The MCM (minichromosome maintenance) proteins of archaea are widely believed to be the replicative DNA helicase of these organisms. Most archaea possess a single MCM orthologue that forms homo-multimeric assemblies with a single hexamer believed to be the active form. In the present study we characterize the roles of highly conserved residues in the ATPase domain of the MCM of the hyperthermophilic archaeon Sulfolobus solfataricus. Our results identify a potential conduit for communicating DNA-binding information to the ATPase active site.
- AAA+ ATPase
- DNA replication
- minichromosome maintenance (MCM)
The homomultimeric MCM (minichromosome maintenance) proteins of archaea have been studied for a number of years. In particular, considerable effort has been put into analyses of the MCMs encoded by Methanothermobacter thermautotrophicum (MthMCM) and Sulfolobus solfataricus (SsoMCM) (for reviews, see [1–4]). In agreement with the proposed role of archaeal MCMs as the replicative helicase, SsoMCM is enriched at replication origins at the time of replication initiation in vivo . Furthermore, in vitro studies have revealed that the recombinant protein forms homohexamers and is capable of directing (d)ATP-dependent melting of partially duplex DNA substrates . Ensemble and single-molecule FRET (fluorescence resonance energy transfer) studies indicate that the enzyme loads on to 3′ to 5′ polarity single-stranded DNA tails with the C-terminal proximal domains facing the duplex DNA [6,7]. This loading results in single-stranded DNA passing through a central pore in the homohexameric MCM ring where it is contacted by basic residues supplied by β-hairpin motifs that line the pore. One β-hairpin is supplied by the N-terminal domains of the protein and a second, the PS1BH (pre-sensor 1 β-hairpin), is found in the C-terminal AAA+ (ATPase associated with various cellular activities) ATPase domain of the MCM [6,8]. Interestingly, the N-terminal domains of MCM, although dispensable for helicase activity in vitro, do play significant roles in modulating substrate choice, enhancing processivity and facilitating inter-subunit communication [8,9]. Furthermore, the N-terminal domains have been shown to interact with the archaeal GINS complex in vivo and in vitro . The C-terminal PS1BH has been shown to be important for DNA binding, and mutation of a single lysine residue at its tip results in a mutant protein with impaired DNA binding, an ATPase activity that is no longer stimulated by DNA and complete loss of helicase activity . In addition, studies of the MthMCM have implicated a second conserved element within the AAA+ domain of the MCM in DNA melting. Deletion of this module, the H2I, a β–α–β structure in the AAA+ fold, resulted in elevated DNA binding and DNA-stimulated ATPase activity, but complete loss of helicase activity . A 4.35 Å (1 Å=0.1 nm) resolution structure of a monomeric form of SsoMCM has been reported . The PS1BH and H2I are adjacent to one another and modelling of a hexameric form of the protein predicts that they point in towards the central cavity of the hexameric ring. Importantly, evidence has been obtained that indicates that both PS1BH and H2I undergo re-positioning during the ATP utilization cycle of the helicase [9,11]. These observations are in agreement with structural studies of the distantly related Superfamily 3 helicases. More specifically, studies of the SV40 LTag (simian virus 40 large tumour antigen) and BPV E1 (bovine papilloma virus E1) helicases have revealed that the analogous PS1BH hairpin in these helicases undergo movements of up to 17 Å during ATP binding, hydrolysis and release by the helicase, serving as the power stroke to drive the helicase along the DNA [13–15]. In MCM, the PS1BH of a given subunit is close to a conserved loop (the ACL) in the N-terminal domains of the adjacent protomer within the hexameric assembly. The ACL plays a key role in communicating information between protomers within the hexamer . Thus, repositioning of the PS1BH during the reaction cycle may serve dual roles; to act as the power stroke and to propagate co-operative interactions between SsoMCM monomers.
Analyses of MCM with mutations in residues predicted to play roles in the ATPase activity of the protein, coupled with mutant doping studies, have revealed that MCM has a rather unusual form of limited co-operativity between protomers . Furthermore, these studies revealed an unusual architecture for the AAA+ ATPase domain of the helicase. Typical AAA+ proteins function as multimers with the ATPase active site formed at the interface between protomers . Thus both subunits contribute residues to the active site. One subunit acts as a ‘cis’ site and contributes side chains such as the Walker A (Lys346 in SsoMCM), Walker B (Asp404) and Sensor 1 (Asn488) residues. However, the neighbouring protomer also plays an active role and contributes ‘trans’-acting residues, such as the arginine finger (Arg473 in SsoMCM). ATP binding, hydrolysis and release can thus affect repositioning of adjacent subunits. In addition, intra-subunit motions are mediated by repositioning of the two principal domains of the AAA+ site, the lid and base domains. The Sensor 2 motif normally plays a key role in this intra-molecular repositioning event. Interestingly, however, MCM has a highly unusual configuration of the Sensor 2 residue (Arg560 in SsoMCM). As initially predicted bioinformatically, and subsequently verified biochemically and structurally, an α-helical insertion preceding the Sensor 2 motif in MCM results in Sensor 2 acting in trans rather than the usual cis configuration [12,16,17]. Additionally, complementation data suggested that MCM may possess a number of additional trans-acting residues, clustered around the base of the PS1BH . Precedent for multiple trans-acting residues has been found in the distantly related viral Superfamily 3 helicases such as SV40 LTag and BPV E1. However, whether the candidate trans-acting residues in MCM do indeed directly co-ordinate ATP, as in the SF3 helicases, or whether they have an indirect role via facilitating the positioning of bona fide trans-acting residues is unknown.
In the present study we address the roles of a cluster of highly conserved residues that occupy a pivotal position in SsoMCM, lying at the juxtaposition of the PS1BH and the post-Sensor 2 α-helix. We identify Arg329 as a key residue in the communication between the DNA-binding site of SsoMCM and the trans component of the ATPase active site.
Site-directed mutagenesis and protein purification
Site-directed mutagenesis was performed using the single primer site-directed mutagenesis method (Stratagene). The primers used for mutagenesis are shown in Table 1. Proteins were purified as described previously . Briefly, proteins were expressed in the Escherichia coli strain Rosetta (DE3). Transformants (1 litre) were grown to A600=0.6 before being induced with 1 mM IPTG (isopropyl β-D-thiogalactopyranoside) for 3 h. Cells were harvested by centrifugation and resuspended in 25 ml of 1× TBS (10 mM Tris/HCl, pH 8.0, and 150 mM NaCl) containing 14 mM 2-mercaptoethanol and one tablet of Complete EDTA-free Protease Inhibitor Cocktail (Roche). Cells were lysed using a French Press (Thermo Spectronic) followed by heat treatment at 75 °C for 20 min and clarification by centrifugation for 15 min at 17000 rev./min in a Beckman JA-25.50 rotor. MCM proteins were first purified over a 5 ml HiTrap Heparin HP column (GE Healthcare) using an ÄKTA Prime system (GE Healthcare). After the column was pre-equilibrated with 1× TBS, the sample was injected and unbound proteins were flushed with 5 cv (column volumes) of 1× TBS. With all the steps run at a flow rate of 2.5 ml/min, proteins were eluted with a 15 cv gradient of 150 mM to 1 M NaCl in 10 mM Tris/HCl, pH 8.0. Fractions (5 ml) were collected and analysed by SDS/PAGE and Coomassie Blue staining. Protein-containing fractions were pooled and concentrated to ≤5 ml using a Vivaspin 20 with a 30 kDa MWCO (molecular-mass cut-off) PES (polyethersulfone; Sartorius Stedim Biotech). The concentrated proteins were then purified over a Hi-load 26–60 Superdex 200 prep-grade column (Amersham Biosciences) using an ÄKTA Purifier system (GE Healthcare). The column was first equilibrated with 1× TBS before the sample was injected. Proteins were eluted with 1 cv (≈350 ml) of 1× TBS and were collected after 100 ml of elution. Fractions (5 ml) were collected and analysed by SDS/PAGE and Coomassie Blue staining. Protein-containing fractions were pooled and further purified over a Mono Q 5/50 GL column (GE Healthcare) using an ÄKTA Purifier system (GE Healthcare). The column was first equilibrated with 1× TBS, the sample was then injected and unbound proteins were flushed with 2 cv of 1× TBS at a flow rate of 1 ml/min. To elute proteins, a 15 cv gradient from 150 mM to 1 M NaCl in 10 mM Tris/HCl, pH 8.0, was applied. Samples were collected in 0.5 ml fractions and were analysed by SDS/PAGE and Coomassie Blue staining. Protein-containing fractions were pooled and concentrated to ≤0.5 ml using a Vivaspin 2 with a 30 kDa MWCO PES (Sartorius Stedim Biotech).
EMSA (electrophoretic mobility-shift assay)
The Y-shaped DNA substrate was prepared by radiolabelling the oligonucleotide [(T)44GCTCGTGCAGACGTCGAGGTGAGG-ACGAGCTCCTCGTGACCACG] with [γ-32P]ATP. This radiolabelled molecule was then annealed to its complementary strand [CGTGGTCACGAGGAGCTCGTCCTCACCTCGACG-TCTGCACGAGC(T)44] before the resultant partial duplex molecule probe was purified via recovery from a gel slice following electrophoresis on SDS/PAGE (8% gel).
DNA binding was performed by mixing 10 nM labelled Y-shaped DNA substrate with the indicated concentration of protein in 50 mM potassium acetate, 20 mM Tris-acetate, 10 mM magnesium acetate, 1 mM DTT (dithiothreitol), pH 7.9, and 5% glycerol. Binding reactions were incubated at 50 °C for 15 min prior to electrophoresis on a 6% polyacrylamide gel (17 cm×17 cm×1.5 mm) in 1× TBE (45 mM Tris/borate and 1 mM EDTA) at 10 V/cm. Gels were dried and analysed by phosphorimagery (Fujifilm FLA-5000). Quantification was performed by using the Aida Image Analyzer software (Raytest Isotope Messgerate).
ATP hydrolysis was performed in a 20 μl reaction volume containing 50 mM potassium acetate, 20 mM Tris-acetate, 10 mM magnesium acetate, 1 mM DTT, pH 7.9, 1 mM ATP (Sigma), 25 nCi of [γ-32P]ATP and 1 μM protein. Where DNA was added, the Y-shaped substrate described above was added to a concentration of 1 μM. Reactions were incubated at 60 °C for 30 min before termination by the addition of 1% SDS. Samples were then analysed using TLC by spotting 2 μl of each sample on CEL 300 PEI TLC plates (Macherey-Nagel), which was subsequently developed with 0.8 M lithium chloride in 1 M formic acid buffer. Analysis was performed by phosphorimagery (Fujifilm FLA-5000) and quantification using Aida Image Analyzer software (Raytest Isotope Messgerate).
DNA unwinding was performed in a 20 μl reaction volume containing 50 mM potassium acetate, 20 mM Tris-acetate, 10 mM magnesium acetate, 1 mM DTT, pH 7.9, 1 nM Y-shaped DNA probe (prepared as described above), 10 mM ATP, 2 μg of BSA and the indicated amount of proteins. Reactions were incubated at 65 °C for 45 min before addition of 20 μl of stop buffer (100 mM EDTA, 2% SDS, 0.1% Bromophenol Blue and 40% glycerol). Samples were electrophoresed on an 8% polyacrylamide gel (17 cm×17 cm×1.5 mm) in 1×TBE and 0.2% SDS. Gels were run for 90 min at 12 V/cm before drying and analysis by phosphorimagery (Fujifilm FLA-5000). Quantification was performed using Aida Image Analyzer software (Raytest Isotope Messgerate).
ATPase complementation assay
Reactions were preformed as described previously . The indicated mutant proteins (1 μM) were mixed for 15 min prior to ATPase assays.
RESULTS AND DISCUSSION
Examination of the sequence conservation of archaeal and eukaryotic MCMs with reference to the 4.35 Å crystal structure (PDB code 3F9V) of SsoMCM reveals an intriguing juxtaposition of secondary structure elements within SsoMCM. Although the low resolution of the available structure precludes accurate positioning of side chains, it is apparent that the helix following the Sensor 2 arginine residue (Arg560) lies on top of the base of a feature described by Chen and colleagues as EXT-HP (external hairpin) . Additionally, the EXT-HP precedes the first β-sheet of the AAA+ domain, this β-sheet is followed by the essential Walker A motif (including Lys346). Examination of sequence conservation reveals that there is a cluster of highly conserved residues in the vicinity of the base of the EXT-HP (Figure 1 and Supplementary Figure S1 at http://www.BiochemJ.org/bj/436/bj4360409add.htm). We reasoned that this cluster might have roles in influencing the ATPase activity of the MCM. In principle, this hypothetical influence could be effected by modulation of either the cis or trans properties of the MCM. In the case of the cis scenario, the residues at the base of the EXT-HP could influence the positioning of the residues of the Walker A motif. Alternatively, the trans-acting activity of the MCM subunit could be altered, by virtue of the residues at the base of the EXT-HP, influencing the positioning of the overlying α-helix and thereby transducing effects to the Sensor 2 Arg560 residue.
Furthermore, the base of the EXT-HP is additionally juxtaposed to the base of the PS1BH. Thus we speculated that there might be a route of communication from the PS1BH, via the EXT-HP to the AAA+ active site.
Previous work from our laboratory had addressed the roles of two adjacent glutamine residues (Gln423 and Gln424) at the base of PS1BH, these residues were found to act as trans residues and to be absolutely required for ATPase and helicase activity of MCM; mutation of these residues to alanine abrogated both activities. In addition, we have revealed that Arg331 (at the base of the EXT-HP) was similarly essential for helicase and ATPase activities and also appeared to play a trans role . However, the molecular basis of this phenotype was not resolved in our previous work.
In the present study, we have made a further seven mutations to four conserved residues. These include E422A and E422R; Glu422 is an extremely highly conserved residue that precedes the previously described Gln423Gln424. We also generated R329E; Arg329 is a conserved basic residue at the base of the EXT-HP. In addition, we made R331I and R331K mutants, as well as targeting Leu565, a conserved hydrophobic residue in the helix following the Sensor 2 residue (Arg560); modelling of this residue suggests that it is in close physical proximity to the Arg331–Leu565 residue and it was mutated to aspartate or lysine.
The resultant mutant proteins were purified and found to elute as hexamers from gel-filtration columns (results not shown). We next subjected them to tests of their biochemical activities.
First we tested DNA-binding activity, using a model Y-shaped DNA substrate in EMSAs. As can be seen in Figure 2 and Supplementary Figure S2 (at http://www.BiochemJ.org/bj/436/bj4360409add.htm), all mutant proteins retained the ability to bind DNA, albeit with affinities somewhat altered from the wild-type protein. More specifically, the E422A and R331K mutant MCMs showed up to 2-fold higher affinity for the DNA substrate than the wild-type MCM; MCM (L565D), MCM (R331I) and MCM (E422R) had the same affinity as the wild-type MCM and MCM (R329E) and MCM (L565K) had up to 2-fold lower affinity for DNA.
Next, we tested the ability of the mutants to hydrolyse ATP in the presence and absence of DNA (Figure 3 and Supplementary Figure S3 at http://www.BiochemJ.org/bj/436/bj4360409add.htm). As we have reported previously, the ATPase activity of wild-type SsoMCM is modestly stimulated (approximately 2-fold) by the presence of DNA. The MCM (R331K), MCM (E422A) and both MCM (L565D) and MCM (L565K) mutants demonstrated ATPase properties similar to wild-type protein, whereas MCM (R331I) and MCM (E422R) lost the ability to hydrolyse ATP. Interestingly, the R329E mutant MCM had significantly reduced ATPase activity and this was no longer stimulated by the addition of DNA. As MCM (R329E) displayed a 2-fold reduced DNA-binding affinity when compared with the wild-type protein, we note that the concentration of DNA added to these assays was 1 μM, a 13-fold higher concentration than the Kd measured in Figure 2. Therefore we think it is unlikely that the non-responsiveness of this mutant to DNA was a simple consequence of having modestly lowered DNA-binding activity compared with the wild-type enzyme.
Having established the DNA-binding and ATPase activities of the mutant MCMs, we next assayed their ability to affect DNA melting using Y-shaped oligonucleotide substrates (Figure 4 and Supplementary Figure S4 at http://www.BiochemJ.org/bj/436/bj4360409add.htm). In agreement with the inability of the MCM (R331I) and MCM (E422R) proteins to hydrolyse ATP, these mutants also lacked detectable helicase activity. The MCM (R331K) mutant showed helicase activity at approximately half the level of wild-type and, interestingly, the MCM (R329E) mutant showed similar activity to MCM (R331K), despite its low and non-DNA-stimulated ATPase activity. Although the E422R mutant MCM had no detectable helicase activity, the potentially less invasive mutation to alanine, E422A, resulted in essentially wild-type activity. Finally, mutation of Leu565 to either an aspartate or lysine residue (L565D or L565K) significantly stimulated the helicase activity of the resultant mutant protein when compared with wild-type. Significantly, these mutants do not show any elevation of either DNA-binding or ATPase rates, suggesting that these alterations enhance the coupling between ATP hydrolysis and helicase activity.
The reduction of the ATPase activities of the E422R and R329E mutant MCMs could be due to the wild-type residues contributing to the ATPase active site. As detailed above, the ATPase site is generated by the juxtaposition of two MCM subunits with co-ordinating residues being donated by both subunits; following previous convention we have termed these cis-acting residues [such as Walker A (Lys346), Walker B (Asp404) and Sensor 1 (Asn488) residues] or trans-acting residues, the classic example being the arginine finger (Arg473) and additionally, in the case of MCM, Sensor 2 (Arg560) and other basic residues . We have shown previously that, although mutation of cis or trans residues affects ATPase activity in homomultimers, mixing a cis mutant and a trans mutant can regenerate activity . This arises because the cis mutant has a wild-type trans site and conversely the trans mutant has a wild-type cis site. Appropriate apposition of subunits of the two mutant types can thus reconstitute a functional active site. We therefore tested the effect of mixing either E422R or R329E mutant with MCM containing the Walker A lysine cis mutant K346E or a second mutant MCM with an isoleucine substitution of the previously identified trans residue Arg331. As can be seen in Figure 5(A), mixing MCM (E422R) with MCM (K346E) restored ATPase activity, indeed levels were significantly higher than wild-type. In contrast, mixing MCM (E422R) with MCM (R331I) did not restore ATPase activity. Similarly, the MCM (R329E) mutant is complemented by the cis mutant MCM (K346E) but not by the trans mutant MCM (R331I) (Figure 5B). Taken together, these results reveal that both MCM (E442R) and MCM (R329E) mutants have specific impairments in the trans component of the ATPase active site.
Perhaps the simplest explanation for the ‘trans phenotype’ of these residues would be that the residues' side chains directly co-ordinate ATP in the active site. This, however, does not appear to be the case for Glu422, as the MCM (E422A) mutant protein, although having a modestly elevated affinity for DNA, has essentially wild-type ATPase and helicase activities. This therefore indicates that the side chain of Glu422 is not directly involved in co-ordinating the ATP moiety.
The E422R and R329E mutant SsoMCMs, although both acting as trans mutants, show distinct behaviours. SsoMCM (E422R) has essentially wild-type DNA binding, but no detectable ATPase or helicase activities. In contrast, SsoMCM (R329E), although less active than the wild-type protein, nevertheless retains helicase activity despite its weak ATPase activity no longer being stimulated by the addition of DNA. Given the relatively close juxtaposition of Glu422 and Arg329 in the structure of SsoMCM, we tested whether an E422R/R329E double mutant could restore activity. However, in agreement with the distinct properties of the individual mutations, the double mutant, similarly to the E422R single mutant, lacked detectable ATPase and helicase activities (results not shown).
Our studies above of the Arg331 residue in the EXT-HP confirm our previous data  that the basic side chain of this residue is of pivotal importance for the ATPase and helicase activities of SsoMCM. Notably, however, mutation of this residue to isoleucine has no effect on the DNA-binding activity of the enzyme. Interestingly, the similarly highly conserved Arg329 shows a different phenotype, a charge-reversal mutation of this residue to glutamate has a modest effect on DNA binding, but renders the SsoMCM's ATPase activity non-responsive to the addition of DNA. This indicates that this mutation is defective in transducing information from the DNA-binding site of the helicase to the ATPase active site. The behaviour of this mutant is superficially reminiscent of that of the T364E mutation of the glutamate switch [18,19]. The glutamate switch is a conserved residue in all clades of AAA+ ATPases that plays a key role in aligning the catalytic glutamate residue in the Walker B motif for activation of a water molecule during ATP hydrolysis [18,19]. Similarly to the R329E mutant, the glutamate switch mutant SsoMCM (T364E) can still bind DNA, yet displays lowered helicase activity and has reduced ATPase activity that is no longer stimulated by DNA. However, Thr364 exerts its influence via the cis-acting Walker B motif and thus contrasts with the trans-acting Arg329. Given the proximity of the base of the Arg329-containing EXT-HP and the α-helix preceding the trans-acting Sensor 2 Arg560, it is tempting to speculate that Arg329 may influence the position of the Sensor 2 residue. Furthermore, the EXT-HP is additionally juxtaposed to the DNA-binding PS1BH. In this regard, it may be significant that we have previously demonstrated that mutation of a conserved lysine residue (Lys430) located at the tip of the PS1BH results in an enzyme that, similar to the R329E MCM, shows lowered basal ATPase activity that is no longer stimulated by DNA .
Taken together, these observations lead us to propose that Arg329 plays a key role in transducing positional information from the DNA-binding PS1BH to the trans-acting face of the ATPase active site, an event at the heart of the mechanochemistry of the MCM helicase.
Stephen Bell and Li Phing Liew designed and interpreted the experiments, and wrote the manuscript. Li Phing Liew performed the experiments.
The work was funded by the BBSRC (Biotechnology and Biological Sciences Research Council) [grant number BB/G0006512/1].
Abbreviations: AAA+, ATPase associated with various cellular activities; BPV, E1, bovine papilloma virus E1; cv, column volume; DTT, dithiothreitol; EMSA, electrophoretic mobility-shift assay; EXT-HP, external hairpin; MCM, minichromosome maintenance; MthMCM, Methanothermobacter thermautotrophicum MCM; MWCO, molecular-mass cut-off; PES, polyethersulfone; SsoMCM, Sulfolobus solfataricus MCM; PS1BH, pre-sensor 1 β-hairpin; SV40, LTag, simian virus 40 large tumour antigen
- © The Authors Journal compilation © 2011 Biochemical Society