|Home | About | Journals | Submit | Contact Us | Français|
Replication proteins encoded by nonconjugative plasmids from the hyperthermophilic archaea of the order Sulfolobales show great diversity in amino acid sequence. We have biochemically characterized ORF735, a replication protein from pSSVi, an integrative nonconjugative plasmid from Sulfolobus solfataricus P2. We show that ORF735 is a DNA helicase of superfamily 3. It unwound double-stranded DNA (dsDNA) in a 3′-to-5′ direction in the presence of ATP over a wide range of temperatures, from 37°C to 75°C, and possessed DNA-stimulated ATPase activity. ORF735 existed in solution as a salt-stable dimer and was capable of assembling into a salt-sensitive oligomer that was significantly larger than a hexamer in the presence of a divalent cation (Mg2+) and an adenine nucleotide (ATP, dATP, or ADP) or its analog (ATPγS or AMPPNP). Both N-terminal and C-terminal portions of ORF735 (87 and 160 amino acid residues, respectively, in size) were required for protein dimerization but dispensable for the formation of the higher-order oligomer. The protein unwound DNA only as a large oligomer. Yeast two-hybrid and coimmunoprecipitation assays revealed that ORF735 interacted with the noncatalytic subunit of host primase. These findings provide clues to the functional role of ORF735 in pSSVi DNA replication.
DNA helicases are ubiquitous motor proteins that utilize the energy of nucleotide triphosphate hydrolysis to translocate along and unwind the duplex DNA in DNA replication, transcription, recombination, and repair (30). These enzymes often exist as oligomers such as hexamers or dimers. Hexameric DNA helicases are widely found in cellular forms of life as well as viruses and plasmids (13, 44, 45, 48, 54). Both bacterial helicases (represented by Escherichia coli DnaB) (5) and eukaryotic helicases (represented by mammalian MCM) (21, 57) have been extensively studied. Electron microscopy (EM) and image studies have revealed a common ring-shaped structure for all known hexameric helicases (36). This ring-like structure allows the enzyme to encircle the DNA and translocate in a processive fashion.
Most extrachromosomal genetic elements encode their own DNA helicases (13, 44, 45, 48, 54). Our knowledge of these helicases has been derived primarily from studies of bacterial and eukaryotic viruses and plasmids. These genetic elements are replicated through the concerted action of self-encoded and recruited host replication proteins. Many viral or plasmid helicases serve more functions than DNA unwinding. For example, the simian virus 40 (SV40) large T antigen is responsible for the recognition as well as the unwinding of the viral replication origin (9-11). The bifunctional T7 gp4 protein possesses both helicase and primase activities, and its C-terminal helicase domain interacts with T7 DNA polymerase to coordinate helicase and polymerase activities (34).
Replicative helicases from the archaea, the third domain of life, have attracted much attention in the past decade. Since the first report of the biochemical properties of an archaeal MCM protein (Methanobacterium thermoautotrophicum MCM) in 1999 (25), much progress has been made in elucidating the structure and function of archaeal MCM proteins (41). On the other hand, studies of DNA helicases encoded by the extrachromosomal genetic elements from the archaea are lagging behind. Archaea are known to carry a diverse array of plasmids and viruses. A large number of plasmids and viruses have been isolated from hyperthermophilic archaea of the genera Sulfolobus and Acidianus (38, 58). DNA sequence analyses have revealed that all of these plasmids and some of the viruses encode putative helicase proteins. The pRN plasmid family, which includes a group of nonconjugative plasmids from Sulfolobus and Acidianus, encodes a novel RepA protein represented by ORF904 of pRN1 from Sulfolobus islandicus. ORF904 is the only replication protein from a Sulfolobus plasmid that has been extensively studied so far (28, 29). The protein possesses a novel prim-pol (primase/polymerase) domain in the N-terminal part and a helicase domain of superfamily 3 (SF3) in the C-terminal part. The prim-pol domain has primase and polymerase activities, both of which have been characterized. It was also reported previously that ORF904 had a weak unwinding activity with 3′-to-5′ polarity, but a detailed characterization of the helicase activity of the protein has not yet been reported (28).
Plasmid pIT3 (39), isolated from a Sulfolobus solfataricus strain from Italy, and three plasmids, pTAU4, pORA1, and pTIK4 (59), obtained from Sulfolobus neozealandicus from New Zealand, encode replication proteins different from ORF904. The replication proteins from pORA1 and pIT3 harbor the N-terminal prim-pol domain (16, 40). However, the remainder of the proteins are not clearly related to the C-terminal part of ORF904. Plasmid pTAU4 encodes an MCM protein homolog and contains no prim-pol domain (16). The N-terminal 120-amino-acid (aa) extension of the pTIK4 protein shows a significant sequence match to some bacterial DNA replicases, and it also lacks a prim-pol domain (16). It is noteworthy that replication proteins from these cryptic plasmids all contain the AAA+ ATPase region comprising the Walker motifs, which are also found in SF3 helicases (27). Taken together, these observations suggest that the replication proteins may function in DNA unwinding and play roles in the initiation of plasmid DNA replication. However, little is currently known about the mechanisms of DNA replication in Sulfolobus plasmids.
We have previously isolated integrative nonconjugative plasmid pSSVi from S. solfataricus P2 (52). Like pSSVx, pSSVi is capable of packaging into a spindle-like viral particle and spreading with the help of the Sulfolobus spindle-shaped virus SSV1 or SSV2. The pSSVi DNA is 5,740 bp in size and contains eight open reading frames (ORFs). The largest ORF (ORF735) of pSSVi encodes a putative replication protein. ORF735 is only distantly related to putative replication proteins from known nonconjugative Sulfolobus plasmids. It lacks the prim-pol domain but has an SF3 helicase domain. Here we report the biochemical characterization of ORF735. This protein assembles into a salt-sensitive higher-order oligomer significantly larger than a hexamer in the presence of a divalent cation and an adenine nucleotide and unwinds DNA as a higher-order oligomer. In addition, we demonstrated that ORF735 interacts with host primase.
The putative helicase gene (orf735) was amplified by PCR using pSSVi DNA as the template (for primer sequences, see Table Table1)1) and cloned between the NdeI and XhoI sites of plasmid pET-30a(+) (Novagen) to yield pET30a-ORF735, which encodes a fusion protein containing the full-length helicase and a His tag (LEHHHHHH) at the C terminus (predicted molecular mass [MM] of 85.7 kDa).
Expression vectors for deletion mutants of ORF735 were constructed by preparing inserts by PCR using pET30a-ORF735 as the template (for primer sequences, see Table Table1)1) and ligating the inserts into the NdeI and XhoI sites of pET-30a(+). Constructs for single point mutants of ORF735 were prepared by overlap extension PCR (20) using pET30a-ORF735 as the template (for primer sequences, see Table Table1).1). Sequences of the inserts in all constructs were verified by DNA sequencing.
E. coli BL21-CodonPlus (DE3)-RIL cells (Novagen) were transformed with the expression vector for the wild-type or mutant ORF735 protein, and the resulting transformant was grown at 37°C in 1.5 liters of LB medium containing 50 μg/ml kanamycin. When the culture reached an optical density at 600 nm (OD600) of 0.6, protein synthesis was induced by the addition of isopropyl-β-d-thiogalactoside to 1 mM. Following incubation at 37°C for an additional 2.5 h, cells were harvested by centrifugation and resuspended in 20 ml buffer A (20 mM Tris-HCl [pH 8.0], 500 mM KCl, and 10% glycerol) supplemented with a mixture of protease inhibitors (0.5 mM phenylmethylsulfonyl fluoride [PMSF], 1 μg/ml benzamidine, 1 μg/ml aprotinin, and 1 μg/ml leupeptin). The cells were then sonicated on ice. The lysate was centrifuged for 30 min at 20,000 × g at 4°C. The supernatant was subjected to heat treatment at 70°C for 15 min. After centrifugation for 30 min at 20,000 × g at 4°C, the supernatant was loaded onto a Hitrap chelating column (1 ml; Amersham Biosciences) preequilibrated in buffer A and eluted with a 0 to 250 mM imidazole gradient in buffer A (30 ml). Peak fractions were pooled and dialyzed for 3 h against buffer B (20 mM Tris-HCl [pH 8.0], 100 mM NaCl, 0.1 mM EDTA, 0.1 mM dithiothreitol [DTT], and 10% glycerol), and the above-described buffer was changed at intervals of 1 h. The sample was then passed through a 0.22-μm filter (Millipore) and loaded onto a Mono Q 5/50 GL column (1 ml; Amersham) preequilibrated in buffer B. The column was washed with 10 volumes of buffer B and eluted with a 100 to 600 mM NaCl gradient in buffer B (15 ml). Peak fractions were pooled and dialyzed overnight against buffer C (25 mM HEPES-NaOH [pH 7.5] and 20% glycerol). The dialyzed sample was aliquoted and stored at −80°C. Protein concentrations were determined by the Lowry method using bovine serum albumin (BSA) as the standard (31).
Oligonucleotides of 62 and 72 nucleotides (nt) in length (for sequences, see Table Table1)1) were synthesized for the preparation of helicase substrates. The 62-mer oligonucleotide was fully complementary to M13mp19 single-stranded DNA (ssDNA), while the 72-mer oligonucleotide base paired, in the middle region, with M13mp19 ssDNA, forming a poly(dT)15 tail on either side. The oligonucleotides were labeled with [γ-32P]ATP using T4 polynucleotide kinase, and the labeled oligonucleotides were purified by phenol-chloroform extraction and ethanol precipitation. To prepare helicase substrates, mixtures containing an oligonucleotide and M13mp19 ssDNA (molar ratio of 1 to 1.5) were incubated for 3 min at 100°C and subsequently allowed to cool slowly to room temperature.
The substrate used to determine directionality was prepared by cleaving a partial duplex formed by annealing M13mp19 ssDNA with the 32P-labeled 62-mer oligonucleotide with HincII, resulting in a linear, blunt-ended template with a 38-nt 32P-labled oligonucleotide and a 24-nt unlabeled oligonucleotide annealed to the 5′ and 3′ ends, respectively, of the M13 ssDNA. Subsequently, the 3′ end of the 24-nt oligonucleotide was extended 1 nucleotide with Klenow fragment and [α-32P]dGTP, yielding a linear M13 ssDNA molecule with radiolabeled oligonucleotides annealed to both ends.
The standard reaction mixture (20 μl) for DNA helicase assays contained 25 mM HEPES-NaOH (pH 7.5), 25 mM NaCl, 3 mM MgCl2, 1 mM DTT, 0.1 mg/ml BSA, 2 mM ATP, ~1 nM 32P-labled substrate (~1,000 cpm), and specified amounts of ORF735. Reaction mixtures were incubated for 30 min at 60°C, and reactions were stopped by the addition of 5 μl of a stop solution (50 mM EDTA, 0.5% SDS, 30% glycerol, 0.1% bromophenol blue, and 0.1% xylene cyanol FF) to the mixture. The samples were then subjected to electrophoresis in a nondenaturing 12% polyacrylamide gel in 1× Tris-borate-EDTA (TBE) buffer. The gel was exposed to X-ray film. The reaction products were quantitated by ImageQuant Storm PhosphorImager analysis (Amersham Biosciences), and any free oligonucleotide in the absence of enzyme was subtracted.
The standard reaction mixture (20 μl) for ATPase assays contained 25 mM HEPES-NaOH [pH 7.5], 3 mM MgCl2, 1 mM DTT, 0.1 mg/ml BSA, 100 μM [γ-32P]ATP (1 μCi), 1 ng/μl M13 ssDNA, and specified amounts of ORF735. Reaction mixtures were incubated for 30 min at 60°C, and reactions were stopped by the addition of 10 mM EDTA to the mixtures. Aliquots (1 μl) of the reaction mixtures were subjected to thin-layer chromatography on a polyethyleneimine-cellulose plate (Macherey-Nagel) in 1 M formic acid-0.5 M LiCl. ATP hydrolysis was quantitated by PhosphorImager analysis. The amount of spontaneously hydrolyzed ATP was determined by using blank reactions without enzyme and subtracted.
The standard reaction mixture (20 μl) for gel mobility shift assays contained 25 mM HEPES-NaOH (pH 7.5), 25 mM NaCl, 3 mM MgCl2, 1 mM DTT, 0.1 mg/ml BSA, 2 mM ATP, 1 nM 32P-labeled 62-mer oligonucleotide (~2,000 cpm), and specified amounts of protein. After incubation for 10 min at 60°C, the reaction mixtures were subjected to electrophoresis in a 4% polyacrylamide gel containing 3 mM MgCl2 in 0.25× TBE buffer. The gel was exposed to X-ray film and analyzed by using a PhosphorImager apparatus.
Samples (200 μl) were loaded onto a Superdex 200 column (10/300 GL; Amersham Biosciences) equilibrated in a solution containing 25 mM HEPES-NaOH (pH 7.5), 1 mM DTT, 150 mM NaCl, and 10% glycerol and run at 0.3 ml/min at 15°C with a fast protein liquid chromatography (FPLC) (Äkta) system (Amersham Biosciences). The column was calibrated with the following gel filtration markers: ferritin (440 kDa), alcohol dehydrogenase (150 kDa), BSA (66 kDa), and ovalbumin (44 kDa).
ORF735 (~29 pmol) was incubated for 30 min at 37°C in a solution containing 25 mM HEPES-NaOH (pH 7.5), 25 mM NaCl, 3 mM MgCl2, and 2 mM ATP in a total volume of 10 μl. Glutaraldehyde was then added to 0.1%. After 15 min at 37°C, the sample was mixed with a 1/10 volume of a loading solution (20% Ficoll, 0.1 M EDTA, and 0.25% bromophenol blue) and electrophoresed through a 3 to 20% gradient polyacrylamide gel (acrylamide-bis ratio of 80:1) in 50 mM Tris-glycine (pH 8.8) overnight at 140 V (9).
Yeast two-hybrid assays were performed by using Matchmaker two-hybrid system 3 (Clontech). The orf735 gene was amplified by PCR from pSSVi DNA (for primer sequences, see Table Table1)1) and cloned into the NcoI/XhoI sites of the pGADT7 vector, yielding pGADT7-ORF735. The genes encoding the two subunits of the primase from S. solfataricus P2 (Pri1 and Pri2) were amplified by PCR (for primer sequences, see Table Table1)1) and cloned into the EcoRI/BamHI sites of the pGBKT7 vector, yielding pGBKT7-Pri1 and pGBKT7-Pri2, respectively. Saccharomyces cerevisiae strain PJ69-4A was cotransformed with a pGADT7-derived plasmid and a pGBKT7-derived plasmid according to the manufacturer's instruction. For each combination of plasmids, four independent cotransformants were streaked onto synthetic dropout (SD) medium lacking Leu and Trp and incubated for 48 h at 30°C. For plating, cells were collected and diluted to an OD600 of 0.1 with sterile double-distilled water (ddH2O), and aliquots (5 μl) of the cell dilution were spotted on SD medium lacking Leu and Trp as well as SD medium lacking Leu, Trp, and His plus 1 mM 3-aminotriazole (3-AT). The plates were incubated for 3 days at 30°C.
Polyclonal antibodies against ORF735 and Pri2 were raised in rabbit and mouse, respectively, by using recombinant proteins as the antigens. To determine the interaction between ORF735 and Pri2 in vivo, cultures of S. solfataricus P2 and S. solfataricus P2 harboring SSV2 and pSSVi were grown to an OD600 of ~1.0. Cells were harvested and resuspended in a 0.5% volume of a solution containing 20 mM Tris-HCl (pH 8.0), 0.1 mM EDTA, 1 mM DTT, and 10% (wt/vol) glycerol supplemented with protease inhibitors (0.5 mM PMSF, 1 μg/ml benzamidine, 1 μg/ml aprotinin, and 1 μg/ml leupeptin). The cells were lysed by sonication on ice. The lysates were centrifuged at 20,000 × g for 30 min at 4°C. The supernatant was subjected to coimmunoprecipitation with anti-ORF735 antibodies according to the manufacturer's instructions (Clontech). The presence of Pri2 in the immunoprecipitates was detected by immunoblotting with anti-Pri2 antibodies.
ORF735 (GI:77176919) is the largest open reading frame in the pSSVi genome and accounts for about half the size of the plasmid. It shares no significant similarity at the amino acid sequence level with replication proteins encoded by known nonconjugative plasmids from the Sulfolobales, such as ORF866 from pORA1, ORF1053 from pTIK4, ORF904 from pRN1, and ORF759 from pTAU4. Furthermore, the protein lacks the prim-pol domain found at the N terminus of the RepA proteins of pRN plasmids (28, 29).
However, BLAST searches identified an AAA+ ATPase domain (amino acid residues ~200 to 420) and a helix-turn-helix (HTH) domain (amino acid residues ~486 to 554) in ORF735 (Fig. (Fig.1A).1A). The AAA+ ATPase domain shows significant sequence matches to SF3 helicase domains (Fig. (Fig.1B).1B). SF3 helicases (15, 26) are known to be encoded only by small DNA and RNA viruses as well as their prophage remnants in cellular genomes (22). This protein superfamily contains only three conserved motifs: the Walker A motif, which is involved in the phosphate binding of ATP (43, 51); the Walker B motif, which is responsible for metal binding and ATP catalysis (7); and motif C (26), which interacts with the terminal phosphate of ATP and thereby senses nucleotide binding and hydrolysis (17). The three conserved motifs were all found in ORF735. This prompted us to speculate that ORF735 was an SF3 helicase. It was noted that the residue at the last position in the Walker B motif (residue 308) in ORF735 is a Gln instead of an acidic amino acid residue (Glu or Asp) typically found at this position. Presumably, Glu307 serves as the key residue at the active site in ORF735. Sequence alignments also revealed the absence of a B′ motif (26) in the protein. In addition, Arg372 appeared to be a putative Arg finger (Fig. (Fig.1B),1B), which is highly conserved among members of this protein superfamily (18). The HTH domain is known to be involved in DNA-protein interactions (4). All genome-sequenced archaea encode a large number of HTH-containing proteins, and the winged HTH (wHTH) class is the most abundant class of HTH domains in these organisms (2). The wHTH domain, in addition to three core helices, contains a C-terminal β-hairpin termed the wing (8). As shown by sequence alignment, the HTH domain of ORF735 belongs to the wHTH class (Fig. (Fig.1C1C).
To examine the activities of ORF735, the gene encoding the protein was cloned and overexpressed in E. coli cells, and the recombinant protein was purified (Fig. (Fig.2).2). Purified ORF735 was assayed for helicase, DNA polymerase, and primase activities. The protein was active in assays of DNA unwinding (Fig. (Fig.3)3) but showed no DNA polymerase or primase activity (data not shown). Therefore, we conclude that ORF735 is a DNA helicase.
Despite its hyperthermophilic origin, ORF735 was able to release the oligonucleotide from the duplex substrate within a wide range of temperatures, from 37°C to 75°C (see Fig. S1 in the supplemental material). The activity of the protein was significantly reduced at 20°C. The rate of the helicase reaction was about twice as fast at 60°C as that at 37°C (Fig. (Fig.4).4). Like other helicases, ORF735 required a divalent cation (Mg2+, Mn2+, or Ca2+) for its helicase activity, with Mg2+ being the preferred cofactor (Fig. S2). The helicase activity of the protein was sensitive to salt and inhibited by ≥100 mM NaCl (Fig. S3).
The helicase activity of ORF735 also depended on the presence of ATP. The protein was optimally active at 2 mM ATP (data not shown). Among other tested nucleotides, only dATP could substitute for ATP to a limited extent (Fig. (Fig.5).5). Interestingly, the helicase activity was abolished in the presence of AMPPNP and greatly diminished but not completely abolished in the presence of ATPγS. Similar observations were reported previously for the MCM proteins from S. solfataricus (6) and M. thermoautotrophicum (46). These data suggest that ATP binding and hydrolysis were required for the unwinding reaction of the protein.
The directionality of DNA unwinding by ORF735 was determined by using a single-stranded DNA template with radiolabeled oligonucleotides annealed to both ends (Fig. (Fig.6A).6A). As shown in Fig. Fig.6B,6B, the oligonucleotide annealed to the 5′ end of the template was released more efficiently by the protein than that annealed to the 3′ end of the template, indicating that ORF735 translocated along the template in a 3′-to-5′ direction.
ORF735 was able to unwind double-stranded DNA (dsDNA) with a single-stranded DNA tail (Fig. (Fig.4).4). Moreover, similar unwinding efficiencies were observed whether the DNA template contained one or two ssDNA tails (data not shown). Unlike SV40 LT and BPV E1 (10, 11, 55), ORF735 was unable to bind stably and initiate the unwinding of a fully complementary dsDNA (see Fig. S4 in the supplemental material). Furthermore, no unwinding of native pSSVi DNA by the enzyme was detected by using a method described previously by Dean et al. (10) (data not shown).
ORF735 showed DNA-stimulated ATPase activity. The level of ATPase activity of the protein was low in the absence of DNA but was drastically increased in the presence of either dsDNA or ssDNA. Single-stranded DNA was more effective than dsDNA in stimulating the ATPase activity of the protein (Fig. (Fig.7A).7A). This contrasts with the finding that the ATPase activity of ORF904 from pRN1 was stimulated more strongly by dsDNA than by ssDNA (28). The effect of DNA on the ATPase activity of ORF735 was concentration dependent (Fig. (Fig.7B).7B). The ATPase activity was most strongly stimulated by DNA at low DNA concentrations (<5 nM). The level of stimulation decreased and eventually leveled off as the DNA concentration increased.
During the purification of recombinant ORF735, the protein was occasionally found to be partially degraded. A major degradation product contained residues 88 to 575, as revealed by N-terminal amino acid sequencing and matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) analysis and was denoted Hel88-575. Two truncated versions of the protein, i.e., Hel88-575 and Hel88-735, were constructed (Fig. (Fig.1A1A and and2).2). The subsequent purification of recombinant Hel88-575 yielded yet another major degradation product, which was identified as Hel177-575. Recombinant Hel177-575 was subsequently prepared (Fig. (Fig.2).2). The mutant protein Hel88-515, which lacks helix 3 and the wing elements of the wHTH domain, was also constructed (Fig. (Fig.2).2). Attempts to prepare a soluble mutant protein lacking the entire wHTH domain were unsuccessful.
The truncated mutant proteins were tested for helicase activity (Fig. (Fig.8).8). Both Hel88-735 and Hel88-575 were as active in DNA unwinding as wild-type ORF735, whereas Hel177-575 showed little helicase activity even at very high concentrations (e.g., 300 nM). It appears that neither the N-terminal nor the C-terminal portion (87 and 160 amino acid residues in size, respectively) of the protein was required for the helicase activity of the protein, while the region from residue 88 through residue 176 was indispensable for the activity. On the other hand, the helicase activity of Hel88-515 was much lower than that of the wild-type protein, suggesting that the wHTH domain plays an important role in DNA unwinding by the protein. Archaeal MCM proteins are known to contain a distinct miscellaneous HTH (mHTH) domain at the C terminus (2). MCM proteins from S. solfataricus and M. thermoautotrophicum have been studied by deletion analysis (3, 23). Unlike Hel88-515, the MCM mutant proteins lacking the mHTH domain showed helicase activity similar to or higher than that of the wild-type proteins.
To test whether Glu307 serves as a key residue in the Walker B motif and Arg372 serves as an Arg finger, two single-point-mutant proteins, the E307A and R372I mutants, were also made (Fig. (Fig.2).2). The ATPase and helicase activities of the two point mutants were measured (Fig. (Fig.9).9). The E307A mutant was inactive in DNA unwinding and was partially active in the ATPase assay. The R372I mutant showed very low levels of ATPase activity and no helicase activity. This finding agrees with the suggestion that Arg372 is an arginine finger in ORF735 since it was reported previously that a mutation of the arginine finger in an AAA+ protein resulted in a significant deficiency in the ATPase activity of the protein (1, 33). The residual ATPase activities of the two mutant proteins did not result from a contaminating ATPase since the substitution of an alanine for the conserved lysine residue in the Walker A motif of the two mutants completely abolished the ATPase activities of the resultant mutant proteins (K265A/E307A and K265A/R372I), which were prepared in the same manner as that used for the single mutants (see Fig. S5 in the supplemental material).
The E307A and R372I mutants were also compared with the wild-type protein for their affinities for ssDNA. Both wild-type and mutant proteins were unable to bind ssDNA in the absence of a nucleotide and showed similar affinities for ssDNA in the presence of ATP (data not shown). Surprisingly, the ssDNA binding by the wild-type and mutant proteins was affected differently when ATP was replaced with ATPγS or ADP (Fig. (Fig.10).10). The affinity of the wild-type protein for ssDNA was moderately reduced in the presence of ATPγS and drastically decreased in the presence of ADP. The E307A mutant bound to ssDNA as tightly in the presence of ATPγS or ADP as in the presence of ATP, whereas the R372I mutant showed a very low affinity for ssDNA in the presence of ATPγS or ADP.
ORF735 eluted as a dimer on a gel filtration column (see Fig. 12A). The dimerization of the protein was not sensitive to salt (up to 0.5 M NaCl) and was independent of Mg2+ and nucleotides (data not shown). To learn more about the ability of ORF735 to oligomerize, we carried out a nondenaturing gradient gel electrophoresis assay (Fig. (Fig.11).11). In this assay, ORF735 was first incubated for 30 min at 37°C under various conditions and then cross-linked with 0.1% glutaraldehyde. The cross-linked products were subjected to electrophoresis in a polyacrylamide gel. As shown in Fig. 11A, at least seven protein bands were observed in the presence of both Mg2+ and ATP. The slow-migrating smear with an average molecular mass of greater than 669 kDa accounted for the bulk of the input protein. Based on their molecular weights, the six well-resolved bands below the smear were presumably the monomeric, dimeric, trimeric, tetrameric, pentameric, and hexameric forms of the protein, respectively. It appears that the protein was cross-linked efficiently into higher-order oligomers, most of which were substantially larger than a hexamer. The apparent heterogeneity of the large cross-linked products may have resulted from different cross-linking patterns. When either Mg2+ or ATP was omitted from the assay, only monomers, dimers, and trimers were observed. Intriguingly, ATPγS, AMPPNP, ADP, or dATP was able to substitute for ATP in the assay, whereas other deoxynucleoside triphosphates (dNTPs) and ribonucleoside triphosphates (rNTPs) were not (Fig. 11B). Therefore, we conclude that the formation of a higher-order ORF735 oligomer depends on the presence of Mg2+ and an adenine nucleotide but not on the hydrolysis of the nucleotide.
Assembly of the higher-order oligomer appeared to involve weak ionic protein-protein interactions, which were destabilized by 100 mM NaCl (data not shown). Apparently, protein-protein interactions responsible for the formation of a dimer are different from those for the assembly of a higher-order oligomer. Attempts to characterize the higher-order oligomer by gel filtration, analytical ultracentrifugation, or cryo-electron microscopy using high protein concentrations (up to 10 μM) in the presence of Mg2+ and ATP were unsuccessful. It is possible that ORF735 formed the higher-order oligomer under still-higher protein concentrations. It was reported previously that the greater-than-hexameric forms of the MCM from S. solfataricus were detectable only at a very high protein concentration (25 μM) (3). Similarly, the MCM protein from M. thermoautotrophicum was also found to exist primarily as dodecamers only at high protein concentrations (47). However, we were unable to perform the assays at higher ORF735 concentrations because the protein formed aggregates under these conditions.
We then determined the ability of the ORF735 mutant proteins to form oligomers by both gel filtration and native gradient gel electrophoresis. The two point mutants (E307A and R372I) came off the Superdex 200 column as dimers and were able to form higher-order oligomers under the same conditions as those used for the wild-type protein (data not shown), ruling out the possibility that the loss of helicase activity in the E307A and R372I mutants (Fig. (Fig.9A)9A) was due to the failure of the two proteins to form a dimer or higher-order oligomer.
The truncated mutant proteins all eluted as a single peak on the gel filtration column (Fig. 12A). Hel88-735 (76 kDa) peaked at a position corresponding to an estimated molecular mass (MM) of 120 kDa, suggesting that the loss of 87 amino acid residues at the N terminus destabilized the dimer such that the protein existed in an equilibrium between the monomeric and dimeric forms. Hel88-575 (57.5 kDa) was eluted at an estimated MM of 62 kDa and, therefore, was unable to form a dimer. These results suggest that the N- and C-terminal portions of ORF735 are involved in the dimerization of the protein. As expected, Hel177-575 (47 kDa) and Hel88-515 (50 kDa) were also defective in forming a dimer.
When analyzed by chemical cross-linking followed by nondenaturing gradient gel electrophoresis, Hel88-735, Hel88-575, and Hel88-515 were able to assemble into higher-order oligomers in the presence of Mg2+ and ATP (Fig. 12B). Intriguingly, Hel88-515 appeared to be more readily cross-linked into higher-order oligomers than the wild-type and other mutant proteins. In contrast, Hel177-575 was cross-linked only into dimers and trimers even in the presence of Mg2+ and ATP. Therefore, neither the N-terminal nor the C-terminal portion of ORF735 was required for the oligomerization of the protein, although they were involved in the dimerization of the protein. Presumably, residues 88 to 176 were involved in the assembly of the higher-order ORF735 oligomer.
Viruses and plasmids are known to encode one or a few replication proteins that often function in the initiation stage of replication, such as origin recognition, DNA unwinding, and primer synthesis, etc., and serve to recruit the host DNA replication machinery to complete their genome replication. Virus- or plasmid-host protein interactions are involved in the recruitment of host replication proteins. We employed yeast two-hybrid assays to identify possible interactions between the ORF735 and S. solfataricus P2 replication proteins. Among the tested host replication proteins, the noncatalytic subunit (Pri2), but not the catalytic subunit (Pri1), of primase was found to interact with ORF735 (Fig. 13A). To confirm this observation, we coinfected S. solfataricus P2 cells with SSV2 and pSSVi. Cells in the early exponential growth phase were harvested and lysed. The cell extract was subjected to immunoprecipitation with antibodies against ORF735 (anti-ORF735). Pri2 was readily coprecipitated with ORF735 (Fig. 13B). In a control experiment, where S. solfataricus P2 cells were not infected, Pri2 was not precipitated with anti-ORF735, excluding the possibility that the antibodies interacted nonspecifically with the extract. Based on these results, we conclude that ORF735 interacts specifically with the noncatalytic subunit of host primase in S. solfataricus.
Replication proteins encoded by nonconjugative plasmids from the Sulfolobales show remarkable diversity in amino acid sequence, in striking contrast to the conservation in genome organization among these plasmids. In this study, we have demonstrated that ORF735, a replication protein from pSSVi, is an SF3 helicase. This is the first report of the biochemical characterization of a helicase from Sulfolobus nonconjugative plasmids. ORF735 forms a salt-stable dimer in the absence of a divalent cation or nucleotide and assembles into a higher-molecular-weight, salt-sensitive oligomer in the presence of both Mg2+ and an adenine nucleotide. The higher-order oligomer was substantially larger than a hexamer and was possibly a double hexamer. The ORF735 helicase appears to function only as a higher-order oligomer, as indicated by the following observations. First, both the formation of the higher-order oligomer and DNA unwinding by ORF735 depended on the presence of Mg2+ and ATP (or dATP) and were salt sensitive. Second, the deletion mutants Hel88-575 and Hel88-515, which were able to form a higher-order oligomer but not a dimer, showed helicase activity, whereas Hel177-575, which was unable to form a higher-order oligomer, was inactive in DNA unwinding. In addition, Hel88-735, Hel88-575, and Hel88-515 all bound ssDNA as well as the wild-type protein, whereas Hel177-575 showed little affinity for ssDNA in the presence of Mg2+ and ATP (data not shown). Therefore, we conclude that ORF735 binds and unwinds DNA as a higher-order oligomer.
However, the nucleotide requirement of ORF735 in protein oligomerization differed from that in DNA unwinding. While the binding of an adenine nucleotide or its analog (ADP, ATPγS, or AMPPNP) is sufficient for protein oligomerization, the hydrolysis of an adenine nucleoside triphosphate (ATP or dATP) is required for the helicase activity. This is consistent with the findings that DNA helicases, such as T7 gp4 (19, 35), T4 gp41 (13), and SV40 LT (42, 50), formed hexamers in the presence of NTP, nucleoside diphosphate (NDP), NMPPNP, NMPPCP, or NTPγS but depended on NTP for the helicase activity.
Oligomeric ORF735 appeared to be able to undergo conformational changes, which were probably accompanied by changes in the affinity of the protein for ssDNA. Wild-type ORF735 and the E307A and R372I mutants were all able to form higher-order oligomers in the presence of Mg2+ and one of the following nucleotide cofactors: ATP, ATPγS, and ADP. Wild-type ORF735 bound to ssDNA tightly in the presence of ATP and weakly in the presence of ADP, suggesting that oligomeric ORF735 may exist in a conformation displaying either a high or low affinity for ssDNA. Several helicases have been shown to interact more strongly with DNA in the presence of NTP than in the presence of NDP. For example, T7 gp4 hexamers formed in the presence of dTTP were capable of binding ssDNA, whereas those formed in the presence of dTDP did not bind DNA (37). The affinity of DnaB for ssDNA was 3 orders of magnitude higher in the presence of ATP than in the presence of ADP (24). Therefore, we propose that there exists in ORF735 an ATP/ADP-dependent molecular switch, which, by controlling the conformational changes in the protein oligomer, couples ATP hydrolysis cycles to the translocation of the helicase along the DNA (Fig. (Fig.14).14). In this explanation, ORF735 exists as a stable dimer that is incapable of oligomerization in the absence of a nucleotide. In the presence of ATP, the protein undergoes conformational changes upon binding to the adenine moiety of ATP and assembles into a higher-order oligomer. The ATP-bound higher-order oligomer exists in a high-affinity conformation and binds tightly to ssDNA. Once ATP is hydrolyzed, the ADP-bound higher-order oligomer converts into a low-affinity conformation and displays a low affinity for ssDNA. This low-affinity conformation facilitates the transient release of the enzyme from DNA. Upon the subsequent rebinding of ATP, the enzyme will rebind the DNA tightly at a different position. Conceivably, a helicase will be unable to move along and unwind DNA if it stays tightly bound to the DNA in the presence of either ATP or ADP. Indeed, the E307A mutant, which showed similarly high affinities for ssDNA in the presence of ATP, ATPγS, or ADP, completely lost helicase activity, although it remained partially active as an ATPase. Since the E307A mutant remained capable of assembling into a higher-order oligomer and ssDNA binding, it was active in nucleotide binding. These data are consistent with previous observations that a mutation of the key residue in the Walker B motif of T7 gp4, E. coli rho, or RuvB impaired nucleotide hydrolysis but not nucleotide binding (12, 32, 53). Glu307 is the conserved residue in the Walker B motif of ORF735 and therefore presumably acts as the catalytic residue in ATP hydrolysis (7). Although the mutation of Glu307 did not impair nucleotide binding, the lack of a functional side chain probably resulted in the failure of the ATP/ADP-dependent molecular switch to distinguish between ATP and ADP, preventing the higher-order oligomer from undergoing conformational changes. Therefore, the mutant protein became stuck in the high-affinity conformation whether it was bound by ATP, ADP, or ATPγS.
On the other hand, the ATP/ADP switch appeared to function normally in the R372I mutant, which bound ssDNA in the presence of ATP but not in the presence of ADP. Arg372 is a putative arginine finger. Structural studies have shown that an arginine finger, first identified in GTPase-activating proteins (56), acts in trans, contributing its functional side chain to the ATPase site and contacting the γ-Pi of the nucleotide bound to the adjacent subunit, thereby influencing the rate of ATP hydrolysis (14, 48). The mutation of this residue resulted in a defect in the ATP hydrolysis of the protein. Therefore, although R372I was capable of undergoing conformational changes, it was defective in helicase activity.
Unlike ORF904 from pRN1, ORF735 showed no DNA polymerase or primase activity. Conceivably, the protein may serve to recruit host DNA replication proteins to effect the replication of the plasmid. Indeed, we found that ORF735 was able to interact with the noncatalytic subunit of host primase. This interaction may couple the synthesis of Okazaki fragments on the lagging strand with the progressing replication fork of the plasmid DNA, as was shown previously for the interaction between DnaG and DnaB in bacteria (49). However, the steps prior to DNA unwinding by ORF735 are unclear. We were unable to detect a site-specific binding or unwinding of the pSSVi DNA by the protein. Additional replication proteins or accessory proteins may be required for the recognition and localized unwinding of the replication origin of the plasmid. It is also possible that additional proteins required for the preunwinding steps are provided by the helper virus since pSSVi appeared to exist as a free plasmid in the host cell only in the presence of a fusellovirus (52).
This work was supported by grant 2004CB719603 from the National Basic Research Program of China, grants 30730003 and 30621005 from the National Natural Science Foundation of China, and grant KSCX2-YW-G-023 from the Chinese Academy of Sciences.
Published ahead of print on 29 January 2010.
†Supplemental material for this article may be found at http://jb.asm.org/.