The Cdc45·Mcm2–7·GINS Protein Complex in Trypanosomes Regulates DNA Replication and Interacts with Two Orc1-like Proteins in the Origin Recognition Complex

Abstract

Accurate DNA replication requires a complex interplay of many regulatory proteins at replication origins. The CMG (Cdc45·Mcm2–7·GINS) complex, which is composed of Cdc45, Mcm2–7, and the GINS (Go-Ichi-Ni-San) complex consisting of Sld5 and Psf1 to Psf3, is recruited by Cdc6 and Cdt1 onto origins bound by the heterohexameric origin recognition complex (ORC) and functions as a replicative helicase. Trypanosoma brucei, an early branched microbial eukaryote, appears to express an archaea-like ORC consisting of a single Orc1/Cdc6-like protein. However, unlike archaea, trypanosomes possess components of the eukaryote-like CMG complex, but whether they form an active helicase complex, associate with the ORC, and regulate DNA replication remains unknown. Here, we demonstrated that the CMG complex is formed in vivo in trypanosomes and that Mcm2–7 helicase activity is activated by the association with Cdc45 and the GINS complex in vitro. Mcm2–7 and GINS proteins are confined to the nucleus throughout the cell cycle, whereas Cdc45 is exported out of the nucleus after DNA replication, indicating that nuclear exclusion of Cdc45 constitutes one mechanism for preventing DNA re-replication in trypanosomes. With the exception of Mcm4, Mcm6, and Psf1, knockdown of individual CMG genes inhibits DNA replication and cell proliferation. Finally, we identified a novel Orc1-like protein, Orc1b, as an additional component of the ORC and showed that both Orc1b and Orc1/Cdc6 associate with Mcm2–7 via interactions with Mcm3. All together, we identified the Cdc45·Mcm2–7·GINS complex as the replicative helicase that interacts with two Orc1-like proteins in the unusual origin recognition complex in trypanosomes.

Introduction

Maintenance of genomic stability requires high fidelity chromosomal DNA replication. To ensure that chromosomal DNA is replicated only once per cell cycle, eukaryotes have evolved a delicate replication licensing system that promotes the assembly of the minichromosome maintenance (Mcm)² protein complex at replication origins during late mitosis and G₁ phase but prevents DNA re-replication after S phase in a single cell cycle (1, 2). The replication licensing system is well understood in budding yeast and is likely conserved throughout evolution (1). During G₁ phase of the cell cycle, replication origins are bound by the well conserved origin recognition complex (ORC) consisting of six related AAA+ ATPases, Orc1–Orc6. ORC is then bound by Cdc6, another AAA+ ATPase, which subsequently recruits Cdt1 and Mcm2–7 onto origins to form the pre-replicative complex (pre-RC). Assembly of the pre-RC, known as replication licensing, occurs only during G₁ phase and is tightly regulated (2).

After the pre-RC is assembled on replication origins, initiation of DNA replication is triggered by S-phase cyclin-dependent kinase (S-CDK) and Dbf4-dependent kinase Cdc7 (3). Phosphorylation of Sld2 and Sld3 by S-CDK (4, 5) generates binding sites for Dpb11, a BRCA1 C terminus domain-containing protein (6). Sld2 is part of the so-called pre-loading complex (pre-LC) that includes Dpb11, the GINS complex consisting of Sld5 and Psf1 to Psf3 (7–9), and the leading strand polymerase pol ϵ (10). Sld3 associates closely with Cdc45, and the latter is recruited onto origins via interactions with the Mcm2–7 complex (11–13). Dpb11 apparently bridges Sld3·Cdc45 with Sld2 and thus recruits the pre-LC to origins (3). Dbf4·Cdc7 phosphorylates the N-terminal tails of several subunits in the Mcm2–7 complex, which promotes the formation of the Cdc45·Mcm2–7 complex by alleviating an inhibitory activity in the N-terminal serine/threonine-rich domain of Mcm4 (14). Cdc45 is also an in vitro target of Dbf4·Cdc7, but the in vivo effect of Cdc45 phosphorylation remains to be addressed (15). Once the pre-LC is loaded onto origins, Cdc6 and/or Cdt1 are phosphorylated by S-CDK and degraded by the 26 S proteasome, whereas Dpb11, Sld2, and Sld3 are likely dissociated from the replisome (3). The Mcm2–7 complex then forms the so-called CMG complex with Cdc45 and the GINS complex, which moves with replication forks and likely functions as the replicative helicase to unwind duplex DNA (16–19). Although there are distinctions in the regulation of DNA replication between simple and complex eukaryotic organisms, the fundamental mechanism of DNA replication is believed to be well conserved among all eukaryotes.

The archaeal DNA replication machinery bears striking similarity to that of eukaryotes, but in contrast to eukaryotes, most archaeal organisms contain only one Mcm protein that forms a homohexameric ring, a few (1–3) Orc1/Cdc6-like proteins (20), and two GINS-like proteins, Gins51 and Gins23, both of which are poorly conserved homologs of eukaryotic GINS proteins (21). Gins51 resembles Sld5 and Psf1 and possesses a conserved A-domain (α-helix) in the N terminus and a B-domain (β-strands) in the C terminus, whereas Gins23 exhibits sequence identity to Psf2 and Psf3 and possesses the A- and B-domains in the reverse order (22). Orc1/Cdc6 is able to bind to the origin of replication and recruits the Mcm homohexamer to the origin (23). Most archaeal species apparently lack homologs of Cdt1, Cdc45, CDK, Dbf4, and Cdc7, and therefore, regulation of replication initiation in archaea may be less complicated than that in eukaryotes.

Regulation of DNA replication in Trypanosoma brucei, an early branched unicellular eukaryote and a parasitic protozoan that causes sleeping sickness in humans, appears to be different from that in archaea and other eukaryotes. Trypanosomes express a single Orc1/Cdc6-like protein that associates with chromatins throughout the cell cycle and is essential for DNA replication (24), indicating the presence of an archaea-like ORC in trypanosomes. However, the trypanosome genome encodes six distinct Mcm proteins (Mcm2–7), implying the potential formation of a eukaryote-like heterohexameric Mcm2–7 complex. Similar to most archaeal organisms, trypanosomes do not express a homolog of Cdt1.³ Further, among the 10 cyclins and 11 Cdc2-related kinases (CRKs) identified in the trypanosome genome (25), none of them functions as bona fide S-phase cyclin and CDK (26–30). Moreover, no homologs of Cdc7 and its partner, Dbf4, as well as Sld2, Sld3, and Dpb11, were identified in the trypanosome genome.³ These observations suggest that regulation of DNA replication initiation in trypanosomes is likely through a mechanism distinct from that in most eukaryotes. It is therefore of paramount interest to dissect the mechanism of DNA replication initiation in trypanosomes because the outcome from these studies might shed novel light on the evolution of the regulatory machinery of DNA replication from simple unicellular eukaryotes to complex multicellular organisms.

In this study, we characterized the CMG complex and explored its association with the origin recognition complex as the first step of our long term goal aimed at delineating the regulatory pathway that controls DNA replication in T. brucei. We identified all six components of the Mcm2–7 complex, all four components of the GINS complex, and Cdc45 and demonstrated the formation of the CMG complex in vitro and in vivo. Further, we showed that the CMG complex, but not the Mcm2–7 complex, possesses in vitro DNA helicase activity and is essential for DNA replication. Finally, we identified a novel Orc1-like protein, Orc1b, and established its interactions with Orc1/Cdc6 and Mcm3. These findings identified the Cdc45·Mcm2–7·GINS complex as an essential replicative helicase and an unusual origin recognition complex containing two Orc1-like proteins, Orc1/Cdc6 and Orc1b, in trypanosomes.

EXPERIMENTAL PROCEDURES

Trypanosome Cell Culture

The procyclic form of T. brucei strain 427 was cultured at 27 °C in SDM-79 medium supplemented with 10% fetal bovine serum (Atlanta Biologicals, Inc). Procyclic 29-13 cell line (31) was cultivated in SDM-79 medium containing 10% fetal bovine serum, 15 μg/ml G418 (Clontech), and 50 μg/ml hygromycin B (Invitrogen). Cells were routinely diluted once the density reached 5 × 10⁶/ml.

RNA Interference

A 400–500-bp sequence from the N-terminal portion of each of the 11 subunits of the CMG complex was PCR-amplified from genomic DNA and cloned into the pZJM vector (32). The resulting constructs were linearized by restriction digestion with NotI and electroporated into the 29-13 cell line according to our previous procedures (33, 34). Successful transfectants were selected with 2.5 μg/ml phleomycin and cloned by limiting dilution. To induce RNAi, 1.0 μg/ml tetracycline was added to the culture medium, and cell growth was monitored daily by counting the cell number with a hemocytometer.

Epitope Tagging of Endogenous Proteins in the Procyclic Form of T. brucei

A C-terminal fragment of each of the 11 CMG subunits was cloned into the pC-EYFP-Neo vector, which was obtained by replacing the PTP module in the pC-PTP-Neo vector (35) with the enhanced yellow fluorescence protein (EYFP) and into the pC-3HA-Bla vector. The resulting constructs were transfected into the wild-type 427 cell line. Correct in situ tagging of one of the two alleles was confirmed by PCR and subsequent sequencing. Stable transfectants were selected under 40 μg/ml G418 or 10 μg/ml blasticidin and cloned by limiting dilution.

Purification of GST-His Fusion Proteins for in Vitro DNA Helicase Activity Assay

Full-length coding sequences of Mcm2–7, Cdc45, and GINS proteins were each cloned into a modified pGEX-4T-3 vector by adding a His₇ tag at the C terminus of each protein in addition to the N-terminal GST tag. A factor Xa recognition site was introduced upstream of the histidine tag. Mutation of the conserved lysine residue in the Walker A motif of individual Mcm proteins was performed by site-directed mutagenesis (Stratagene). Recombinant proteins were expressed in Escherichia coli BL21 Rosetta cells and purified through the column of nickel-agarose and subsequently the column of glutathione-Sepharose 4B beads. Purified recombinant proteins were dialyzed against 50 mm Tris-Cl and 50 mm NaCl. For reconstitution of the CMG complex, both the GST tag and the His₇ tag were removed by thrombin protease and factor Xa (Sigma-Aldrich) from each of the CMG proteins with the exception of Mcm2 in which the His tag was retained for Western blot analysis after glycerol gradient centrifugation.

Glycerol Gradient Sedimentation Analysis

A mixture of the CMG proteins (100 ng for each protein) was loaded on top of 4 ml of 15–35% glycerol gradient and run for 12 h on a Beckman Coulter Optima TL ultracentrifuge at 4 °C. After completion of the centrifugation, 200-μl fractions were collected from the top of the tube and analyzed for the protein content with anti-His antibody to detect His-tagged Mcm2.

Substrate Preparation for DNA Helicase Activity Assay

The DNA helicase substrate used in this study is a partial DNA duplex consisting of a γ-³²P-labeled DNA oligonucleotide (30-mer) annealed to the M13mp18 single-stranded (ss) DNA. The 30-mer oligonucleotide (TTTTTTAGGATCCCCGGGTACCGATTTTTT) with a six T-base overhang at both ends was labeled at the 5′-end with [γ-³²P]ATP by T4 polynucleotide kinase. Subsequently, the labeled oligonucleotide was incubated with 2 μg of M13mp18 ssDNA in annealing buffer (40 mm Tris-HCl, pH 7.5, 10 mm MgCl₂, 50 mm NaCl, and 1 mm DTT). The mixture was heated at 95 °C for 5 min and allowed to cool down slowly at room temperature. The substrate was then purified by gel filtration through a Sepharose 4B column. Fractions with the highest radioactivity were pooled and used for in vitro helicase activity assay.

In Vitro DNA Helicase Activity Assay

In vitro DNA helicase assay measures the displacement of ³²P-labeled DNA oligonucleotide from the partial DNA duplex catalyzed by trypanosome Mcm proteins and the CMG complex. The assay was performed in a 10-μl reaction mixture consisting of 50 mm Tris-HCl, pH 8.0, 1 mm ATP, 1.5 mm MgCl₂, 50 mm KCl, 8 mm DTT, 4% (w/v) sucrose, 80 μg/ml BSA, ∼1 ng of ³²P-labeled substrate (2,000–3,000 cpm), and purified recombinant proteins. The reaction mixture was incubated at 37 °C for 1 h and then separated on a 12% native polyacrylamide gel before exposing the dried gels to x-ray films.

Northern Blot

Total RNA was purified from T. brucei cells with the TRIzol reagent (Invitrogen). Northern blot was performed as described previously (36). Briefly, total RNA was separated on a formaldehyde-agarose gel and transferred onto a nitrocellulose membrane in 20× SSC (150 mm NaCl and 0.15 mm sodium citrate). Northern hybridization was carried out overnight at 42 °C in 50% formamide, 6× SSC, 0.5% SDS, and 5× Denhardt's solution with 0.1 mg/ml salmon sperm DNA. Blots were washed in 2× SSC and 0.1% SDS for 30 min, 1× SSC and 0.1% SDS for 30 min, and 0.5× SSC and 0.1% SDS for 30 min and then exposed to x-ray films.

GST Pulldown and Western Blot

Mcm2–7, GINS proteins, Orc1/Cdc6, and Orc1b were each fused with the GST tag at the N terminus and a His₇ tag at the C terminus, expressed in E. coli BL21 Rosetta strain, and purified through the column of nickel-agarose. The eluate was incubated with glutathione-Sepharose 4B beads, and after a thorough wash, the beads were incubated at room temperature for 30 min with T. brucei lysate expressing HA-tagged proteins. T. brucei lysate was prepared by incubating the cells with trypanosome immunoprecipitation buffer (25 mm Tris-Cl, pH 7.6, 100 mm NaCl, 1 mm DTT, 1% Nonidet P-40, and protease inhibitor mixture) on ice for 30 min and clearing by centrifugation (34). The beads were then washed five times with immunoprecipitation buffer, and proteins were eluted by boiling the beads in 1× SDS-PAGE sampling buffer. Eluted proteins were then fractionated in SDS-PAGE, transferred onto a PVDF membrane, and immunoblotted with anti-HA mAb as described (37). GST was also purified from E. coli and incubated with T. brucei lysate to serve as the negative control. After ECL development, the blots were stained with Coomassie Blue to detect purified recombinant GST fusion proteins for loading control.

BrdU Incorporation and Detection

BrdU incorporation and detection were carried out according to a previously published protocol (29). Briefly, control and RNAi-induced cells (3 or 5 days) were incubated with 20 mm BrdU and incubated for 6 h. Cells were harvested, fixed in ethanol, and incubated with anti-BrdU antibody (Sigma-Aldrich; diluted 1:400) at room temperature for 1 h. After three washes, cells were incubated with FITC-conjugated goat anti-mouse IgG (diluted 1:400) at room temperature for another hour. Cells were then washed three more times, mounted in VECTASHIELD mounting medium containing DAPI (Vector Laboratories), and examined under a fluorescence microscope.

Fluorescence Microscopy and Immunofluorescence Microscopy

Procyclic cells expressing endogenously EYFP-tagged Mcm2–7 and GINS proteins were harvested by centrifugation at 2,000 rpm for 5 min, washed once in PBS, and fixed in 4% paraformaldehyde. The fixed cells were washed with PBS, re-suspended in PBS, and adhered on poly-l-lysine-treated coverslips. The slides were mounted in VECTASHIELD mounting medium containing DAPI and examined under a fluorescence microscope according to our previous procedures (38).

Cells expressing HA-tagged Cdc45 were cultured in the absence or presence of 0.1 μg/ml leptomycin B for 4 and 8 h. Cells were harvested by centrifugation, washed once in PBS, and fixed in 4% paraformaldehyde. The fixed cells were washed with PBS, suspended in PBS, and adhered to poly-l-lysine-treated coverslips. Cells were treated with blocking buffer (1% BSA, 0.1% Triton X-100 in PBS) for 1 h at room temperature and then incubated with FITC-conjugated anti-HA monoclonal antibody (Sigma-Aldrich) diluted in PBS containing 1% BSA. After three washes with wash buffer (0.1% Triton X-100 in PBS), slides were mounted in VECTASHIELD mounting medium containing DAPI (Vector Laboratories) and examined with a fluorescence microscope.

Yeast Two-hybrid Assay

Full-length coding sequences of Mcm2–7, Cdc45, GINS subunits, Orc1/Cdc6 and Orc1b were each cloned into pGADT7 vector for expression of proteins fused to the Gal4 activation domain (prey) or into pGBKT7 vector for expression of proteins fused to the Gal4 binding domain (bait). Yeast strains AH109 (mating type a) and Y187 (mating type α) were transformed with prey or bait plasmids, respectively. Strains carrying different combinations of bait and prey were generated by mating the haploids in YPDA medium for 24 h at 30 °C followed by plating on SD-Leu-Trp plates to select for the presence of both plasmids. Each combination strain was then spotted in three 10-fold serial dilutions onto SD-Leu-Trp and SD-His-Leu-Trp plates. The latter plate selects for clones carrying the bait protein and the prey protein that interacts with the bait protein (34).

Co-immunoprecipitation

To make endogenously PTP-tagged Mcm3, Psf3, and Orc1/Cdc6 cell lines, C-terminal fragments of Mcm3, Psf3, and Orc1/Cdc6 were each cloned into the pC-PTP-Neo vector (35), and the resulting plasmids were linearized and transfected into the 427 cell line. Successful tagging of target proteins was confirmed by Western blot with anti-ProtC antibody (Roche Applied Science) that specifically recognizes the protein C module in the PTP epitope. Subsequently, Cdc45, Psf3, Orc1/Cdc6, and Orc1b were each cloned into the pC-3HA-Bla vector, and the resulting constructs were transfected into the pC-Mcm3-PTP-Neo cell line. Similarly, Mcm3 and Cdc45 were endogenously tagged with a triple HA epitope in the pC-Psf3-PTP-Neo cell line, and Orc1b was endogenously tagged with a triple HA epitope in the pC-Orc1/Cdc6-PTP-Neo cell line.

The double transfectants were lysed in trypanosome immunoprecipitation buffer and incubated with IgG-Sepharose beads (Invitrogen) at 4 °C for 1 h. The beads were washed six times with immunoprecipitation buffer and boiled in 1× SDS-PAGE sampling buffer. Immunoprecipitates were separated on SDS-PAGE and detected by Western blot with anti-HA antibody and anti-ProtC antibody, respectively.

Protein Stability Assay

Trypanosome cell lines expressing endogenously HA-tagged Cdc45, Mcm3, Orc1/Cdc6, or Psf3 were incubated with 100 μg/ml cycloheximide (Acros Organics), and time course samples were collected for Western blot analysis. TbAUK1 was included as the control.

RESULTS

Identification of the CMG Complex in Trypanosomes

To identify the components of the CMG complex in trypanosomes, we searched the trypanosome genome database with yeast and human CMG proteins as queries and identified all six subunits of the Mcm2–7 complex (Mcm2, Tb11.02.5730; Mcm3, Tb927.2.3930; Mcm4, Tb11.01.4070; Mcm5, Tb11.02.3270; Mcm6, Tb11.01.3510; Mcm7, Tb11.01.7810), all four subunits of the GINS complex (Sld5, Tb927.3.4810; Psf1, Tb09.160.3540; Psf2, Tb11.01.2230; Psf3, Tb927.4.4680), and Cdc45 (Tb11.02.0310) (supplemental Fig. 1). All six Mcm proteins possess the well conserved ATPase domain that contains a Walker A motif involved in ATP binding, a Walker B motif that orients the nucleophilic water molecule, and an arginine finger motif known to make contact with the γ-phosphate of ATP (39). In addition, a zinc finger motif, which facilitates the formation of the double-hexamer of the Mcm2–7 complex (39), was identified in the N-terminal tails of five Mcm proteins except Mcm3 (supplemental Fig. 1), suggesting that trypanosome Mcm proteins may also be capable of forming a double-hexamer.

The Cdc45 family proteins are known to lack any motifs, but trypanosome Cdc45 exhibits an overall sequence identify of ∼20–25% to yeast and human Cdc45 proteins (data not shown). A recent study suggested that Cdc45 possesses a domain that is homologous to the bacterial RecJ protein, a 5′-3′ exonuclease involved in DNA repair, recombination, and replication (40), and trypanosome Cdc45 also appears to possess this conserved domain in its N terminus (data not shown). Similar to the GINS proteins from other eukaryotes, the four trypanosome GINS proteins are distantly related to each other, but they exhibit an overall sequence identity of ∼20–30% to their respective GINS homologs from yeast and humans. The four trypanosome GINS proteins can be grouped into two types based on the arrangement of the two conserved domains, the A-domain consisting of mainly α-helices and the B-domain consisting of β-strands. Sld5 and Psf1 possess a large A-domain in their N termini and a small B-domain in the C termini, whereas Psf2 and Psf3 possess the reverse order of the two domains (supplemental Fig. 1).

To investigate the potential interactions among Mc2–7, Cdc45, and GINS proteins, a yeast two-hybrid assay was performed, and the results were summarized in Fig. 1A with the detailed results shown in supplemental Fig. 2. Four Mcm proteins, Mcm2, Mcm3, Mcm5, and Mcm6, make contact with each of the other five Mcm proteins. However, Mcm4 and Mcm7 only interact with a few Mcm proteins; Mcm4 interacts with Mcm3, Mcm5, and Mcm6, whereas Mcm7 interacts with Mcm2, Mcm3, and Mcm5. No interaction between Mcm4 and Mcm7 was detected. Each of the four GINS proteins makes contact with other GINS proteins, suggesting the formation of a tight GINS complex. Additionally, each of the four GINS proteins also interacts with at least one Mcm protein; Sld5 interacts with all six Mcm proteins, whereas Psf1 interacts with Mcm3, Mcm5, and Mcm6. Psf2 and Psf3 only make contact with Mcm5 (Fig. 1A). Apparently, Mcm5 interacts with all four subunits of the GINS complex, and Sld5 makes contact with all Mcm proteins, suggesting that the two proteins are likely located in the interface between Mcm2–7 and the GINS complex.

FIGURE 1.

Identification of the CMG complex in T. brucei. A, interactions among the 11 subunits of the CMG complex. Black lines indicate the protein-protein interactions detected by yeast two-hybrid assays, and gray lines indicate the protein-protein interactions detected by GST pulldown assays. B, interactions of Cdc45, Mcm2–7, and GINS proteins in vivo in trypanosomes. PTP-tagged Mcm3 and Psf3 were each immunoprecipitated, and the immunoprecipitates were blotted with anti-HA antibody to detect co-precipitated 3HA-tagged proteins. IP, immunoprecipitation; IB, immunoblotting.

To examine whether Cdc45 interacts with Mcm2–7 and GINS proteins in vitro, GST pulldown was carried out. We found that all four GINS proteins and three Mcm proteins, Mcm2, Mcm4, and Mcm5, were able to pull down Cdc45 from the cell lysate (Fig. 1A and supplemental Fig. 3), indicating that Cdc45 does interact with the Mcm2–7 complex and the GINS complex. This result agrees with the recent report that Cdc45 makes extensive contacts with the GINS complex and interacts with Mcm2 and Mcm5 in the Mcm2–7 complex (41), although our data showed that Cdc45 also interacts with Mcm4 (Fig. 1A and supplemental Fig. 3).

Finally, to investigate whether the CMG complex is formed in vivo in trypanosomes, we performed co-immunoprecipitation assays. Mcm3 was tagged at one of its endogenous locus with a PTP epitope (35) at the C terminus, and in the same background, Cdc45 and Psf3 were each tagged endogenously with a triple HA epitope. Similarly, Psf3 was tagged endogenously with a PTP tag, and Mcm3 and Cdc45 were each tagged with a triple HA epitope in the same cell line. Precipitation of Mcm3-PTP through binding to the IgG-Sepharose column was able to co-precipitate both Cdc45-3HA and Psf3-3HA as detected by Western blot with anti-HA antibody (Fig. 1B, left panel). Similarly, precipitation of Psf3-PTP was able to bring down both Mcm3-3HA and Cdc45-3HA (Fig. 1B, right panel). These results indicate the formation of the CMG complex in vivo in trypanosomes.

Biochemical Characterization of Mcm2–7 and the CMG Complex

To investigate whether the CMG complex possesses DNA helicase activity, we purified individual CMG proteins expressed in bacteria, reconstituted the complex in vitro, and tested the in vitro activity toward unwinding duplex DNA. Each of the 11 CMG proteins was purified to near homogeneity (Fig. 2A). We next tested whether the CMG complex can be reconstituted. All 11 purified proteins were mixed and loaded onto a glycerol gradient for centrifugation. Fractions were then collected and examined by Western blot with anti-His antibody against the 7×His tag at the C terminus of Mcm2, and the band intensity was measured and plotted as a histogram. As shown in Fig. 2B, a single peak was detected between fractions 10 and 13 with an estimated molecular mass of ∼700 kDa (Fig. 2B). This is in agreement with the calculated molecular mass (741 kDa) of the CMG complex, suggesting that the CMG complex was formed in vitro.

FIGURE 2.

In vitro helicase activity assay with purified recombinant CMG proteins. A, purification of recombinant GST-His fusion CMG proteins. Individual CMG proteins were fused with the GST at the N terminus and a His₇ tag at the C terminus, expressed in E. coli, and purified to near homogeneity. B, glycerol gradient centrifugation to monitor the reconstituted CMG complex. Fractions were collected and analyzed by Western blot with anti-His antibody to detect His-tagged Mcm2. The intensity of the Mcm2-His protein band was measured and plotted. C, in vitro helicase activity assay. Purified recombinant proteins were incubated either individually or in combination with the circular substrate, a ³²P-labeled 30-mer oligonucleotide annealed to an M13 single-stranded circle. Positions of the double-stranded substrate and displaced oligonucleotide are indicated with arrows. The asterisks indicate ³²P-labeling of the 5′ end of the substrate. Boiled and Mock represent the control lanes with heat denatured substrate or without protein, respectively.

Individual Mcm proteins were tested for their potential helicase activity, and we found that Mcm4 alone was able to catalyze the unwinding of the circular DNA substrate in vitro (Fig. 2C). Mutation of the conserved lysine residue (Lys-428) in the Walker A domain of Mcm4 to alanine completely abolished this activity (Fig. 2C). To investigate whether the Mcm2–7 complex has in vitro helicase activity, we first tried to decrease the amount of Mcm4 protein used in the helicase assay experiment to minimize the helicase activity of Mcm4 so that the activity of Mcm2–7 complex can be distinguished from the activity contributed by Mcm4 alone. We found that when the amount of Mcm4 protein was decreased from 1.5 μg to 100 ng, the helicase activity of Mcm4 was not detectable (Fig. 2C). When all six Mcm proteins were mixed together, no helicase activity was detected either (Fig. 2C). We then examined the helicase activity of the reconstituted CMG complex and found that ATP-dependent release of ³²P-labeled oligonucleotide from the circular DNA substrate was detected (Fig. 2C), suggesting that association with Cdc45 and GINS proteins activates the in vitro helicase activity of Mcm2–7. This observation is consistent with the previous report that the Mcm2–7 complex has negligible helicase activity toward the circular DNA substrate, but the activity of Mcm2–7 is significantly enhanced when the CMG complex is formed (17). Our data suggest that the CMG complex, but not the Mcm2–7 complex, acts as the replicative helicase in trypanosomes.

Subcellular Localization of CMG Proteins during the Cell Cycle

As essential DNA replication factors, components of the CMG complex are loaded onto replication origins during G₁ phase but are exported out of the nucleus after DNA replication to prevent re-assembly of the replication complex (42, 43). To determine the subcellular localization of the CMG complex during different stages of the cell cycle in T. brucei, each of the 11 CMG proteins was tagged at the C terminus with EYFP or HA epitope and expressed from their respective endogenous locus in the procyclic form of T. brucei. All six Mcm proteins were found to be confined to the nucleus throughout the cell cycle from G₁ to telophase (Fig. 3), which resembles the subcellular distribution of Orc1/Cdc6 in trypanosomes (24) but differs from yeast Mcm2–7 proteins that are exported out of the nucleus after DNA replication (42, 43), presumably as one of the three independent mechanisms for preventing DNA re-replication in yeasts (44). However, the constant association of trypanosome Mcm2–7 proteins with the nucleus throughout the cell cycle suggests that such a control mechanism is likely absent from trypanosomes.

FIGURE 3.

Mcm proteins are confined to the nucleus throughout the cell cycle in T. brucei. A, localization of Mcm2-EYFP fusion protein during different stages of the cell cycle. DIC, differential interference contrast. B, localization of EYFP-tagged Mcm3–7 proteins at G₁ phase and telophase. Images showing the nuclear localization of these proteins during other stages of the cell cycle were not presented.

Like the Mcm2–7, all four GINS proteins were also localized to the nucleus throughout the cell cycle from G₁ to telophase (Fig. 4A), indicating that the Mcm2–7 complex and the GINS complex are likely associated with each other throughout the cell cycle. However, the GINS complex in yeasts and humans appears to be dissociated from the Mcm2–7 complex by moving from the nucleus to the spindle at metaphase and anaphase and then to the midzone during telophase and cytokinesis (45). GINS proteins appear to interact genetically with the chromosomal passenger complex (CPC) to regulate chromosome segregation (45). Given that the trypanosome GINS complex remains in the nucleus during mitosis, it may also play a role in chromosome segregation like its counterparts in yeasts and humans.

FIGURE 4.

Subcellular localizations of GINS proteins and Cdc45 in T. brucei. A, localization of EYFP-tagged Sld5, Psf1, Psf2, and Psf3 at G₁ phase and telophase. Images showing the nuclear localization of these proteins during other stages of the cell cycle were not presented. DIC, differential interference contrast. B, localization of Cdc45-3HA during different stages of the cell cycle. C, subcellular localization of Cdc45-3HA in the absence (−LMB) and presence (+LMB) of leptomycin B, an inhibitor of the nuclear export factor CRM1/exportin 1.

Cdc45, however, displayed a subcellular localization pattern different from Mcm2–7 and GINS proteins (Fig. 4B). It started to appear in the nucleus at G₁ phase and was significantly enriched in the nucleus during S and G₂ phases, but it was not detected in the nucleus thereafter (Fig. 4B). Instead, it appeared to spread throughout the cytoplasm in mitotic cells (Fig. 4B). During trypanosome mitosis, the nuclear envelope remains intact (46), and nuclear export of proteins is mediated by the nuclear export factor CRM1/exportin 1 (47). To test whether Cdc45 is exported out of the nucleus after G₂ phase, we treated the cells with leptomycin B, an inhibitor of CRM1/exportin 1 (48), and examined the effect on Cdc45 localization in mitotic cells. We found that although Cdc45 remained in the cytoplasm in the control mitotic cell, Cdc45 was only detected in the two nuclei of the mitotic cell treated with leptomycin B (Fig. 4C). We then searched for potential nuclear export signal sequence in Cdc45 and other CMG proteins and found that only Cdc45 possesses a putative nuclear export signal from residues 337–345 (LFLLRHLSL; the conserved leucine residues are in bold) (supplemental Fig. 1), which resembles the nuclear export signal consensus sequence (ΦX_2–3ΦX₂ΦXΦ; Φ represents a hydrophobic residue, and X represents any residue) (49). These observations suggest that Cdc45 is exported out of the nucleus to the cytoplasm, likely through the CRM1/exportin 1-mediated pathway.

To confirm that Cdc45 is not degraded, we monitored the turnover of Cdc45. Endogenous Cdc45 was tagged with a triple HA epitope, and cells were incubated with cycloheximide. Time course samples were then collected, and Western blot was performed. Cdc45-3HA level was constant in all time points after cycloheximide treatment. Similarly, levels of Orc1/Cdc6-3HA, Mcm3-3HA, and Psf3-3HA were also constant during the 8 h of cycloheximide treatment (Fig. 5). As a control, the TbAUK1-3HA level started to decrease after cycloheximide treatment for 2 h (Fig. 5). These results suggest that unlike TbAUK1, which undergoes degradation during the cell cycle,³ all three proteins from the CMG complex (Cdc45, Mcm3, and Psf3) and Orc1/Cdc6 are not subject to degradation.

FIGURE 5.

Stability of Cdc45, Orc1/Cdc6, Mcm3, and Psf3 proteins. Cells expressing endogenously 3HA-tagged proteins were treated with cycloheximide, and time course samples after cycloheximide treatment were collected for Western blot with anti-HA antibody. The 3HA-tagged TbAUK1 was included as the control.

Due to tight association of the Mcm2–7 complex and the GINS complex with the nucleus throughout the cell cycle in trypanosomes (Figs. 3 and 4), the exclusion of Cdc45 from the nucleus may provide a means to prevent the formation of an active CMG helicase complex and re-replication of chromosomal DNA. Together, these observations identified a subcellular localization pattern of the CMG complex in trypanosomes that is different from that in yeast and metazoans, which may have important implications on replication licensing in trypanosomes (see “Discussion” below).

RNAi Silencing of CMG Genes in the Procyclic Form

To investigate the function of the CMG complex in trypanosomes, RNAi was performed in the procyclic form of T. brucei. RNAi was induced by adding 1.0 μg/ml tetracycline to the cell culture, and cells before and after RNAi induction were harvested for Northern blot to monitor the change in the mRNA level. As shown in the insets of Fig. 6, except for Cdc45 RNAi, which only reduced the Cdc45 mRNA level to ∼40%, RNAi induction resulted in a significant decrease (>90%) of mRNA level in all other RNAi cell lines. As a consequence, knockdown of most CMG genes, except for Mcm4, Mcm6 and Psf1, led to growth inhibition (Fig. 6). In particular, RNAi of Mcm3 or Cdc45 resulted in drastic growth inhibition and eventual cell death after 4 days, whereas RNAi of Mcm2 or Psf2 only slightly inhibited cell growth (Fig. 6). The lack of growth inhibition by RNAi of Mcm4, Mcm6, and Psf1 suggests that there is likely an excess of these proteins and that knockdown is insufficient to reveal their essential function.

FIGURE 6.

RNAi silencing of the 11 subunit genes of the CMG complex in the procyclic form of T. brucei. The procyclic cells harboring the respective RNAi constructs were cultivated without (− Tet) or with (+ Tet) tetracycline and monitored for cell growth. The insets show the levels of mRNA, monitored by Northern blot, in cells before (−) and after (+) RNAi for 2 days. Total RNA was stained with ethidium bromide and included as the loading control.

To characterize the effect of RNAi-mediated silencing of individual CMG genes on cell cycle progression, trypanosome cell lines before and after RNAi induction were examined under a fluorescence microscope to count the number of nuclei (N) and kinetoplasts (K). In each of the eight RNAi cell lines that exhibited growth inhibition (Fig. 6), RNAi led to a gradual decrease of 1N1K cells from ∼75 to 30–40%, which was accompanied by a gradual increase of zoid cells (0N1K) from less than 1 to ∼40% of the total population (supplemental Fig. 4A), indicating a severe interference of the nuclear cycle. These phenotypes are similar to that caused by Orc1/Cdc6 RNAi (24) and cyclin knockdowns (26) in trypanosomes. In addition to zoid production, cells with abnormal and irregularly shaped nuclei were also increased from 0 to 20–30% in each of the four Mcm RNAi cell lines, whereas only a few of such cells were produced by RNAi silencing of Cdc45 and the three GINS genes (supplemental Figs. 4A and 5). The nucleus in these cells appears to contain many punctate dots or mis-segregated chromosomes as stained by DAPI (supplemental Fig. 5). These observations suggest that Mcm proteins might also be involved in maintaining the structural integrity of the nucleus during DNA replication.

The CMG Complex Is Required for DNA Replication

To examine whether RNAi of individual CMG genes affects DNA replication, a BrdU incorporation assay was carried out in uninduced control and RNAi cells after tetracycline induction for 3 or 5 days. Those cells that had undergone vigorous DNA replication incorporated BrdU into their DNA and were detected by anti-BrdU immunostaining (supplemental Fig. 4B, Control), whereas those cells that had not undergone DNA replication or were defective in DNA replication due to RNAi of the CMG genes did not incorporate BrdU into the DNA and therefore were not stained by anti-BrdU antibody (supplemental Fig. 4B, Mcm3 RNAi). In the asynchronous wild-type trypanosome culture, BrdU-positive cells constitute ∼30% of the total cell population (supplemental Fig. 4C). In the eight RNAi cell lines that exhibited growth defects upon RNAi (Fig. 6), the percentage of BrdU-positive cells decreased to 2–8% of the total cell population (supplemental Fig. 4C). These results suggest that RNAi of each of the eight CMG genes inhibited DNA replication.

Identification of the Origin Recognition Complex and Its Association with the CMG Complex

The ORC in eukaryotes consists of six distinct Orc proteins (Orc1–Orc6) (50), whereas the archaeal ORC is formed by only one Orc1/Cdc6-like protein (20). The pre-RC in eukaryotes is composed of the heterohexameric ORC, two replication licensing factors, Cdc6 and Cdt1, and the heterohexameric Mcm2–7 complex (2), whereas the pre-RC in archaea consists of a single Orc protein, Orc1/Cdc6, and a homohexameric Mcm complex (20, 51). The composition of the ORC and the pre-RC in trypanosomes is still not known. A single Orc1/Cdc6 is found to associate with the chromatin and regulates DNA replication in T. brucei (24), and our present study has identified a heterohexameric Mcm2–7 complex (see above). It appears that trypanosome pre-RC is likely formed by Orc1/Cdc6 and Mcm2–7. However, a careful search of the trypanosome genome database using the six Orc proteins from yeast and humans as the bait identified another Orc1-like protein that is bigger than Orc1/Cdc6 but is significantly smaller than the Orc1 homologs from yeast and animals (Fig. 7A). This novel Orc1-like protein, which we named Orc1b (Tb09.160.0830), possesses well conserved Walker A and Walker B motifs of a typical AAA-type ATPase domain but lacks the arginine finger motif (Fig. 7A). Additionally, the Walker A and Walker B motifs in Orc1b are further separated by an insertion of ∼120 amino acids (Fig. 7A). This raised the question of whether Orc1b possesses ATPase activity. Recombinant Orc1b was expressed and purified from E. coli, and in vitro ATPase activity assay showed that Orc1b does not possess detectable ATPase activity (data not shown). However, whether Orc1b in trypanosomes possesses ATPase activity is still not known, and due to its tight association with Orc1/Cdc6 and Mcm3 (see below), an in vitro ATPase activity assay with immunoprecipitated Orc1b will not be able to distinguish the activity of Orc1b from the activity of co-precipitated Orc1/Cdc6 and/or the Mcm2–7 complex and therefore was not performed.

FIGURE 7.

Interactions among Mcm3, Orc1/Cdc6, and a novel Orc1-like protein, Orc1b. A, identification of a novel Orc1-like protein, Orc1b, in trypanosomes. The upper panel shows the schematic representation of the structure of Orc1b and Orc1/Cdc6 from trypanosomes and Orc1 from the budding yeast. Alignment of Walker A motif (A), Walker B motif (B), and arginine finger (R) in the ATPase domain of Orc1b, Orc1/Cdc6, and yeast Orc1 is presented in the lower panel. Tb, T. brucei; Sc, Saccharomyces cerevisiae; BAH, bromo-adjacent homology. A. thaliana, Arabidopsis thaliana; H. sapiens, Homo sapiens. aa, amino acids. B, interaction of Orc1/Cdc6 with Orc1b detected by in vitro GST pulldown. IB, immunoblotting. C, interactions of Mcm3 with Orc1/Cdc6 and Orc1b detected by in vitro GST pulldown assays. D, interaction of Orc1/Cdc6 and Orc1b in vivo in trypanosomes. PTP-tagged Orc1/Cdc6 was brought down by IgG-Sepharose beads, and the immunoprecipitates were blotted with anti-HA antibody to detect 3HA-tagged Orc1b. IP, immunoprecipitation. E, interactions of Mcm3 with Orc1/Cdc6 and Orc1b in vivo in trypanosomes. PTP-tagged Mcm3 was precipitated with IgG-Sepharose beads, and the immunoprecipitates were blotted with anti-HA antibody to detect 3HA-tagged Orc1/Cdc6 and Orc1b.

To investigate whether Orc1b is a component of the ORC and whether Orc1b and Orc1/Cdc6 interact with the CMG complex, GST pulldown was carried out, and the results indicated that Orc1b interacts with Orc1/Cdc6 in vitro (Fig. 7B). Additionally, both GST-Orc1/Cdc6 and GST-Orc1b were able to bring down Mcm3-3HA from the cell lysate (Fig. 7C), suggesting that the two Orc1-like proteins interact with Mcm3 in vitro. Finally, to test whether Orc1/Cdc6, Orc1b, and Mcm3 interact with each other in vivo in trypanosomes, co-immunoprecipitation was carried out. A double transfectant was generated by tagging the endogenous Mcm3 with a PTP epitope and tagging the endogenous Orc1/Cdc6 or Orc1b with a triple HA tag. Similarly, a double transfectant was created by tagging the endogenous Orc1/Cdc6 with the PTP epitope and tagging the endogenous Ocr1b with a triple HA tag. Precipitation of Orc1/Cdc6-PTP was able to pull down Orc1b-3HA (Fig. 7D), and precipitation of Mcm3-PTP brought down both Orc1/Cdc6-3HA and Orc1b-3HA (Fig. 7E), indicating in vivo interactions among Orc1/Cdc6, Orc1b, and Mcm3 in trypanosomes.

Collectively, these results suggest that trypanosome ORC comprises at least two Orc1-like proteins, Orc1/Cdc6 and Orc1b. In addition, through associations with Mcm3 in the Mcm2–7 complex, Orc1/Cdc6 and Orc1b may recruit the Mcm2–7 complex to origins to form the pre-RC in trypanosomes.

DISCUSSION

In this study, we identified the Cdc45·Mcm2–7·GINS complex as the replicative helicase and an unusual origin recognition complex containing two Orc1-like proteins in trypanosomes. The identification of Cdc45·Mcm2–7·GINS complex indicates that trypanosomes possess the DNA helicase machinery typical of a eukaryote. However, trypanosomes may contain an ORC that is different from all eukaryotes studied so far. However, this unusual ORC is likely able to recruit the Mcm2–7 complex to replication origins, which is achieved through interactions between Mcm3 and the two Orc1-like proteins (Fig. 7). Based on our current understanding of DNA replication initiation in other eukaryotes (52) and the identification of the above mentioned protein complexes in trypanosomes, we found that DNA replication initiation in T. brucei shares some common features with that in yeast and animals but also exhibits several distinct features. During G₁ phase of the cell cycle, replication origins are bound by the ORC consisting of Orc1/Cdc6, Orc1b, and likely additional ORC proteins. Given the absence of Cdc6 and Cdt1 homologs in T. brucei, the Mcm2–7 complex is likely recruited onto replication origins through Mcm3-mediated binding to both Orc1-like proteins, which leads to the formation of the pre-RC. This complex, however, does not possess DNA helicase activity. When cells start to enter S phase, the GINS complex and Cdc45 are loaded onto origins through interactions with the Mcm2–7 complex, which leads to the formation of the CMG complex that possesses DNA helicase activity and can unwind duplex DNA (Fig. 2). After G₂ phase, Cdc45, but not the Mcm2–7 complex as in yeast, is exported out of the nucleus to the cytoplasm, which leads to the dissociation of the CMG complex and therefore prevents DNA re-replication.

Although the fundamental principle of DNA replication initiation in trypanosomes is similar to that in other eukaryotes, the mechanism and regulation of DNA replication initiation in trypanosomes are likely different. The trypanosome genome does not encode Dbf4 protein and its kinase partner, Cdc7, which is known to phosphorylate all Mcm proteins but Mcm5 (53, 54). Phosphorylation of the N-terminal tails of Mcm2, Mcm4, and Mcm6 by Dbf4·Cdc7 appears to promote the stable formation of the Cdc45·Mcm2–7 complex (13), and strikingly, phosphorylation of the N-terminal tail of Mcm4 by Dbf4·Cdc7 alleviates an inhibitory activity in Mcm4 (14). In cells lacking the inhibitory N-terminal tail of Mcm4, Dbf4·Cdc7 becomes dispensable for DNA replication initiation (14). The N-terminal tails of eukaryotic Mcm4 proteins are very rich (29% in yeast Mcm4 and 24% in human Mcm4) in serine and threonine residues, some of which are targets of Dbf4·Cdc7 and S-CDK. However, the N-terminal tail of trypanosome Mcm4 contains only 8% of serine and threonine residues, and therefore, it likely does not possess an inhibitory activity that has to be relieved by Dbf4·Cdc7-mediated phosphorylation as in other eukaryotes. In this regard, trypanosomes may not need Dbf4·Cdc7 kinase.

The N-terminal tail of yeast Mcm4 is also phosphorylated by S-CDK (55), and in the absence of Dbf4·Cdc7, phosphorylation by S-CDK is crucial for DNA replication initiation (14, 56). Like all the Mcm4 proteins in other eukaryotes, trypanosome Mcm4 also possesses the consensus CDK phosphorylation sites ((S/T)PX(K/R; the serine/threonine residues in bold indicate the phosphorylation site) (56)) at Thr-10 and Ser-20 in its N-terminal tail, suggesting that trypanosome Mcm4 is a potential CDK substrate. However, due to the lack of S-phase cyclin and CDK homologs in trypanosomes, we investigated the potential interaction of Mcm4 with CRK1 and CRK2, both of which are required for G₁/S transition in trypanosomes (29), and found that neither protein kinase interacts with Mcm4,³ indicating that Mcm4 is unlikely a substrate of CRK1 and CRK2. It should be noted that although Mcm4 is phosphorylated by S-CDK in yeasts, the principal targets of S-CDK in yeasts are Sld2 and Sld3, which, when phosphorylated by S-CDK, associate with a BRCA1 C terminus domain-containing protein, Dpb11 (6), and recruit Cdc45 and the GINS complex to replication origins (57, 58). Therefore, the formation of the Cdc45·Mcm2–7·GINS complex is dependent on S-CDK-mediated phosphorylation of Sld2 and Sld3. Nevertheless, trypanosomes do not have Sld2 and Sld3 homologs.³ It remains unclear how CMG complex formation in trypanosomes is regulated.

One of the interesting findings from this study is the confinement of the Mcm2–7 complex to the nucleus throughout the cell cycle (Fig. 3). This suggests that after DNA replication, the complex is not exported out of the nucleus. In contrast, yeast Mcm2–7 is gradually exported from the nucleus to the cytoplasm during S phase and excluded from the nucleus at G₂ and M phases (42, 43); this is promoted by the mitotic cyclin·CDK complex Clb·Cdc28 through phosphorylating a novel nuclear export signal sequence adjacent to the nuclear localization signal sequence on Mcm3 (59). It has been established that nuclear exclusion of the Mcm2–7 complex constitutes one of the three independent mechanisms for preventing DNA re-replication in yeasts (44) and many other systems (39). However, a scan of the amino acid sequence of trypanosome Mcm3 failed to identify nuclear export signal and nuclear localization signal as well as any consensus CDK phosphorylation site (data not shown). These observations suggest that mitotic CDK-mediated regulation of nucleocytoplasmic transport of Mcm2–7 is likely not present in trypanosomes and that prevention of DNA re-replication in trypanosomes may be accomplished through a mechanism different from that in yeast and animals (see below).

The exclusive localization of trypanosome GINS proteins to the nucleus throughout the cell cycle (Fig. 4A) contrasts significantly with the subcellular distribution of GINS proteins in yeast and human cells where the GINS protein Psf2 localizes to the nucleus in interphase and is enriched in part on the spindle in metaphase and anaphase and in the midzone during cytokinesis (45). The GINS complex in yeast and humans appears to play crucial roles in mitosis and cytokinesis by interacting genetically with CPC (45). In trypanosomes, RNAi silencing of Sld5, Psf2, and Psf3 each inhibited DNA replication (supplemental Fig. 4, B and C), and due to this inhibition, any potential defect in mitosis will not be detected because the cells are not allowed to progress beyond the S phase. Trypanosomes possess a novel CPC that displays a dynamic subcellular localization during mitosis and cytokinesis and is essential for chromosome segregation and cytokinesis (33, 34, 38). It will be interesting to test whether this novel CPC also interacts with the GINS proteins and whether the GINS proteins play any role in mitosis in trypanosomes.

The presence of two Orc1-like proteins in the origin recognition complex appears to be unique to trypanosomes and has never been found in any other eukaryotes previously. However, it remains unknown whether trypanosome ORC contains six different subunits like the ORC in other eukaryotes. Because trypanosomes were branched early in evolution and possess a number of regulatory pathways that are not shared by any other eukaryotes, it is very likely that the trypanosome ORC comprises divergent Orc subunits. Further experiments are needed to affinity-purify the Orc complex from trypanosomes.

Although trypanosome Orc1/Cdc6 exhibits sequence homology to both Orc1 and Cdc6 from yeast and metazoans and can functionally rescue yeast Cdc6 deletion (24), Orc1b is only homologous to Orc1 proteins from other eukaryotes (Fig. 7). This raised the possibility that Orc1/Cdc6 in trypanosomes could function as the replication licensing factor, i.e. Cdc6, rather than as a component of the origin recognition complex, i.e. Orc1. Nevertheless, trypanosome Orc1/Cdc6 associates with the chromatin throughout the cell cycle (24) and is stable (Fig. 5), indicating that it is not subject to proteasome-mediated degradation or nuclear export after S phase. Therefore, it may not function as a bona fide licensing factor like Cdc6 and Cdt1 in other eukaryotes, both of which are known to be either degraded after S phase or excluded from the nucleus to prevent the replication origins from relicensing (60–65). Importantly, this regulation of Cdc6 and/or Cdt1 protein level in the nucleus represents another of the three mechanisms that are employed by eukaryotic cells to prevent re-replication after S phase (44). In contrast, Orc proteins remain associated with the chromatin after DNA replication, but they are inactivated by mitotic CDK-mediated phosphorylation, which constitutes the third mechanism for preventing DNA re-replication in yeasts (44). Strikingly, neither Orc1/Cdc6 nor Orc1b from trypanosomes is a potential substrate of the mitotic CDK homolog CRK3 because of the lack of canonical CDK phosphorylation sites and direct interactions with CRK3 in vitro and in vivo.³ This implies that the two Orc1-like proteins are likely not regulated by CRK3. Taken together, the apparent lack of regulation of Orc1/Cdc6 protein level in the nucleus (24) and no nuclear exclusion of the Mcm2–7 complex after S phase (Fig. 3) suggest the presence in trypanosomes of a replication licensing system different from that in yeasts and metazoans. Our studies suggest that nuclear exclusion of Cdc45 (Fig. 4, B and C) appears to constitute one mechanism for preventing DNA re-replication in trypanosomes, but whether additional control mechanisms are involved remains to be determined.