Site-directed mutagenesis reveals roles for conserved amino acid residues in the hexameric DNA helicase DnaB from Bacillus stearothermophilus

Abstract

Site-directed mutagenesis studies on conserved amino acid residues within motifs H1, H1a, H2 and H3 of the hexameric replicative helicase DnaB from Bacillus stearothermophilus revealed specific functions associated with these residues. In particular, residues that coordinate a bound Mg²⁺ in the active site (T217 and D320) are important for the function of the enzyme but are not required for the formation of stable hexamers. A conserved glutamic acid (E241) in motif H1a is likely to be involved in the activation of a water molecule for in line attack on the γ-phosphate of the bound nucleotide during catalysis. A conserved glutamine (Q362) in motif H3 acts as a γ-phosphate sensor and mediates the conformational coupling of nucleotide- and DNA-binding sites. The nature of the residue at this position is also important for the primase-mediated activation of DnaB, suggesting that primase uses the same conformational coupling pathway to induce its stimulatory effect on the activity of DnaB. Together, these mutations reveal a conservation of many aspects of biochemical activity in the active sites of monomeric and hexameric helicases.

INTRODUCTION

DnaB is the main replicative DNA helicase in bacteria utilising the energy from ATP binding and hydrolysis to separate the strands of the DNA duplex in advance of the DNA polymerase at the replication fork (1). The functional form of the enzyme is hexameric, as was shown by protein crosslinking (2), sedimentation equilibrium analysis (3,4) and gel filtration studies (4). It has been reported that Mg²⁺ ions are crucial for the stability of the hexamer formed by the Escherichia coli DnaB because in their absence hexamers dissociate to form trimers (3). However, more recent data contradict these earlier findings showing that Mg²⁺ ions do not significantly affect the stability of the E.coli (5) or Bacillus stearothermophilus (4) DnaB hexamers.

The physiological functions of DnaB are modulated by the interaction with DNA under the control of ATP binding and hydrolysis and with other proteins such as the replication initiator protein DnaA (6), the helicase loaders DnaC (7) and DnaI (8), the primase DnaG (4,9,10) and the DNA polymerase III holoenzyme (11,12). These interactions are instrumental in regulating enzyme activity within the primosomal and replisomal complexes. For example, prior to initiation of DNA replication, the E.coli helicase loader (DnaC) interacts with the replicative helicase (DnaB) in equimolar quantities forming a DnaB₆–DnaC₆ complex (13,14). The DnaB₆– DnaC₆ complex interacts with the oriC–DnaA nucleoprotein complex forming a preinitiation complex, and subsequently transfers the helicase onto the DNA in a process that involves ATP hydrolysis (14) and release of DnaC (15).

In the bacteriophage λ system the viral analogue of the E.coli DnaC replication protein, the λP protein, also interacts with the bacterial DnaB helicase. However, unlike DnaC (which stimulates the DnaB-mediated priming reaction), λP suppresses the ATPase and helicase activities of DnaB (16). In this case the preinitiation complex contains the phage O initiator protein, λP and an inactived DnaB helicase. The helicase is then loaded onto the DNA with the concomitant disassembly of the preinitiation complex and the activation of DnaB, by the action of a molecular chaperone system (17).

In the replisomal complex, the helicase, primase and polymerase are components of a dynamic multiprotein complex in which the functions of all three enzymes are coordinated during lagging strand synthesis at the replication fork (11,12). The interaction of the primase with the helicase is mediated by a 16 kDa C-terminal domain of the primase (4,12), and this interaction results in a stimulation of the priming, ATPase and helicase activities of the complex (4,10,18). The protein–protein interface (4,19) and the molecular details of this interaction are still unknown.

In this study, we have employed site-directed mutagenesis to investigate the roles of conserved amino acid residues at the molecular level, thus enabling us to elucidate certain aspects of the molecular mechanism of action of the hexameric replicative helicase DnaB from B.stearothermophilus. We confirm the important catalytic roles of the conserved lysine (K216) and threonine (T217) of motif H1 (equivalent to the Walker A motif) as well as the conserved aspartic acid (D320) of motif H2 (equivalent to the Walker B motif). We provide direct evidence that residues which coordinate the Mg²⁺ ion, although essential for activity, are not essential for the formation of stable hexamers of B.stearothermophilus DnaB, confirming our previous observations (4). In addition, we show that a conserved glutamate (E241) residue is important for the activity of DnaB and is likely to be the water activating residue in the active site of the enzyme. Taken together, these data reveal a conservation of function for residues in the active sites of monomeric and hexameric helicases as well as the recombination protein RecA. Finally, we report that the nature of the amino acid residue at position 362 in motif H3 affects the DNA-binding ability of the enzyme in response to nucleotide binding, providing evidence for a link between the DNA- and ATP-binding sites. The nature of this residue also affects the ability of DnaB to be stimulated by the primase, DnaG, suggesting a common conformational pathway for activation of DnaB by both primase and DNA.

MATERIALS AND METHODS

Site-directed mutagenesis

Site-directed mutagenesis was performed by ‘splicing by overlap extension’ as described elsewhere (20,21). The final PCR products carrying the appropriate mutations were gel purified and then cut with suitable restriction endonucleases (AatII and HindIII). The restriction fragments carrying the mutations were used to replace the equivalent fragments in the dnaB gene. Mutants were selected by detecting the appearance or disappearance of restriction sites. The K216A, T217A, E241A and D320N mutations create additional NlaIV, BbvI, DdeI and AseI restriction sites, respectively, whereas the Q362A and Q362K mutations abolish a PvuII site (Table 1). No restriction site was created or abolished by the D320A mutation and consequently successful D320A mutants were selected by sequencing the appropriate dnaB gene region. All mutants were sequenced in order to verify the mutations and the absence of additional mutations which might have been introduced by the PCRs. Table 1 shows the sequences of the mutagenic oligonucleotides.

Table 1. Oligonucleotides used for mutagenesis.

The mutations are indicated in bold.

Protein purification

All mutant DnaB proteins were purified using the purification procedure for the wild-type DnaB protein as described before (22). The P16 fragment of DnaG was purified as described elsewhere (4). Protein samples were analysed by SDS–PAGE in 10% resolving gels with 4% stacking gels (23). Gels were stained with Coomassie Brilliant Blue and destained in 10% (v/v) acetic acid, 25% (v/v) methanol. All proteins were >99% pure as estimated by Coomassie-stained SDS–PAGE gels.

ATPase assays

The ATPase activity of the wild-type and mutant DnaB proteins was assayed by linking it to NADH oxidation (24). Reactions were carried out in the ATPase reaction mixture [20 mM Tris pH 7.5, 50 mM NaCl, 10 mM MgCl₂, 5 mM DTT, 24 nM enzyme and 1.2 µM single-stranded (ss)DNA cofactor] at varying ATP concentrations in a total reaction volume of 1 ml. Because of its low ATPase activity, reactions using the K216A mutant were carried out at 240 nM protein and 3.6 µM ssDNA cofactor. The ssDNA cofactor used was a synthetic 81mer oligonucleotide, the sequence of which can be found elsewhere (25). Reactions were initiated by the addition of the protein. For the T217A, D320A and D320N mutants, ATPase assays were carried out at 843 nM concentrations, but no activity could be detected. The effect of P16 on the ATPase activity of the wild-type and mutant DnaB proteins was studied by preincubating 24 pmol (240 pmol for K216A) of DnaB with 160 pmol (1.6 nmol for K216A) of P16 in a total volume of 10 µl in a buffer containing 20 mM Tris pH 7.5, 150 mM NaCl, 1 mM EDTA and 10% glycerol at room temperature for 30 min, prior to adding them to the ATPase reaction mixture. The final concentrations of DnaB and P16 in the reaction mixture were 24 and 160 nM, respectively (with the exception of K216A, 0.24 and 1.6 µM, respectively). Although all reaction rates are expressed as ATP molecules hydrolysed per second per DnaB monomer, it should be noted that it is unlikely that all monomers in the hexameric complex will be hydrolysing ATP at the same rate. All protein concentrations quoted above refer to monomers and all DNA concentrations are expressed in terms of molecules of DNA.

Helicase assays

The 3′–5′ tailed DNA substrate used in the helicase assays was prepared as described elsewhere (25). One molecule of DNA substrate is defined as one molecule of M13mp18 ssDNA with one molecule of oligonucleotide annealed to it. DnaB (16 nM wild type, 25 nM Q362 mutant) was incubated in the helicase reaction mixture (20 mM Tris pH 7.5, 50 mM NaCl, 20 mM MgCl₂, 10% glycerol, 5 mM DTT and 0.5 nM DNA substrate) at 50°C for 10 min. Reactions were initiated by addition of ATP to a final concentration of 3 mM and terminated at 5 min intervals by the addition of helicase stop buffer (0.4% w/v SDS, 40 mM EDTA, 8% v/v glycerol, 0.1% w/v bromophenol blue). The effect of P16 on the helicase activity of the wild-type and mutant DnaB proteins was studied by adding a large excess of P16 (8 µM) to the helicase reaction mixture. P16 was preincubated with the appropriate amount of DnaB protein in 50 mM Tris pH 7.5, 1 mM EDTA, 10% glycerol, at room temperature for 15 min prior to adding the proteins to the helicase reaction mixture. Displaced oligonucleotide was separated from annealed oligonucleotide through a 10% non-denaturing polyacrylamide gel at constant voltage (130 V). Gels were dried and exposed to Fuji RX X-ray film. Quantitative analysis was performed with a phosphorimager and ImageQuant software (Molecular Dynamics). Helicase activity was determined as a percentage of radioactively labelled oligonucleotide displaced from the M13 ssDNA. All protein concentrations quoted above refer to monomers.

Gel shift assays

ssDNA binding reactions were carried out in a total volume of 20 µl, using purified wild-type and mutant DnaB proteins at various concentrations (0.046, 0.092, 0.184, 0.368, 0.736, 1.472, 2.94 and 5.89 µM) in a binding buffer (20 mM Tris pH 7.5, 50 mM NaCl, 5 mM MgCl₂, 4 mM DTT, 10% glycerol and 0.875 nM of the ssDNA substrate) in the presence of ADPNP (2.5 mM). The radioactively labelled single-stranded oligonucleotide used in these assays was a single-stranded 66mer: 5′-ATGACCATGATTACGAATTCGAGCTTTTTTTTTGGGGATCCTCTAGAGTCGACCTGCAGGCATGCA-3′.

The double-stranded (ds)DNA binding reactions for the Q362 mutants were carried out using the same oligonucleotide as above with its complementary strand to construct the dsDNA substrate. All other experimental details are as described for the ssDNA substrate. Binding reactions were carried out at room temperature for 20 min. Protein– oligonucleotide complexes were separated from unbound oligonucleotide by electrophoresis through 8% polyacrylamide, 5% glycerol non-denaturing gels, in 0.5× TBE buffer supplemented with 10 mM MgCl₂, at 100 V at room temperature. Following electrophoresis, gels were dried under vacuum and exposed to Fuji RX X-ray film.

Analytical gel filtration

Analytical gel filtration was performed using a prepacked Superdex S200 (HR 10/30) size exclusion column (Pharmacia Biotech). Purified protein (0.25 mg) was loaded onto the column that was equilibrated previously in running buffer (50 mM Tris pH 7.5, 2 mM EDTA and 200 mM NaCl). Buffer was run through the column at 0.5 ml/min and 0.5 ml fractions were collected. The absorbance at 280 nm was monitored and samples taken from the peaks were analysed by SDS–PAGE, as described earlier. DnaB–P16 complexes were preformed by mixing DnaB with a large excess of P16 (a six times molar excess with reference to the monomers), followed by incubation at room temperature for 10 min, prior to loading onto the column. The column was calibrated with molecular weight markers between 178 and 669 kDa (data not shown).

RESULTS

Structure and amino acid sequence comparisons

Amino acid sequence comparisons of the DnaB sequences from a wide variety of bacterial species (Fig. 1) showed that all DnaB proteins have Walker A and B motifs characteristic of NTP-binding/hydrolysing enzymes (26). Previous structural studies have reported structural homology between the hexameric helicase from bacteriophage T7 (27,28), the monomeric DNA helicase PcrA (29) and the recombination protein RecA (30), despite the limited sequence homology between these proteins. The structural homology is not confined just to the overall fold of the proteins but extends to their active sites, where important amino acid residues are spatially, rather than sequentially, conserved (29,30). Using the structures as a guide, several of these conserved residues in the active site of RecA can be aligned with equivalent conserved residues in DnaB, so we wished to determine whether the conservation of structure and function at the active site of RecA and PcrA could be extended to include the hexameric helicases such as DnaB.

Figure 1.

Amino acid sequence comparisons between the B.stearothermophilus DnaB (BstDnaB) and DnaB proteins from different bacterial species, as well as the homologous hexameric T7 gene 4 helicase, F₁ATPase, the RecA recombination protein and the monomeric PcrA helicase. Conserved motifs are labelled and shown in bold. The amino acid residues targeted for mutagenesis in this study are denoted by an asterisk.

In the active sites of these enzymes, a conserved threonine residue from the Walker A motif is responsible for metal ion coordination in both RecA (T73) and PcrA (T38) proteins (and many other NTPases) and a conserved lysine which is located adjacent to this threonine in the Walker A motif interacts with the bound ATP and is crucial for ATPase activity (21,30,31). The equivalent residues are easily identified in B.stearothermophilus DnaB, as K216 and T217 within the conserved motif H1 (equivalent to the Walker A motif).

A conserved glutamate residue (E96 in RecA, E224 in PcrA) is responsible for activating the attacking water molecule (21,30). In PcrA, E224 is part of the conserved helicase motif II and is situated adjacent in both sequence and space to a conserved aspartate residue (D223) which coordinates the bound Mg²⁺ ion (21,31). A conserved aspartate residue in RecA (D144) also coordinates the bound metal ion in the active site, but in this case, although D144 and E96 are not adjacent in the primary sequence, they are in close spatial proximity in the active site of the enzyme (30). As there is no glutamate residue adjacent to D320 in motif H2 (analogous to the Walker B motif) in the primary sequence of DnaB, we proceeded to identify conserved glutamate residues between motifs H1 and H2 by using amino acid sequence comparisons (Fig. 1). Such comparisons identified a glutamate residue (E241) in motif H1a which is conserved among a wide variety of hexameric helicases. The crystal structures of T7 gene 4 helicase (27,28) confirm that the position of this residue in the active site would be consistent with this function.

Finally, a conserved glutamine (Q194) in the active site of RecA was shown to mediate ATP binding/hydrolysis-induced conformational changes (32). An equivalent conserved glutamine (Q254) in the active site of PcrA plays a similar role (33). Sequence analysis identified a conserved glutamine (Q362) residue in motif H3 which we thought might serve an analogous function in DnaB (Fig. 1). Interestingly, this residue is a histidine in T7 gene 4 helicase.

We used site-directed mutagenesis to examine the roles of these conserved residues in DnaB and to compare these results with similar studies on RecA and/or PcrA.

Helicase activities of mutant DnaB proteins

The mutant proteins K216A, T217A, E241A, D320A and D320N had no detectable helicase activity but the Q362A and Q362K mutant proteins retained some activity (Fig. 2). In order to determine the reasons for the reduced activity of the mutant proteins, we investigated a number of biochemical properties that contribute to the helicase activity.

Figure 2.

Helicase assays for wild-type and mutant DnaB proteins. The gels show time course reactions for wild-type and Q362 mutant proteins, as indicated. All reactions were carried out at 50°C, as described in Materials and Methods. Lanes 1 and 2 show boiled and annealed DNA substrates, respectively. Samples were loaded immediately after termination of the reaction onto a continuously running gel to prevent reannealing of substrate, which is why the displaced oligonucleotide appears to be retarded with time.

ATPase activities of mutant DnaB proteins

The ATPase activity of wild-type DnaB from B.stearothermo philus does not obey simple Michaelis–Menten kinetics (4). This peculiar kinetic behaviour is likely to be the result of cooperative conformational changes of the hexameric ring, induced by ATP binding/hydrolysis. All of the mutant DnaB proteins have reduced ATPase activities compared to the wild-type enzyme (Fig. 3).

Figure 3.

A comparison of the ATPase profiles of the mutant and wild-type DnaB proteins (A–E, as indicated) in the presence of ssDNA cofactor. (F) All the data in a combined graph for direct comparison (open squares, wild-type DnaB; filled diamonds, Q362A; open circles, Q362K; open triangles, E241A; filled circles, K216A). Details of the experiments are given in Materials and Methods. All rates are expressed as ATP molecules hydrolysed per second per monomer of DnaB. No ATPase activity could be detected for the T217A, D320A and D320N mutants.

The E241A and K216A mutant proteins have severely reduced ATPase activity compared to the wild-type enzyme (Fig. 3). Mutation of the residues that coordinate the Mg²⁺ ion in the active site (T217A, D320A and D320N) resulted in proteins with no detectable ATPase activity (Fig. 3), even when the concentration of Mg²⁺ was increased to 50 mM in the reaction mixtures (data not shown). The ATPase activity of the wild-type DnaB was similar at Mg²⁺ concentrations between 10 and 50 mM (data not shown).

The shape of the ATPase activity profiles of the Q362 and K216A mutant proteins were different to that of the wild-type enzyme. The Q362 mutant proteins also had markedly reduced ATPase activities (Fig. 3). In the wild-type enzyme, the ATPase activity appears to follow Michaelis–Menten kinetics up to 0.5–0.75 mM ATP, reaching an apparent plateau. However, as the ATP concentration is increased further, there is a sharp increase followed by rapid inhibition of the ATPase activity (4). The altered ATPase activity profile observed in the Q362 mutants is not the result of an inherent inability of these mutant proteins to form hexamers, since these mutants form stable hexamers as shown by gel filtration studies (Fig. 4). One possible explanation for this different behaviour is that the Q362A/K and K216A mutant proteins cannot undergo the same conformational changes induced by ATP binding/hydrolysis that presumably occur in wild-type DnaB. If this were the case, then the interactions between these mutant enzymes and effector proteins might also be affected. Previously we have shown that a fragment (P16) of the primase (DnaG) is sufficient for primase-mediated stimulation of the ATPase and helicase activities of DnaB in B.stearothermophilus (4). Thus we investigated the effects of the mutations on the P16-mediated stimulation of the ATPase activity.

Figure 4.

Analytical gel filtration studies showing that mutant Q362K and K216A proteins physically interact with the P16 fragment of the primase, DnaG. The elution profiles of the Q362K, K216A and wild-type DnaB proteins are shown in (A), (C) and (F), respectively. The elution profiles of the complexes of Q362K, K216A and wild-type DnaB with P16 are shown in (B), (D) and (G), respectively. Experimental details are described in Materials and Methods. Samples from the peaks were analysed by SDS–PAGE, as shown in (E) (mutant proteins) and (H) (wild-type DnaB). In (E) lanes 1 and 7 show the broad range molecular weight markers from Bio-Rad. Their molecular weights (in kDa) are indicated by arrows on the left of the picture. Lanes 2 and 4 show samples taken from the Q362K and K216A peaks, whereas lanes 3 and 5 show samples taken from the peaks representing the Q362K/P16 and K216/P16 complexes, respectively. Lane 6 shows a sample from the P16 peak. In (H) the numbers of the lanes correspond to the numbers of the peaks in (F) and (G). The relative positions of DnaB and the P16 proteins in (E) and (H) are indicated by arrows.

Stimulation of ATPase activity by P16

Quantitative analysis of stimulation by P16 could only be undertaken for those mutants that retained detectable ATPase activity. Although the ATPase activity of the E241A mutant DnaB is reduced relative to that of the wild-type enzyme, it can still be activated by P16 in a similar manner (Fig. 5). The Q362A mutant is also activated by P16 but the Q362K and K216A mutant proteins seem to have lost the ability to be activated by P16 (Fig. 5), although the greatly reduced ATPase activity in the K216A mutant makes the measurements difficult. There are at least two possible explanations for the lack of stimulation of ATPase by P16 in the mutant proteins: either that they cannot interact with P16, or they cannot undergo P16-mediated conformational changes. Gel filtration studies rule out the former suggestion, since both K216A and Q362K mutant proteins form hexamers which can physically interact with P16, as shown in Figure 4. This leaves us with the conclusion that the proteins are unable to undergo the appropriate conformational changes that are required for activity in the wild-type enzyme. This proposition was tested further by looking at the affinity of the protein for DNA (see below).

Figure 5.

The effect of P16 on the ATPase profile of mutant and wild-type DnaB proteins. Experimental details are given in Materials and Methods. Reactions were carried out in the absence (circles) and presence (squares) of P16. Rates are expressed as ATP molecules hydrolysed per second per monomer of DnaB.

Finally, addition of P16 or full-length DnaG did not result in recovery of detectable ATPase activity for the T217A, D320A and D320N mutant enzymes, despite the fact that they can physically interact with P16 as shown by gel filtration (data not shown).

Stimulation of helicase activity by P16

In the absence of P16, the E241A mutant protein has a lower ATPase activity than that of Q362K (Fig. 3). However, in the presence of P16, its ATPase activity is stimulated to a level comparable to that seen for the Q362K protein (Fig. 5). Since there is a correlation between ATPase and helicase activities in the wild-type and most of the mutant proteins, we expected that stimulation of the ATPase activity on binding P16 might also stimulate helicase activity, as observed for the wild-type protein (4). Therefore, in the presence of P16, the helicase activity of E241A should be comparable to that of Q362K. However, this was not the case. Even in the presence of excess P16, no helicase activity could be detected (data not shown). This observation implies that the E241A mutation is an ‘uncoupling’ mutation, since, despite having a reasonable ATPase activity that can be stimulated further by the binding of P16 (Fig. 5), the protein has lost all measurable helicase activity.

In contrast, for both of the Q362 mutant proteins, stimulation of helicase activity by P16 was correlated with the degree of stimulation of ATPase activity induced by P16 (Figs 5 and 6).

Figure 6.

The effect of P16 on the helicase activities of mutant and wild-type DnaB enzymes. All experimental details are described in Materials and Methods. Filled squares and triangles represent reactions carried out in the absence and presence of P16, respectively. Helicase activity is determined as a percentage of radioactively labelled oligonucleotide displaced from M13 ssDNA with time. The concentrations of wild-type (16 nM) and Q362 mutant (25 nM) enzymes used in these reactions are equivalent to 1360 and 870 nt of M13 ssDNA per DnaB hexamer, respectively. Helicase activities for the mutants K216A, T217A, E241A, D320A and D320N were non-detectable (data not shown).

Binding of DnaB mutants to ssDNA

Gel shift assays revealed that most of the mutants bound to ssDNA in the presence of ADPNP but with affinities somewhat affected to different extents depending on the specific mutation (Figs 7 and 8). The E241A and D320A proteins exhibited the best ssDNA binding with their affinities reduced by ∼3–4-fold. The D320N mutant also bound to ssDNA but with its affinity reduced by ∼15-fold. The K216A and T217A mutations affected the ssDNA-binding ability of the protein markedly, with very weak binding for the former and almost no detectable binding for the latter.

Figure 7.

(A) Gel shift assays using mutant (E241A, D320A, D320N, K216A and T217A) and wild-type DnaB proteins. Binding reactions were carried out at protein concentrations (expressed as monomers) of 0.046, 0.092, 0.184, 0.368, 0.736, 1.472, 2.94 and 5.89 µM (lanes 1–8, respectively), using a radioactively labelled single-stranded 66mer synthetic oligonucleotide (0.875 nM). All reactions were carried out in the presence of 2.5 mM ADPNP. Other experimental details are described in Materials and Methods. (B) The results are summarised in the form of a table. K_d values are defined as the concentration of protein (expressed as monomers) required to shift half of the DNA substrate. Nd refers to ‘non-determinable’ because of very low binding affinity.

Figure 8.

Gel shift assays using single- (A) and double-stranded (B) synthetic oligonucleotides. A comparison between wild-type, Q362A and Q362K DnaB proteins. (A) ssDNA binding reactions for the Q362 mutants were carried out under the conditions described in Figure 7, using 0.092, 0.184, 0.368, 0.736, 1.472, 2.94 and 5.89 µM (lanes 1–7, respectively) protein concentrations (expressed as monomers), using a radioactively labelled single-stranded 66mer synthetic oligonucleotide (0.875 nM) in the presence or absence of ADPNP (2.5 mM), as indicated. (B) dsDNA binding reactions for the Q362 proteins were carried out as described in (A) but in this case a dsDNA substrate (0.875 nM) was used in the presence or absence of ADPNP (2.5 mM), as indicated.

For the Q362 mutations, however, the situation was different. Although the Q362A mutation did not result in defective ssDNA binding (Fig. 8A), surprisingly, the Q362K mutation produced a protein that bound ssDNA in the absence of ADPNP, with a significantly higher affinity than that of the wild-type enzyme or any of the other mutants (Fig. 8A). Moreover, addition of ADPNP to the binding mixture did not increase its avidity for ssDNA, but instead resulted in a reduction in its binding affinity.

Effect of Q362K mutation on binding to dsDNA

The Q362K mutation also affected the ability of the enzyme to bind dsDNA. Wild-type (Fig. 8) and all of the other mutants (data not shown) bind dsDNA weakly in comparison to ssDNA. The presence of ADPNP in the binding mixture slightly enhanced binding to dsDNA (Fig. 8). In stark contrast to the wild-type enzyme, the Q362K mutant binds dsDNA very efficiently, albeit with an affinity less than that for the ssDNA (Fig. 8A and B). Furthermore, the affinity of the Q362K protein for dsDNA is reduced in the presence of ADPNP, in a manner similar to that observed for binding of ssDNA in this mutant (Fig. 8). It appears that the Q362K mutation has altered the conformation of the hexamer from a ‘low’ to a ‘high’ affinity state for DNA binding in the absence of bound nucleotide and binding of ADPNP seems partly to reverse this ‘conformational’ change. In contrast, binding of ADPNP appears to have the reverse effect for the wild-type (and all of the other mutant proteins), with the nucleotide complex having a higher affinity for DNA (Fig. 8). The nature of the mutation at this position seems to be important as the Q362A mutant was similar to the wild-type protein (Fig. 8A and B) and did not exhibit any of the Q362K DNA-binding properties.

DISCUSSION

Structural and mechanistic comparisons with T7 helicase, RepA, PcrA and RecA

At present there is no high resolution crystal structure available for the helicase domain of DnaB. There are, however, crystal structures for the homologous helicase domain of the replicative helicase-primase of bacteriophage T7 (27,28) and of the hexameric helicase RepA (29). We can identify equivalent amino acid residues from sequence homology between these proteins (Fig. 1). The crystal structures of T7 helicase show how conserved amino acid residues are spatially arranged around the bound ATP in the nucleotide-binding site. Structural and mechanistic comparisons can therefore be made and these are useful in determining the biochemical roles of the mutated residues in DnaB.

In one of the crystal structures of the T7 helicase there is a magnesium ion bound at the metal-binding site (28). S319 (equivalent to T217 in DnaB) contacts the bound cation directly in the ‘active’ conformation. The T217A mutation in DnaB resulted in a protein that lacks ATPase and helicase activities. D424 (equivalent to D320 in DnaB) of motif H2 also coordinates the bound cation. In PcrA helicase, an equivalent conserved aspartic acid (D223) of motif II has been shown to coordinate the metal ion in the active site (21,32). The involvement of the conserved aspartic acid of motif H2 in metal ion coordination is consistent with our mutagenesis studies showing that the D320A and D320N mutations result in DnaB proteins which lack ATPase or helicase activities. A similar D424N mutation in T7 helicase also resulted in a protein with virtually abolished ATPase activity and no detectable helicase activity (34). The fact that all of our mutations of the residues that coordinate the bound metal ion resulted in totally inactive proteins which could still form stable hexamers provides direct evidence that, at least for B.stearothermophilus DnaB, the cation is needed for catalysis but is not required for hexamer formation. Although it has been reported previously that magnesium was required for hexamer formation for E.coli DnaB (3), more recent evidence suggests that the E.coli enzyme may be more similar to B.stearothermophilus DnaB in this regard (5).

It is not immediately obvious why the T217A mutation has eliminated the ability of the enzyme to bind ssDNA and dsDNA either in the presence or absence of ADPNP. A conformational switch linking magnesium binding, via motifs I and II, to binding of DNA has been suggested for PcrA, although in this case the equivalent mutation (T38A) did not abolish the ssDNA binding ability of the protein, but instead reduced the ssDNA-mediated stimulation of the ATPase activity (21). ATP binding is essential for DNA binding and mutations of residues involved in ATP binding in the active site will be expected to affect ATP binding and therefore also DNA binding. Our data show that mutations of K216, D320 and T217, all of which are thought to participate in ATP binding, have affected DNA binding.

The T7 helicase crystal structures show that the highly conserved lysine (K318) of the Walker A motif contacts a non-bridging oxygen of the β-phosphate, suggesting that this lysine is an important catalytic residue. This is supported by biochemical evidence which shows that the K318A mutation results in a 200-fold decrease in nucleotide hydrolysis and complete loss of helicase activity in this enzyme (34,35). Our mutagenesis results on DnaB also confirm these findings since the equivalent K216A mutation results in a protein with severely reduced ATPase activity and no detectable helicase activity.

A glutamic acid is thought to activate the attacking water molecule during nucleotide hydrolysis in PcrA (21), Rep (36) and NS3 (37) helicases as well as in the recombination protein RecA (31). A conserved glutamic acid occupies an equivalent spatial position in T7 and RepA helicases and could be involved in water activation during nucleophilic attack on the γ-phosphate of the bound nucleotide (27,28). From amino acid sequence comparisons we identified the equivalent highly conserved glutamic acid (E241) in DnaB (Fig. 1). The E241A mutation resulted in severely reduced ATPase activity and no detectable helicase activity, with only a minor alteration of the DNA binding affinity. The results of our mutation are consistent with the role of water activator.

Conformational coupling of the nucleotide- and DNA-binding sites is mediated by a conserved glutamine

The PcrA helicase and the RecA protein have been suggested previously to employ a similar energy coupling mechanism, in which they utilize a conserved glutamine residue as γ-phosphate sensor, thus coupling ATP hydrolysis-induced conformational changes to strand separation and strand exchange, respectively (31,33,38). Analogous coupling mechanisms seem to operate in other nucleotide-hydrolysing enzymes and this theme has been suggested to be of wide applicability to other diverse groups of NTP-utilising enzymes (31,38). In common with PcrA and RecA proteins, DnaB (22) and Rep (36) also possess a highly conserved glutamine residue (Q362 and Q254, respectively), which we considered likely to participate in coupling conformational changes to ATP binding/hydrolysis. Interestingly, the equivalent residue in T7 helicase is a histidine (H465) in motif H3 (Fig. 1). In the crystal structures of the T7 helicase, H465 can be seen to be hydrogen bonded to the γ-phosphate of the bound nucleotide in a position consistent with participation in the conformational coupling of the nucleotide- and DNA-binding sites (27,28). An analogous histidine residue (H179) is also present in the hexameric RepA helicase (29). Our mutagenesis results show that the nature of the mutation at this position affects the ability of the DnaB protein to bind both ssDNA and dsDNA in response to nucleotide binding (Figs 7 and 8). In the absence of nucleotide, both wild-type DnaB and the Q362A mutant proteins bind either ssDNA or dsDNA weakly, but in the presence of ADPNP the affinity of both proteins for ssDNA and dsDNA is increased significantly (Figs 7 and 8). A similar observation has been reported for the T7 helicase, which interacts with ssDNA more tightly in its dTTP-bound form (39,40). Intriguingly, the opposite seems to be the case in the Q362K mutant DnaB protein. This protein binds ssDNA and dsDNA quite well in the absence of nucleotide whereas in the presence of ADPNP its binding affinity for both ssDNA and dsDNA is markedly reduced (Fig. 8).

One interpretation of these data is that although the Q362A mutant protein is unable to undergo the conformational changes related to ATP binding/hydrolysis that occur in the wild-type enzyme (indicated by its different ATPase activity profile), it retains the ability to go through the conformational changes that regulate DNA binding (manifested by its unaltered response to ADPNP but good affinity for DNA). In contrast, although the Q362K mutant also fails to undergo the conformational changes related to ATP binding/hydrolysis (again shown by its different ATPase activity profile), in this case the mutation has also caused a conformational change that has profoundly altered the DNA-binding properties of the hexamer. Subsequent binding of the ADPNP in the active site apparently results in additional changes that reduce the affinity of the mutant protein for both ssDNA and dsDNA.

These data suggest that the conserved Q362 residue plays a pivotal role in a ‘conformational coupling pathway’, which couples the nucleotide- and DNA-binding sites. Q362 is probably acting as a γ-phosphate sensor in the active site sensing nucleotide binding/hydrolysis and forming a conformational switch analogous to that reported for the PcrA and RecA proteins (31,38), although the details of how this conformational switch is operating in DnaB require further investigation. Furthermore, in view of the sensitivity of DnaB to the nature of the residue at this position, it is interesting that this residue is a histidine in the T7 gene 4 and RepA helicases.

The primase-induced stimulation of DnaB activity is mediated via the same conformational coupling pathway as the nucleotide-induced modulation of DNA binding

The nature of the residue at the 362 position also seems to determine the ability of the enzyme to be stimulated by the primase. Although the Q362A mutant protein retains its ability to be stimulated by P16, this is not the case for the Q362K mutant. Two possibilities for this behaviour are that either the hexamer formed by the Q362K mutant protein has adopted a different conformation and hence fails to bind the primase or, alternatively, that the stimulatory pathway is somehow altered such that binding of the nucleotide in the active site now causes a different effect. Gel filtration studies reveal that the former suggestion is unlikely because the Q362K mutant forms stable hexamers and is able to bind P16 in a manner similar to that of the wild-type DnaB protein (Fig. 4). Instead, it is more likely that the nature of the amino acid residue at this position determines the primase-induced effect on the activity of DnaB. These data also imply that the primase utilises the same ‘conformational coupling pathway’, which couples the nucleotide- and DNA-binding sites, in order to stimulate the activity of DnaB, and that this pathway has been disrupted in the Q362K mutant.

Conservation of function in the active sites of different helicase families and RecA

The ATP-binding domains of RecA, DnaB-family hexameric helicases and superfamily I helicases, such as PcrA, share a similar protein fold (27–31). There is also a considerable degree of conservation of spatial organisation of critical residues in the active sites of these enzymes despite the poor conservation of residues in the linear sequences of the proteins. Thus the same protein scaffold can provide similar residues that function in similar ways and which, although they originate from different places in the linear sequence of the proteins, are positioned similarly in the active site. We illustrate some examples of this conservation in the work presented herein, but this is unlikely to be the whole story. For example, it has been shown that a conserved arginine residue contributes to catalysis in PcrA (21), with a similarly positioned arginine being present in the active site of T7 helicase (23). In PcrA, this arginine arises from a different protein domain to that which contains the ATP-binding site, while in T7 helicase the residue comes from a different protein subunit. In RecA, it has been proposed that an arginine residue also contributes in this way (D.Camerini-Otero, personal communication), although there is not yet a structure of the RecA–ATP complex to allow us to visualise this interaction directly. Consequently, it appears that structural and sequential homologies between these enzymes give rise to the common features of coupling of DNA binding to hydrolysis of ATP that are conserved across this functionally diverse group of proteins.