To date, methanogens are the only group within the archaea where firing DNA replication origins have not been demonstrated in vivo. In the present study we show that a previously identified cluster of ORB (origin recognition box) sequences do indeed function as an origin of replication in vivo in the archaeon Methanothermobacter thermautotrophicus. Although the consensus sequence of ORBs in M. thermautotrophicus is somewhat conserved when compared with ORB sequences in other archaea, the Cdc6-1 protein from M. thermautotrophicus (termed MthCdc6-1) displays sequence-specific binding that is selective for the MthORB sequence and does not recognize ORBs from other archaeal species. Stabilization of in vitro MthORB DNA binding by MthCdc6-1 requires additional conserved sequences 3′ to those originally described for M. thermautotrophicus. By testing synthetic sequences bearing mutations in the MthORB consensus sequence, we show that Cdc6/ORB binding is critically dependent on the presence of an invariant guanine found in all archaeal ORB sequences. Mutation of a universally conserved arginine residue in the recognition helix of the winged helix domain of archaeal Cdc6-1 shows that specific origin sequence recognition is dependent on the interaction of this arginine residue with the invariant guanine. Recognition of a mutated origin sequence can be achieved by mutation of the conserved arginine residue to a lysine or glutamine residue. Thus despite a number of differences in protein and DNA sequences between species, the mechanism of origin recognition and binding appears to be conserved throughout the archaea.

INTRODUCTION

DNA replication in archaea employs enzymes that are more similar to those utilized by eukaryotes than bacteria in the equivalent process. In general, the archaeal proteins are simplified versions of their eukaryotic counterparts [1,2]. For example, in eukaryotes a six-subunit ORC (origin recognition complex) binds to origins of replication and acts as a ‘landing pad’ for the assembly of pre-replication and pre-initiation proteins such as Cdc (cell division cycle) 6 and MCM (minichromosome maintenance). The closest homologues of ORC proteins in archaea are the Orc/Cdc6-like proteins, which share homology with eukaryotic Cdc6 and the C-terminal portion of ORC1. Most archaea possess two Orc/Cdc6 genes, although there are a number of exceptions; Methanopyrus kandleri, Methanocaldococcus jannashii and Methanococcus maripaludis have only distantly related Cdc6 homologues [35], species of the Sulfolobus genus possess three [6], whereas Halobacterium sp NRC-1 contains ten Cdc6-like genes [7].

Structures of Cdc6 proteins from Pyrobaculum aerophilum [8] and Aeropyrum pernix [9] show that archaeal Cdc6 proteins consist of three domains. The C-terminal domain forms a WHD (winged helix domain) and binds DNA in a sequence-specific manner via a HTH (helix-turn-helix motif) [6,10,11]. Comparison of archaeal HTH domain sequences reveals that ‘Class I’, but not ‘Class II’ Orc/Cdc6 proteins, possess a conserved motif on the second helix consistent with a DNA-recognition helix [9]. Classical WHD proteins recognize DNA sequences by inserting the recognition helix into the major groove of DNA, while the β-hairpin ‘wing’ interacts with the phosphodiester backbone [12]. Compatible with this, mutation of Arg334 and Arg335 residues in the putative recognition helix of MthCdc6-1 strongly reduced DNA binding, whereas mutations in positively charged residues of the MthCdc6-1 ‘wing’ (Arg358, Lys360, Arg362) affected DNA binding only modestly [10]. Recent data have shown that multiple copies of the A. pernix Cdc6-1/ORC1 dimer recognize and bind four A. pernix ORB (origin recognition box) elements, causing a specific change in the topology of the DNA and partial unwinding of an AT-rich region, as judged by sensitivity to DNase I and P1 endonuclease [13]. Class II (Orc2/Cdc6-2) proteins display only partial origin-sequence-specific DNA binding and do not possess conserved sequences within the recognition helix, but rather show conservation in the wing motif of the protein which is implicated in a different mechanism of DNA binding [9].

GC skew and Z-curve analyses have been used to predict origins of replication for a number of archaeal species, including M. thermautotrophicus and Methanosarcina mazei [14,15], although these techniques have not been useful in other archaea such as Methanococcus jannaschii or Sulfolobus solfataricus [15,16]. Whether or not predictions exist, experimental data are still required to confirm that putative origins act as sites of replication initiation in vivo. Functional archaeal replication origins have been experimentally verified by a number of methods [6,7,1720]. Unusually for prokaryotes with circular chromosomes, a number of archaea including both of the Sulfolobus species examined have been shown to possess multiple origins of replication [6,18,20,21]. To date, methanogens are the only group within the archaea where replication origins have not been experimentally demonstrated. The presence of ORB repeats is specific to archaeal replication origins. ORBs of 36 bp were first described in S. solfataricus near an AT-rich region that might act as a site for initial duplex melting. Several ORB sequences are typically located upstream of Class I Orc/Cdc6 genes and ORBs were identified in most sequenced archaea apart from methanogens [6]. In vitro Cdc6 proteins specifically recognized S. solfataricus ORB (SsORB) sequences. This interaction is believed to be essential for DNA duplex melting and origin firing. Mutation of three consecutive bases in either of two positions in the SsORB consensus abolished binding of SsCdc6-1 protein to the mutated sequence [6].

A cluster of repeats was identified in the upstream region of the MthCdc6-1 gene [14] and was shown to contain a ‘mini-ORB’ sequence of 5′-TTACA(g/c)TTGAA(a/c)N-3′ [10]. This sequence is a minimized version of the ORB sequences found in other archaea. MthCdc6-1 shows 40-fold higher binding affinity for mini-ORB sequences than for a random DNA sequence of the same length and co-operative binding to a 181 bp duplex containing three mini-ORB sequences. The affinity of MthCdc6-1 for the triple repeat is ∼40-fold higher than that for a single site and 10-fold higher than expected for DNA containing three independent ORB sites [10].

In the present study we show that the region predicted to contain the origin of replication in M. thermautotrophicus does indeed function as an origin in vivo. By presenting mutated ORB sequences to MthCdc6-1 in a previously described DNA filter-binding assay [10] we show that a single guanine that is invariant among all archaeal origin sequences described to date is critical for Cdc6 binding. In addition, we show that a number of single point mutations within the mini-ORB consensus have little or no effect on MthCdc6-1 binding in vitro, but that the DNA sequence required for tight binding of MthCdc6-1 extends beyond the 13 bp mini-ORB consensus previously reported. We have generated mutant Cdc6 proteins that confirm that an R334A mutation results in loss of origin-sequence binding. In addition, we have shown that an R334K Cdc6 variant shows promiscuous binding of origin sequences where the conserved guanine has been changed and that an R334Q variant displays altered origin sequence specificity. The R334Q protein no longer binds the wild-type origin sequence, but recognizes sequences where the guanine has been replaced by a cytosine or adenosine. Our results suggest that a conserved mechanism is employed by the Class I archaeal Cdc6/ORC proteins in specifically recognizing replication origin sequences.

EXPERIMENTAL

M. thermautotrophicus growth

Cultures of M. thermautotrophicus DSM1053 were grown in defined mineral-salt medium [22] buffered with 50 mM Mops (pH 7.4 at 20 °C) at 58 °C under strictly anaerobic conditions in a 3 litre vessel (Applikon) sparged with 240 ml/min H2/CO2 and mixed at 650 rev./min. Approx. 107 cells (equivalent to ∼100 μl of a culture at D=0.4) were aerobically fixed by the addition of 1 ml of ice-cold 77% ethanol and analysed by flow cytometry as described previously [23]. Flow cytometry was performed with an A40 miniFCM (Apogee Flow Systems).

Replication intermediate analysis

A 4 kb HindIII fragment containing the putative origin of replication from M. thermautotrophicus was cloned into pUC118 to produce pJC064 (‘b’ in Figure 1A). The adjacent 5.29 kb HindIII fragment (upstream of MthCdc6-1) was cloned into pUC118 to produce pJC070 (‘a’ in Figure 1A). Two 2.8 kb fragments were PCR-amplified from M. thermautotrophicus genomic DNA using the primer pairs: 5′-GATATCTCATCTATCCTTGGGAGTGG-3′ and 5′-CACCCTAAACTTGACAGGCTAATGG-3′, 5′-GTTCCACGTTTATACAGGATGCAAGG-3′ and 5′-CTTAAGGAACTCCTGGAAAATGGAGG-3′. The purified products of these PCR reactions were used as a template to generate a control probe to the 5.7 kb HindIII fragment downstream of MthCdc6-1 (‘c’ in Figure 1A). M. thermautotrophicus cells were harvested aerobically into 133 mM EDTA/33% glycerol, rapidly cooled in dry ice/ethanol and recovered by centrifugation (6500 g for 10 min at 4 °C). Cell pellets were lysed by three rounds of freezing in liquid nitrogen and grinding in a pestle and mortar. The lysate was diluted 4-fold [20 mM Tris/HCl (pH 8.0), 5 mM EDTA and 10% sucrose] and incubated at 60 °C for 30 min in 2% SDS. Genomic DNA was recovered by propan-2-ol precipitation, RNase digestion and two rounds of phenol/chloroform extraction before being purified by caesium chloride gradient ultracentrifugation (55000 rev./min for 22 h at 16 °C). Genomic DNA was dialysed against HindIII restriction buffer containing EDTA before being digested with HindIII and MgCl2 was added. Replication intermediates were enriched, separated and visualized as previously described [24] using the probes described above.

The M. thermautotrophicus origin of replication is functional in vivo

Figure 1
The M. thermautotrophicus origin of replication is functional in vivo

(A) The genomic region of M. thermautotrophicus showing the presence of the fragments used to probe the gels in (BD) (marked as a–c respectively), location of nearby genes and the relative position and orientation of ORB sequences. An A/T-rich region that could be the site of initial unwinding is indicated. DP1, DNA polymerase II small subunit. (B) Two-dimensional gel probed with a genomic fragment upstream of the putative replication origin showing only a Y-arc consistent with the presence of a fork. (C) Two-dimensional gel probed with a genomic fragment containing the putative origin of replication. Bubble arc is indicated with an arrow. (D) Two-dimensional gel probed with a genomic fragment downstream of the putative replication origin showing only a Y-arc consistent with the presence of a fork.

Figure 1
The M. thermautotrophicus origin of replication is functional in vivo

(A) The genomic region of M. thermautotrophicus showing the presence of the fragments used to probe the gels in (BD) (marked as a–c respectively), location of nearby genes and the relative position and orientation of ORB sequences. An A/T-rich region that could be the site of initial unwinding is indicated. DP1, DNA polymerase II small subunit. (B) Two-dimensional gel probed with a genomic fragment upstream of the putative replication origin showing only a Y-arc consistent with the presence of a fork. (C) Two-dimensional gel probed with a genomic fragment containing the putative origin of replication. Bubble arc is indicated with an arrow. (D) Two-dimensional gel probed with a genomic fragment downstream of the putative replication origin showing only a Y-arc consistent with the presence of a fork.

Cloning, mutagenesis and isolation of MthCdc6-1

MthCdc6-1 (ORF1412) was PCR-amplified from genomic DNA and cloned into pQE30 (Qiagen) to produce pJC062. pJC062 was used as a template to produce R334A, R334K and R334Q MthCdc6-1 variants using the QuikChange® site-directed mutagenesis kit (Stratagene). Cultures were grown overnight in 2× LB (Luria–Bertani) medium without IPTG (isopropyl β-D-thiogalactoside) induction and MthCdc6-1 protein was purified as previously described [10] with the following modifications: MthCdc6-1 protein was immobilized on Talon Affinity Resin (BD Biosciences) pre-equilibrated with lysis buffer [40 mM Tris/HCl (pH 8.0), 500 mM sodium chloride, 20% glycerol, 10 mM imidazole, 1 mM 2-mercaptoethanol, 1 mM PMSF and 1 μM pepstatin]. The bound protein was washed with 10 column volumes of lysis buffer containing 75 mM imidazole before MthCdc6-1 was eluted [with 40 mM Tris/HCl (pH 8.0), 400 mM potassium acetate, 20% glycerol, 125 mM imidazole, 1 mM 2-mercaptoethanol, 1 mM PMSF and 1 mM pepstatin] and desalted using a NAP-25 column (GE Biosciences). Aliquots of MthCdc6-1 were snap frozen in liquid nitrogen and stored at −80 °C. The protein purity was checked by SDS/PAGE.

DNA-binding assays

Protein–DNA binding was assessed using a modification of the method described previously [10]. Oligonucleotides labelled with [γ-32P]ATP and T4 polynucleotide kinase were annealed to a complementary oligonucleotide. MthCdc6-1 was desalted into binding buffer [40 mM Bis-Tris propane (pH 7.0), 70 mM potassium acetate, 20% glycerol, 3 mM DTT (dithiothreitol), 2 mM MgCl2 and 0.1 mg/ml BSA] using a NAP-10 column (GE Biosciences). Reaction mixtures of 100 μl containing 0–1.2 mM MthCdc6-1 protein, 0.8 nM substrate and binding buffer were assembled in microtitre plates and assayed as described [10]. Binding constants for the wild-type Cdc6 protein and ORB8 DNA sequence were calculated by curve-fitting a simple independent binding-sites model equation to experimental data in SigmaPlot [10,25]. The resulting Bmax value (65%) was then used to constrain fitting of the same equation to all other data. Where DNA binding was strongly disrupted we observed a poor fit of the equation and Kd (app) was not determined (n.d.).

RESULTS

GC skew methods have predicted a single origin of replication from the genomic sequence of M. thermautotrophicus [14]. Consistent with this notion, a number of direct and inverted repeats have been identified in this region, although the number and exact sequence of the repeats varies between reports [10,14]. We observed 11 repeats of the previously reported consensus 5′-TTACA(c/g)TTGAA(a/c)N-3′ [10], where the repeats are regularly spaced some 49–53 nucleotides apart, possibly suggesting a phased binding of MthCdc6-1 (see Supplementary Figure 1 at http://www.BiochemJ.org/bj/409/bj4090511add.htm). In addition to the mini-ORB sequence, we noticed that all 11 sequences contained an additional conserved region 3′ to the mini-ORB consensus of 5′-NN(a/t)(t/c)(c/g)NCNC-3′ (see Supplementary Figure 1).

Analysis of the putative origin-containing region was performed by probing genomic DNA isolated from exponentially growing M. thermautotrophicus cells and separated by neutral/neutral (N/N) two-dimensional agarose gel electrophoresis [26]. A functional origin of replication was detected on hybridization of the putative origin sequence using a 4 kb probe (Figure 1), as indicated by the bubble arc structure (indicated by the arrow in Figure 1C) and diminished Y-arc characteristic of a firing origin of replication. The discontinuity between the top of the bubble arc and the Y-arc (the so-called ‘bubble-to-Y transition’) is also highly characteristic of a firing origin of replication. In order to demonstrate that the bubble structure was specific to the fragment probed, the blot was stripped and hybridized to ∼5 kb probes specific for adjacent fragments (Figures 1B and 1D). Both probes yielded a complete Y-arc, and the absence of any bubble arc structures, indicating that these regions of DNA contained only fork structures and suggesting that DNA replication in M. thermautotrophicus was bidirectional and proceeds via the movement of a pair of replication forks (Figure 1). The two-dimensional analysis consistently yielded the same results when using DNA isolated from a number (n=5) of different exponentially growing cultures (e.g. Figure 2). We took additional samples from an actively growing culture of M. thermautotrophicus to examine the initiation of replication as cells entered stationary phase (Figure 2). Growth stage was confirmed by flow cytometry (Figures 2A, 2C and 2E) before replication intermediates were analysed. We observed a decrease in the frequency of replication initiation from stationary-phase samples (Figure 2F), but replication did not appear to completely cease. These results are typical of an asynchronous microbial culture, where a small population of cells can continue to proliferate even under limiting growth conditions. Our results are also consistent with the growth characteristics of methanogens, which are not known to produce a dormant stage or sporulate, as they normally exist in hydrogen-limited environments which would result in slow but continuous (‘cryptic’) growth. Although GC skew analysis predicted only a single origin of replication in the genome of M. thermautotrophicus [14], we also probed an 11.5 kb DNA locus around the open reading frame encoding the MthCdc6-2 gene (chromosomal position 1460708–1472211 bp). Our results showed no bubble structures in this DNA region (results not shown). These results indicate that there is no replication origin near the MthCdc6-2 gene.

Replication initiation decreases, but does not cease, as M. thermautotrophicus cells enter stationary phase

Figure 2
Replication initiation decreases, but does not cease, as M. thermautotrophicus cells enter stationary phase

(A, C and E) Flow cytometric analyses of cells taken at different stages of growth show a characteristic reduction in average genome content [23]. (B, D and F) Two-dimensional gels of DNA samples taken from the corresponding growth stages show that initiation frequency decreases, but is not completely lost, in stationary phase. Bubble arcs are indicated by arrows.

Figure 2
Replication initiation decreases, but does not cease, as M. thermautotrophicus cells enter stationary phase

(A, C and E) Flow cytometric analyses of cells taken at different stages of growth show a characteristic reduction in average genome content [23]. (B, D and F) Two-dimensional gels of DNA samples taken from the corresponding growth stages show that initiation frequency decreases, but is not completely lost, in stationary phase. Bubble arcs are indicated by arrows.

The sequence requirements of MthCdc6-1 DNA binding were examined using a filter-binding assay [10]. Duplexes of 34 bp of ORBs 8, 9, 10 or 11 containing the previously described mini-ORB consensus sequence [10] plus flanking regions were incubated with purified MthCdc6-1 (Figure 3A). ORB8 DNA binding and apparent dissociation constants [Kd (app)] for MthCdc6-1 (Table 1) were comparable with those published previously, even though the maximal protein concentration used was lower than in those experiments [10]. Interestingly, variations in the sequence of the regions flanking the ORB consensus had a small but measurable effect on DNA binding (Figure 3A and Table 1). In contrast with DNase I footprinting experiments where SsCdc6-1 was able to recognize ORB sequences from both Pyrococcus and Halobacterium [6], reactions containing MthCdc6-1 and ORB sequences from P. abyssi or S. solfataricus resulted in reduced binding (Figure 3A). A 34 bp duplex of random sequence was not bound by MthCdc6-1, but the addition of increasing amounts of this and other competitor DNA into reactions containing the origin sequence resulted in more specific origin binding (Figure 3B and Table 1). These results were obtained using two different DNA duplexes of random sequence, indicating that the observation was not a sequence-based artefact, although it does not exclude the possibility of an effect of the filter-binding assay.

Table 1
Kinetic data for the interaction between MthCdc6-1 and different ORB sequences

Kd (app) values for the binding of MthCdc6-1 to different Orb-sequence-containing DNA duplexes. Values were derived using a simple independent binding site model equation with Bmax fixed to the value obtained for Orb8 (65%). ORB8: upper case bases indicate consensus sequence residues, lower case bases indicate the remaining (non-conserved) sequence. Other sequences: upper case indicates a variation in sequence from ORB8, • indicates identity with ORB8. In all cases, R2≥0.98. n.d. indicates that the Kd (app) for these samples was not determined. Pa, Pyrococcus abyssi; Ss, Sulfolobus solfataricus.

Name Sequence Kd (app) (μM) 
ORB8 5′-catggtcagaTTACAcTTGAAatggatgtCtCcc-3′ 0.35 
ORB9 GCA••ATTTT•••••••••••••TC••CC•••AT 0.43 
ORB10 ATCCAGAT•G•••••••••••••A••••••C•A• 0.62 
ORB11 ATAA•AG•AT•••••••••••••CT•CCA•A••• 0.68 
ORB8 In the presence of 10-fold excess random competitor 0.30 
ORB8 In the presence of 100-fold excess random competitor 0.23 
ORB8 In the presence of 1000-fold excess random competitor 0.24 
ORB8 In the presence of 1000-fold excess random competitor two 0.21 
SM1 •••••••••••C•••••••••••••••••••••• 0.34 
SM2 •••••••••••••T•••••••••••••••••••• 0.51 
SM3 •••••••••••••••T•••••••••••••••••• 0.48 
SM4 ••••••••••••••••C••••••••••••••••• 0.64 
SM5 ••••••••••••••G••••••••••••••••••• 0.43 
SM6 ••••••••••••••••••C••••••••••••••• n.d. 
SM7 ••••••••••••••••••A••••••••••••••• n.d. 
DM1 •••••••••••••TG••••••••••••••••••• n.d. 
DM2 •••••••••••••••••••GG••••••••••••• n.d. 
DM3 ••••••••••••••••••AG•••••••••••••• n.d. 
DM4 •••••••••••••••••••••••••••••T•T•• n.d. 
DM5 •••••••••••••••••••••••••G•A•••••• n.d. 
3′ end •••••••••••••••••••••TAATGCCATGGAA n.d. 
PaORB CAAC•C•G••C••G•G••••••A•ACC•TGGGGG 7.41 
SsORB TTTACCTTAAGTTC•C••G•G••••C •AA GGGGT n.d. 
Random ACGTACTGACCAGTTGAGTTCTAATGCCATGGAA n.d. 
Name Sequence Kd (app) (μM) 
ORB8 5′-catggtcagaTTACAcTTGAAatggatgtCtCcc-3′ 0.35 
ORB9 GCA••ATTTT•••••••••••••TC••CC•••AT 0.43 
ORB10 ATCCAGAT•G•••••••••••••A••••••C•A• 0.62 
ORB11 ATAA•AG•AT•••••••••••••CT•CCA•A••• 0.68 
ORB8 In the presence of 10-fold excess random competitor 0.30 
ORB8 In the presence of 100-fold excess random competitor 0.23 
ORB8 In the presence of 1000-fold excess random competitor 0.24 
ORB8 In the presence of 1000-fold excess random competitor two 0.21 
SM1 •••••••••••C•••••••••••••••••••••• 0.34 
SM2 •••••••••••••T•••••••••••••••••••• 0.51 
SM3 •••••••••••••••T•••••••••••••••••• 0.48 
SM4 ••••••••••••••••C••••••••••••••••• 0.64 
SM5 ••••••••••••••G••••••••••••••••••• 0.43 
SM6 ••••••••••••••••••C••••••••••••••• n.d. 
SM7 ••••••••••••••••••A••••••••••••••• n.d. 
DM1 •••••••••••••TG••••••••••••••••••• n.d. 
DM2 •••••••••••••••••••GG••••••••••••• n.d. 
DM3 ••••••••••••••••••AG•••••••••••••• n.d. 
DM4 •••••••••••••••••••••••••••••T•T•• n.d. 
DM5 •••••••••••••••••••••••••G•A•••••• n.d. 
3′ end •••••••••••••••••••••TAATGCCATGGAA n.d. 
PaORB CAAC•C•G••C••G•G••••••A•ACC•TGGGGG 7.41 
SsORB TTTACCTTAAGTTC•C••G•G••••C •AA GGGGT n.d. 
Random ACGTACTGACCAGTTGAGTTCTAATGCCATGGAA n.d. 

Specific DNA binding by MthCdc6-1

Figure 3
Specific DNA binding by MthCdc6-1

(A) Variations in the Orb sequences found in M. thermautotrophicus have a small effect on the extent of MthCdc6-1 binding. Despite similarities in sequence, ORB sequences from other species are not bound by MthCdc6-1. Pa, Pyrococcus abyssi; Ss, Sulfolobus solfataricus. (B) Addition of increasing quantities of competitor (random sequence) DNA results in more specific binding of MthCdc6-1 to Orb8 duplex. (C) Single-point mutations within the Orb8 sequence has little effect on Cdc6–Orb binding other than the mutation of a conserved guanine to cytosine in SM6, which abrogates Orb binding. (D) Mutation of pairs of nucleotides within the mini-ORB consensus affects the ability of MthCdc6-1 to effectively bind DNA. Mutation of nucleotides 3′ to the mini-ORB consensus also compromises the ability of MthCdc6-1 to effectively bind DNA. Substrate sequences and Kd (app) values (where appropriate) are listed in Table 1. All points are the mean of three independent reactions. Error bars indicate ±1 S.D.

Figure 3
Specific DNA binding by MthCdc6-1

(A) Variations in the Orb sequences found in M. thermautotrophicus have a small effect on the extent of MthCdc6-1 binding. Despite similarities in sequence, ORB sequences from other species are not bound by MthCdc6-1. Pa, Pyrococcus abyssi; Ss, Sulfolobus solfataricus. (B) Addition of increasing quantities of competitor (random sequence) DNA results in more specific binding of MthCdc6-1 to Orb8 duplex. (C) Single-point mutations within the Orb8 sequence has little effect on Cdc6–Orb binding other than the mutation of a conserved guanine to cytosine in SM6, which abrogates Orb binding. (D) Mutation of pairs of nucleotides within the mini-ORB consensus affects the ability of MthCdc6-1 to effectively bind DNA. Mutation of nucleotides 3′ to the mini-ORB consensus also compromises the ability of MthCdc6-1 to effectively bind DNA. Substrate sequences and Kd (app) values (where appropriate) are listed in Table 1. All points are the mean of three independent reactions. Error bars indicate ±1 S.D.

In order to determine more exactly the sequence requirements of MthCdc6-1 for specific ORB binding, we challenged the wild-type Cdc6 protein with additional 34 bp duplexes containing point mutations in the consensus sequence. Interestingly, we found that most single mutations within the consensus sequence appeared to have little effect on DNA binding (Figure 3C and Table 1). The exceptions to this were SM6 and SM7 (where SM is single mutation), where mutations at the same G (to C or A respectively) within the mini-ORB consensus sequence effectively resulted in a complete loss of binding of MthCdc6-1 (Figure 3C and Table 1). Two double mutations (DM1 and DM3) within the consensus sequence also resulted in abrogation of binding (Figure 3D), one of which, DM3, overlapped with SM6/SM7. A third double mutation in the consensus sequence, DM2, resulted in a significant decrease in total DNA binding by MthCdc6-1. Consistent with the extended MthORB consensus sequence that we observed (above), a similar partial inhibition of MthCdc6-1 binding was observed with duplexes where double mutations were made 3′ to the consensus sequence (Figure 3D and Table 1), implying that these 3′ bases were also important for MthCdc6-1–ORB interactions. This was confirmed by replacing 13 bps of sequence 3′ to the original consensus sequence with a random sequence, which resulted in an almost complete loss of MthCdc6-1 binding (Figure 3D and Table 1).

In order to test the hypothesis that the conserved guanine identified in our experiments interacted directly with the conserved Arg334 previously characterized [10], we generated and purified three MthCdc-1 variants, R334A, R334K, R334Q, with the selection of amino acids being based on well-characterized bidentate protein–DNA interactions [27] (Figure 4). These proteins, along with the wild-type protein, were subjected to in vitro DNA-binding assays in the presence of ORB8, SM6 (G to C), SM7 (G to A) and random sequence duplexes (Figure 5). As previously reported [10], the R334A mutation in MthCdc6-1 disrupted the interaction of Cdc6 with ORB8. R334A also showed poor binding of SM6, SM7 and the random sequence (Figure 5A). The R334K variant of Cdc6 showed a modest decrease in binding of ORB8 compared with wild-type, but was also able to bind the mutated origin sequences SM6 and SM7 (Figure 5B). The R334Q variant was also able to bind SM6 and SM7, but in contrast with R334K or wild-type Cdc6, this protein did not bind the ORB8 sequence, thus showing an altered DNA-sequence-binding specificity compared with the wild-type MthCdc6-1 (Figure 5C).

Variants in MthCdc6-1 may affect Cdc6–Orb interactions

Figure 4
Variants in MthCdc6-1 may affect Cdc6–Orb interactions

(A) Coomassie-Blue-stained SDS/PAGE gel of purified wild-type and variant MthCdc6-1 proteins. The position and size of molecular–mass markers are indicated on the left-hand side. (B) Probable non-disruptive interactions between amino acid side chains and DNA via the major groove (based on [27]). From top to bottom the interactions illustrated are arginine–guanine, lysine–guanine and glutamine–adenosine.

Figure 4
Variants in MthCdc6-1 may affect Cdc6–Orb interactions

(A) Coomassie-Blue-stained SDS/PAGE gel of purified wild-type and variant MthCdc6-1 proteins. The position and size of molecular–mass markers are indicated on the left-hand side. (B) Probable non-disruptive interactions between amino acid side chains and DNA via the major groove (based on [27]). From top to bottom the interactions illustrated are arginine–guanine, lysine–guanine and glutamine–adenosine.

Altered specificity of binding can be engineered into MthCdc6-1

Figure 5
Altered specificity of binding can be engineered into MthCdc6-1

(A) Confirmation that the introduction of the R334A mutation into MthCdc6-1 disrupts Orb8 as well as all other DNA binding. (B) A MthCdc6-1 R334K variant shows slightly reduced affinity for Orb8, but promiscuous binding to the SM6 and SM7 substrates. (C) The R334Q variant of MthCdc6-1 cannot bind Orb8, but is able to bind DNA duplexes containing the SM6 and SM7 mutations, demonstrating an altered specificity for DNA sequences for this protein.

Figure 5
Altered specificity of binding can be engineered into MthCdc6-1

(A) Confirmation that the introduction of the R334A mutation into MthCdc6-1 disrupts Orb8 as well as all other DNA binding. (B) A MthCdc6-1 R334K variant shows slightly reduced affinity for Orb8, but promiscuous binding to the SM6 and SM7 substrates. (C) The R334Q variant of MthCdc6-1 cannot bind Orb8, but is able to bind DNA duplexes containing the SM6 and SM7 mutations, demonstrating an altered specificity for DNA sequences for this protein.

DISCUSSION

DNA replication in the archaea is interesting both for understanding this fundamental process in the third domain of life and as a way of probing the mechanisms that underlie eukaryotic replication. The replication initiation process is still only partially understood in archaea, and well-characterized model systems are required for mechanistic studies. Within the archaea, the replication proteins of M. thermautotrophicus are among some of the best characterized to date. M. thermautotrophicus possesses a strikingly organized array of ORB sequences. Evenly spaced repeats flanking both sides of an AT-rich region upstream of the MthCDC6-1 gene (Figure 1A) were first reported to contain twelve 13-mers with significant homology with repeats found by GC skew analysis in Pyrococcus species [14], although we find only eleven (see Supplementary Figure 1 at http://www.BiochemJ.org/bj/409/bj4090511.add.htm). Although the repeats of P. furiosus were identified as homologous with the functional ORBs identified in Sulfolobus those of M. thermautotrophicus were not [6]. Limited homology between the mini-ORB sequences of M. thermautotrophicus and those of Sulfolobus and Pyrococcus has since been noted [10]. The present study supports the notion that the active origin of replication in M. thermautotrophicus contains eleven full-length ORB sequences that are homologous with those previously identified in other archaea. Our results provide the first experimental confirmation that the putative site of DNA replication initiation in M. thermautotrophicus is in fact a functional replication origin.

As previously reported [9,10], we have shown that MthCdc6-1 can bind to ORB8 sequences in vitro. We have additionally shown that this interaction is improved in the presence of competitor DNA. One possible explanation for this observation is that MthCdc6-1 contains two separate DNA-binding sites, and that binding of non-specific DNA to one site causes some change in the protein that stabilizes the binding of the specific DNA at the other site. We have determined that changes to the DNA sequences 3′ to the mini-ORB consensus affect MthCdc6-1 binding. Replacement of the entire 3′ region of the duplex used for DNA binding resulted in a complete loss of DNA binding, suggesting that the originally reported mini-ORB consensus is not sufficient for effective Cdc6–ORB interactions. Mutations located outside of the mini-ORB sequence in the conserved 3′ residues (DM4 and DM5) did not completely disrupt binding, although they did have a significant effect, supporting the notion that specific contacts between MthCdc6-1 and the 3′ sequence could be crucial for MthORB binding. The limited conservation in this region of the ORB suggests that the 3′ sequence may play a role in enhancing Cdc6–DNA binding, or stabilization of the protein–DNA complex, rather than in sequence-specific recognition by MthCdc6 protein. It should be noted that the sequences 3′ to the mini-ORB identified in the present study were present in the original MthCdc6-1 DNA-binding studies [10], and can also be identified in a synthetic triple repeat of the mini-ORB that was shown to be bound by MthCdc6-1 [11] (see Supplementary Figure 2 at http://www.BiochemJ.org/bj/409/bj4090511add.htm). Thus our novel findings are completely consistent with other studies in this species and we are confident that the 3′ sequences identified by our experiments form part of the essential MthORB sequence required for MthCdc6-1 binding, although their contribution has not previously been appreciated. ORB sequences from different archaeal species show weaker homology in this region and the MthORB diverges greatly from the consensus. We believe that the differences in this part of the ORB sequence are probably sufficient to explain the differences between the SsCdc6-1 and MthCdc6-1 proteins with regard to their abilities to bind ORBs from other species.

All of the double mutations that we tested affected the total amount of DNA bound by MthCdc6-1. DM1 (CT to TG) showed a very strong reduction in binding that did not occur when either of these mutations was tested alone, suggesting that these bases may be involved in a co-operative interaction with Cdc6 protein residues. Two of the seven single-point mutations that we tested in the consensus sequence had a significant effect on Cdc6-1 binding (SM6, G to C, and SM7, G to A), and the location of this guanine residue corresponds directly to one of the ‘triple mutant’ residues shown to disrupt Sulfolobus Cdc6-1–ORB interactions [6]. Interestingly, a guanine corresponding to the residue mutated in SM6 is completely conserved in the ORB sequences of all archaeal species where an origin has been described to date (Figure 6A). Furthermore, guanine, but not cytosine, can form two hydrogen bonds with arginine via the major groove of a double helix without disrupting base pairing [27,28]. Given that the R334A mutation in MthCdc6-1 has previously been shown to abolish ORB binding [10], the most obvious mechanism by which the SM6 mutation might effect changes in MthCdc6-1 binding is if this base interacts with Arg334 via hydrogen bonding. Such an interaction could be key either in the initial positioning of MthCdc6-1 on the DNA, or in stabilizing the protein–DNA interaction once the specific site has been recognized. Arg334 is also completely conserved in the Class I archaeal Cdc6 proteins known to bind to origin sequences, but not in the Class II Cdc6 variants that do not (Figure 6B). Our mutagenesis of MthCdc6-1 has confirmed that Arg334 is essential for the Cdc6 interaction with guanine. Mutating this amino acid to a lysine residue reduced the specificity of binding and enabled Cdc6 to bind to a wider range of origin sequences. Most interestingly, we were able to show that a MthCdc6-1 variant containing a glutamine residue at position 334 displayed altered sequence specificity as it will no longer bind to the wild-type guanine-containing ORB8 sequence, but is now able to recognize cytosine or adenine in its place, albeit with a lower affinity than we observe for the wild-type interaction. Thus it appears that the potential to form hydrogen bonds through a major groove/recognition helix guanine–arginine interaction, as described in the present study, is part of a generally conserved mechanism employed in replication origin binding by all archaeal ORC/Cdc6 proteins. Variation in the surrounding DNA sequences and changes to the amino acid sequence of the recognition helix in the WHD of Class I Cdc6 proteins could thus explain the sequence specificity observed in Cdc6-1 proteins and ORBs from different archaeal species.

A conserved mechanism employed in replication origin binding by all archaeal ORC/Cdc6 proteins?

Figure 6
A conserved mechanism employed in replication origin binding by all archaeal ORC/Cdc6 proteins?

(A) The SM6 guanine (highlighted in black) is conserved in all archaeal origins characterized to date, as shown by an alignment of the consensus sequences of mini-ORB regions from a number of species (based on [10]). (B) ClustalW alignment of archaeal Cdc6 proteins in the region of Arg334 (highlighted in black) showing that this amino acid is completely conserved in all Class I archaeal Cdc6 proteins, but is not present in Class II Cdc6 proteins. Mth, M. thermautotrophicus, Sso, S. solfataricus, Afu, A. fulgidus, Ape, A. pernix, Pfu, P. furiosus.

Figure 6
A conserved mechanism employed in replication origin binding by all archaeal ORC/Cdc6 proteins?

(A) The SM6 guanine (highlighted in black) is conserved in all archaeal origins characterized to date, as shown by an alignment of the consensus sequences of mini-ORB regions from a number of species (based on [10]). (B) ClustalW alignment of archaeal Cdc6 proteins in the region of Arg334 (highlighted in black) showing that this amino acid is completely conserved in all Class I archaeal Cdc6 proteins, but is not present in Class II Cdc6 proteins. Mth, M. thermautotrophicus, Sso, S. solfataricus, Afu, A. fulgidus, Ape, A. pernix, Pfu, P. furiosus.

While this manuscript was under review, crystal structures of S. solfataricus Cdc6-1 and Cdc6-3 forming a heterodimer bound to ori2 DNA [29] and that of A. pernix Cdc6-1 bound to an origin sequence [30] have been published. Both papers provide structural data fully consistent with our conclusions. In particular, Gaudier et al. [30] clearly show the essential interaction through the major groove between a guanine (Gd10) in the ORB and an arginine (Arg345) in Cdc6-1 of A. pernix, which correspond to those identified in our system (and predicted from our results). In addition the structures indicate that the less-conserved 3′ ORB sequence is important for DNA binding and suggest that there are two independent DNA-binding sites that interact with each other, clearly supporting our present results concerning non-sequence-specific competitor. The fact that our results are completely consistent with the findings of these crystal structures serves to reinforce the conserved nature of these interactions across all archaea. We have created models where substitution of a lysine or glutamine residue for Arg345 in the A. pernix structure was visualized (see Supplementary Figure 3 at http://www.BiochemJ.org/bj/409/bj4090511.add.htm; thanks to Dale Wigley for providing the crystal co-ordinates before release and Fred Anston for modelling assistance). This has allowed us to visualize the conserved hydrogen bonds of the lysine substitution, and the potential of a glutamine residue to disrupt this interaction and contribute to the binding phenotypes observed for our mutant MthCdc6-1 proteins.

We thank Alison Walters, Cyril Sanders and Marjan van der Woude for comments on the manuscript and Mrs Magdaléna Országhová for excellent technical assistance. This work was funded by a BBSRC (Biotechnology and Biological Sciences Research Council) David Phillips Research Fellowship (J. P. J. C.), grant APVT-51-024904 from the Science and Technology Assistance Agency (Slovak Republic) and a Royal Society International Short Visit/Fellowship (A. I. M.).

Abbreviations

     
  • Cdc

    cell division cycle

  •  
  • DM

    double mutation

  •  
  • HTH

    helix-turn-helix motif

  •  
  • ORB

    origin recognition box

  •  
  • ORC

    origin recognition complex

  •  
  • SM

    single mutation

  •  
  • WHD

    winged helix domain

References

References
1
Bernander
R.
The archaeal cell cycle: current issues
Mol. Microbiol.
2003
, vol. 
48
 (pg. 
599
-
604
)
2
Edgell
D. R.
Doolittle
W. F.
Archaea and the origin(s) of DNA replication proteins
Cell
1997
, vol. 
89
 (pg. 
995
-
998
)
3
Slesarev
A. I.
Mezhevaya
K. V.
Makarova
K. S.
Polushin
N. N.
Shcherbinina
O. V.
Shakhova
V. V.
Belova
G. I.
Aravind
L.
Natale
D. A.
Rogozin
I. B.
, et al. 
The complete genome of hyperthermophile Methanopyrus kandleri AV19 and monophyly of archaeal methanogens
Proc. Natl. Acad. Sci. U.S.A.
2002
, vol. 
99
 (pg. 
4644
-
4649
)
4
Zhang
R.
Zhang
C. T.
Identification of replication origins in the genome of the methanogenic archaeon, Methanocaldococcus jannaschii
Extremophiles
2004
, vol. 
8
 (pg. 
253
-
258
)
5
Zhang
R.
Zhang
C. T.
Identification of replication origins in archaeal genomes based on the Z-curve method
Archaea
2005
, vol. 
1
 (pg. 
335
-
346
)
6
Robinson
N. P.
Dionne
I.
Lundgren
M.
Marsh
V. L.
Bernander
R.
Bell
S. D.
Identification of two origins of replication in the single chromosome of the archaeon Sulfolobus solfataricus
Cell
2004
, vol. 
116
 (pg. 
25
-
38
)
7
Berquist
B. R.
DasSarma
S.
An archaeal chromosomal autonomously replicating sequence element from an extreme halophile, Halobacterium sp
strain NRC-1. J. Bacteriol.
2003
, vol. 
185
 (pg. 
5959
-
5966
)
8
Liu
J.
Smith
C. L.
DeRyckere
D.
DeAngelis
K.
Martin
G. S.
Berger
J. M.
Structure and function of Cdc6/Cdc18: implications for origin recognition and checkpoint control
Mol. Cell
2000
, vol. 
6
 (pg. 
637
-
648
)
9
Singleton
M. R.
Morales
R.
Grainge
I.
Cook
N.
Isupov
M. N.
Wigley
D. B.
Conformational changes induced by nucleotide binding in Cdc6/ORC from Aeropyrum pernix
J. Mol. Biol.
2004
, vol. 
343
 (pg. 
547
-
557
)
10
Capaldi
S. A.
Berger
J. M.
Biochemical characterization of Cdc6/Orc1 binding to the replication origin of the euryarchaeon Methanothermobacter thermoautotrophicus
Nucleic Acids Res.
2004
, vol. 
32
 (pg. 
4821
-
4832
)
11
Kasiviswanathan
R.
Shin
J. H.
Kelman
Z.
DNA binding by the Methanothermobacter thermautotrophicus Cdc6 protein is inhibited by the minichromosome maintenance helicase
J. Bacteriol.
2006
, vol. 
188
 (pg. 
4577
-
4580
)
12
Gajiwala
K. S.
Burley
S. K.
Winged helix proteins
Curr. Opin. Struct. Biol.
2000
, vol. 
10
 (pg. 
110
-
116
)
13
Grainge
I.
Gaudier
M.
Schuwirth
B. S.
Westcott
S. L.
Sandall
J.
Atanassova
N.
Wigley
D. B.
Biochemical analysis of a DNA replication origin in the archaeon Aeropyrum pernix
J. Mol. Biol.
2006
, vol. 
363
 (pg. 
355
-
369
)
14
Lopez
P.
Philippe
H.
Myllykallio
H.
Forterre
P.
Identification of putative chromosomal origins of replication in Archaea
Mol. Microbiol.
1999
, vol. 
32
 (pg. 
883
-
886
)
15
Zhang
R.
Zhang
C. T.
Single replication origin of the archaeon Methanosarcina mazei revealed by the Z curve method
Biochem. Biophys. Res. Commun.
2002
, vol. 
297
 (pg. 
396
-
400
)
16
Zhang
R.
Zhang
C. T.
Multiple replication origins of the archaeon Halobacterium species NRC-1
Biochem. Biophys. Res. Commun.
2003
, vol. 
302
 (pg. 
728
-
734
)
17
Maisnier-Patin
S.
Malandrin
L.
Birkeland
N. K.
Bernander
R.
Chromosome replication patterns in the hyperthermophilic euryarchaea Archaeoglobus fulgidus and Methanocaldococcus (Methanococcus) jannaschii
Mol. Microbiol.
2002
, vol. 
45
 (pg. 
1443
-
1450
)
18
Lundgren
M.
Andersson
A.
Chen
L.
Nilsson
P.
Bernander
R.
Three replication origins in Sulfolobus species: synchronous initiation of chromosome replication and asynchronous termination
Proc. Natl. Acad. Sci. U.S.A.
2004
, vol. 
101
 (pg. 
7046
-
7051
)
19
Matsunaga
F.
Forterre
P.
Ishino
Y.
Myllykallio
H.
In vivo interactions of archaeal Cdc6/Orc1 and minichromosome maintenance proteins with the replication origin
Proc. Natl. Acad. Sci. U.S.A.
2001
, vol. 
98
 (pg. 
11152
-
11157
)
20
Norais
C.
Hawkins
M.
Hartman
A. L.
Eisen
J. A.
Myllykallio
H.
Allers
T.
Genetic and physical mapping of DNA replication origins in Haloferax volcanii
PLoS Genet.
2007
, vol. 
3
 pg. 
e77
 
21
Robinson
N. P.
Bell
S. D.
Extrachromosomal element capture and the evolution of multiple replication origins in archaeal chromosomes
Proc. Natl. Acad. Sci. U.S.A.
2007
, vol. 
104
 (pg. 
5806
-
5811
)
22
Nölling
J.
Frijlink
M.
de Vos
W. M.
Isolation and characterisation of plasmids from different strains of Methanobacterium thermoformicicum
J. Gen. Microbiol.
1991
, vol. 
137
 (pg. 
1981
-
1986
)
23
Majernik
A. I.
Lundgren
M.
McDermott
P.
Bernander
R.
Chong
J. P.
DNA content and nucleoid distribution in Methanothermobacter thermautotrophicus
J. Bacteriol.
2005
, vol. 
187
 (pg. 
1856
-
1858
)
24
Liang
C.
Weinreich
M.
Stillman
B.
ORC and Cdc6p interact and determine the frequency of initiation of DNA replication in the genome
Cell
1995
, vol. 
81
 (pg. 
667
-
676
)
25
Clarke
A. R.
Engel
P. C.
Enzymology
LabFax
1996
BIOS Scientific Publisher and Academic Press
(pg. 
203
-
204
)
26
Brewer
B. J.
Fangman
W. L.
The localization of replication origins on ARS plasmids in S. cerevisiae
Cell
1987
, vol. 
51
 (pg. 
463
-
471
)
27
Luscombe
N. M.
Laskowski
R. A.
Thornton
J. M.
Amino acid-base interactions: a three-dimensional analysis of protein–DNA interactions at an atomic level
Nucleic Acids Res.
2001
, vol. 
29
 (pg. 
2860
-
2874
)
28
Bloomfield
V. A.
Crothers
D. M.
Tinoco
I.
Jr
Nucleic Acids: Structures, Properties, and Functions
2000
Sausalito, CA
University Science Books
29
Dueber
E. L.
Corn
J. E.
Bell
S. D.
Berger
J. M.
Replication origin recognition and deformation by a heterodimeric archaeal Orc1 complex
Science
2007
, vol. 
317
 (pg. 
1210
-
1213
)
30
Gaudier
M.
Schuwirth
B. S.
Westcott
S. L.
Wigley
D. B.
Structural basis of DNA replication origin recognition by an ORC protein
Science
2007
, vol. 
317
 (pg. 
1213
-
1216
)

Supplementary data