Purification and Characterization of theSchizosaccharomyces pombe Origin Recognition Complex

The origin recognition complex (ORC) plays a central role in the initiation of DNA replication in eukaryotic cells. It interacts with origins of DNA replication in chromosomal DNA and recruits additional replication proteins to form functional initiation complexes. These processes have not been well characterized at the biochemical level except in the case of Saccharomyces cerevisiae ORC. We report here the expression, purification, and initial characterization of Schizosaccharomyces pombeORC (SpORC) containing six recombinant subunits. Purified SpORC binds efficiently to the ars1 origin of DNA replication via the essential Nterminal domain of the SpOrc4 subunit which contains nine AT-hook motifs. Competition binding experiments demonstrated that SpORC binds preferentially to DNA molecules rich in AT-tracts, but does not otherwise exhibit a high degree of sequence specificity. The complex is capable of binding to multiple sites within the ars1 origin of DNA replication with similar affinities, indicating that the sequence requirements for origin recognition in S. pombe are significantly less stringent than in S. cerevisiae. We have also demonstrated that SpORC interacts directly with Cdc18p, an essential fission yeast initiation protein, and recruits it to the ars1 origin in vitro. Recruitment of Cdc18p to chromosomal origins is a likely early step in the initiation of DNA replication in vivo. These data indicate that the purified recombinant SpORC retains at least two of its primary biological functions and that it will be useful for the eventual reconstitution of the initiation reaction with purified proteins.

The origin recognition complex (ORC) plays a central role in the initiation of DNA replication in eukaryotic cells. It interacts with origins of DNA replication in chromosomal DNA and recruits additional replication proteins to form functional initiation complexes. These processes have not been well characterized at the biochemical level except in the case of Saccharomyces cerevisiae ORC. We report here the expression, purification, and initial characterization of Schizosaccharomyces pombe ORC (SpORC) containing six recombinant subunits. Purified SpORC binds efficiently to the ars1 origin of DNA replication via the essential Nterminal domain of the SpOrc4 subunit which contains nine AT-hook motifs. Competition binding experiments demonstrated that SpORC binds preferentially to DNA molecules rich in AT-tracts, but does not otherwise exhibit a high degree of sequence specificity. The complex is capable of binding to multiple sites within the ars1 origin of DNA replication with similar affinities, indicating that the sequence requirements for origin recognition in S. pombe are significantly less stringent than in S. cerevisiae. We have also demonstrated that SpORC interacts directly with Cdc18p, an essential fission yeast initiation protein, and recruits it to the ars1 origin in vitro. Recruitment of Cdc18p to chromosomal origins is a likely early step in the initiation of DNA replication in vivo. These data indicate that the purified recombinant SpORC retains at least two of its primary biological functions and that it will be useful for the eventual reconstitution of the initiation reaction with purified proteins.
In bacteria, bacteriophage, and animal viruses the initiation of DNA replication takes place at defined nucleotide sequences known as origins of replication (1). Initiator proteins bind to such origin sequences and promote the biochemical steps leading to the establishment of replication forks. The interaction of initiator proteins with origins is less well understood in the case of eukaryotic cells where initiation of DNA replication occurs at multiple sites along chromosomal DNA (2). The best characterized eukaryotic chromosomal origins of replication are those of the budding yeast Saccharomyces cerevisiae. Like the origins of prokaryotes and animal viruses, budding yeast origins are modular in nature and are composed of several short, well defined sequence blocks distributed over a region of ϳ100 -150 bp (3)(4)(5). The most highly conserved sequence block of budding yeast origins is the A domain which contains an essential 11-bp ARS consensus sequence. An additional, less well conserved sequence block, referred to as the B domain, serves to enhance the efficiency of origin utilization (3)(4)(5). The six-subunit S. cerevisiae Origin Recognition Complex (ScORC) 1 binds specifically to the ARS consensus sequence in a reaction requiring ATP (6). Genetic and biochemical studies have established that ORC plays a central role in the initiation of DNA replication and that it functions, at least in part, to recruit essential replication factors to origins of DNA replication to form the pre-replication complex (7)(8)(9)(10)(11). One such factor is Cdc6p which is required, together with ScORC, to load the MCM complex, a putative DNA helicase, onto DNA (10, 12) (for review, see Ref. 2).
Homologues of ScORC subunits have been identified in a variety of eukaryotic species including humans (13). In addition, protein complexes containing ORC-related subunits have been identified in extracts of Xenopus laevis eggs, Drosophila melanogaster embryos, Schizosaccharomyces pombe cells, and human HeLa cells (14 -17). Thus, ORC has been highly conserved during evolution, suggesting the existence of common mechanisms for initiating DNA replication in all eukaryotes. However, it is not yet clear whether the interaction of ORC with origins of DNA replication in other species is similar to that in S. cerevisiae. Indeed, there is considerable evidence that the initiation of DNA replication in metazoans can occur at many sites within broad replication zones, suggesting that the sequence requirements for initiation may be more relaxed than in budding yeast (18,19).
In previous studies we and others have shown that fission yeast origins of DNA replication differ dramatically from their budding yeast counterparts (20 -22). Fission yeast origins have a minimal size of 500 to 1000 bp and are very rich in AT base pairs. However, they do not share a common consensus sequence comparable to the ARS consensus sequence of S. cerevisiae replication origins. In addition, S. pombe origins are characterized by a high degree of functional redundancy. Sequence blocks that are important for origin function appear to be composed of smaller AT-rich sequence elements that can be deleted individually without significantly affecting origin activity (23). Several genetic properties of S. pombe origins can be rationalized by the finding that the S. pombe homologue of one of the ORC subunits (SpOrc4p) contains an N-terminal DNAbinding domain consisting of nine AT-hook motifs (24). We have suggested that binding of the N-terminal domain of SpOrc4p to appropriately spaced AT-tracts serves to tether the ORC complex to S. pombe origins of DNA replication. Consistent with this possibility, we have demonstrated that the isolated SpOrc4 subunit can bind to DNA containing a known S. pombe origin (24). However, the DNA binding properties of the SpORC holo-complex have not been reported.
To understand origin recognition and the assembly of initiation complexes in S. pombe, it will be essential to characterize the biochemical properties of SpORC. For this purpose, we have expressed all six SpORC subunits in insect cells using the baculovirus system, and we have purified the complex to near homogeneity. The purified SpORC binds with high affinity to a known origin of DNA replication in S. pombe (ars1). Our data indicate that SpORC recognizes multiple sites within ars1 DNA, consistent with the hypothesis that origin selection in S. pombe is less sequence-specific than in S. cerevisiae. The binding of SpORC to ars1 DNA is mediated by the N-terminal domain of the SpOrc4 subunit. We have demonstrated that this domain is essential for the viability of S. pombe. Finally, we have shown that SpORC interacts directly with Cdc18p, a key regulator of the initiation of DNA replication. This interaction recruits Cdc18p to origin DNA, which is a likely early step in the initiation of chromosomal DNA replication. Thus, the expression and purification of SpORC should facilitate biochemical analysis of the initiation reaction.

EXPERIMENTAL PROCEDURES
Cloning of S. pombe orc and cdc18 Genes into Baculovirus Vectors-The nomenclature for S. pombe orc genes follows the suggestion in Ref. 30. To construct a baculovirus expression vector containing the orc1 ϩ gene, the full-length gene was synthesized by PCR using oligonucleotides 5Ј-TCGCGGATCCTGAACTTCTCTTAATATGCCTAGAAGA-AAG-3Ј and 5Ј-CCGTAACTGCAGTTATGCTATCCCAGCAAG-3Ј as primers and pGEX4T-1-orc1 as template (25). The PCR product was digested with BamHI and PstI and cloned into the corresponding sites in the plasmid pFastBac1 (Invitrogen) to yield pFastBac1-SpOrc1.
The orc2 ϩ and orc3 ϩ genes were inserted into the same pFastBac Dual vector (Invitrogen). The orc2 ϩ gene was amplified by using oligonucleotides 5Ј-TCGCGGATCCTGAACTTCTCTTAATATGCCTAGA-AGAAAG-3Ј and 5Ј-CCGTAACTGCAGTTATGCTATCCCAGCAAG-3Ј as primers and S. pombe genomic DNA as template. The resulting PCR product was digested with BamHI and PstI and cloned into the corresponding sites in the plasmid pFastBac Dual vector to yield pFastBac Dual-orc2. The orc3 ϩ gene was amplified by using oligonucleotides 5Ј-CCGAGAATTCTAATATCAAATTAATATGTCAGCAATAC-TACAA-3Ј and 5Ј-GCATTCCTGCAGTTACTGATGATAAATAGT-3Ј as primers and pET19b-orc3 (kindly provided by Dr. J. Hurwitz, Sloan-Kettering Institute) as template. The PCR product was cloned into the EcoRI and PstI sites of pFastBac1, resulting in pFastBac1-Sporc3. The pFastBac1-Sporc3 was linearized with RsrII and the termini were filled-in with T4 DNA polymerase. The orc3 ϩ gene was released by digestion with SphI and cloned between the SmaI and SphI sites of pFastBac Dual-Sporc2 to yield pFastBac Dual-Sporc2-Sporc3.
The N-terminal FLAG-tagged Cdc18p was constructed as follows. The coding sequence of the FLAG tag was first created by annealing two oligonucleotides (5Ј-GGCTTGCCACCATGGATTACAAGGATGACGAC-GATAAGGGATCCT-3Ј and 5Ј-TCGAAGGATCCCTTATCGTCGTCAT-CCTTGTAATCCATGGTGGCA-3Ј) and then ligated the product into the BamHI and BssHII sites of pFastBac1 to yield pFastBac1-FLAG. The S. pombe cdc18 ϩ gene (26) was generated by PCR and cloned into the StuI and NotI sites of pFastBac1-FLAG.
All constructs were confirmed by DNA sequencing. Recombinant viruses were generated and amplified according to the manufacturer's instructions (Bac-to-BacTM Baculovirus Expression Systems, Invitrogen).
Antibody Production-For generation of antibodies, the SpORC subunits were expressed in bacteria as recombinant fusion proteins tagged with either the His 6 -epitope or the glutathione S-transferase moiety. For SpOrc1p, a truncated gene containing N-terminal residues 1-295 was amplified and cloned into the BamHI site of pGEX2T (Amersham Bioscience). For SpOrc2p, a C-terminal segment of gene containing residues 239 -534 was amplified and cloned into the BamHI and SalI sites of pGEX 4T-1 (Amersham Bioscience). For SpOrc3p, a truncated gene containing N-terminal residues 1-344 was amplified by using oligonucleotides 5Ј-CCGCGGATCCGATGTCAGCAATACTACAA-3Ј and 5Ј-CCGAAAGCTTCATTTAAATAATCCAGACAAG-3Ј as primers and pFastBac1-Sporc3 as template. The PCR product was then cloned into the BamHI and HindIII sites of the pET21b expression vector (Novagen). For SpOrc5p, the C-terminal segment containing residues 221-454 was amplified by using oligonucleotides 5Ј-CCGCGGATCCAT-GGAGTGTTTTTGGAGTGCA-3Ј and 5Ј-CCGAAAGCTTCTATCCCGC-CAAATAACTAT-3Ј as primers and pFastBac1-Sporc5 as template. The PCR product was then inserted between the BamHI and HindIII sites of the pET21b expression vector. For SpOrc6p, the pET19b-orc6 containing the full-length orc6 ϩ gene was used to express recombinant protein. Recombinant glutathione S-transferase fusion proteins were expressed in bacteria and purified on a glutathione column according to the manufacturer's instructions (Amersham Bioscience). All His 6tagged proteins were expressed in bacteria and purified by nickelagarose (Ni-agarose) chromatography according to instructions provided by the manufacturer (Qiagen). The purified proteins were used as antigens to immunize rabbits (Covance Research Products, Denver, PA).
Expression and Purification of SpORC and Cdc18p in Insect Cells-Sf9 insect cells were cultured at 27°C in Grace's medium supplemented with 10% fetal bovine serum. For expression of recombinant SpORC, Sf9 cells (2 ϫ 10 6 cells/ml) were co-infected with five recombinant baculoviruses expressing all six subunits at a multiplicity of infection of 2-5. After 48 h cells were harvested and washed once with ice-cold phosphate-buffered saline (10 mM phosphate buffer, pH 7.3, 140 mM NaCl, 2.7 mM KCl) and collected by centrifugation. Infected cell pellets from 250-ml cultures were lysed in 50 ml of L buffer (20 mM Tris-HCl, pH 6.8, 0.4 M sorbitol, 150 mM KOAc2, 5 mM MgCl 2 , 5 mM MgSO 4 , 1% Triton X-100) on ice for 5 min. The suspension was centrifuged at 1,400 ϫ g for 5 min at 4°C, and the resulting chromatin-enriched pellet was extracted with 12 ml of E buffer (50 mM Hepes pH 7.5, 2.5 mM MgCl 2 , 500 mM NaCl, 10% glycerol, 1% Triton X-100) on ice for 30 min. The suspension was then centrifuged at 100,000 ϫ g for 30 min at 4°C. The supernatant was collected and incubated with 0.25 ml of anti-HA antibody-conjugated agarose (F7-agarose) (Santa Cruz Biotech) at 4°C for 2 h. The beads were washed once with 25 ml of E buffer and 1 ml of F buffer (50 mM Hepes pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 5 mM MgOAc2, 10% glycerol) twice. The F7-agarose-bound proteins were eluted by incubation with 0.8 ml of F buffer containing 2 mg/ml HA peptide overnight at 4°C. This step yielded a nearly homogeneous 2 R. Y. Chuang and T. Kelly, unpublished data.
SpORC which was used for most experiments. In some cases (e.g. Fig. 1) the protein was further purified by Ni-agarose affinity chromatography. The eluate from the antibody affinity step was collected and incubated with 0.2 ml of Ni-agarose for 1 h at 4°C. The resin was washed with 5 ml of F buffer and 1 ml of E buffer. The SpORC was eluted with E buffer containing 200 mM imidazole.
The expression and purification of FLAG-Cdc18p protein was performed by methods identical to those used for SpORC except as follows. The infected cells from 180 ml of culture were lysed in 25 ml of L buffer. After centrifugation, the chromatin pellets were extracted with 8 ml of E buffer for 30 min at 4°C. After centrifugation at 100,000 ϫ g for 30 min, the supernatant was incubated with 0.3 ml of FLAG antibody beads (Sigma-Aldrich) at 4°C for 2 h. The Cdc18p was eluted from the antibody beads with 0.7 ml of F buffer containing 1 mg/ml FLAG peptide (Sigma-Aldrich). All buffers used for protein purification contained the following protease inhibitors: leupeptin (10 g/ml), aprotinin (10 g/ml), soybean trypin inhibitor (2 g/ml), bestatin A (10 g/ml), N-tosyl-L-phenylalanine chloromethyl ketone (20 g/ml), 4-(2-aminoethyl)-benzenesulfonylfluoride hydrochloride (AEBSF) (2 mM).
For immunoprecipitation experiments, chromatin-enriched extracts were prepared as described above, and SpORC was purified by Niagarose chromatography. After elution with imidazole, the eluate was incubated with anti-SpOrc5p antibody cross-linked to protein A-Sepharose (Amersham Bioscience) as described (17). After 2-3 h incubation at 4°C, the resin was washed three times with 1 ml of E buffer. The immunoprecipitated proteins were separated by 10% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and analyzed by either silver staining or immunoblotting.
DNA Binding Assays-For the DNA binding experiments of Fig. 2, a 1.15-kb DNA fragment containing the ars1 origin was synthesized by PCR using 5Ј-biotinylated oligonucleotide, B-5Ј-CAAGGTTTTGCATA-GAATCC-3Ј and 5Ј-GAATTCGAGTCTAACTCCTT-3Ј as primers and pRC20 (21) as template. The PCR product was then coupled to streptavidin-conjugated magnetic beads (Dynalbeads M-280 Streptavidin; Dynal Corp.) according to the manufacturer's instructions. The binding assays were performed by incubating 2 pmol of purified SpORC with beads containing 1 pmol of ars1 DNA in 20 l of F buffer with 1 mg/ml bovine serum albumin at 4°C for 1 h. Where indicated, 1 mM ATP was included in the reaction mixtures. Beads were washed three times with 200 l of F buffer containing 0.1% Nonidet P-40. The bound proteins were released by the addition of 1 ϫ SDS loading dye and separated by 10% SDS-PAGE, followed by Western blotting analysis with antibodies against SpORC subunits (1, 2, 3, 5, and 6). Anti-HA antibody, 12CA5, (Roche Molecular Biochemicals), was used for detecting recombinant HA-His 6 -SpOrc4p.
The immunoprecipitation (McKay) DNA binding assays of Fig. 3, B and C, were performed as described (27) with the following modifications. A 1.5-kb DNA fragment containing the 1.2-kb ars1 origin and 0.3-kb of plasmid DNA was excised from pRC20 with SapI and BamHI. This fragment was further digested with HindIII and XbaI, yielding four DNA fragments: 577, 424, 214, and 324 bp. The first three fragments were derived from ars1 and the fourth from the pBluescript KS sequence (Stratagene). The DNA fragments were labeled with Klenow enzyme (New England Biolabs Inc.) in the presence of dGTP, dCTP, dTTP, and [ 32 P]ATP (3000 Ci/mmol) at 25°C for 1 h. The labeled fragments were purified by Sephadex G-25 chromatography. SpORC (or SpOrc4p) bound to F7-agarose beads via the HA epitope tag was incubated with 32 P-labeled DNA fragments in 30-l reaction mixtures containing 50 mM Hepes pH 7.5, 1 mM EDTA, 1 mM EGTA, 5 mM MgOAc2, 10% glycerol, 0.1 mg/ml poly(dG)-poly(dC) (Amersham Bioscience), 0.05% Nonidet P-40, and NaCl at the indicated concentration. Incubation was carried out at 25°C for 30 min. The beads were washed with 500 l of reaction buffer without poly(dG)-poly(dC) three times. Bound DNA was released in 1% SDS, 10 mM EDTA and separated on a 1.5% agarose gel.
The immunoprecipitation (McKay) DNA binding assays of Fig. 3D were performed as described (27) with the following modifications. The ars1, ars3001 (28), and ars3002 (22) origin fragments and the control fragment, containing nucleotides 2114 -2916 of the SpOrc4 open reading frame, were generated by PCR and inserted into pBluescript KS. After excision from purified plasmid DNA the fragments were labeled with 32 P as described above. Binding reactions (40 l) contained 12.5 mM Hepes pH 7.5, 5 mM EGTA, 2.5 mM MgOAc2, 0.66 mg/ml poly(dI-dC)-(dI-dC), 2 mg/ml bovine serum albumin, 2.5 mM dithiothreitol, 10 ng of SpOrc4 (tagged with HA3), and 18 ng of [ 32 P]DNA. After incubation at room temperature for 10 min, the DNA-protein complexes were collected by incubation with F7 anti-HA-agarose beads for 10 min.
Bound radioactive DNA was released in 1% SDS, 10 mM EDTA and eparated on a 1.5% agarose gel. Radioactivity was quantified in a PhosphorImager.
For filter binding assays, 200-bp DNA fragments spanning the ars1 origin were synthesized by PCR. The ars1-1 fragment was labeled at terminal restriction sites with [␣-32 P]dATP using the Klenow fragment of DNA polymerase I. Synthetic deoxyribonucleotide polymers were purchased from Amersham Bioscience. Competition binding reactions (30 l) contained 50 mM Hepes pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 0.1% BSA, 0.4 nM radioactive ars1-1 DNA, ϳ5 nM SpORC (estimated by silver staining), and the indicated concentrations of unlabeled competitor DNA. After incubation at 25°C for 30 min the reaction mixtures were passed through nitrocellulose filters (HA, Millipore Corp.). The filters were washed, dried, and counted in a scintillation spectrometer.
SpORC and Cdc18p Interaction-SpORC (ϳ2.4 g) was bound to F7-agarose beads in the presence or absence of 2 mg/ml HA peptide. FLAG-Cdc18p, purified as described above, was incubated with SpORC bound F7-agarose in the binding buffer (50 mM Hepes pH 7.5, 200 mM NaCl, 10% glycerol, and 1% Triton X-100) at 4°C for 4 h. After washing with binding buffer, the beads were treated with 1% SDS at 65°C for 10 min to release bound proteins. Proteins were resolved by 10% SDS-PAGE and detected by immunoblotting with antibodies against SpORC and the FLAG epitope.
For analysis of SpORC-Cdc18p-ars1 ternary complexes, 2 pmol of SpORC was incubated with magnetic beads containing 1 pmol of bound ars1 for 1 h at 4°C as described above. The beads were washed with binding buffer, and 1 pmol of FLAG-Cdc18p was added in the presence or absence of 1 mM ATP. The reaction mixture was incubated at 4°C for 1 h. Beads were then washed three times with 200 l of F buffer containing 0.1% Nonidet P-40 and re-suspended in 1 ϫ SDS loading dye. The bound FLAG-Cdc18p was separated by 10% SDS-PAGE and detected by Western blotting analysis using anti-FLAG antibody-conjugated horseradish peroxidase (Sigma-Aldrich).
The plasmid pArg3HA was derived from the plasmid pSLF272 (31) by replacing the ura4 ϩ marker and nmt1 promoter DNA fragment with the S. pombe arg3 ϩ gene. The latter was synthesized by PCR with oligonucleotides 5Ј-TTCCCCCGGGCAGTATTTCATCAACGTAC-3Ј and 5Ј-TTCCCCCGGGATCTATCAATGAGTTTACG-3Ј as primers and the plasmid paR3 (kindly provided by Dr. S. Waddell, Marie Curie Research Institute, UK) as template. The 2.9-kb orc4 ϩ gene was cloned into the XhoI/NotI sites of pArg3HA, yielding pArg3HA-orc4 ϩ . The 3.9-kb coding region of full-length orc4 ϩ plus 1-kb of upstream DNA was cloned into the XhoI/NotI sites of pArg3HA, yielding pArg3HA-prm-orc4. The orc4 gene lacking the N-terminal AT-hook domain (orc4-⌬N) (24) was cloned by replacing the wild-type orc4 ϩ gene in pArg3HA-prm-orc4 to yield pArg3HA-prm-C-orc4. To determine whether the orc4 gene lacking the N-terminal deletion could rescue an S. pombe strain lacking SpOrc4p, the YRC23 strain was transformed with plasmids pArg3HA, pArg3HA-orc4, pArg3HA-prm-orc4, or pArg3HA-prm-C-orc4. Transformed cells were selected on Edinburgh minimal media-arg-ura-leu plates. Colonies were then streaked out on Edinburgh minimal mediaarg-ura-leu plates in the presence or absence of thiamine (5 g/ml) and grown at 30°C for 2-3 days.

Reconstitution and Purification of Recombinant SpORC-To
characterize the biochemical properties of SpORC, we expressed all six subunits of SpORC in insect cells using the baculovirus expression system. To facilitate the purification of the protein complex, hexahistidine (His 6 ) and hemagglutinin (HA) epitope tags were added to the N terminus of SpOrc4p. We have previously observed by rescue of an orc4 deletion that the SpOrc4p with a His 6 -epitope tag in this position is functional in vivo (data not shown). Insect cells were co-infected with recombinant baculoviruses encoding each of the SpORC subunits. Since SpOrc4p contains nine copies of the AT-hook-DNA binding motif, we expected that the recombinant SpORC complex would be tightly associated with chromosomal DNA (24). Therefore, the first step in the purification procedure was to prepare a chromatin-enriched fraction in a buffer containing 1% non-ionic detergent. We recovered greater than 90% of the expressed SpOrc1-SpOrc5 and about half of the SpOrc6p in the chromatin fraction. To find conditions for release of SpORC in a soluble form, the chromatin fraction was extracted with buffers containing increasing concentrations of NaCl. We observed that more than 80% of SpORC was released from the chromatin-enriched fraction in 0.5 M NaCl. This observation is consistent with analysis of the stability of complexes between DNA and purified SpORC (see below) and with previous observations (16).
The solubilized SpORC was purified by sequential affinity chromatography on F7 anti-HA antibody-agarose and Ni-agarose. The eluate from the Ni-agarose affinity matrix was analyzed by SDS-PAGE and silver staining (Fig. 1A). Six major polypeptides with mobilities consistent with the calculated molecular weights of SpOrc1p-SpOrc6p were observed. The identity of the six polypeptides was confirmed by Western blotting with specific antibodies against each SpORC subunit (Fig. 1B). The SpOrc2 subunit consisted of a doublet band (Fig. 1, A and  B). The upper band of the doublet is likely a phosphorylated form of the protein that accumulates in the G 2 and M phases of the cell cycle (32,33).
Additional fractionation experiments were performed to verify the association of the SpORC subunits. For example, chromatin extracts were subjected to two sequential steps of affinity purification, one specific for HA-His 6 -SpOrc4p (Ni-agarose) and one specific for SpOrc5p (immunoprecipitation with anti-SpOrc5p antibodies). As shown in Fig. 1, C and D, all six subunits were recovered by this procedure. In addition, the six SpORC subunits co-eluted when subjected to Mono-S column chromatography (data not shown). We conclude that the recombinant SpORC represents a holo-complex of all six subunits.
Interaction of SpORC with Origin DNA-We have previously shown that the isolated SpOrc4 subunit binds to S. pombe ars1 DNA in vitro via its N-terminal AT-hook domain (24). We have hypothesized that the AT-hook domain serves to tether the SpORC holo-complex to origins of DNA replication. To investigate this possibility we analyzed the DNA binding activity of purified SpORC using several different assays. In the first assay, the fission yeast ars1 origin of replication was synthesized by PCR using a 5Ј-biotin labeled primer. The labeled DNA was then immobilized on streptavidin-conjugated magnetic beads. After incubation with purified SpORC, the ars1 beads were washed, and the bound proteins were eluted and analyzed by SDS-PAGE and immunoblotting with specific antibodies against each subunit of SpORC (Fig. 2). All six subunits were recovered in the bound fraction with similar efficiencies (ϳ50% in this experiment). Unlike S. cerevisiae ORC, SpORC bound to origin DNA in the absence of ATP. Little, if any, SpORC bound to streptavidin beads in the absence of ars1 DNA.
To begin to explore the specificity of the interaction of SpORC with DNA, we performed binding assays using the method developed by McKay (27). In this assay, radioactively labeled DNA fragments were incubated with purified HA-His 6 -SpORC, and the resulting protein-DNA complexes were col-lected with F7 anti-HA antibody-agarose beads. We made use of three non-overlapping fragments of 0.2, 0.4, and 0.5 kb spanning the ars1 origin and a 0.3-kb non-ars1 control frag-

FIG. 1. Reconstitution of a six-subunit recombinant SpORC in insect cells. Sf9 insect cells were co-infected with baculoviruses ex-
pressing the S. pombe Orc1, Orc2, Orc3, Orc5, Orc6, and HA-His 6 -Orc4 subunits. A and B, SpORC was purified from a chromatin-enriched extract by anti-HA antibody affinity chromatography and Ni-agarose chromatography as described under "Experimental Procedures." The purified SpORC was analyzed by SDS-PAGE and the individual subunits were detected by silver staining (A) or immunoblotting with subunit-specific antibodies (B). C and D, SpORC was purified by Niagarose affinity chromatography by means of the His 6 epitope tag present in the SpOrc4 subunit. The partially purified complex was incubated with protein A beads conjugated to either control preimmune or anti-SpOrc5p antibodies. The bound polypeptides were resolved by SDS-PAGE and detected by silver staining (C) or by immunoblotting with subunit-specific antibodies (D). The asterisk indicates an unknown protein of 35-40 kDa that is absent in the most highly purified preparations of SpORC.

FIG. 2.
SpORC holo-complex binds to S. pombe ars1 origin DNA. Purified SpORC was incubated with ars1 DNA conjugated to magnetic beads or with magnetic beads alone in the presence or absence of 1 mM ATP as indicated. The bound polypeptides were analyzed by SDS-PAGE and detected by immunoblotting with antibodies specific for each SpORC subunit. ment (Fig. 3A). All of the fragments derived from ars1 possess AT contents in excess of 65% which is typical of S. pombe ars elements. The AT content of the control non-ars DNA is ϳ47%. At relatively low ionic strengths, all four DNA fragments were bound by the purified SpORC (Fig. 3B). At 350 mM NaCl, the binding of the control fragment was completely eliminated and only the fragments derived from ars1 showed significant binding. At 500 mM NaCl, the binding of ars1 fragments was abolished. These data indicate that the purified SpORC has significant nonspecific DNA binding activity, but the affinity of the protein for the AT-rich fragments derived from ars1 is significantly greater than that for control DNA with a lower AT content. Interestingly, under the conditions of our experiments the stability of DNA-SpORC complexes formed with the three different origin fragments were almost indistinguishable. This result indicates that SpORC can bind to multiple sites within ars1 and is consistent with our previous genetic data suggesting that the origin is composed of redundant elements (21).
The DNA binding properties of the purified six subunit SpORC are similar to those that we have previously reported for the isolated SpOrc4 subunit (24). Fig. 3C shows a direct comparison of the DNA binding properties of the purified SpORC and SpOrc4p. The fractions of the various fragments bound and the salt sensitivity of binding were indistinguishable for the isolated SpOrc4p and the holo-complex. These results are consistent with the hypothesis that the N-terminal domain of SpOrc4p plays a central role in targeting SpORC to origins of DNA replication (24).
We also employed the McKay assay to study the binding of SpOrc4 to additional S. pombe origins of DNA replication. As shown in Fig. 3D, SpOrc4 bound to ars3001 and ars3002 to a significantly greater extent than to the control non-ars DNA fragment. As in the case of ars1 the AT contents of ars3001 and ars3002 are quite high (74 and 80%, respectively), while the AT content of the non-ars fragment derived from S. pombe coding sequence is 60%.
To explore the specificity of SpORC binding in a more quantitative fashion, we carried out a series of competition filter binding experiments in which we measured the ability of various unlabeled DNA molecules to compete for SpORC binding to a 200-bp radioactive DNA fragment (ars1-1) derived from the minimal ars1 origin (Fig. 4). In the first series of experiments the competing DNAs were synthetic polymers with different compositions. We observed that both poly(dA-dT)-(dA-dT) and poly(dA)-(dT) were extremely effective as competitors. In fact, the data in Fig. 4A indicate that the affinity of SpORC for these two AT-containing polymers is significantly greater than its affinity for ars1-1 DNA itself. In contrast, poly(dG-dC)-(dG-dC) or poly(dG)-(dC) were extremely poor competitors, indicating that the affinity of SpORC for GC tracts is orders of magnitude less than for AT tracts.
To determine whether SpORC binds preferentially to a specific region of ars1, we generated a series of overlapping unlabeled 200-bp DNA fragments spanning the complete origin (Fig. 4B). An equal amount of each fragment was tested for its ability to compete with radioactive ars1-1 DNA for binding to a fixed quantity of SpORC. Strikingly, each of the eight ars1 fragments competed effectively for SpORC binding to the radioactive ars1-1 fragment. Moreover, the extent of competition was quite similar for all of the fragments, ranging from 43% (ars1-1) to 68% (ars1-7). These data demonstrate that SpORC is capable of binding to multiple sites within ars1 with similar affinities. Based upon a simple equilibrium model we estimate that the relative affinities of SpORC for the eight ars1 fragments do not differ by more than a factor of 2-3. Thus, the high affinity of SpORC for ars1 and other S. pombe origins is probably due to the cumulative effect of many potential AT-richbinding sites. This situation is markedly different from S. cerevisiae where in most cases a single ScORC-binding site predominates.
FIG. 3. SpOrc4p tethers SpORC to origin DNA. A, fragments of ars1 used in DNA binding experiments. The DNA fragments, ars1-I, ars1-II, and ars1-III, were derived from the S. pombe ars1 origin by restriction digestion and end-labeled with [ 32 P]dATP as described under "Experimental Procedures." A control non-ars fragment of 0.32 kb was prepared in a similar fashion. The shaded region indicates the minimal ars1 origin of DNA replication defined by genetic experiments (21). B, a mixture of the four 32 P-labeled DNA fragments was incubated with purified SpORC immobilized on F7 anti-HA-agarose beads or with control beads alone. The binding reactions, containing the indicated concentrations of NaCl, were incubated at room temperature for 30 min. The bound DNA was eluted with SDS and analyzed by agarose gel electrophoresis, followed by autoradiography. C, similar DNA binding assays were performed with either the SpORC holo-complex (left) or purified SpOrc4 subunit (right) in reaction mixtures containing 250 or 350 mM NaCl. D, purified SpOrc4, tagged with the HA3 epitope, was incubated with radioactive DNA fragments containing the S. pombe origins ars1, ars3001, ars3002, or a control non-ars DNA segment. DNA-protein complexes were collected by immunoprecipitation with F7 anti-HA-agarose beads. The bound radioactive DNA was released by incubation in 1% SDS, fractionated by agarose gel electrophoresis, and quantified in a PhosphorImager.

The N-terminal Domain of SpOrc4p Is Essential for Viability in S. pombe-The N-terminal AT-hook domain present in
SpOrc4p has not been observed in ORCs from other species. Given that our DNA binding data suggested a role for the N-terminal domain in origin selection, it was important to determine whether the domain is essential for SpORC function. For this purpose, we asked whether a form of SpOrc4p lacking the N-terminal domain could replace the wild-type protein in vivo. We made use of a haploid S. pombe strain in which the chromosomal copy of orc4 ϩ was deleted and viability was maintained by a plasmid expressing SpOrc4p under the control of the nmt1 promoter. This strain is non-viable in the presence of thiamine which strongly represses transcription from the nmt1 promoter. We introduced into this strain a plasmid carrying a mutant orc4 gene (orc4-⌬N), lacking the N-terminal segment which encodes all nine AT-hooks, but retaining the C-terminal segment which is homologous to Orc4 in other species. The mutant protein was expressed under control of the orc4 ϩ promoter. As shown in Fig. 5, the mutant gene was unable to rescue the viability of the test strain when expression of the wild-type orc4 ϩ was repressed in the presence of thiamine. Similar results were obtained after introduction of control plasmids carrying either no inserted orc4 ϩ gene or the wild-type orc4 ϩ gene lacking its promoter. In contrast, the wild-type orc4 ϩ gene under the control of its own promoter readily rescued the viability of the test strain in the presence of thiamine. Thus, the N-terminal domain of SpOrc4p is essential, presumably because it is required to target SpORC to origins.
Interaction of SpORC with Cdc18p-There is evidence that one major function of ORC may be to recruit a regulator of DNA replication, Cdc6p/Cdc18p, to origins of DNA replication. Biochemical studies in budding yeast and X. laevis are consistent with the hypothesis that ORC interacts with Cdc6p/Cdc18p and is essential for the association of Cdc6p/Cdc18p with origin DNA (7,11,34). In our initial experiments to explore the possible interaction between S. pombe Cdc18p and ORC, we tagged Cdc18p at its N terminus with the FLAG epitope and expressed it in the baculovirus expression system. Previous studies have demonstrated that epitope tags placed at the N terminus of Cdc18p do not interfere with its function in vivo (35). The recombinant FLAG-Cdc18p was extracted from a chromatin-enriched fraction and purified by affinity chromatography. Analysis of the purified protein by SDS-PAGE revealed a major band (or in some cases a closely spaced doublet band) at the expected molecular weight for FLAG-Cdc18p (Fig.  6A). HA-His 6 -SpORC was purified as described above and immobilized by incubation with F7 anti-HA-agarose beads. Control beads were prepared by incubation of HA-tagged SpORC with anti-HA-agarose in the presence of an excess of HA peptide. Analysis by Western blotting demonstrated that the SpORC holo-complex bound to anti-HA-agarose in the absence, but not in the presence, of HA peptide (Fig. 6B). When FLAG-Cdc18p was incubated with the immobilized SpORC, about 25% of the input protein bound to the beads (Fig. 6C). Under the same conditions, the binding of FLAG-Cdc18p to the control beads was not detectable, indicating that the Cdc18p efficiently forms specific complexes with SpORC. To exclude the unlikely possibility that the observed interaction of Cdc18p and SpORC was mediated by contaminating DNA, we repeated the binding assay in the presence of ethidium bromide and obtained similar results (data not shown). Thus, the interaction between Cdc18p and SpORC is likely to be direct.
As a first step toward reconstitution of the S. pombe initiation complex with purified proteins, we studied the interactions FIG. 4. Specificity of SpORC binding. A, fixed quantities of SpORC and radioactive ars1-1 DNA were incubated in the presence of various amounts of competitor DNA as indicated. The fraction of the input radioactive ars1-1 DNA bound to SpORC was measured by the nitrocellulose filter binding assay. B, a series of eight overlapping 200-bp fragments spanning the ars1 origin was generated by PCR. Each fragment was tested for its ability to compete for binding to SpORC. For this purpose, a constant amount of each competitor fragment (4 nM) was incubated with fixed quantities of SpORC (ϳ5 nM) and radioactive ars1-1 DNA (0.4 nM) as above. The control reaction contained no competitor DNA. The extent of binding of the radioactive ars1-1 DNA was measured by the nitrocellulose filter binding assay.
FIG. 5. The N-terminal AT-hook domain of SpOrc4p is essential for cell viability. An S. pombe strain (YRC23) expressing the S. pombe orc4 ϩ gene under the control of the repressible nmt1 promoter, was transformed with the indicated plasmids carrying derivatives of the orc4 ϩ gene or with vector alone. Transformed colonies were selected and streaked onto plates without (A) or with (B) added thiamine to repress the expression of the endogenous orc4 ϩ gene. The orc4-⌬N plasmid contains the wild-type orc4 ϩ promoter and a mutant orc4 ϩ gene that lacks the N-terminal AT-hook domain. Two control plasmids containing the wild-type orc4 ϩ gene were also tested. One plasmid (orc4 ϩ ) contains the wild type orc4 ϩ promoter and gene, while the other contains the wild-type orc4 ϩ gene without a functional promoter.
of SpORC, Cdc18p, and ars1 origin DNA (Fig. 7). A 1.2-kb DNA fragment containing the complete functional ars1 origin was end-labeled with biotin and bound to magnetic beads. The resulting ars1 beads were incubated with a purified SpORC as described in the experiment of Fig. 2. The ars1-SpORC binary complexes were then incubated with purified FLAG-Cdc18p and bound FLAG-Cdc18p was detected by immunoblotting with anti-FLAG antibodies. In this experiment, the molar ratio of SpORC:Cdc18p:ars1 was approximately 2:1:1. We observed that about 25% of the added FLAG-Cdc18p bound to the ars1-SpORC beads. None of the added FLAG-Cdc18p bound to beads alone. However, a small, but significant, fraction of the FLAG-Cdc18p bound to ars1 DNA in the absence of SpORC. Thus, our data indicate that Cdc18 has some intrinsic capacity to bind to DNA by itself, but that the association of Cdc18 with origin DNA is greatly enhanced by SpORC. This finding is consistent with the hypothesis that SpORC mediates the recruitment of Cdc18p to origin DNA. Importantly, formation of the ternary complex of ars1, Cdc18, and SpORC did not require the presence of ATP (Fig. 7). It has been reported that the interaction of S. cerevisiae Cdc6p with ScOrc1p requires an intact ATP-binding site (36). However, it has also been shown that the Cdc6p containing alterations in either the Walker A or Walker B motifs binds efficiently to the complete six-subunit ScORC, suggesting that there may be Cdc6p-ScORC interactions that are independent of ATP (11). DISCUSSION To understand the initiation of DNA replication at the molecular level, it will be necessary to reconstitute the initiation complex with purified proteins. Toward this end we have purified recombinant S. pombe ORC and characterized two of its primary functions, origin binding and recruitment of Cdc18p. The SpORC was produced by co-expression of recombinant subunits in the baculovirus expression system. We have developed a simple affinity purification scheme to obtain SpORC from chromatin extracts of infected insect cells. This scheme relies on epitope tags placed at the N terminus of SpOrc4p where they do not interfere with SpORC function in vivo. We have obtained highly purified SpORC and demonstrated that it consists of a stable complex of six subunits analogous to ORCs of budding yeast, X. laevis, D. melanogaster, and Homo sapiens (6,14,15,17).
Except for the well studied case of S. cerevisiae, the nature of origins of DNA replication in eukaryotic cells remains poorly understood. Previous genetic studies have shown that S. pombe origins differ significantly from those of S. cerevisiae, suggesting that there may be substantial differences in the mechanisms of origin recognition employed by the two species (20,21,37). S. pombe origins are large, highly AT-rich sequences composed of functionally redundant elements. Importantly, they lack a common consensus sequence analogous to the ARS consensus sequence element recognized by ScORC. These properties have raised the question of whether origin recognition in fission yeast is less specific than in budding yeast. Our studies of the interaction of purified SpORC strongly suggest that this is the case. The binding of SpORC to ars1 DNA is governed by the SpOrc4 subunit, as demonstrated by the fact that the binding of the isolated SpOrc4 subunit is indistinguishable from that of the complete SpORC holo-complex. We have previously shown that the N-terminal domain of SpOrc4p, which contains nine AT-hook motifs, is necessary and sufficient for DNA binding by the isolated subunit (24), and the present study has shown that this domain is essential for viability of fission yeast. The binding of SpORC to ars1 origin DNA appears to have a relatively low level of sequence specificity. The protein is capable of binding to multiple sites within ars1 DNA with similar affinity. Although the protein binds with the highest affinity to AT-rich DNA, we also observed detectable binding to a control DNA with significantly lower AT content than that generally found in S. pombe origins. Our data are consistent with a model in which SpORC is targeted preferentially to long AT-rich regions within the S. pombe chromosomal DNA without regard to the specific nucleotide sequence of such regions. We expect that such regions would contain multiple and potentially overlapping binding sites for SpORC as in the case of ars1. One consequence of this model is that SpORC would be expected to be targeted preferentially to non-coding regions of the genome which have a higher average AT content than coding sequences. Obviously, our data do not rule out the possibility that the pattern of SpORC binding and initiation of DNA replication in vivo may be modulated by other factors such as chromatin organization and local transcriptional activity (38). Nevertheless, it is likely that origin recognition in S. pombe is significantly less constrained by primary nucleotide sequence than in S. cerevisiae. It is possible that the same is true for metazoans, but further work will be required to assess this possibility (18,19).
It has been demonstrated that binding of ScORC to origin FIG. 6. SpORC interacts with the Cdc18p replication regulator. SpORC was incubated with F7 anti-HA-agarose beads in the presence (control beads) or absence (SpORC beads) of an excess of HA peptide. The resulting beads were incubated with purified FLAG-Cdc18p, and the proteins bound to the beads were eluted with SDS. A, FLAG-Cdc18p, purified by affinity chromatography and analyzed by SDS-PAGE followed by silver staining. B, SDS-PAGE of eluate from SpORC beads and control beads analyzed by immunoblotting with antibodies specific for SpORC subunits. C, SDS-PAGE of eluate from SpORC beads and control beads analyzed by immunoblotting with anti-FLAG antibodies specific for Cdc18. FIG. 7. SpORC facilitates the binding of Cdc18p to origin DNA. SpORC was incubated with ars1 DNA conjugated to magnetic beads to form binary protein-DNA complexes as described in the legend to Fig. 2. The complexes were incubated with FLAG-Cdc18p in the presence or absence of ATP as indicated. Beads alone or ars1 beads lacking SpORC served as controls. The bound proteins were analyzed by SDS-PAGE followed by immunoblotting with anti-FLAG antibody to detect FLAG-Cdc18p.
DNA is strictly dependent upon binding of ATP to Orc1, and it seems likely that ATP binding and hydrolysis plays an important role in the assembly and function of the initiation complex (6,39). The binding of SpORC to ars1 DNA does not require ATP, but it is possible that there are additional interactions between SpORC and DNA that are dependent on ATP. Such interactions could be difficult to detect in the presence of the strong ATP-independent binding mediated by SpOrc4p, especially if they are weak or transient in nature. One function of the N-terminal domain of SpOrc4p may be to facilitate additional protein-DNA interactions by tethering the complex to origins, thereby increasing its local concentration in the vicinity of the DNA. While such interactions may not directly contribute to origin selection, they may be critical for the downstream steps in initiation of DNA replication. This possibility is currently under investigation.
Cdc6p/Cdc18p plays a key role in the initiation of S. pombe DNA replication and appears to be essential for the loading of the putative MCM helicase at origins (40,41). It has been assumed that Cdc18p is targeted to origins via interactions with ORC, but there have been few biochemical studies to probe this interaction directly. In vitro studies with X. laevis egg extracts have shown that ORC is required for the association of Xenopus Cdc6p with chromatin (7). More recently, biochemical studies with purified S. cerevisiae proteins have demonstrated interactions between Cdc6p and ScORC that alter the conformation of ScORC and modulate its DNA binding properties (11). Our data indicate that purified S. pombe ORC interacts specifically and efficiently with Cdc18p in the presence or absence of origin DNA. This finding differs from the results of studies in S. cerevisiae demonstrating that the Cdc6p-ScORC interaction is stabilized by the presence of origin DNA (11). This difference could be due to species variation or may simply reflect differences in experimental conditions, since special washing conditions were required to observe origin dependence of the Cdc6p-ORC interaction in S. cerevisiae, and the dependence was not observed in other studies (34,36). Neither the binary SpORC-Cdc18p complex nor the ternary SpORC-Cdc18p-ars1 complex requires ATP for formation. As noted above, the possible role of ATP in the interaction between S. cerevisiae Cdc6p and ScORC is not yet clear, although it has been demonstrated that mutations in the nucleotide-binding motifs of Cdc6p do not abolish the interaction in vitro (11).
It appears that Cdc18p has some affinity for ars1 DNA in the absence of SpORC. This observation may be related to the recent finding that S. cerevisiae Cdc6p has an intrinsic nonspecific DNA binding activity (42), but further work will be required to assess this possibility, since we cannot yet completely rule out the presence of a contaminating DNA-binding protein in our purified Cdc18p. In any case, the association of Cdc18p with ars1 DNA is greatly enhanced by the presence of SpORC, consistent with the hypothesis that ORC functions to recruit Cdc18p to origins via direct interactions. The ability to form ternary complexes of SpORC, Cdc18p, and ars1 in vitro should facilitate the further biochemical analysis of the assembly of initiation complexes.