J Biol Chem, Vol. 275, Issue 20, 15049-15059, May 19, 2000

The Intrinsic DNA Helicase Activity of Methanobacterium thermoautotrophicum H Minichromosome Maintenance Protein^*

David F. Shechter, Carol Y. Ying^§, and Jean Gautier^¶

From the Department of Genetics and Development, the Department of Dermatology, and the  Integrated Program in Cellular, Molecular and Biophysical Studies, Columbia University, New York, New York 10032

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Minichromosome maintenance proteins (MCMs) form a family of conserved molecules that are essential for initiation of DNA replication. All eukaryotes contain six orthologous MCM proteins that function as heteromultimeric complexes. The sequencing of the complete genomes of several archaebacteria has shown that MCM proteins are also present in archaea. The archaea Methanobacterium thermoautotrophicum contains a single MCM-related sequence. Here we report on the expression and purification of the recombinant M. thermoautotrophicum MCM protein (MtMCM) in both Escherichia coli and baculovirus-infected cells. We show that purified MtMCM protein assembles in large macromolecular complexes consistent in size with being double hexamers. We demonstrate that MtMCM contains helicase activity that preferentially uses dATP and DNA-dependent dATPase and ATPase activities. The intrinsic helicase activity of MtMCM is abolished when a conserved lysine in the helicase domain I/nucleotide binding site is mutated. MtMCM helicase unwinds DNA duplexes in a 3'  5' direction and can unwind up to 500 base pairs in vitro. The kinetics, processivity, and directionality of MtMCM support its role as a replicative helicase in M. thermoautotrophicum. This strongly suggests that this function is conserved for MCM proteins in eukaryotes where a replicative helicase has yet to be identified.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Minichromosome maintenance proteins (MCMs)¹ were first identified independently through two different genetic screens designed to isolate mutants that either lose the ability to replicate a plasmid containing a single origin of replication (mcm mutants) (1) or cell cycle mutants (cdc) that arrest at the G₁/S border (cdc mutants) (2).

A large number of genetic and biochemical experiments in yeast, Drosophila, Xenopus, and mammals have demonstrated that MCM proteins are essential for initiation of DNA replication in eukaryotes (3-9). Six orthologous MCM proteins have been identified in all eukaryotes studied (reviewed in Refs. 10-12). In yeast, all six proteins are essential, and null mutations in any of the six genes result in cell cycle arrest prior to the onset of DNA replication (4, 6, 9, 13-19). MCM proteins are a component of the prereplicative complex whose activation is a prerequisite to DNA replication initiation (20) and requires the ordered assembly of the origin recognition complex, Cdc6, MCM, and Cdc45 proteins on the chromatin. Experiments in Xenopus cell-free extracts using specific immunodepletions of origin recognition complex, Cdc6, or MCM proteins have demonstrated the sequential nature of these assembly steps where Cdc6 assembles on origin recognition complex-bound chromatin, then allowing the loading of MCM proteins (21-23).

Experiments in a variety of model systems have indicated that all six MCM subunits are present in MCM complexes, suggesting that MCMs might exist as heterohexameric complexes in the cell in a 1:1 stoichiometry (reviewed in Ref. 12). However, biochemical purification and extraction procedures have also revealed the existence of subcomplexes within this hexameric complex consisting of a tightly bound core complex (MCM4, -6, and -7) more loosely associated with MCM2 and a second complex composed of MCM3 and -5 (24-29). These observations have raised the possibility that several MCM complexes with potentially different functions might co-exist in the cell.

MCM proteins share extensive similarities within a central core domain. All MCM proteins contain a putative ATPase domain within this core, which resembles that found in DNA helicases. In particular, the highly conserved Walker A and Walker B domains, which have been implicated in ATP binding and ATP hydrolysis, respectively, are found in all MCM proteins (30). Walker A and B domains correspond to the conserved regions I and II of the helicase superfamily (31).

A variety of observations have suggested that MCM proteins could function as a replicative helicase either during DNA replication initiation or for both the initiation and elongation steps. 1) Helicases are DNA-dependent NTPases, using the energy of hydrolysis to transiently unwind DNA. The Walker domains are therefore key features of helicases (31). 2) Loading MCM proteins onto the chromatin requires the Cdc6 loading factor, because the loading of the bacterial DnaB helicase requires DnaC (reviewed in Ref. 32). 3) MCM protein complexes purified from Schizosaccharomyces pombe exhibited ring-shaped structures when observed by electron microscopy. Some of these structures seemed to display a ringlike 6-fold symmetry reminiscent of that observed with some DNA helicases (24). 4) Purified MCM complexes composed of MCM4, MCM6, and MCM7 contained weak helicase activity in vitro, restricted to the unwinding of very short oligonucleotides (33, 34). On the other hand, the hexameric MCM complex purified from S. pombe did not exhibit any helicase activity (24). In addition, the MCM2 subunit has been reported to display inhibitory activity in vitro toward the helicase activity of the MCM4, -6, and -7 complex (35).

Studies on the localization of MCM proteins during the process of replication have also provided conflicting data. Chromatin precipitation studies in budding yeast supported the idea that some MCM proteins move along the replication fork as DNA replication proceeds, suggesting that if MCM proteins are a helicase they could be involved in an initial unwinding step as well as during elongation (36, 37). Immunofluorescence studies in other organisms, however, have indicated that MCM proteins do not colocalize with the sites of incorporation of nucleotides during replication, suggesting that the role of MCM proteins might be restricted to an initiation event (38-40).

Helicases are enzymes that unwind the nucleic acid duplex and are essential for many cellular processes such as replication, repair, or transcription. They can process the unwinding of DNA with specific directionality and are referred as 3'  5' or 5'  3' helicases. Helicases are generally oligomeric, usually dimers or hexamers. It has been demonstrated that several of the helicases implicated in DNA replication such as SV40 or polyoma T antigen have a hexameric structure (41, 42). The crystal structure of the replicative hexameric T7 DNA helicase protein 4 was recently solved, suggesting that progression along the DNA leading to the unwinding of the duplex used a corkscrew-like inchworm mechanism (43). The directionality of the helicase is not conferred by the dimeric or hexameric structure, since both specificities have been described with each type of structure (reviewed in Ref. 44). Despite the presence of a large number of putative helicase sequences in the eukaryotic genome (134 putative helicase sequences have been identified in S. cerevisiae), no replicative helicase has yet been identified in eukaryotes (45).

The complete sequence of an archaebacterium genome (Methanobacterium thermoautotrophicum) revealed that archaea contained putative open reading frames sharing homology with different components of the prereplicative complex, namely origin recognition complex, Cdc6, and MCM (46). Furthermore, putative origins of replication have been recently identified in archaea (47). In addition to the fact that archaea possess homologues of most of the essential genes involved in DNA replication (48), this observation strengthened the idea that archaea possess features that are more closely related to eukaryotic than prokaryotic DNA replication, supporting the hypothesis that the divergence between archaea and eukaryotes occurred later than the divergence between archaea and prokaryotes. The subsequent sequencing of other archaea genomes confirmed these findings. Interestingly, two of the archaea genomes sequenced (M. thermoautotrophicum and Archaeoglobus fulgidus) contained a single open reading frame related to MCM proteins (49), while Methanococcus jannaschii contained four MCM-related sequences (50). Nothing is yet known about the biochemical function of these MCM-related proteins in archaea. In particular, it is not known whether they might be involved in DNA replication.

As a step toward better understanding the biochemical role of MCM proteins and as an attempt to clarify some of the discrepancies regarding their possible biochemical function, we decided to take advantage of the presence of a single MCM protein in M. thermoautotrophicum, since it opened the possibility to test the biochemical function of this protein in a simpler system where the potential inhibitory effect of other subunits would not be relevant.

Here we report the expression and the purification of the M. thermoautotrophicum MCM-related ORF proteins from Escherichia coli and baculovirus-infected cells. We show that the purified protein multimerizes in complexes consistent with being double hexamers. We demonstrate that MtMCM is a DNA helicase that preferentially uses dATP and that nucleotide binding is essential for activity. We also show that MtMCM is a DNA-dependent dATPase and ATPase. MtMCM works catalytically to unwind DNA duplexes in a 3'  5' direction and can unwind up to 500 base pairs in vitro. The kinetics, processivity, and directionality of the enzymatic activity all support the idea that MtMCM can work as a replicative helicase.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Nucleotides, Oligonucleotides, and Other Reagents-- [-³²P]dATP (3000 Ci/mmol) and [-³²P]ATP (6000 Ci/mmol) were purchased from Amersham Pharmacia Biotech. dNTPs and ATP were from APBiotech, while the other nucleotides were from Roche Molecular Biochemicals. Oligonucleotides were purchased from Sigma Genosys. T4 polynucleotide kinase was from New England Biolabs, and the Klenow enzyme was from Roche Molecular Biochemicals.

Cloning of M. thermoautotrophicum MCM-- MtMCM was cloned from M. thermoautotrophicum genomic DNA (a generous gift of Dr. Reeve). To facilitate purification of MtMCM protein, sequences encoding a 6-histidine tag were added to the C terminus of the MtMCM gene by PCR. The MtMCM gene was amplified by PCR using Vent DNA polymerase (New England Biolabs) with CCCCTCTAGAAAACCTATAAATCATATGATGAAAACCGTGGATAAGAGC as the 5' primer (the engineered XbaI and NdeI restriction sites, respectively, are underlined) and GGGGCTCGAGTTAGTGATGGTGATGGTGATGCAGACTATCTTAAGGTATCCC as the 3' primer (XhoI site is underlined). The PCR product was subcloned into the XbaI and XhoI sites of the baculovirus expression vector pFASTBAC (Life Technologies, Inc.) and sequenced using an Applied Biosystem Automated Sequencer. Subsequently, MtMCM was subcloned into the NdeI and XhoI sites of the E. coli expression vector pET21b (Novagen). MtMCM recombinant baculovirus was generated based on the BAC-TO-BAC Baculovirus Expression System (Life Technologies, Inc.).

MtMCM was mutated at lysine 325 to alanine by PCR using Vent DNA polymerase (NEB) with GGGCGGCCCAGCAGGCCATAA as the 5' primer (the SfiI site is underlined) and GCGCGGTACCCTTACCGCTGGTGTATATCCCCCTGGGGGCCAGCTTTGAGACGTACTTGAGCATCTGTGACGCACCGATACCGGGGTCCCC as the 3' primer (the KpnI site, the modified codon (lysine  alanine), and a silent mutation in the SmaI site are underlined, respectively). The PCR product was subcloned back into the MtMCM-pET21b vector at the unique SfiI and KpnI sites present in the MtMCM sequence. The mutation was detected by loss of the SmaI site and confirmed by sequencing.

Expression and Purification of MtMCM Protein-- For expression of MtMCM protein in E. coli, the pET21b vector containing the MtMCM gene was transformed into DE3(BL21)pLysS cells (Novagen) and grown from an overnight culture in 5.5 liters of LB medium containing 50 µg/ml carbenicillin and 34 µg/ml chloramphenicol at 37 °C and 375 rpm. When the cells reached an A₆₀₀ of 0.6, protein expression was induced by the addition of isopropyl-1-thio--D-galactopyranoside to 0.6 mM. The cells were grown for an additional 3 h and harvested by centrifugation. The cells were resuspended in a small volume of 50 mM Tris-HCl, pH 7.5, 10% sucrose, and protease inhibitors (100 µg/ml phenylmethylsulfonyl fluoride, 0.2 µg/ml aprotinin, 0.2 µg/ml leupeptin) and frozen at 70 °C immediately. The cells were then thawed and sonicated on ice and spun in an SS34 rotor at 16,000 rpm for 15 min. The supernatant was removed and precipitated with ammonium sulfate first at 20% and then at 45% saturation. The 45% pellet was resuspended in and dialyzed overnight against 25 mM Tris-HCl, pH 7.8, 50 mM NaCl, 10% glycerol, 5 mM imidazole, and protease inhibitors. Following the increase of imidazole concentration to 50 mM, the protein was loaded onto a pre-equilibrated 10-ml Ni²⁺-nitrilotriacetic acid-agarose (Qiagen) column and eluted with a 50-300 mM imidazole gradient. The peak pool of protein was dialyzed overnight against 25 mM Tris-HCl, pH 7.8, 100 mM NaCl, 10% glycerol, 1 mM dithiothreitol, 1 mM EDTA, and protease inhibitors. This pool was then loaded onto a 10-ml Source Q (APBiotech) column pre-equilibrated in the above buffer. The column was washed with 3 volumes of buffer and then eluted with a 100-ml 100-450 mM NaCl gradient. The peak fractions were pooled, and the protein concentration (2.0 mg/ml) was determined by an A₂₈₀ measurement using the calculated extinction coefficient of 29,570 liters mol¹ cm¹.

For eukaryotic expression of MtMCM protein, Sf9 insect cells (Invitrogen) were infected with the recombinant baculovirus at a multiplicity of infection of 5 and harvested 96 h postinfection. The native recombinant MtMCM protein in infected cell lysate was purified on a Ni²⁺-nitrilotriacetic acid-agarose column and eluted with 350 mM imidazole and then dialyzed against 20 mM Hepes-KOH, pH 7.4, 80 mM NaCl, 1 mM dithiothreitol. The protein concentration (0.5 mg/ml) was determined by the Bradford method (Bio-Rad) using bovine serum albumin as a standard.

For expression of K325A-MtMCM, the mutant vector was transformed and expressed in E. coli and purified as described for the wild type, with the exception of the Source Q step.

Preparation of DNA Helicase Substrate-- The substrate used in the standard DNA helicase reaction was prepared by annealing a 5'-end-labeled 63-mer oligonucleotide (5'-TGCCAAGCTTGCATGCCTGCAGGTCGACTCTAGAGGATCCCCGGGTACCGAGCTCGAATTCGT-3' (the HincII site is underlined)) to M13mp18 single-stranded DNA (New England Biochemicals). The oligonucleotide was labeled in a reaction (20 µl) containing 18 pmol of 63-mer, 1× polynucleotide kinase buffer (New England Biochemicals), 20 µCi of [-³²P]ATP, and 15 units of T4 kinase for 30 min at 37 °C. The labeled oligonucleotide was separated from free nucleotide in a G-25 spin column (Roche Molecular Biochemicals), and the specific activity of the labeled DNA was determined by liquid scintillation counting (usually 1500 cpm/fmol). The helicase template was prepared by annealing the oligonucleotide to M13 in the following conditions: a 32-µl reaction volume containing 2 pmol of M13 DNA and 4 pmol of labeled oligonucleotide, 100 mM NaCl, 25 mM Tris-HCl, pH 7.8, and 15 mM MgCl₂. The reaction was layered with mineral oil, placed in a heat block at 100 °C, and then slowly cooled to room temperature. The annealed substrate was separated from unannealed oligonucleotide via gel filtration through a 5-ml Biogel A-15M (Bio-Rad) column in a buffer containing 25 mM Tris-HCl, pH 7.8, and 100 mM NaCl. Six-drop fractions were collected, and the peak fractions in the void volume were pooled. The yield of template was calculated using the specific activity of the oligonucleotide.

The template used to determine directionality was prepared by digesting the standard template with HincII, resulting in a linear, blunt-ended template with a 38-nucleotide oligonucleotide annealed to the 3'-end of the M13 molecule and a 25-nucleotide ³²P-labeled oligonucleotide annealed to the 5'-end. Subsequently, the 3'-end of the 38-mer was extended 2 nucleotides with the Klenow enzyme and [-³²P]dATP, resulting in a linear template with short radiolabeled double-stranded regions at each end.

Helicase Assay-- The standard helicase assay reaction (20 µl) contained 25 mM Tris-HCl, pH 7.8, 2.5 mM dATP, 5 mM MgCl₂, 1 mM dithiothreitol, 200 µg/ml bovine serum albumin, and either 5 or 10 fmol of helicase substrate. For incubations at 60 °C, the reaction was incubated in a heated-top PCR machine (PE9600) in order to prevent evaporation. The time course reactions were incubated layered with mineral oil in a heat block. The reactions were incubated for 30 min and stopped by the addition of 5 µl of a 5× stop solution (final concentrations: 0.5% SDS, 40 mM EDTA, 0.5 mg/ml proteinase K, 20% Ficoll, and 0.01% bromphenol blue) and then run on a 10% polyacrylamide TBE gel containing 0.1% SDS at a constant voltage of 150 V, soaked in 20% trichloroacetic acid, and exposed to a PhosphorImager screen (Molecular Dynamics). The reaction products were quantitated, and any free oligo in the absence of enzyme was subtracted.

dATPase/ATPase Assay-- The standard assay reaction (10 µl) contained 25 mM Tris-HCl, pH 7.8, 100 µM dATP, 5 nM [-³²P]dATP, 5 mM MgCl, 1 mM dithiothreitol, and 200 µg/ml bovine serum albumin. Incubations were performed as described for the helicase assay and stopped by the addition of 1 µl of 0.5 M EDTA. The reaction products (NTP, NDP, P_i) were separated by running 1 µl of the reaction on a polyethyleneimine-cellulose TLC plate (Selecto Scientific) in 0.8 M acetic acid and 0.8 M lithium chloride. The plate was air-dried and exposed to a PhosphorImager for imaging and quantification. Any background [-³²P]dADP released in the absence of enzyme was subtracted. The molar values were calculated by quantification of 1 µl of the reaction spotted on the TLC plate after the run.

Gel Filtration/Size Analysis-- A Superose 6 (APBiotech) column was equilibrated and run on an FPLC system in 25 mM Tris-HCl, pH 7.8, 10% glycerol and the indicated salt or EDTA concentration. The column was run at 0.4 ml/min at 4 °C. Injections were made in a volume of 200 µl. Fractions (14 drops, about 385-400 µl) were collected after 7 ml had flowed through the column after injection of the sample. Gel filtration markers (Bio-Rad), when indicated, were run in the same buffer conditions as the experiment.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

M. thermoautotrophicum H Contains a Single MCM Homologue

The complete sequence of M. thermoautotrophicum revealed the presence of one ORF (ORF 1770) that was similar to MCM protein sequences. When this ORF was compared with the entire protein data base using the BLAST software (51), it appeared that MtMCM was most closely related to the MCM4 branch of the MCM protein family (10, 12). Using the ClustalW 1.8 alignment program (52), we have compared the MtMCM protein sequence with the fission yeast Cdc21/MCM4 and Xenopus MCM4 sequences. This alignment is shown in Fig. 1. MtMCM lacks the N-terminal region present in all the eukaryotic proteins. MtMCM was found to be 35% identical with both S. pombe and Xenopus laevis sequences over the aligned regions lacking the N termini. Over the same region, the MtMCM protein was 49 and 48% similar to the S. pombe and X. laevis sequences, respectively. The highest homology was found in the core region of the proteins encompassing the putative nucleotide binding regions and ATPase motifs. Interestingly, the N-terminal region of eukaryotic MCM4 proteins contains putative consensus sites for phosphorylation by cyclin-dependent kinase (see Fig. 1) (12). These sites are absent from the MtMCM protein. We have underlined the Walker A and Walker B domains of the protein sequences, which also correspond to the conserved domains I and II (31) found in all known helicases (Fig. 1).

View larger version (91K): [in this window] [in a new window]   Fig. 1.   Multiple sequence alignment of MtMCM, SpMCM4, and XlMCM4 proteins. M. thermoautotrophicum H ORF 1770 was translated and named MtMCM. A ClustalW 1.8 alignment was made with S. pombe Cdc21/MCM4 (SpMCM4) and with X. laevis MCM4 (XlMCM4). Black boxes, amino acid identities; gray boxes, amino acid similarities. The Walker A and Walker B domains of the nucleotide binding region are underlined in black. The position of the lysine that was mutated to an alanine (K325A) is indicated above the MtMCM sequence. Asterisks indicate the positions of cyclin-dependent kinase putative phosphorylation sites in the yeast and Xenopus sequences.

Purification of MtMCM

In order to determine the role of MtMCM, we isolated the cDNA encoding for MtMCM from M. thermoautotrophicum genomic DNA (a generous gift from Dr. J. Reeve) and inserted it into two different expression vectors as fusion proteins containing a C-terminal His₆ tag (see "Experimental Procedures"). The protein was expressed in E. coli and in insect cells using a baculovirus expression system. In both cases, the protein was soluble and was purified under native conditions. Following a three-step purification procedure (see "Experimental Procedures"), the protein synthesized in E. coli migrated as a single band with an apparent molecular mass of 75 kDa, consistent with its predicted molecular mass of 76.5 kDa (Fig. 2A, left panel). The identity of the protein was further confirmed by Western blotting detection using an anti-His₆ tag antibody (data not shown and Fig. 3B). We also expressed MtMCM in E. coli in which the conserved lysine within the Walker A domain was replaced by an alanine, referred to as K325A-MtMCM (see "Experimental Procedures" and Fig. 1). The Coomassie stain of the SDS-PAGE of the three proteins is shown in Fig. 2A. No other bands were detected on the gels. MtMCM protein expressed from insect cells migrated as a doublet in which both polypeptides contained the His tag, since they were both recognized by an anti-His₆ tag antibody (data not shown).

View larger version (27K): [in this window] [in a new window]   Fig. 2.   MtMCM protein purification and helicase activity. MtMCM protein was overexpressed in and purified from both E. coli and baculovirus-infected insect cells. The K325A-MtMCM was overexpressed in and purified from E. coli, as described under "Experimental Procedures." A, Coomassie-stained SDS-PAGE gels showing the purified proteins as indicated. B, the MtMCM proteins (40 ng each, 26.5 nM monomer) were assayed for helicase activity, as described under "Experimental Procedures." Helicase activity was measured by the ability of 40 ng of each protein (26.5 nM monomer) to melt a 63-mer oligonucleotide from single-stranded M13 (10 fmol (0.5 nM) template) in the presence of 2.5 mM dATP and 5 mM MgCl2 for 30 min at 60 °C.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   MtMCM protein forms a stable high molecular weight complex. MtMCM protein was run on a native gradient polyacrylamide gel and through a Superose 6 gel filtration column in order to analyze its native conformation. A, the wild type protein was run through a native gradient (4-29% polyacrylamide) gel in parallel with markers (thyroglobulin (669 kDa), ferritin (440 kDa), catalase (232 kDa), lactase dehydrogenase (140 kDa), and albumin (67 kDa)) and stained with Coomassie Blue. A semilog plot of the migration distance of the markers allowed for calculation of an approximate molecular mass for the complex of 900-950 kDa. B, the wild type protein produced in E. coli and in insect cells was run through a native gradient gel and immunoblotted with an anti-His6 antibody. C, 150 µg of wild type MtMCM protein was injected onto a Superose 6 column equilibrated with 125 mM NaCl. Aliquots (20 µl) of the collected fractions were run on an SDS-PAGE gel and stained with Coomassie Blue, revealing a single peak of protein elution. Markers (thyroglobulin (670 kDa), -globulin (158 kDa), ovalbumin (44 kDa), myoglobin (17 kDa), and vitamin B12 (1.35 kDa)) were run through the column in the same buffer conditions. A semilog plot of the elution peak of the markers based on the A280 profile allowed for calculation of an approximate molecular mass for the complex of 900-950 kDa. D, 150 µg of wild type MtMCM protein was injected onto a Superose 6 column equilibrated with 750 mM NaCl. E, 150 µg of wild type MtMCM protein was injected onto a Superose 6 column equilibrated with 15 mM EDTA. F, 150 µg of K325A-MtMCM protein was injected onto a Superose 6 column equilibrated with 125 mM NaCl. C-F, the peak elution positions of the markers are noted with arrows above the gels.

MtMCM Protein Contains Intrinsic Helicase Activity That Requires an Intact Nucleotide-binding Domain

MCM proteins have been proposed to function as helicases during DNA replication. However, whether the helicase activity was intrinsic to the MCM proteins in eukaryotes has been somewhat controversial. In order to demonstrate unequivocally that the MtMCM protein has an intrinsic helicase activity and that this activity is not due to a contaminating activity, we assayed the protein purified from both prokaryotic and eukaryotic expression systems. By assaying for helicase activity in both baculovirus and E. coli expressed MtMCM, we ruled out the argument that the helicase activity is due to a contaminating activity that copurifies with the protein. Furthermore, we introduced a lysine to alanine point mutation in the MtMCM protein within the Walker A domain (K325A-MtMCM). All Walker A domains contain a conserved lysine shown to be essential for nucleotide binding.

We then assayed the three different purified proteins shown in Fig. 2A for DNA helicase activity. In this assay, a radiolabeled 63-mer oligonucleotide was annealed to single-stranded M13 DNA, and the helicase activity was measured by the ability to melt or unwind the labeled oligonucleotide from M13 circular DNA (see "Experimental Procedures"). Both wild-type MtMCM proteins, expressed in either E. coli or in baculovirus-infected cells, displayed DNA helicase activity, while no helicase activity was detected in the K325A-MtMCM protein (Fig. 2B). This demonstrates unambiguously that the helicase activity was indeed intrinsic to MtMCM protein and not due to a contaminating protein co-purifying with MtMCM. Furthermore, these data indicate that this is a Walker A type helicase, requiring a nucleotide triphosphate for activity.

MtMCM Proteins Assemble into Large Complexes That Contain the Helicase Activity

Eukaryotic MCM proteins have been shown to assemble into large multimeric complexes. In order to assess the size of the MtMCM proteins purified under native conditions, we used nondenaturing gradient gel electrophoresis (Fig. 3A), a method that proved to be very valuable to resolve MCM protein complexes (53, 54), or gel filtration followed by SDS-PAGE (Fig. 3C). With either method, all of the purified MtMCM was determined to exist as a large complex (Fig. 3). The estimated molecular mass of this complex was determined to be 900-950 kDa following nondenaturing gel electrophoresis or gel filtration. This size is consistent with MtMCM forming a dodecamer that would have a predicted mass of 918 kDa. Both the baculovirus and E. coli expressed MtMCM protein assembled into complexes of similar size (Fig. 3B). The K325A-MtMCM protein also assembled into complexes of similar size (Fig. 3, C and F).

The stability of this large complex was tested by flowing the wild-type protein expressed in E. coli through a Superose 6 column in the presence of 125 mM NaCl, 750 mM NaCl, or 15 mM EDTA. In all three conditions, the vast majority of the protein eluted as a complex of about 900 kDa, and no monomeric form of MtMCM was observed. In the presence of either 750 mM NaCl or 15 mM EDTA, we observed a slight shift toward the lower molecular mass that could be consistent with a fraction of the dodecameric form dissociating into single hexamers (Fig. 3, D and E). In these conditions, we did not observe the dissociation of the complexes into monomeric MtMCM. Also, these studies demonstrated that the absence of helicase activity in the K325A-MtMCM protein (Fig. 2B) was not due to the inability of the protein to form large molecular weight complexes, since K325A-MtMCM protein was found to exist exclusively as a complex of ~900 kDa (Fig. 3F).

We next tested the fractions across the Superose-6 purification profile (Fig. 4A) for both helicase (Fig. 4B) and DNA-stimulated dATPase (Fig. 4C) activities (see "Experimental procedures"). The peak of helicase and dATPase activities primarily coincided with the protein elution profile of the MtMCM protein complex, in fractions 10-18. In this experiment, we analyzed the ability of MtMCM to displace a ³²P-labeled 63-mer labeled oligonucleotide from an M13 ssDNA circle. DNA-stimulated ATPase activity also coincided with the protein profile of MtMCM (data not shown).

View larger version (24K): [in this window] [in a new window]   Fig. 4.   Co-elution of helicase and NTPase activity with the high molecular weight MtMCM complex on a Superose 6 column. 150 µg of MtMCM protein was loaded onto a Superose-6 sizing column as described for Fig. 2C. A, every third fraction was run on an SDS-PAGE gel (20 µl aliquot) and stained with Coomassie. The Load lane contained 4 µg of MtMCM protein. The peak stained fractions coincided with the peak on the recorded A280 reading (not shown). B, every third fraction (1 µl) was subjected to a helicase assay (5 fmol of template), as described under "Experimental Procedures." The Substrate lane contained 5 fmol of the template without any added protein, incubated in the same conditions as the rest of the fractions. The Boiled lane contained 5 fmol of template heated to 100 °C for 5 min. C, quantification of the helicase (solid line) and dATPase (dotted line) assays shown along with the position of the peak elution of the molecular weight markers, as determined by the A280 profile.

Biochemical Properties of the MtMCM Helicase

Nucleotide Requirements-- All helicases contain a conserved nucleotide-binding motif. Helicase-mediated DNA unwinding might require both nucleotide binding and hydrolysis. We tested the helicase activity of the MtMCM protein in the presence of a variety of nucleotides and nucleotide analogues. As shown in Fig. 5, only ATP and dATP were able to efficiently support DNA helicase activity catalyzed by MtMCM. All other nucleotides and deoxynucleotides tested were not efficient co-factors. Interestingly, the MtMCM DNA helicase activity was abolished in the presence of AMPPNP and was greatly diminished but not completely abolished in the presence of ATPS, suggesting that both ATP binding and hydrolysis are required for the unwinding reaction. We found that dATP worked more efficiently than ATP in our helicase assays (Fig. 5); therefore, most of the experiments presented were carried out using dATP. No ATPase activity was detected in K325A-MtMCM (data not shown), which also lacked detectable helicase activity.

View larger version (10K): [in this window] [in a new window]   Fig. 5.   Properties of MtMCM helicase: nucleotide effects. The bar graph shows nucleotide energy requirements for MtMCM helicase activity. Each reaction contained 5 fmol of helicase substrate, 26.5 nM MCM monomer, and a 2.5 mM concentration of the indicated NTP. The activity is represented as the total amount of oligonucleotide melted from M13. The experiments using dATP and ATP were performed in triplicate.

Enzyme Titration and Kinetics-- We then determined the minimal concentration of enzyme sufficient for unwinding activity by titrating down the amount of MtMCM protein at given concentrations of dATP (2.5 mM) and oligonucleotide substrate (0.25 nM). The helicase activity was found to be mostly linear over the range of 0.5-35 nM (Fig. 6, A and B).

View larger version (41K): [in this window] [in a new window]   Fig. 6.   Titration of MtMCM protein in the helicase assay and temperature effect upon the kinetics of MtMCM helicase activity. MtMCM protein was titrated in the helicase assay by serially diluting the protein in 3-fold steps from 167 ng/ml (110 nM) to 0.4 ng/ml (0.5 nM). 1 µl of each dilution was assayed for helicase activity in a volume of 20 µl with 10 fmol (0.25 nM) of the standard substrate for 30 min. A, helicase activity of the titrated protein. B, quantification of the helicase activity, performed in duplicate, plotted versus the concentration of the MtMCM monomer. C and D, the kinetics of MtMCM helicase activity was examined at three temperatures: 21, 37, and 60 °C. In each case, 85 fmol of helicase template (0.5 nM) and 344 ng of MtMCM (26.5 nM) were incubated in the standard reaction buffer in a volume of 170 µl. At the indicated times, a 20-µl sample was removed into 5 µl of the stop solution. C, the time course at the various temperatures, showing the melting of the oligonucleotide from the M13 ssDNA. The Substrate lane contained 10 fmol of template without any added enzyme incubated in the same conditions as the rest of the points. The Boiled lane contained 10 fmol of template heated to 100 °C for 5 min, fully denaturing the template. D, quantification of the helicase assays at 21 °C (), 37 °C (), and 60 °C ().

Since M. thermoautotrophicum is a thermophilic microorganism with an optimal growth temperature of 65 °C (55), we examined the kinetics of helicase activity at different temperatures. Using a 63-mer oligonucleotide annealed to M13, we determined the kinetics of DNA unwinding at 21, 37, and 60 °C for a given amount of enzyme (26.5 nM). We used 60 °C as the highest temperature for helicase assays to circumvent possible thermal melting of the oligonucleotide at higher temperatures. As seen in Fig. 6, C and D, a dramatic temperature effect was observed with no melting occurring at 21 °C and a 7-fold lower helicase activity seen at 37 °C compared with 60 °C. Interestingly, a lag period with no unwinding was consistently observed during the first minutes of the assay.

DNA Requirements for NTPase-- Our data show that MtMCM helicase activity requires nucleotide binding and hydrolysis that is DNA-dependent. We analyzed the DNA requirement for this nucleotide hydrolysis. The data for dATPase activity are shown in Fig. 7. Similar data were obtained for ATP hydrolysis (data not shown). The quantification of dATP released in the presence or in the absence of DNA was determined. Although some weak dATPase activity was observed at high MtMCM protein concentration in the absence of DNA (Fig. 7D), a 7-fold increase in dATPase was observed in the presence of either ssDNA or dsDNA (Fig. 7D). At low molarity of enzyme (25 nM), in the linear range of activity, this difference was even more dramatic (Fig. 7B). Both ssDNA and dsDNA were found to be equally effective at stimulating the dATPase activity when assayed for its kinetics or in a titration assay.

View larger version (51K): [in this window] [in a new window]   Fig. 7.   DNA dependence of MtMCM dATPase activity. MtMCM protein was titrated in the dATPase assay by serially diluting the protein in 3-fold steps from 167 ng/ml (220 nM) to 0.4 ng/ml (1 nM). Each concentration was assayed in the presence of various DNAs: single-stranded M13mp18, double-stranded plasmid DNA, and a 63-mer oligonucleotide. A, dATPase activity of the titrated protein (1 µl) incubated in a 10-µl reaction volume in the presence of 31.25 ng of M13mp18, 5 ng of 63-mer, 31.25 ng of plasmid, or no DNA. B, quantification of the molar quantity of hydrolyzed dATP in the presence of M13 (), 63-mer (), plasmid (), or no DNA (). C, time course of MtMCM dATPase activity. 120 ng of MtMCM protein (20 nM) was incubated in an 80-µl reaction volume, as described under "Experimental Procedures," in the presence of 250 ng of M13mp18, 10 ng of 63-mer, 250 ng of plasmid, or no DNA. At the times indicated, 10-µl aliquots were removed to 1 µl of 0.5 M EDTA to stop the reaction. D, quantification of the molar quantity of hydrolyzed dATP in the presence of M13 (), 63-mer (), plasmid (), or no DNA ().

Directionality-- Two classes of DNA helicase have been described that translocate either in the 5'  3' direction, such as DnaB (56), or that translocate in the 3'  5' direction, such as the hexameric SV40 large T antigen helicase (57). We determined the directionality of the MtMCM helicase by using asymmetrical templates. In this experiment, a 5'-labeled 63-mer oligonucleotide containing a blunt-end restriction site was annealed to M13. The annealed template was digested and then labeled at its 3'-end, giving rise to an asymmetrical template containing both a 3'-labeled and a 5'-labeled oligonucleotide of 40 and 25 nucleotides in length, respectively (see Fig. 8A and "Experimental Procedures"). Fig. 8B shows that only the oligonucleotide labeled at its 5'-end was melted, establishing that the enzyme was moving in the 3'  5' direction on the translocating strand, assuming that MtMCM loads on the ssDNA side.

View larger version (35K): [in this window] [in a new window]   Fig. 8.   Directionality and processivity of the MtMCM helicase activity. MtMCM protein was assayed for directionality by the use of a modified template and for processivity by using an extended template. A, the directionality template was prepared as shown and as described under "Experimental Procedures." B, the directionality template (5 fmol) was incubated at 60 °C in the absence of added protein (Substrate lane) and with the addition of 40 ng of MtMCM (26.5 nM monomer). The positions of the two possible melted products are indicated on the right. C, the standard template was extended with Klenow in the presence of all four dNTPs for the times indicated. Those progressively longer templates were used to assay MtMCM activity. The lanes labeled Substrate pool and Boiled were pools of these extended templates from the same experiment, incubated without protein and heated to 100 °C, respectively.

Processivity-- In a previous study using purified eukaryotic MCM proteins, helicase activity was described as having very low processivity and only unwinding oligonucleotides of 20-mer in length (33). We felt it was very important to establish the processivity of the MtMCM helicase to assess whether it could act as a replicative helicase. In the experiments described above, we have used synthetic oligonucleotides of various lengths, up to 63-mer, that were efficiently melted by the MtMCM helicase. In order to estimate the maximum oligonucleotide length that could be processed by the enzyme, we prepared templates of variable length in which the labeled nucleotide was extended by Klenow DNA polymerase. Our results showed that using 40 ng of enzyme, oligonucleotides of at least 500 base pairs were melted (Fig. 8C). This high processivity is compatible with MtMCM being a replicative helicase.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In this report, we show that the single MCM protein from a thermophilic microorganism has intrinsic helicase and DNA-dependent NTPase activities and that the enzymatic activities reside in a large multimeric complex.

Sequence analysis of several archaea genomes has shown that they contain different numbers of MCM-related sequences (10, 12), with at least two species, M. thermoautotrophicum and A. fulgidus, containing a single MCM ORF. In both cases this putative MCM protein is most closely related to the eukaryotic MCM4 protein (see Fig. 1 and data not shown). MCM4 is a target for multiple cell cycle-regulated phosphorylation (53, 58), and all eukaryotic MCM4s contain multiple putative cyclin-dependent kinase phosphorylation sites clustered in the N-terminal region of the protein (Fig. 1 and data not shown). Interestingly, none of the archaea sequences that we have analyzed (M. thermoautotrophicum, A. fulgidus, M. jannaschii, Aeropyrum pernix, and Sulfolobus solfataricus) contain cyclin-dependent kinase phosphorylation sites. One interpretation of these observations is that during evolution MCM-dependent DNA replication appeared before cell cycle regulation by cyclin-dependent kinases, since archaea do not contain cyclin-dependent kinases.

MtMCM as a Replicative Helicase-- The most important finding in this report is that a single MCM protein contains intrinsic DNA helicase activity. This was demonstrated by showing that highly purified MtMCM protein overexpressed from either a prokaryotic or a eukaryotic expression system displayed helicase activity that co-purified with the protein. In addition, we establish unambiguously that the helicase activity is intrinsic to the MtMCM protein by showing that a single amino acid mutation in the Walker A domain of the MtMCM protein completely abolishes the MtMCM-associated helicase activity.

This is in agreement with previous findings that MCM proteins, purified from mammalian cells, have associated helicase activity (33). However, the helicase activity of mammalian MCM proteins exhibited properties different from the MtMCM helicase. Mammalian MCM proteins have been reported to melt only short oligonucleotides (less than 30-mer), while we show that MtMCM protein can melt oligonucleotides of at least 500 nucleotides in length. Additionally, MtMCM is able to displace a 63-mer oligonucleotide when present at equimolar ratio with the oligonucleotide (considering that MtMCM forms hexamers or double hexamers), while the recombinant hexameric MCM proteins from mammals were reported to melt a 17-mer oligonucleotide only when present in a 40-fold molar excess of hexamers to the oligonucleotide (34).

We find that MtMCM helicase unwinds the DNA in the 3'  5' direction, similarly to the mammalian MCM proteins. This is the directionality of SV40 large T antigen and polyoma PyV large T antigen, two viral hexameric DNA replicative helicases (42, 57, 59).

The data presented in this report strongly support the idea that MtMCM could function as a replicative helicase because of its high processivity and its catalytic properties. Two recent reports on the characterization of the MtMCM-associated helicase activity (60, 61) describe observations that are generally similar to those reported here (see below). In particular, our observation that MCM helicase can unwind up to 500 base pairs is entirely consistent with it being a replicative helicase requiring such high processivity.

The fact that a single amino acid mutation in the conserved lysine within the Walker A domain completely abolishes helicase activity strongly suggests that nucleotide binding is essential for helicase activity. The Walker A domain of the NTPase region has been shown to be important for nucleotide binding in several enzymes (62-65), while the B domain was shown to be important for nucleotide hydrolysis by site-directed mutagenesis or through structural studies (63-65). Interestingly, our analysis of nucleotide requirements for MtMCM helicase activity suggests that nucleotide binding and nucleotide hydrolysis might play distinct functions. Indeed, the helicase activity of the wild-type MtMCM protein was completely abolished in the presence of AMPPNP, a nonbinding ATP analogue, while some low helicase activity was observed in the presence of ATPS, an ATP analogue that can bind but is not hydrolyzed. This concurs with the previously proposed mechanism of helicase activity in which nucleotide binding promotes first a conformational change followed by a subsequent change associated with nucleotide hydrolysis (43). We also observe that MtMCM uses both ATP and dATP, with a significant preference for dATP. Another helicase from calf thymus has also been shown to use dATP preferentially over other nucleotides (66).

Mutation in the conserved Lysine residue in the Walker A domain of mouse MCM6 (34) resulted in a MCM complex with residual helicase activity. In order to detect any possible residual helicase activity in K325A-MtMCM, we tested very high concentrations of protein (up to 300 nM) but never observed helicase activity (data not shown). This difference might be due to the fact that in the mouse MCM complex both MCM4 and MCM7 had intact Walker A domains that could account for the residual helicase activity.

Our data strengthen the previous hypothesis that MCM proteins are a helicase and could be a replicative helicase. The current data on mammalian MCM proteins indicate that only a subset of the MCM proteins (33) display helicase activity in vitro (MCM4, -6, and -7), while MCM2 is inhibitory (35). In this context, the essential in vitro function of MCM3 and MCM5 has yet to be determined.

MtMCM Complex Formation and Assembly-- Another important observation is that all of the MtMCM protein assembles into large oligomeric complexes. We did not observe the presence of monomeric MtMCM following purification of the overexpressed protein in insect cells or in bacteria. Additionally, K325A-MtMCM was also found entirely in a complex. This differs from the work by Kelman et al. (60), who observed monomeric MtMCM. This discrepancy could be due to the position of the His₆ tag, which we positioned at the C terminus of the protein as opposed to the N-terminal His₁₀ tag used by Kelman et al. (60). The other difference is that they purified the protein following its denaturation by urea, while we purified the native protein.

Our observations of the molecular mass of the MtMCM complex in low salt conditions, using two different methods, show that the protein is present in a dodecameric form, consistent with being double hexamers. These hexamers appear very stable, since they do not dissociate in NaCl up to 750 mM NaCl. In addition, the stability of this oligomeric complex does not require divalent cations, since it is stable in the presence of EDTA. However, we observe a slight broadening of the elution peak of the complex in the presence of EDTA or high salt that could reflect the transition of a fraction of the protein from a dodecameric or double hexamer form to a hexameric form. The structure of the hexameric T7 helicase was recently solved, showing that the monomers arrange into a helical filament resembling a ringlike structure when viewed along the axis of the filament (43). Although there is no information about the possible structure of the MtMCM helicase, if it were such a helical oligomeric filament, it is easy to imagine that it could exist either as a single or a double hexamer moving along the DNA.

Several observations suggest that assembly and multimerization of MtMCM are essential steps for activity. In time course experiments, we always observe a lag phase between the onset of the reaction and the actual detection of helicase activity (Figs. 6D and 7D). This could reflect the necessity for disassembly of a preformed complex and its reassembly around the DNA. Additional support for this idea comes from competition experiments in which we added unlabeled template along with the labeled template either at the start of the helicase assay or 5 or 10 min following initiation of the assay. While adding an excess of unlabeled competitor template at 10 min had little effect on the amount of labeled template melted by the helicase, adding the competitor template at 5 or 0 min resulted in up to an 80% reduction of the amount of labeled template melted by the helicase (data not shown).

We have also observed by gel shift assay that MtMCM can directly bind to a 63-mer oligonucleotide when present in a 3-8-fold excess of the double hexameric form over the oligonucleotide (data not shown), suggesting that the interaction between the MtMCM complex and the DNA is a stable one, in agreement with previous work showing the DNA binding ability of MCM proteins (33, 60). The assembly of the complex on the DNA is probably an essential step for the unwinding of the DNA, since in addition to bringing the enzyme in contact with its substrate, the NTPase activity of the enzyme is stimulated. This stimulation might correspond to some conformational changes in the complex. We show that both ssDNA and dsDNA equally stimulate the NTPase activity. Whether the stimulation by dsDNA is due to single strand regions in the duplex arising at 60 °C or true stimulation by dsDNA remains to be clarified.

	ACKNOWLEDGEMENT

We thank Dr. G. Freyer for helpful discussions and comments on the manuscript.

	FOOTNOTES

^* This work was supported in part by American Cancer Society Grant RPG-99-040-01-CCG (to J. G.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Supported by a Cancer Biology Training Grant predoctoral fellowship.

^¶ To whom correspondence should be addressed: Dept. of Genetics and Development, Columbia University, VC15-1526, 630 W. 168th St., New York, NY 10032. Tel.: 212-305-9586; Fax: 212-305-7391; E-mail: jg130@columbia.edu.

Published, JBC Papers in Press, March 9, 2000, DOI 10.1074/jbc.M000398200

	ABBREVIATIONS

The abbreviations used are: MCM, minichromosome maintenance protein; MtMCM, M. thermoautotrophicum MCM protein; ORF, open reading frame; PCR, polymerase chain reaction; PAGE, polyacrylamide gel electrophoresis; ssDNA, single-stranded DNA; dsDNA, double-stranded DNA; ATPS, adenosine 5'-O-(thiotriphosphate); AMPPNP, 5'-adenylyl-,-imidodiphosphate.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES