Double Hexamer Disruption and Biochemical Activities of Methanobacterium thermoautotrophicum MCM*

Methanobacterium thermoautotrophicum MCM (mtMCM) is a helicase required for DNA replication. Previous electron microscopy studies have shown mtMCM in several oligomeric forms. However, biochemical studies suggest that mtMCM is a dodecamer, likely a double hexamer (dHex). The crystal structure of the N-terminal fragment of mtMCM reveals a stable dHex architecture. To further confirm that the dHex is not an artifact of crystal packing of two hexamers, we investigated the relevance of the dHex by disrupting the hexamer-hexamer interactions seen in the crystal structure via site-directed mutagenesis and examining various biochemical activities of the mutants in vitro. Using a combination of biochemical and structural assays, we demonstrated that changing arginine to alanine at amino acid position 161 or the insertion of a six-aminoacid peptide at the hexamer-hexamer interface disrupted dHex formation and produced stable single hexamers (sHex). Furthermore, we showed that the sHex mutants retained wild-type level of ATPase and DNA binding activities but had decreased helicase activity when compared with the wild type dHex protein. These biochemical properties of mtMCM are reminiscent of those of SV40 large T antigen, suggesting that the dHex form of mtMCM may be the active helicase for DNA unwinding during the bidirectional DNA replication.

mtMCM complex assembly and function in the DNA replication process as a replication helicase. Results of a crystallographic study have revealed that the N-terminal portion of mtMCM (N-mtMCM) assembles as a double hexamer (dHex) (11).
The idea that an active replication helicase has a dHex structure is intriguing. The head-to-head (N terminus to N terminus) conformation of two hexamers, as observed in the N-mtMCM dHex structure, seems awkward at first. Nevertheless, the dHex architecture may make sense from an organizational standpoint inside a cell. One might imagine a fixed point of replication where the helicase pulls double-stranded DNA (dsDNA) toward it from both directions, thereby eliminating the need to coordinate the movement of the vast number of essential proteins involved in replication. Indeed, replication foci attached to the nucleomatrix have been observed in cells (12).
A viral replicative helicase, SV40 large T antigen (LTag), is a functional homolog of the cellular MCM. LTag also functions as a helicase for unwinding replication forks. The single hexameric form (sHex) of LTag has helicase activity in vitro; however, the helicase activity of dHex is ϳ10 -15 times higher than that of the sHex form (13,14). Ample evidence suggests that LTag dHex formation is essential for SV40 DNA replication (15)(16)(17). The high-resolution structures of the helicase domain of LTag reveal the presence of a large chamber inside the helicase domain with outlet side channels. These structural features, together with previous biochemical/genetic data prompted us to promote a looping model for dsDNA unwinding by LTag dHex (18 -20) in which strand separation occurs within the large chamber when DNA is pulled inside the dHex helicase and the separated single-stranded DNA (ssDNA) extrudes from the side channels. This process is triggered by ATP binding and hydrolysis. With the availability of the multiple crystal structures of LTag in different nucleotide binding states, we now have a good understanding of the mechanisms by which the ATP binding and hydrolysis trigger the conformational changes of LTag hexameric helicase that couple the energy of ATP hydrolysis to the dsDNA unwinding (18,21).
To date, the mtMCM is the only cellular replicative helicase reported to form a dHex structure (7)(8)(9)11), 3 although a recent model for another archaeal MCM is suggested to also function as a dHex (23). The availability of the N-mtMCM crystal structure and the detailed interactions known to occur at the hexamer-hexamer interface present a unique opportunity for investigating the in vitro biochemical activity of the dHex and sHex forms of mtMCM through structure-guided mutagenesis. Here we report the results of the mutational and functional studies. Specifically, we generated mutants designed to disrupt the hexamerhexamer interactions and obtained stable sHex mutants. In addition, we compared the biological activities of these sHex mutants to those of the * This work is supported by Grant R01-AI055926 from the National Institutes of Health and an ACS grant (to X. S. C.), the Spanish Ministerio de Educació n y Ciencia (BFU2004-00217/BMC), Comunidad Autó noma de Madrid (07B-0032; GR/SAL/0645/ 2004), National Institutes of Health (1R01HL70472), European Union (FP6 -502828), and Fondo de Investigaciones Sanitarias (G03-185) (to J. M. C.), and Fundació n BBVA (2004X578) (to C. S. M.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondence should be addressed: wild type (wt) dHex forms, including ATPase, DNA binding, and helicase activities.
Protein Purification and Native Gel Shift Assay-All proteins were expressed in Escherichia coli cells at 24°C overnight using isopropyl 1-thio-␤-D-galactopyranoside at a concentration of 0.2 mM. Cell pellets were lysed by sonication in purification buffer (50 mM Tris, pH 8.0, 250 mM NaCl, 10 mM DTT) containing the protease inhibitors pepstatin, leupeptin, and phenylmethylsulfonyl fluoride. The GST-MCM fusion protein in the supernatant was purified using glutathione resin (Amersham Pharmacia Biotech) at 4°C. The mtMCM proteins were cleaved from the glutathione S-transferase fusion using thrombin and then further purified by gel filtration FPLC using a pre-equilibrated preparation grade Superose 6 column (16/60 cm) or preparation grade Superdex-200 column (16/60 cm). Peak fractions were collected and concentrated using Amicon Ultra (Millipore) centrifuge devices and quantified first using a Bio-Rad protein assay and then by Coomassie Blue staining of SDS-PAGE gels. For native gel shift assays, 6% polyacrylamide gels were electrophoresed at 4°C in Tris/glycine buffer (25 mM Tris, pH 8.0, 250 mM glycine) for 60 min at 200 volts.
Electron Microscopy-wt and mutant N-mtMCM (100 and 250 g/ml in 50 mM Tris⅐HCl, pH 8.0, 250 mM NaCl, 10 mM DTT) were incubated for 1 h at 37°C. FL-mtMCM mutants were diluted to 20 g/ml in 50 mM Tris⅐HCl, pH 8.0, 500 mM NaCl, 3 mM DTT. All samples were adsorbed to glow-discharged, collodion/carbon-coated copper grids and stained with 2% uranyl acetate. Grids were examined in a JEOL 1200 EX-II transmission electron microscope at 80 kV and 60,000ϫ magnification. Low dose micrographs on Kodak SO-163 plates were digitized in a Zeiss SCAI scanner with a pixel size of 7 m, then binned to 21 m (3.5 Å in the sample). Image processing was carried out with XMIPP (24,25). Manually picked particles were extracted in 64 ϫ 64 pixel frames, normalized, centered, and aligned in two dimensions using a reference-free method (26). Classification was performed on the centered particles using neural network-based techniques (27). Images in the resulting groups were realigned before calculating averages. Twodimensional resolution was calculated with the spectral signal to noise ratio method with the threshold set at 4 (28). 6-Fold symmetry was imposed on the end-on view average images, as this was the predominant symmetry in their rotational spectra.
ATPase Assay-10-l reactions were assembled on ice containing 50 mM NaCl, 5 mM MgCl 2 , 50 mM Tris, pH 7.8, 1 mM DTT, 0.1 mg/ml bovine serum albumin, [␣-32 P]ATP (Amersham, 3000 Ci/mmol), and various amount of protein. After incubation at 50°C for 30 min, the reactions were stopped by adding 10 mM EDTA. A 5-l aliquot from each reaction was applied to a prewashed PEI cellulose TLC plate (EMD Chemicals Inc.), dried, and run for 2 h in 2 M acetic acid and 0.5 M LiCl. Plates were dried, autoradiographed using phosphorimaging plates, and quantified.
DNA Binding Assays-All primers (Operon) were resuspended in 10 mM NaOH and purified using a Mono Q 10/100 (Amersham Pharmacia Biotech) column by FPLC. Purified DNA was concentrated and quantified using A 260 . For the double strand assay, complementary primers were mixed, boiled for 10 min in 900 ml of water, and allowed to cool to room temperature overnight. The annealed dsDNA was then purified using FPLC over a Superdex-75 16/60 (Amersham Pharmacia Biotech) column in 50 mM NaCl. The DNA used for the single strand binding was 5Ј-AAAGCGCTGACCTATCGCGTATAGCTCGAGGA-3Ј and the complementary oligonucleotides used for the double strand assay were 5Ј-GCGCTGACCTATCGACCTATACGGTTAGCC and 5Ј-GGCT-AACCGTATAGGTCGATAGGTCAGCGC. The primers were labeled using T4 polynucleotide kinase (NEB) and [␥-32 P]ATP (Amersham, 3000 Ci/mmol). The labeled primer was purified using Microspin G-25 columns (Amersham Pharmacia Biotech) and stored at 4°C. 20-l reactions were assembled at room temperature with 50 mM NaCl, 5 mM MgCl 2 , 50 mM Tris, pH 7.8, 1 mM DTT, 0.1 mg/ml bovine serum albumin, 15 nM DNA, and various amount of mtMCM protein. Reactions were incubated at 25°C for 30 min and then 5 l of loading buffer (50% glycerol, 0.1% xylene cyanol, and 0.1% bromphenol blue) were added. Reactions were loaded onto a 2% agarose gel and electrophoresed for 1 h at 80 volts in 1ϫ TBE buffer. Gels were dried and exposed to film and phosphorimaging plates for quantification.
Helicase Assay-The helicase assay was performed essentially as described in Ref. 10. Briefly, a 60-nucleotide primer, 5Ј-TTTTTTTTT-TTTTTTTTTTTTTTTCGCGCGGGGAGAGGCGGTTTGCGTA-TTGGGCGCC (Operon), was labeled with [␥-32 P]ATP and then annealed to M13mp18 (NEB) ssDNA in a 2:1 ratio in 300 mM NaCl and 20 mM Tris, pH 8.0, resulting in a 34-base-pair dsDNA region and a 26-nucleotide 3Ј-overhang. The dsDNA substrate was then purified using 700 l of Sephacryl 300HR resin in a 2-ml glass column (Bio-Rad). 20-l reactions containing 20 mM Tris, pH 7.8, 10 mM MgCl 2 , 1 mM DTT, 5 mM ATP, 0.1 mg/ml bovine serum albumin, 0.25 nM dsDNA substrate, and various amount of mtMCM protein were assembled on ice and then incubated at 50°C for 30 min, followed by the addition of 5ϫ stop solution (10 mM EDTA, 0.5% SDS, 0.1% xylene cyanol, 0.1% bromphenol blue, and 50% glycerol). The reactions were electrophoresed for 50 min at 150 volts on a 12% acrylamide gel in 1ϫ TBE on ice. The gel was dried and then exposed to film and phosphorimaging plate for quantification.

RESULTS
Mutations, Protein Expression, and Purification-The sHex mutants that specifically disrupt dHex formation were designed based on the crystal structure of the N-mtMCM (11). The interface between the two hexamers in the dHex structure of N-mtMCM is extensive and is mediated by both main chain packing and side chain interactions through the amino acids located within the zinc binding domain (Fig. 1A). One of the amino acid residues, Arg-161 (Fig. 1B), forms five hydrogen bonds (H-bonds) at the interface. Because there are 12 Arg-161 residues in a dHex complex, Arg-161 is predicted to play an important role in the hexamer-hexamer interactions.
To disrupt formation of the dHex, we generated two mutants: a point mutation, R161A, and a six-amino-acid insertion mutant (6Ins) in which the amino acid sequence GSGSGG was inserted between residues Cys-158 and Gly-159. The six-amino-acid peptide in the 6Ins mutant was designed to be positioned at the interface between the two hexamers, providing a short bulge of six amino acids that should hinder the ability of Arg-161 to H-bond to the opposing hexamer, thereby achieving an effect similar to that of the R161A mutation. R161A and 6Ins were constructed not only in the context of the complete protein, FL-mtMCM, but also in the N-terminal construct, N-mtMCM, used for crystallographic studies. Because N-mtMCM alone forms dHex, the parallel mutations on N-mtMCM can provide an additional line of evidence for disruption of the dHex formation. The six constructs used for this study are listed in Fig. 1C. All six proteins were expressed and purified to near homogeneity using the conditions described herein and as shown in Fig. 2, A and B. Disruption of dHex by Mutations-To evaluate the ability of the R161A and 6Ins mutations to disrupt dHex formation in both the N-mt-MCM and FL-mtMCM constructs, we used a combination of three methods, gel filtration chromatography, native gel shift, and electron microscopy (EM), to analyze the oligomeric states of the proteins.
Using gel filtration chromatography, we found that the elution profile of wt N-mtMCM had a peak with an apparent molecular mass of ϳ380 kDa, consistent with the size of a dodecamer (or dHex). In contrast, the elution peaks for the two mutants of N-mtMCM displayed a clear shift toward an apparent molecular mass of ϳ190 kDa (Fig. 2C), consistent with the size of a hexamer. Similarly, the elution peak for FL-mtMCM had an apparent molecular mass of ϳ960 kDa, consistent with a dodecamer assembly, whereas the two mutants eluted at peaks with an apparent molecular mass of ϳ450 -500 kDa (Fig. 2D), consistent with sHex. The identity of the proteins from all the peaks was confirmed by SDS-PAGE analysis (Fig. 2, A and B). The wt proteins of both FL-mtMCM and N-mt-MCM behaved as dodecamers and the mutant proteins as hexamers.
To further analyze the disruption of dHex assembly in mutants, proteins from all six mtMCM constructs were subjected to a native gel shift assay. Despite the use of a wide variety of buffer and salt conditions, the three FL-mtMCM (wt and two mutants) proteins did not show a distinct band pattern on the native gel shift assays. However, the N-mt-MCM constructs displayed clear band patterns in the native gel shift assays (Fig. 2E). The two mutant proteins migrated faster than the wt (Fig. 2E, lanes 1-3), suggesting that dHex assembly was disrupted. The faster migration of R161A on native gel could also be caused by the elimination of the positively charged arginine residues (6/hexamer). To examine the charge effect on native gel migration, we used another mutant with two arginine residues (Arg-227 and Arg-230) mutated to alanine as a control (Fig. 2E, lane 4). The mutations were not expected to affect dHex formation because both residues are away from the hexamer-hexamer interface. This mutant behaved like a wt dHex in gel filtration (data not shown) and migrated in a manner similar to that of the wt dHex protein in native gel (Fig. 2E), demonstrating that the loss of two arginines in one molecule (or 12/hexamer) was not sufficient to change the migration in the native gel shift assay. Thus, the faster migration of R161A is not the result of charge changes but likely because of disruption of the hexamer-hexamer interactions needed for dHex formation.
We used EM to directly visualize the presence of dHex in the wt and mutant proteins for both FL-mtMCM and N-mtMCM. We will focus on the data of the N-mtMCM proteins here because the availability of the crystal structure of this construct allows the direct comparison of the EM results with the atomic model (Fig. 3). For the wt N-mtMCM, images corresponding to the side view of the dHex (Fig. 3A) and to the end-on view looking down the channel of the complex (Fig. 3B) were observed. However, for the R161A and 6Ins mutants, no dHex side images were present, and somewhat different end-on views were observed (Fig. 3, C and D). Notably, the difference in the chirality of the end-on average images of the wt and the mutants is striking (Fig. 3 compare B with C and D). The end-on view of the wt dHex has no chirality (Fig. 3B), as was the case for an end-on projection of a low pass-filtered map calculated from the dHex crystal structure (Fig. 3E). In contrast, the end-on views of the two mutants have obvious chirality (Fig. 3, C and D), similar to the chirality seen in a projection of the low pass-filtered map of a sHex atomic model from the crystal structure (Fig. 3F). Fig. 3G summarizes the presence or absence of side views and FIGURE 1. The N-mtMCM dHex structure. A, ribbon diagram of the dHex structure viewed from the side. The two independent hexamers are colored yellow and blue. The 12 zinc atoms, each bound by one monomer, are shown as purple spheres. B, close-up view of the hexamer-hexamer interaction interface. The two monomers are colored in gray and cyan for one hexamer and wheat and magenta for the second hexamer. The zinc atoms are shown as green spheres. The Arg-161 (R161) from each monomer (12 in total) has the capacity to form five hydrogen bonds at the double hexamer interface. C, the six constructs used in this study, three for FL-mtMCM and three for N-mtMCM. DECEMBER 23, 2005 • VOLUME 280 • NUMBER 51 chirality in wt and mutant N-mtMCM complexes. These EM results further demonstrated that R161A and 6Ins mutants disrupted dHex formation of the mtMCM complex.

Structure and Function of an Archaeal MCM
ATP Hydrolysis-ATPase assays were done only with FL-mtMCM because the FL constructs contain the ATPase domain and the N-mt-MCM constructs lack this domain. For this reason, the DNA binding and helicase assay described below were also performed only on the FL-mtMCM constructs. The ATPase assays showed that the two mutant FL-mtMCM proteins, R161A-FL and 6Ins-FL, were capable of hydrolyzing ATP (Fig. 4A). Quantification of the assays (Fig. 4B) revealed no significant difference in ATP hydrolysis between the wt and the two mutants.
DNA Binding-The assays of ssDNA and dsDNA binding showed that the mutant FL-mtMCM proteins bound DNA over a range of protein concentrations (Fig. 5, A and B), and quantification of the assays revealed a similar level of DNA binding activity for the wt and the sHex mutants (Fig. 5, C and D).
Helicase Assay-When we examined the helicase activity of the wt and sHex mutants of the FL-mtMCM, all showed helicase activity (Fig.  6A). However, the sHex mutants consistently exhibited lower helicase   All proteins contained UV light-absorbing material at the void volume resulting from large aggregates labeled as peak 1. The peak 2 position of the wt N-mtMCM is consistent with that of a dHex (labeled dH). Both R161A-N and 6Ins-N mutants lacked the dHex peak 2, but had peak 3, which corresponds to the expected position of an sHex (sH). D, Superose-6 gel filtration profile of wt FL-mtMCM, R161A-FL, and Ins6-FL. The wt FL-mtMCM had a peak (peak 1) with an apparent molecular mass consistent with that of the dHex (dH), whereas both mutants had peaks (peak 2) consistent in size with a sHex (sH). Peak 3 for all proteins contained degraded proteins that was well separated from peak 1 and 2. E, native gel shift assay of the wt and mutant N-mtMCM proteins. Lane 1, wt N-mtMCM; lane 2, R161A-N; lane 3, 6Ins-N; lane 4, arginine mutant (R227A and R230A) used as a control for the charge effect on the migration on the native gel shift assay. The wt N-mtMCM and the double arginine mutant control migrated similarly (dH). Both mutants, R161A-N and 6Ins-N, migrated faster than the wt, indicating that both mutations successfully converted dHex to sHex in solution.
activity than wt protein at multiple protein concentrations (Fig. 6B). The activity difference was the greatest at 0.5 nM protein concentration, where both mutants had minimal level of activity (ϳ1.2% released the ssDNA, Fig. 6B), but the wt showed ϳ13 times higher activity (with ϳ16% released ssDNA, Fig. 6B). The difference between mutant and wt helicase activity decreased as the protein concentrations increased over the range between 0.5 and 8.0 nM (Fig. 6B), with up to 5 times lower activity for mutant R161A-FL and up to 2 times lower for mutant 6Ins-FL when compared with the wt helicase activity. The R161A mutation appeared to have a greater effect on helicase activity than the 6Ins, perhaps because of the flexibility of the peptide (GSGSGG) inserted at the hexamer-hexamer interface.

DISCUSSION
By a combination of mutagenesis and biochemical and structural assays, the results in this report unequivocally confirm for the first time that the hexamer-hexamer interactions seen in the x-ray dHex structure of mtMCM is not a crystal packing artifact and that the dHex architecture is functionally relevant. These results are consistent with prior observations that wt mtMCM behaved as an oligomeric complex of dHexs or dodecamers in solution (7)(8)(9) and suggest that the single heptamer (29) or single hexamer (30) observed in previous EM studies may not be the natural oligomeric state under physiological conditions.
Using gel filtration chromatography, native gel shift analysis, and EM, we have demonstrated that the mutations of R161A and 6Ins of both the FL-mtMCM and the N-mtMCM constructs disrupted dHex formation and generated sHex complex. The sHexs of the mutant proteins were stable in solution. Interestingly, residue Arg-161 is not a highly conserved residue among other archaeal MCMs, which could explain why no other archaeal MCMs have been shown to behave like a dHex in solution.
Previous EM studies of the mtMCM protein complex did not observe dHex structure, but only showed heptamer, hexamer, and polymers (22,29,30). Using EM, we have also detected oligomeric forms of 6-and 7-fold rings and an occasional open ring structure. 3 However, we consistently observed dodecamer (dHex) formation in a variety of salt and protein concentrations as assayed using native gel shift, gel filtration chromatography, and by direct EM observation. Moreover, the dHex structure accounted for a large portion of the images detected by EM in those buffers closest to physiological conditions (50 -100 mM NaCl, 50 mM Tris⅐HCl, pH 8.0). 3 Using gel filtration chromatography, we obtained the pure FL-mt-MCM proteins either from the peak corresponding to dHex form for the wt construct or from the peak corresponding to sHex form for the mutant constructs (Fig. 2, C and D). These isolated pure complexes remained in the sHex and dHex peaks, respectively, in a second run of gel filtration analysis. These isolated sHex and dHex complexes were used for functional assays. We found no significant differences in ATPase activity and DNA binding between the sHex mutants and the wt protein. This result is not necessarily surprising, because the ATP binding sites are in the helicase domain, distal to the hexamer-hexamer interface, and the DNA binding areas are located inside the hexameric channel (11).

Structure and Function of an Archaeal MCM
The helicase activity assay revealed that the wt dHex and mutant sHex proteins were all capable of unwinding dsDNA. However, the two sHex mutants consistently displayed lower helicase activity over the range of the tested protein concentrations. The helicase activity difference was the most dramatic at a lower protein concentration of 0.5 nM, at which the wt protein has ϳ13 times higher helicase activity than either of the two mutants (Fig. 6B). Interestingly, SV40 LTag, the viral functional homologue of the MCM complex, also exhibits helicase activity in vitro both as sHex and dHex complexes, and the in vitro helicase activity of the single hexamers is ϳ10 -15 times less than that of the dHex form (13,14). However, LTag dHex formation has been shown to be essential for DNA replication in vivo (15). The determination of whether the dHex structure and the hexamer-hexamer interactions play an essential role for mtMCM function in DNA replication will require in vivo studies.
The in vitro helicase assay also revealed that the two mutants, R161A-FL and 6Ins-FL, had different helicase activity. Mutant R161A-FL consistently displayed lower helicase activity than mutant 6Ins-FL at protein concentrations equal to or above 1.0 nM (Fig. 6B). Because neither mutant could form dHex, we reasoned that the lower helicase activity of R161A could be because of a role of residue Arg-161 beyond mediation of the hexamer-hexamer interaction. An alternative explanation is that mutation of R161A more efficiently disrupts dHex formation than 6Ins. The inserted string of glycine and serine residues was designed to disrupt the hexamer-hexamer interaction but not the protein folding. The flexible nature of the peptide could allow certain forms of hexamer-hexamer interaction.
Another noteworthy observation is that the helicase activity of the mutants approached to that of the wt at higher protein concentration (e.g. at 8.0 nM). A convenient speculation is that at higher protein concentration the sHexs of the mutants might come together to form hexamers to give the higher helicase activity. Unfortunately, we could not use much higher mtMCM protein concentrations for a further helicase activity test, because the higher protein concentration of the wt and mutant proteins started to inhibit the helicase activity, a phenomenon also observed for SV40 large T antigen helicase.
In summary, we successfully generated sHex mutants through structure-guided mutagenesis. Studies of the biochemical activities of the sHex mutants with the wt dHex proteins revealed that they had similar ATPase and DNA binding activities but that the sHex mutants had lower helicase activity than the wt dHex protein. These biochemical results, together other structural and biochemical data (7)(8)(9)(10)(11), suggest that the mtMCM complex may act similarly to SV40 LTag in functioning as a double hexamer for DNA replication.