Cdt1 Forms a Complex with the Minichromosome Maintenance Protein (MCM) and Activates Its Helicase Activity*

Mcm4/6/7 forms a complex possessing DNA helicase activity, suggesting that Mcm may be a central component for the replicative helicase. Although Cdt1 is known to be essential for loading of Mcm onto the chromatin, its precise role in pre-RC formation and replication initiation is unknown. Using purified proteins, we show that Cdt1 forms a complex with Mcm4/6/7, Mcm2/3/4/5/6/7, and Mcm2/4/6/7 in glycerol gradient fractionation through interaction with Mcm2 and Mcm4/6. In the glycerol gradient fractionation, Mcm4/6/7-Cdt1 forms a complex (speculated to be a (Mcm4/6/7)2-Cdt13 assembly) in the presence of ATP, which is significantly larger than the Mcm4/6/7-Cdt1 complex generated in its absence. Furthermore, DNA binding and helicase activities of Mcm4/6/7 are significantly stimulated by Cdt1 protein in vitro. We generd a Cdt1 mutant, which fails to stimulate DNA binding and helicase activities of4/6/7. This mutant Cdt1 showed reduced interaction with Mcm and is deficient in the formation of a high molecular weight complex with Mcm. Thus, a productive interaction between Cdt1 and MCM appears to be essential for efficient loading of MCM onto template DNA, as well as for the efficient unwinding reaction.

To license the origins for DNA replication, multiprotein complexes, called pre-replicative complexes (pre-RCs), 2 need to be formed. Cdc6 and Cdt1 are both required for pre-RC formation (1)(2)(3)(4). They associate with the chromatin in a manner dependent on the origin recognition complex (ORC) bound to the origins throughout the cell cycle and facilitate the loading of the hexameric Mcm2-7 complex onto chromatin at late mitosis (telophase) through early G 1 (5,6). Mcm2-7 are members of the AAA family proteins and share the conserved ATP binding domain (7,8). The Cdt1-interacting protein, Geminin, together with CDK activity, prevents the formation of pre-RC during S-G 2 phases, thereby assuring that the genome is replicated once and only once per cell cycle (9 -15).
In Escherichia coli, the initiator, DnaA, binds to oriC and unwinds the origin. The DnaB helicase complexed with ATPbound DnaC is then recruited to the origin to form a prepriming complex. DnaC, the DnaB-loading factor, forms a tight complex with DnaB protein at a 1:1 molar ratio in the presence of ATP (16 -19). Cryo-electron microscopy studies of the DnaC-DnaB complex have revealed that DnaC polypeptides in the forms of three dumbbell-shaped dimers interlock into one of the faces of the hexameric DnaB helicase (20). Upon loading onto the origin DNA, ATP bound to DnaC is hydrolyzed, resulting in release of DnaC and activation of the DnaB helicase (21).
In eukaryotes, Mcm4/6/7 complexes were found to possess DNA helicase activity in vitro, and it was proposed that MCM may play a role in unwinding of the duplex DNA and progression of the replication forks (22,23). All the six Mcm2-7 polypeptides contain highly conserved DNA-dependent ATPase motifs, and the presence of the intact ATPase motifs in all the six subunits is essential for viability (8,24). All six Mcm subunits are required for elongation of DNA chains in Saccharomyces cerevisiae, supporting the above proposal (5,25). Whereas all Mcm subunits are required for the elongation stage of DNA replication in vivo, DNA helicase activity has been identified only in the Mcm4/6/7 complex (23, 26 -29).
Cdt1 has been shown to be an essential component of the pre-RC, required for the recruitment of the Mcm proteins to origins. It interacts with Mcm (30) and binds both singlestranded and double-stranded DNA (31). In this report, we have purified and characterized mouse Mcm4/6/7 and Cdt1 proteins. We report here that Mcm4/6/7 forms a complex with multiple molecules of Cdt1 in the presence of ATP, and Cdt1 protein significantly stimulates the DNA binding and helicase activities of Mcm4/6/7. We generated a Cdt1 mutant deficient in the activation of Mcm4/6/7 helicase and DNA binding. Results described in this report suggest a role for Cdt1 protein not only in the loading of MCM onto chromatin but also in execution of its helicase activity.

Cloning of Wild-type and Mutant Forms of Cdt1 Gene into a Baculovirus Transfer Vector and an E. coli Expression
Vector-A DNA fragment containing the full-length mouse Cdt1 gene (Gene ID 67177, GI: 31982545) was cloned into the * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. □ S The on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. S1-S3. 1  complexes in infected High 5 cell lysates were purified with consecutive steps involving nickelagarose affinity chromatography, anti-FLAG M2 antibody-agarose affinity chromatography, and glycerol gradient sedimentation as previously described (27).
The highly purified recombinant Cdt1 protein was prepared from insect cells and E. coli as follows. To generate recombinant baculovirus expressing the Cdt1 protein, the recombinant bacmid was constructed and used for transfection of the insect cell, according to the manufacturer's protocol (BAC-TO-BAC baculovirus expression system, Invitrogen). For expression of the Cdt1 protein, High five cells were infected with recombinant baculoviruses and were collected at 60-h post-infection. The recombinant Cdt1 protein in the infected cell lysate was purified as follows. The infected insect cells were washed once in cold phosphate-buffered saline and then suspended in lysis buffer (1 ml per 1.5 ϫ 10 7 High 5 cells) consisting of 10 mM Tris-HCl, pH 7.5, 130 mM NaCl, 1% Triton X-100, 10 mM NaF, 10 mM sodium phosphate buffer, 10 mM Na 4 P 2 O 7 , and Complete protease inhibitor mixture (EDTA-free, Roche Applied Sciences). After incubation for 40 min on ice, insoluble materials were removed by centrifugation at 40,000 rpm (50.2 Ti rotor; Beckman) for 40 min at 4°C. To 1 volume of the clarified lysate, 1:10 volume of Ni-NTA (nitrilotriacetic acid)-agarose (Qiagen) pre-equilibrated with buffer A (50 mM sodium phosphate buffer, pH 6.0, 300 mM NaCl, 10% glycerol, 0.01% Triton X-100, and 0.1 mM phenylmethylsulfonyl fluoride) was added. After incubation at 4°C for 2 h on a rocking platform, the beads were collected and washed six times with 10 bed volumes of buffer A containing 10 mM imidazole. Bound proteins were eluted stepwise in four bed volumes of buffer A containing 200 mM imidazole. The eluted fractions containing Cdt1 was mixed with anti-FLAG M2 Ab-agarose beads (Sigma) equilibrated with buffer A (same as above except that the pH was adjusted at 7.5), and incubated at 4°C for 1 h on a rocking platform. After washing four times with 10 bed volumes of buffer A (pH 7.5), the Cdt1 protein was eluted stepwise with three bed volumes of buffer A (pH 7.5) containing 0.2 mg/ml of the FLAG peptide. Further purification of the Cdt1 protein was carried out by 15-35% or 15-30% linear glycerol gradient (2 ml) centrifugation at 38,000 rpm for 17 h (TLS55 rotor; Beckman). The buffers used in the gradient was 0.15 M NaCl, 20 mM Tris-HCl, pH 7.5, 0.5 mM EDTA, 1 mM DTT, 0.1 mM phenylmethylsulfonyl fluoride, and 0.01% Triton X-100. Fractions (130 l each) were collected and were analyzed by SDS-PAGE, followed by staining with silver.
His-FLAG or T7-tagged Cdt1 was overproduced in the E. coli strain BL21 (DE3) codon plus grown in LB medium with 30 g/ml kanamycin and 50 g/ml chloroamphenicol. Isopropyl-␤-D-thiogalactopyranoside was added to a final concentration of 0.5 mM, and the cells were incubated for an additional 2 h at 28°C. Cells were suspended in lysis buffer and incubated at 4°C for 30 min on a rocking platform after addition of 0.2 mg/ml lysozyme. The suspension was sonicated to reduce viscosity. The procedures for purification using the NTA column, FLAG antibody resin, and glycerol gradient centrifugation were the same as described above. After NTA and FLAG antibody purification, the elute was applied to a Mono Q PC 1.6/5 SMART column (GE Healthcare LifeScience) equilibrated with Buffer Q (25 mM Tris-HCl, pH 7.5, 10% glycerol, 0.5 mM EDTA, 1 mM DTT, and 0.01% Triton X-100) with 60 mM NaCl. The elution was carried out with Buffer Q containing a linear gradient from 0 to 1 M NaCl. Further purification was performed using T7-tag antibody-agarose beads according to the manufacturer's instructions (Novagen). Finally, glycerol gradient centrifugation was carried out.
Reagents-Labeled and unlabeled dNTPs/rNTPs were purchased from GE Healthcare LifeScience. M13mp18 singlestranded circular DNA (ssDNA) and T4 polynucleotide kinase were from TAKARA, and anti-FLAG M2 antibody-agarose and FLAG peptide were from Sigma. Oligonucleotides were purchased from Hokkaido System Science Co., Ltd. (Hokkaido, Japan), and were purified on polyacrylamide gel, followed by purification by Sep-Pak Plus C18 cartridge column (Waters).
Construction of Partial Heteroduplex Substrates for Helicase Assays-The oligonucleotide [5Ј-GTTTTCCCAGTCACGAC-GTTGTAAAACGACGGCCAGT-(dT) 40 -3Ј] (carrying the 40-mer oligo(dT) tail at the 3Ј-end of the hybridizing 37-mer sequence) was first labeled with [␥-32 P]ATP and T 4 polynucleotide kinase. The labeled dT 40 -37-mer (37-mer region hybridizing at positions 6289 -6326 of M13mp18 DNA) was annealed to the single-stranded circular M13mp18 DNA in a reaction mixture containing 20 mM Tris-HCl, pH 7.5, and 10 mM MgCl 2 , which were heated to 95°C, kept at 67°C for 1 h, and then allowed to slowly cool down to 37°C. The labeled substrates were purified by Sepharose CL4B column chromatography (GE Healthcare LifeScience).
Gel Shift and DNA Helicase Assays-In gel shift assays, the 77-mer oligonucleotide [5Ј-GTTTTCCCAGTCACGACGTT-GTAAAACGACGGCCAGT-(dT) 40 -3Ј], the 5Ј-terminus of which was labeled with [␥-32 P]ATP and T 4 polynucleotide kinase, was used as a substrate. Mcm4/6/7 protein was incubated at 30°C for 30 min in reaction mixtures (15 l) containing 25 mM Hepes-NaOH, pH 7.5 or 25 mM Tris-HCl, pH 7.5, 50 mM sodium acetate, 10 mM Mg(CH 3 COO) 2 , 20 mM 2-mercap-toethanol, 0.25 mg/ml bovine serum albumin, 10 mM ATP, and labeled substrates of the amount indicated. After addition of 2 l of 50% glycerol, the reaction mixtures were directly applied on a nondenaturing polyacrylamide gel in 1ϫ Tris borate-EDTA (TBE), and electrophoresis was conducted at room temperature. For DNA helicase assays, the reaction mixtures were incubated at 37°C for 1 h. The reactions were terminated by addition of EDTA (20 mM), SDS (0.1%), and 2 g of proteinase K, and were incubated for an additional 15 min. The samples were separated by electrophoresis on a nondenaturing polyacrylamide gel in 1ϫ TBE.
Immunoprecipitation-Immunoprecipitation experiments were performed using T7-antibody beads (Novengen). Pre-washed T7-antibody beads were mixed with indicated proteins in buffer containing 10 mM Hepes-NaOH, pH 7.5, 100 mM sodium acetate, 2 mM Mg(CH 3 COO) 2 , 2 mM ATP, 10% glycerol, and 0.01% Triton X. After 1 h of incubation, beads were washed with the above buffer, and bound proteins were eluted with 0.1 M citric acid, pH 2.2. The samples were run on SDS-PAGE, followed by silver staining.

Mcm2/3/4/5/6/7, Mcm4/6/7, and Mcm2/4/6/7 Form Complexes with
Cdt1-During the loading and translocation on the template DNA, the Mcm helicase is expected to interact with many factors including Cdt1, a loading factor for Mcm. We examined the interaction between Cdt1 and various forms of Mcm complexes by glycerol gradient centrifugation. As shown in Fig. 1A, the purified Cdt1 protein sedimented predominantly in the fractions 12-13, at a molecular weight of about 70,000. The Mcm4/6/7 complex sedimented at around fractions 5-6, expected for a dimer of two trimers as previously reported (Fig. 1B) (23). When Cdt1 was mixed with Mcm4/6/7, the peak moved to around fraction 4 corresponding to a molecular mass of 670 kDa (Fig. 1C). The Mcm2/3/4/5/6/7 or Mcm2/4/6/7 complex also cosedimented with a portion of Cdt1 in glycerol gradient centrifugation (Fig. 1, D and E).    (23,31). When Cdt1 protein alone was incubated with ssDNA, it generated smeared shifted bands in electrophoretic mobility shift assays on polyacrylamide gel ( Fig. 2A, lane  16). On the other hand, Mcm4/6/7 alone incubated with ssDNA formed a single sharp mobilityshifted band along with fast-migrating smeared bands, which may have arisen from dissociation of the complex during the electrophoresis ( Fig.  2A, lane 2). When Cdt1 was added to the reaction mixtures containing a fixed amount of Mcm4/6/7, the mobility of the complex was retarded, indicative of formation of a trimolecular complex containing Mcm4/6/7, Cdt1, and DNA. The amount of this complex increased, as more Cdt1 protein was added, suggesting a possibility that Cdt1 promotes binding of the Mcm complex to DNA. Quantification indicated that the stimulation was as much as 7-fold ( Fig. 2A, right). Cdt1 did not significantly affect DNA binding of T-antigen helicase under the identical condition ( Fig. 2A,  right). Similar results were obtained in DNA binding assays using Cdt1 purified from E. coli (Fig. 2C).
We also conducted pull-down assays using a biotin-labeled ssDNA to examine the effect of Cdt1     Fig. 2A) were examined for DNA helicase activities (Fig. 2B). Addition of increasing amounts of Cdt1 significantly stimulated the helicase activity of Mcm4/6/7, whereas that of T-antigen helicase was not affected under the same condition (Fig. 2B, lower). Cdt1 stimulated the helicase activity of Mcm4/6/7 as much as 5-fold (lanes 7 or 8 in Fig. 2B). The isolated Mcm4/6/7-Cdt1 complex exhibited more efficient DNA binding and helicase activities than Cdt1-free Mcm4/6/7 complex (supplemental Fig. S2). However, addition of excess amount of Cdt1 led to inhibition of the helicase activity (Fig. 2D). Titration of Cdt1 indicated that addition of more than 18 molecules of the Cdt1 monomer per Mcm4/6/7 hexamer resulted in inhibition. Excess Cdt1 may compete with Mcm for binding to the substrate DNA and titrate out the effective Mcm4/6/7 helicase. These behaviors of Cdt1 are similar to the known effect of DnaC on DnaB helicase activity (21,32,33).
Geminin binds to Cdt1 and inhibits its activity. We purified the wild-type geminin and a mutant geminin lacking the coiledcoiled domain. This mutant geminin does not interact with Cdt1 protein, as judged from immunoprecipitation assays (data not shown). Fig. 3A shows that addition of the wild-type geminin to the helicase reaction in the presence of Cdt1 resulted in strong inhibition of the helicase activity, whereas the mutant geminin did not affect the increase of helicase activity by Cdt1. Because geminin alone did not affect the helicase action of Mcm4/6/7 at the same concentrations (data not shown), almost complete inhibition by geminin (lanes 6 and 7) may be due to the inhibitory effect of the geminin-Cdt1 complex on Mcm helicase. In DNA binding assays, the wild-type geminin, when added to the Cdt1-DNA complex, formed a new complex migrating at a distinct position which represents the Cdt1geminin-DNA ternary complex (Fig. 3B, lanes 5 and 6). When added to the Mcm-Cdt1-DNA complex, it disrupted the slowmigrating complex, regenerating the Mcm-DNA complex (Fig.  3B, lanes 9 and 10). In contrast, the mutant geminin did not affect the Mcm-Cdt1-DNA complex even at a high concentration (Fig. 3B, lanes 11 and 12), in keeping with the preservation of helicase activity under the same conditions. These results clearly indicate that the stimulation of the helicase activity was due to the action of Cdt1 protein. Although Cdt1 stimulated DNA binding activity of Mcm2/3/4/5/6/7 hexamer on singlestranded DNA, the Mcm heterohexamer did not show any DNA helicase activity even in the presence of Cdt1 protein (data not shown).
The above results point to functional similarity between Cdt1 and DnaC proteins, the latter of which binds to DnaB helicase and stimulates its helicase activity. In light of critical roles of ATP in regulation of the DnaC-DnaB complex, we then examined the effect of ATP concentration on the DNA binding and helicase activities of Mcm4/6/7 in the absence and presence of Cdt1 protein (Fig. 4A). The optimum ATP concentration for the helicase reaction in the presence of Cdt1 was 10 mM, but DNA binding of Mcm4/6/7 did not require ATP either in the presence or absence of Cdt1 (Fig. 4A, lower). In contrast, DNA binding was inhibited, as more ATP was added (Fig. 4B).    (Fig. 5A). The peak fractions corresponded to the protein marker thyroglobulin with an estimated molecular mass of about 650 kDa. This value is larger than the Mcm4/6/7 protein alone in the presence of ATP, which sediments at ϳ450 kDa (Fig. 5E). We previously reported that the Mcm4/6/7 in the absence of ATP sedimented at ϳ350 kDa in glycerol gradient and migrates at 550 kDa in native acrylamide gel (22,23). This was confirmed in Fig. 5F. Meanwhile, Cdt1 protein alone peaked at 70 kDa (fraction 12) as expected for a monomer under this condition (Fig.   5G). In the absence of ATP, on the other hand, the Mcm4/6/7-Cdt1 complex sedimented in a rather broad peak at around 350 kDa, smaller than the Mcm-Cdt1 complex in the presence of ATP (Fig. 5B). The Mcm4/6/7-Cdt1 complex formation was confirmed also in the regular native gel or blue native polyacrylamide gel (BN-PAGE, data not shown).
The ratio of the protein amount of Mcm4/6/7 and Cdt1 was 3:1 in the peak fractions (fraction 2) of the glycerol gradient in Fig. 5A. Taking the molecular masses of Cdt1 (70 kDa) and Mcm4/6/7 (300 kDa) into account, a molar ratio of Cdt1 to Mcm is estimated to be ϳ3:2. Thus, in the presence of ATP, about three Cdt1 molecules may bind to the two Mcm4/6/7 trimers (hexamers). In contrast, the Mcm4/6/7-Cdt1 complexes in the absence of ATP was more heterogeneous in size, and the amount of Cdt1 associated with Mcm4/6/7 was lower than the complex in the presence of ATP (Fig. 5B). These data suggest that the Mcm4/6/7-Cdt1 complex exists as a mixture of different forms with different numbers of bound Cdt1 molecules, and that a large complex containing three Cdt1 molecules may be preferentially generated in the presence of ATP.
A Mutant Cdt1 Protein Fails to Activate Mcm4/6/7 Helicase-To identify residues critical for stimulation of MCM helicase, we generated a mutant Cdt1 (KKAA) in which two lysine residues in the 441 KXXXK 445 were change to alanine (Fig. 6A). The Lys 441 residue is conserved from fission yeast to human. We purified the mutant protein after overexpression in insect cells or in E. coli.
We next examined the activity of the mutant protein in DNA binding and helicase assays of Mcm4/6/7. The KKAA mutant showed very little stimulatory effect on the helicase activity of Mcm4/6/7 and only slightly stimulated its DNA binding activity (Fig. 6B). The slow-migrating bands representing the Mcm-Cdt1-DNA complex were not detected and the smeared bands representing Cdt1-DNA increased, suggesting that the KKAA mutant Cdt1 fails to generate a complex with Mcm (Fig. 6B, lower, lane 10). Kinetics of helicase assays with the identical amounts of the wild-type or mutant Cdt1 protein also indicated very little or no stimulatory effect of the mutant protein on the Mcm helicase activity (Fig. 6C). We have also conducted helicase and DNA binding assays with the wild-type and mutant recombinant Cdt1 protein generated in E. coli, and obtained similar results (supplemental Fig. S3). This indicates that the reduced activity of the mutant protein is not due to differential modification (such as phosphorylation).
To clarify the nature of the mutant protein, we examined DNA binding activity and protein-protein interactions. DNA binding activity of the mutant protein in a pull-down assay with biotinylated DNA was comparable to that of the wild-type protein (Fig. 6D). Interaction between T7-tagged Cdt1 and Mcm4/ 6/7 complex was examined with immunoprecipitation with anti-T7-tag antibody. The mutant Cdt1 was found to interact with Mcm4/6/7 with reduced affinity (Fig. 6E), consistent with the results of gel shift assays (Fig. 6B). In the control, binding of the mutant Cdt1 protein to geminin was not affected (Fig. 6F). These results indicate that the KKAA mutant interacts with Mcm with reduced affinity and fails to stimulate DNA binding and helicase activities of Mcm4/6/7. The mutant Cdt1 did not generate the high molecular weight complex as efficiently as the wild-type did even in the presence of ATP (Fig. 5, C and D), and the amount of the cosedimenting Cdt1 polypeptide was significantly reduced, reflecting its reduced affinity to Mcm. The mutant Cdt1 alone behaved as a monomer, as the wild-type did (Fig. 5H). We also detected the Mcm-Cdt1 complex in native polyacrylamide gel electrophoresis, but this complex was not detected with the KKAA mutant protein (data not shown). These results strongly indicate that the mutant Cdt1 fails to generate a helicase-active high molecular weight Mcm-Cdt1 complex because of its reduced affinity to Mcm.

DISCUSSION
DNA helicase plays a central role during the course of DNA replication and loading of the replicative helicase onto a template DNA is a crucial event for initiation of DNA replication. In E. coli, DnaB is the replicative helicase, and DnaC forms a heterohexamer with DnaB (DnaC 6 -DnaB 6 ) to deliver the DnaB helicase onto the unwound single-stranded DNA at the origin (17,34) On the other hand, Mcm4/6/7 helicase was inhibited in the presence of excess amount of Cdt1. Careful titration indicated that the presence of more than 18-fold molar excess Cdt1 molecule per one Mcm4/6/7 hexamer leads to inhibition. Results from glycerol gradient analyses were also consistent with the formation of (Mcm4/6/7) 2 Cdt1 3 complex in the presence of ATP. At the fork, the Mcm4/6/7 protein forms a double hexamer (27,29), and a dimer of (Mcm4/6/7) 2 Cdt1 3 complex may  be generated. Excess free Cdt1 may inhibit the helicase reaction by directly binding to substrate DNA, thus competing out Mcm4/6/7. Consistent with the inhibitory effect of Cdt1 on duplex unwinding, Cdt1 is rapidly released from chromatin once S phase is initiated (36).
Similarity between Cdt1 and DnaC-DnaC of E. coli is a DnaB-loading protein and utilizes ATP binding to assemble the DnaB helicase at the origin. In the presence of ATP, a maximum of six DnaC monomers bind cooperatively to the DnaB hexamer to generate a DnaC 6 -DnaB 6 heterohexamer (16,17,19). Once DnaB is delivered to the origin, DnaC is released from both DnaB and the origin (34,38). DnaC stimulates DnaB helicase activity, and excess DnaC inhibits DnaB helicase (21,37). Together with the fact that ATP can be UV-cross-linked to Cdt1 protein in vitro, 3 these point to a significant similarity between the Cdt1 and DnaC proteins. On the other hand, there is very little similarity on the amino acid level between the two proteins. In E. coli, DNA-dependent ATPase activity of DnaB was reported to be unaffected by dnaC (16), but more recently was reported to be activated by DnaC (37). DNA-dependent ATPase activity of Mcm4/6/7 protein was not affected by Cdt1 protein, and Cdt1 alone exhibited very little or no ATPase activity (data not shown).
Another protein, Cdc6, an ATP-binding protein, is also required for assembly of the Mcm helicase at the origin in vivo. Cdc6 is a member of the AAAϩ family of ATPases, shows sequence similarity to DnaC and was thought to be analogous to the DnaC helicase loader (5,39). It is known that both Cdc6 and Cdt1 proteins are required for association of Mcm with replication origins in vivo (5). Cdt1 interacts with Cdc6 protein in vitro (40). 3 Thus, it is possible that the combined actions of Cdt1 and Cdc6 are required for loading of Mcm, as in the case of Bacillus subtilis DnaB and DnaI, which are a pair of helicase loaders required for the loading of the Bacillus replicative helicase DnaC (41). However, Cdc6 shares strong sequence similarity with Orc1 (42,43), suggesting that Cdc6 may be a part of the ORC initiator assembly on which pre-RC would be assembled. In fact, Stillman's group recently showed in a single-particle reconstruction of electron microscopic images that the ORC-Cdc6 complex forms a ring-shaped structure with dimensions similar to those of the ring-shaped MCM helicase. Thus, Cdc6 could be regarded as an additional subunit of the ORC 3 Z. You   . D, pull-down assay was carried out with 50 ng of Cdt1 (wild type or KKAA) and 50 pmol of biotinylated oligonucleotide in the presence or absence of ATP. The bound (left half) and unbound (right half) proteins are shown. * indicates nonspecific bands. E, purified Mcm4/6/7 (200 ng) was mixed with the same amount (200 ng) of wild-type or KKAA T7-tagged Cdt1, and immunoprecipitation experiments were performed using T7-antibody beads as described under "Experimental Procedures," except that beads were further washed twice with T7-wash buffer (4.29 mM Na 2 HPO 4 , 1.47 mM KH 2 PO 4 , 2.7 mM KCl, 137 mM NaCl, and 0.1% Tween-20, pH 7.3). F, purified geminin (200 ng) was mixed with same amount (200 ng) of wild-type or KKAA T7-tagged Cdt1, and immunoprecipitation experiments were performed using T7-antibody beads as described under "Experimental Procedures." In D, E, and F, the samples were run on 4 -20% SDS-PAGE, and the proteins were detected by silver staining. In E and F: I, input; U, unbound; B, bound. Cdt1 was overexpressed and purified from insect cells (B and C) or from E. coli (D, E, and F). The KKAA mutant did not promote the high molecular weight complex formation with Mcm at a higher concentration and did not stimulate the helicase activity of Mcm4/6/7. The interaction of the mutant Cdt1 with Mcm was significantly reduced, although it interacted with DNA and with geminin as efficiently as wild-type.
complex (43,44). In contrast, the Cdt1 forms a stable complex with Mcm proteins in vivo (30,45) and in vitro (this study). Taken together, we would like to propose the possibility that Cdt1 may be the functional homolog of DnaC protein in E. coli. Future studies on the structure of the Mcm-Cdt1 complex as well as its interaction with ATP would be needed to test this hypothesis.