. (B) pG5λ6.6-coupled beads were incubated in LSS supplemented with WGA (lanes 3 and 4) or the control buffer (lanes 1 and 2) for the indicated time periods. After incubation, the proteins bound to the beads were analyzed by western blotting with the appropriate antibodies as indicated. One microliter each of the extracts was loaded in lanes 1 and 7 in (A).

Figure 5

The effect of linearization of plasmid on pre-RC formation. (A) pG5λ6.6 (11 kb) or pBluescript (3 kb) coupled beads were pre-treated with XbaI to linearize the plasmid (lanes 2 and 4) or with the control buffer (lanes 1 and 3). After treatment, the plasmid was recovered from the beads, separated by agarose gel electrophoresis and stained with ethidium bromide. (B) The same pre-treated plasmid beads as in (A) were incubated in LSS supplemented with 2 μg/ml geminin (lanes 2, 4, 6 and 8) or the control buffer (lanes 1, 3, 5 and 7) for 30 min. The proteins bound to the beads were then analyzed by western blotting. (C and D) pBluescript-coupled beads were incubated in LSS for 30 min. The beads were then washed and treated with XbaI or the control buffer for 15 or 30 min. The plasmid after 15 min treatment with XbaI (C) and the proteins that remained bound to the beads (D) are shown. Lane 1 in (D) corresponds to the beads after 30 min incubation in LSS.