Sequence-independent DNA binding and replication initiation by the human origin recognition complex

Abstract

We report that a highly purified human origin recognition complex (HsORC) has intrinsic DNA-binding activity, and that this activity is modestly stimulated by ATP. HsORC binds preferentially to synthetic AT-rich polydeoxynucleotides, but does not effectively discriminate between natural DNA fragments that contain known human origins and control fragments. The complex fully restores DNA replication to ORC-depleted Xenopus egg extracts, providing strong evidence for its initiator function. Strikingly, HsORC stimulates initiation from any DNA sequence, and it does not preferentially replicate DNA containing human origin sequences. These data provide a biochemical explanation for the observation that in metazoans, initiation of DNA replication often occurs in a seemingly random pattern, and they have important implications for the nature of human origins of DNA replication.

Figure 1.

Purification of recombinant HsORC and DNA binding by HsORC. (A) Composition of purified recombinant HsORC. HsORC was purified from a chromatin-enriched extract by Ni-agarose and anti-HA antibody affinity chromatography. The purified HsORC was analyzed by SDS-PAGE, and the individual subunits were detected by silver staining. (B) Purified HsORC was analyzed by SDS-PAGE and Western blotting with subunit-specific antibodies. (C) Titration of protein. Recombinant HsORC at the indicated concentrations was incubated with 0.5 nM radiolabeled lamin B2 ori II probe in the presence or absence of 1 mM ATP. The amount of radioactive lamin B2 ori II DNA bound by HsORC was determined by the nitrocellulose filter-binding assay. (D) Recombinant HsORC prefers ATP but does not require hydrolysis for DNA-binding activity. Nitrocellulose assays were performed as described in A except that 0.75 ng/μL recombinant HsORC and the indicated concentrations of various nucleotides were used.

Figure 4.

Properties of DNA-affinity-purified HsORC. (A) DsDNA affinity purification of HsORC. HsORC prepared as in Figure 1A was further purified by adsorption to magnetic beads containing immobilized DNA followed by elution with 500 mM NaCl as in Figure 3B. SDS-PAGE was used to analyze 1 μL of input, 2 μL of unbound, and 1 μL of eluted fractions, and the individual subunits were detected by silver staining. (B) Titration of protein. DNA-affinity-purified HsORC at the indicated concentrations was incubated with 0.5 nM radiolabeled lamin B2 ori II probe in the presence or absence of 1 mM ATP. The amount of radioactive lamin B2 ori II DNA bound by HsORC was determined by the nitrocellulose filter-binding assay. (C–E) Competition experiments. DNA-affinity-purified HsORC was incubated with a fixed quantity of radiolabeled lamin B2 ori II probe and various amounts of different competitor DNAs in the absence of ATP. Similar results were obtained in the presence of ATP. (C) Competition assays with synthetic polynucleotides were performed as described for Figure 2A. (D) Competition assays with human lamin B2 origin DNA fragments were performed as in Figure 3A. (E) Competition assays with nonorigin DNAs were performed as in Figure 3C.

Figure 2.

Recombinant HsORC preferentially binds to AT-rich sequences. (A) Competition with synthetic polynucleotide duplex DNA. Fixed amounts of recombinant HsORC and radioactive lamin B2 ori II DNA were incubated with various amounts of competitor polynucleotide duplex DNA in the presence of ATP. The fraction of input radiolabeled lamin B2 ori II DNA bound to HsORC was determined by the nitrocellulose filter-binding assay. The fold competitor is in terms of concentration of nucleotides. (B) Recombinant HsORC binds to poly(dA) · poly(dT). Recombinant HsORC was incubated with oriP DNA conjugated to magnetic beads in the absence (No Comp.) or presence (A/T Comp.) of poly(dA) · poly(dT). The bound proteins were analyzed by SDS-PAGE and detected by Western blotting with antibodies specific for each HsORC subunit. Input material (I), unbound fraction (U), and bound fraction (B) are indicated.

Figure 3.

Recombinant HsORC binds DNA with little sequence specificity. Recombinant HsORC was incubated with a fixed quantity of radiolabeled lamin B2 ori II probe and various amounts of different competitor DNAs in the presence of ATP as indicated. The extent of binding of the radioactive lamin B2 ori II DNA was measured by the nitrocellulose filter-binding assay. As in Figure 2, the fold competitor is in terms of concentration of nucleotides. (A) Competition with human lamin B2 origin DNA. Overlapping fragments that constitute the human lamin B2 origin were generated by PCR and used as competitors. (B) Competition with Chinese hamster DHFR origin DNA. Fragments that span DHFR oriβ were synthesized by PCR and used as competitors. (C) Competition with nonorigin DNA. A fragment of pUC19 and three fragments of pBS (I, II, and III) were used as competitors.

Figure 5.

Purified HsORC supports DNA replication. (A) Sperm chromatin (9000/μL final concentration) was incubated with XlOrc2-depleted egg cytosol (EC; sample c), XlOrc2-depleted EC supplemented with 13.5 ng/μL final concentration HsORC (sample a), or mock-depleted EC (sample b) for 30 min, followed by incubation with NPE. Replication products were separated on a 0.8% agarose gel and visualized by autoradiography. Incorporation of [α-³²P]dATP was measured at 15, 30, and 45 min after addition of NPE. The fraction of input DNA replicated by 45 min (73%) was calculated from the fraction of the endogenous dNTP pool consumed (Blow and Laskey 1986), assuming an endogenous concentration of dATP of 50 μM in egg cytosol and NPE. (B) The radioactivity in each lane of the experiment shown in A was measured using a PhosphorImager, normalized such that the highest value was equal to 100, and graphed.

Figure 6.

HsORC can initiate DNA replication from any DNA sequence. (A) HsORC can replicate a plasmid of prokaryotic composition. pBS (9 ng/μL final concentration) was incubated with XlOrc2-depleted EC (squares), XlOrc2-depleted EC supplemented with 14.5 ng/μL final concentration HsORC (triangles), or mock-depleted EC (circles) for 30 min, followed by incubation with NPE. DNA replication was measured at 15, 30, and 45 min after addition of NPE. (B) HsORC-dependent replication is inhibited by p27^Kip. pBluescript (pBS;9 ng/μL final concentration) was incubated with XlOrc2-depleted EC (squares) or XlOrc2-depleted extract supplemented with 45 ng/μL HsORC for 30 min (circles, triangles). NPE (triangles, squares) or NPE preincubated with 85 ng/μL final concentration p27^Kip (circles) was added to each reaction, and DNA replication was measured 15, 30, and 45 min after addition of NPE. (C) Three nonoverlapping 200-bp DNA fragments (termed a, b, and c) were PCR-amplified from pBluescript and coupled separately to magnetic beads. The beads were incubated for 20 min in XlOrc2-depleted EC supplemented with 10 ng/μL HsORC (lanes 1–3), XlOrc2-depleted EC (lanes 4–6), or mock-depleted EC (lanes 7–9). The beads were then isolated and washed. Bound proteins were separated by SDS-PAGE and transferred to PVDF membrane that was cut into strips and probed with antibodies against HsOrc2, HsOrc5, XlMcm3, and XlCdc6.

Figure 7.

HsORC replicates origin-containing and “originless” plasmids with the same efficiency. (A–C) DNA replication of an origin-containing plasmid (circles) and a control plasmid containing only prokaryotic DNA sequences (squares) was compared separately in mock-depleted egg cytosol (panel I), or XlOrc2-depleted egg cytosol supplemented with different amounts of HsORC (panels II–IV). In panel I, DNA replication of the control plasmid (triangles) and the ori-containing plasmid (diamonds) was also measured using XlOrc2-depleted egg cytosol that was not supplemented with HsORC. (A) Control plasmid, pET24gp4–63;ori plasmid, contains the DHFR oriβ origin of replication. (B) Control plasmid, φpRI10;ori plasmid, contains the lamin B2 origin of replication. (C) Control plasmid, φpRI10;oriplasmid, contains the intergenic region between MCM4 and PRKDC. Plasmid concentration was 9 ng/μL in egg cytosol. (D) DNA replication of the lamin B2 plasmid (B2), an originless control plasmid (B2Δ), or both was examined in mock-depleted (lanes 1–4), XlORC-depleted (lanes 5,6), or XlORC-depleted extract supplemented with HsORC (lanes 7–18) as above. In each case, total plasmid concentration was 10 ng/μL in egg cytosol. At 15 and 45 min, reaction products were separated on an agarose gel, and replication intermediates and products (Walter and Newport 2000), are indicated to the right. The total radioactivity in each lane was determined with a PhosphorImager (see boxes).