The traE Gene of Plasmid RP4 Encodes a Homologue of Escherichia coli DNA Topoisomerase III*

The polypeptide encoded by the plasmid RP4traE gene shows extensive protein sequence similarity toEscherichia coli topB, the gene encoding DNA topoisomerase III (Topo III). The traE gene product has been cloned into a bacteriophage T7-based transient expression system, and the polypeptide has been expressed and purified. The TraE protein exhibits topoisomerase activity similar to that of Topo III. Relaxation is stimulated by high temperature and low concentrations of Mg2+. In addition, similar to E. coli Topo III, the TraE protein is a potent decatenase and can substitute for Topo III activity in vivo. The biochemical properties of the TraE protein in vitro suggest that the protein may be involved in the resolution of plasmid DNA replication intermediates either during vegetative replication or in conjugative DNA transfer. Putative homologues of Topo III have been found to be encoded by other broad host range, conjugative plasmids isolated from both Gram-negative and Gram-positive organisms, suggesting that Topo III-like polypeptides may have an essential role in the propagation of many promiscuous plasmids.

Plasmids are autonomously replicating extrachromosomal elements. In addition to replicating within a single organism, many of these elements possess the ability of intraspecies and interspecies transfer. Hence, many plasmids can be classified as mobile genetic elements. These mobile genetic elements are often clinically relevant as many of these "promiscuous" plasmids can encode polypeptide(s) that confer resistance to single/ multiple antibiotic(s) and heavy metal ion(s). These resistance genes are often contained within a transposon that has integrated into the plasmid molecule (for reviews, see Refs. [1][2][3][4]. Plasmid RP4 is a 60,099-base pair broad host range, transmissible genetic element that belongs to the IncP␣ incompatibility group. The RP4 plasmid genome has been completely sequenced and revealed the presence of 74 genes (5). The majority of DNA metabolic enzymes used by plasmid RP4 are host-encoded; however, the plasmid itself encodes several specialized DNA metabolic enzymes within its genome. These include proteins that are involved in partitioning (6 -8), plasmid transfer (for a review, see Refs. 1 and 9), and DNA repli-cation (for a review, see Ref. 10). Of the transfer genes, the traI gene encodes a specialized DNA topoisomerase activity, the "nickase," which is responsible for catalyzing the single-strand cleavage of plasmid molecules required for the initiation of transfer (11)(12)(13). In addition to the TraI protein, sequence analysis has revealed that the RP4 genome encodes another potential topoisomerase activity. The traE gene encodes a putative polypeptide with extensive protein sequence similarity to Escherichia coli DNA topoisomerase III (Topo III) 1 (14).
As a preliminary step in the elucidation of the role of the TraE protein in plasmid RP4 DNA metabolism, the traE gene was cloned, and its gene product was overproduced and purified. The purified TraE protein has been shown to possess topoisomerase activity and can substitute for Topo III in vivo; therefore, this topoisomerase is a true homologue of E. coli Topo III. An understanding of the role of the TraE protein in plasmid RP4 DNA metabolism may shed light on the role of Topo III in cellular DNA metabolism.

MATERIALS AND METHODS
DNA and Nucleotides-X174 replicative form I DNA (covalently closed, negatively supercoiled DNA molecule(s)) was purchased from Life Technologies, Inc. DNA oligonucleotides were prepared by the University of Maryland Biopolymer Laboratory. Radiolabeled nucleoside triphosphate was purchased from Amersham Corp.
Enzymes and Reagents-Acrylamide, restriction enzymes, and agarose were from Life Technologies, Inc. DE52 and P-11 cellulose were from Whatman. Trypsin inhibitor-agarose and single-stranded DNAcellulose was from Sigma. Sephacryl S-200 was from Pharmacia Biotech Inc.
Protein Determination-Protein concentration was determined by the method of Bradford (15) using a Bio-Rad protein assay kit.
Radiolabeling of Oligonucleotides-Oligonucleotides were 5Ј-end-labeled using bacteriophage T4 polynucleotide kinase (Life Sciences, Inc.) and [␥-32 P]ATP as per the manufacturer's recommendations. The labeled oligonucleotides were fractionated on a polyacrylamide gel. The region containing the labeled oligonucleotide was excised, and the DNA was isolated by direct elution of the fragment into 10 mM Tris-HCl (pH 7.5 at 22°C) and 1 mM EDTA. The radiolabeled oligonucleotides were diluted to a specific activity of 5000 cpm/pmol by the addition of excess unlabeled oligonucleotide.
Construction of the TraE Expression Vector-The gene encoding the TraE protein was amplified by the polymerase chain reaction using the following primer sequences: 5Ј-GAGGGTAGGGGGACATATGCAATT-TGAACG-3Ј (amino-terminal) and 5Ј-ATCATGGATCCCTTTCACTC-CTGGTTGGTG-3Ј (carboxyl-terminal). The amplified product contained an NdeI restriction site (shown in boldface) encompassing the initiation codon of the traE open reading frame and a BamHI site (shown in boldface) after the termination codon of the traE open reading frame. The amplified product was digested with NdeI and BamHI and ligated with NdeI/BamHI-digested plasmid vector pET-3c (pTraE). To ensure that no mutations occurred during the polymerase chain reaction amplification process, the expression vector was also constructed using site-specific mutagenesis. A 3348-base pair SphI fragment of plasmid RP4 that encompasses traE was subcloned into M13mp19 replicative form DNA. Uracil-containing single-stranded DNA was isolated using this construct, and oligonucleotide-directed site-specific mutagenesis (16) was used to create the NdeI and BamHI sites at the beginning and end of the gene, respectively. The mp19 DNA (replicative form I) that contained the altered traE gene was cleaved with NdeI and BamHI, isolated, purified, and ligated with NdeI/BamHI-digested pET-3c. The biochemical properties of the purified TraE protein obtained from either construct were indistinguishable.
Purification of the TraE Protein-The induction of the TraE polypeptide was initiated by infection of the expression strain, harboring the pTraE plasmid DNA, with bacteriophage CE6 (17). Chromatography on trypsin inhibitor-agarose was included in the purification to reduce proteolysis (18). To prevent any contamination of the TraE protein with endogenous Topo III, it was purified from E. coli strain BL21 in which the gene encoding Topo III (topB) had been disrupted (18). The TraE protein was purified by a modification of a previously described protocol that included DE52, Bio-Gel HT, single-stranded DNA-cellulose, and Sephacryl S-200 chromatography (19).
Superhelical DNA Relaxation Assays-Superhelical DNA relaxation reaction mixtures (25 l) contained 40 mM Hepes-KOH buffer (pH 8.0 at 22°C), 1 mM magnesium acetate (pH 7.0), 0.1 mg/ml bovine serum albumin, 40% (v/v) glycerol, 200 ng of X174 replicative form I DNA, and the indicated amount of topoisomerase (20). Reactions were incubated at 52°C for 10 min, and the reaction products were separated on an agarose gel and visualized by staining with ethidium bromide as described previously (19).
oriC DNA Replication Assay-The replication of oriC-containing DNA in vitro was performed as described previously (18). The replication products were separated by agarose gel electrophoresis and visualized by autoradiography (21). The percentage of replication products existing as form II molecules (nicked or gapped circular DNA molecule(s)) was quantified using a Fuji BAS 1000 phosphoimager.
Topoisomerase-induced DNA Cleavage Assay-Reaction mixtures (5 l) contained 40 mM Hepes-KOH buffer (pH 8.0 at 22°C), 0.1 mg/ml bovine serum albumin, 1 mM magnesium acetate (pH 7.0), and 5 pmol of a radiolabeled 45-base oligonucleotide. The oligonucleotide was 5Ј-CAGAATCAGAATGAGCCGCAACT2TCGGGATGAAAATGCTCAC-AAT-3Ј, where the arrow indicates the site of Topo III cleavage. The indicated amounts of Topo III and TraE were incubated for 3 min at 37°C, and the reaction was stopped by the addition of SDS to 2%. The reactions were adjusted to 45% formamide, 10 mM EDTA, 0.025% bromphenol blue, and 0.025% xylene cyanol and heat-denatured for 5 min at 90°C. The reaction products were separated by electrophoresis on a polyacrylamide gel (19:1) containing 50% (w/v) urea. The gels were then dried and autoradiographed.
Bacteriophage P1 Transduction-Bacteriophage P1 lysates were prepared from E. coli strain K38 topB:kan r . This strain contains a topB gene disruption where a kanamycin resistance cassette has been inserted into the EcoRV site within the gene (14). E. coli strain DM750, harboring the appropriate plasmid DNAs, was grown overnight in LB medium (5 ml) containing 200 g/ml ampicillin. The cells were harvested and resuspended in 5 ml of sterile 0.1 M magnesium sulfate and 5 mM calcium chloride. The resuspended cells were then incubated for 20 min at 37°C with gentle agitation. The cells (0.1 ml) were mixed with 0.1 ml of the P1 lysate and allowed to stand for 20 min at 37°C. Sodium citrate (0.2 ml of a 1 M solution) was then added to the mixture, and 0.2 ml of the mixture was distributed onto LB agar plates containing 50 g/ml kanamycin and 100 g/ml ampicillin. The plates were incubated for 24 h at 37°C, and the number of kanamycin/ampicillinresistant colonies was recorded. The mixture was also plated on agar medium containing only ampicillin to determine the total number of transductants in the experiment.

RESULTS
The traE Gene Shows Extensive Protein Sequence Similarity to E. coli topB-A search of the National Center for Biotechnology Information (NCBI) nonredundant data base has revealed that E. coli topB shows extensive protein sequence similarity to the plasmid RP4 traE gene (Fig. 1A). The traE open reading frame is capable of encoding an 82-kDa polypeptide. The similarity between the polypeptides encoded by traE and topB extends through their first 600 amino acid residues and then diverges in the carboxyl-terminal residues. The region of the similarity is identical to that observed between E. coli topB and topA (the gene encoding DNA topoisomerase I (Topo I)) (14,22). The carboxyl-terminal residues of the putative TraE protein contain two potential zinc-finger motifs that show protein sequence similarity to one of the zinc-finger motifs of Topo I (23) and to each other (Fig. 1B).
The Plasmid RP4 TraE Protein Exhibits Superhelical DNA Relaxation Activity-To assess whether the traE gene encoded a DNA topoisomerase activity, the gene was subcloned from plasmid RP4 into plasmid pET-3c, a bacteriophage T7 RNA polymerase-based transient expression plasmid (17). Cells containing the traE expression plasmid pTraE were induced by the addition of bacteriophage CE6, which contains the gene encoding T7 RNA polymerase under the control of the bacteriophage P L promoter. The TraE protein was then purified from the induced cells using the same purification protocol as that used for E. coli Topo III (19). The final preparation contained a polypeptide with the expected molecular mass of 82 kDa and two minor proteolysis products (Fig. 2). The regions of local alignment are shown in the context of the fulllength polypeptides. Colons represent amino acid identity, and periods represent a conservative change. B, protein sequence alignment of the potential zinc-finger motifs of the TraE protein and one of the zincfinger motifs of Topo I. The MacVector protein sequence alignment program was used to align the potential zinc-finger motifs of the TraE protein (traE) with one of the known zinc fingers of E. coli Topo I (topA) (23). The cysteine residues that constitute the zinc-finger motif are shown in boldface.
The purified TraE polypeptide was assayed for superhelical DNA relaxation activity. A titration of TraE revealed that the protein possessed DNA relaxation activity comparable to that of Topo III (Fig. 3). In addition, the superhelical DNA relaxation activity of TraE exhibited virtually the same biochemical properties as that of Topo III. TraE-catalyzed relaxation of superhelical DNA was stimulated by increasing temperatures (Fig. 4, lanes 2-6 and 7-11) and inhibited by increasing concentrations of Mg 2ϩ (compare lanes 2-6 and 7-11).
TraE Protein-catalyzed Cleavage Sites Are Identical to Those of E. coli Topo III-Type 1 DNA topoisomerases act by binding to DNA, making a transient single-stranded break in the DNA, catalyzing a strand passage event, and resealing the transient break (for a review, see Ref. 24). Although E. coli Topo I and Topo III possess a high degree of protein sequence similarity (14), the two enzymes cleave single-stranded DNA substrates at different sites (19,25); therefore, the TraE protein was analyzed for its cleavage site specificity (Fig. 5). In contrast to Topo I (Fig. 5, lanes 8 -10), TraE-catalyzed cleavage of a 45-base oligonucleotide substrate (lanes 5-7) generated a pattern identical to that of Topo III (lanes 2-4). The 45-base substrate contained one predominant Topo III cleavage site and multiple minor sites. TraE protein-catalyzed cleavage of the substrate occurred predominantly at the major Topo III cleavage site (Fig. 5, compare lanes 4 and 5). Since only a small number of Topo III cleavage sites are present on the substrate, it is possible that TraE protein-catalyzed cleavage of single-stranded DNA can occur at sites distinct from those of Topo III. However, at a minimum, the two enzymes do have a subset of cleavage sites in common. Cleavage of the substrate by the TraE protein and Topo I is more efficient than cleavage by Topo III. In addition, with increased levels of TraE protein, a cleavage product accumulates that is identical to that of Topo I (Fig. 5,  compare lanes 7 and 8). This product has also been observed using large amounts of Topo III in a cleavage reaction (26).
The TraE Protein Exhibits Decatenation Activity-It has previously been shown that E. coli Topo III and Topo I catalyze distinctly different reactions in vitro (18). DNA topoisomerase I was very efficient in the relaxation of negatively supercoiled DNA (i.e. removing intramolecular linkages), whereas DNA  12) or 120 fmol of TraE protein (lanes 2-11). Reactions were incubated at 30°C (lanes 2 and 7), 37°C (lanes 3 and 8), 42°C (lanes 4 and 9), 52°C ( lanes 5 and 10), and 65°C (lanes 6 and 11) in the presence of either 5 mM (lanes 1-6) or 1 mM (lanes 7-12) magnesium acetate. The reaction products were electrophoresed on a 1% agarose gel, and the DNA was visualized by staining with ethidium bromide. oc, open circle nicked or gapped circular DNA; sc, negatively supercoiled circular DNA. topoisomerase III was very efficient in the decatenation of multiply interlinked plasmid DNA dimers and the resolution of DNA replication intermediates (i.e. intermolecular linkages). To assess whether the TraE protein was capable of resolving DNA replication intermediates, the enzyme was included in an in vitro replication reaction. The products of the reaction were separated on an agarose gel, and the products were visualized by autoradiography (Fig. 6). Similar to Topo III, the TraE protein was capable of resolving plasmid replication intermediates (Fig. 6, lanes 2-5). The specific activity of the TraE protein was 30% of that of Topo III (Fig. 6, lanes 7-10).
The traE Gene Can Substitute for topB in Vivo-In the course of constructing E. coli strains that lack the activities of the two type 1 topoisomerases, Topo I and Topo III, it has been observed that it is extremely difficult to transduce a topB gene disruption into topA deletion strains that contain gyrase compensatory mutations (i.e. DM750 and DM800 (27,28)); however, in the presence of a Topo III expression plasmid, pDE1 (14), the disruption of the chromosomal copy of topB in these strains is easily accomplished. 2 This is presumably because promiscuous transcription of topB from the expression plasmid provides the cell with sufficient amounts of the enzyme to remain viable in the absence of the chromosomal gene. Western blot analysis of cell extracts prepared from cells harboring plasmid pDE1 reveals clearly detectable amounts of Topo III, whereas cell extracts prepared from cells that do not contain the expression plasmid do not reveal detectable quantities of the enzyme (data not shown). This observation is consistent with the extremely low abundance of Topo III in E. coli (19).
To assess whether the TraE protein can substitute for Topo III in vivo, a P1 transduction experiment was performed in which the DM750 cells (⌬topA) being transduced to topB (using a P1 lysate prepared from cells with a topB:kan r disruption) contained the traE expression vector pTraE (traE in vector pET-3c). As can be seen in Table I, cells harboring the topB expression plasmid pDE1 (topB in vector pET-3c (14)) and the traE expression plasmid pTraE could be transduced to topB: kan r . DM750 cells harboring a topA expression plasmid, pTI1 (topA in vector pET-3c (18)), or plasmids incapable of producing an active Topo III polypeptide (pDE2 contains a topB frameshift mutation (14) in vector pET-3c, and pF328 contains a topB mutation that changes the active-site tyrosine residue to phenylalanine in vector pET-3c (20, 26)) exhibited a 15-63-fold reduced transduction frequency compared with cells containing pTraE and a 44 -179-fold reduced transduction frequency compared with cells harboring pDE1. The observation that the transduction frequency of cells containing pF328 is 48 -136-fold lower than that of cells containing pTraE and pDE1 indicates that the presence of the topoisomerase activity of the TraE protein or Topo III is absolutely essential to observe transduction; however, Topo I activity cannot substitute for either enzyme. These results indicate that the TraE protein is a true functional homologue of Topo III. DISCUSSION The polypeptide encoded by the plasmid RP4 traE gene exhibits extensive protein sequence similarity to E. coli DNA topoisomerase III. The carboxyl-terminal amino acid residues of the TraE protein contain two potential zinc-finger motifs and more closely resemble E. coli DNA topoisomerase I. To ascertain whether the TraE protein was a topoisomerase, the traE gene was cloned into the bacteriophage T7 RNA polymerasebased transient expression vector. The polypeptide was induced, purified to apparent homogeneity, and shown to possess DNA topoisomerase activity in vitro.
The enzyme, unlike E. coli DNA topoisomerase I, exhibits biochemical properties virtually identical to those Topo III. The TraE protein is stimulated by both low concentrations of Mg 2ϩ and high temperature. In addition, TraE protein-catalyzed cleavage of a single-stranded substrate occurs at sites identical to those of Topo III. Perhaps the most distinguishing biochemical property of E. coli Topo III is the ability of the enzyme to catalyze the resolution of DNA replication intermediates in vitro. Topo III and DNA topoisomerase IV are unique among the bacterial topoisomerases in their ability to efficiently decatenate multiply interlinked plasmid DNA dimers and resolve plasmid replication intermediates (19, 29 -31). DNA gyrase is very inefficient at decatenating multiply interlinked plasmid DNA dimers (32), and Topo I has been shown to be incapable of resolving plasmid replication products in vitro (18). Since the TraE protein has features that are common to both Topo I and Topo III, it was of interest determine whether the enzyme was capable of resolving plasmid replication intermediates in vitro. Analysis of the TraE protein using an oriC-based in vitro replication system indicated that the protein could efficiently resolve DNA replication intermediates and that the specific activity of the protein was one-third that of Topo III.
The traE protein can also effectively substitute for Topo III 2 S. Malpure and R. J. DiGate, unpublished data.
FIG. 6. Analysis of products generated by Topo III and the TraE protein during oriC replication in vitro. oriC DNA replication reactions were performed as described previously (21). Reactions  harboring different plasmids Bacteriophage P1 transductions each were performed as described by Miller (42). The transduction frequencies of each plasmid-containing strain are expressed relative to the transduction frequency of the strain without the plasmid. The data represent the average ratio of three different experiments. The transduction frequency of cells containing the topB expression plasmid ranged from 1.8 ϫ 10 Ϫ7 to 16.1 ϫ 10 Ϫ7 . The total numbers of transductants in the three experiments were 1137 (for the strain containing pDE1) and 566 (for the strain containing pTraE). in vivo. The presence of a plasmid that carries the traE or topB gene in cells that contain a deletion of topA is sufficient to allow the disruption of the chromosomal copy of topB. This cannot be efficiently accomplished in the presence of plasmids that encode Topo I or an inactive Topo III polypeptide. Taken together, these data strongly suggest that the TraE protein is a true homologue of Topo III. The role of the TraE protein in plasmid RP4 DNA metabolism is unclear. The genes encoding the essential enzymes for plasmid replication have been mapped and do not include traE (33)(34)(35)(36). The gene is located in the primase gene cluster (5) and is embedded in a region that contains genes essential for plasmid transfer; however, traE has not been found to be essential for this process (36 -38). The finding that the TraE protein is a true homologue of E. coli Topo III may explain why no function has been ascribed to the traE gene. The majority of studies concerning both RP4 plasmid replication and conjugative transfer have used E. coli as the host organism. In addition, many organisms may possess Topo III-like activities. It is not surprising that the role of TraE remains undefined when the host(s) contains a homologous enzyme that possesses analogous properties; however, although the TraE protein can substitute for Topo III in vivo, it remains to be seen whether the converse is also true. To address this issue, our laboratory is in the process of constructing a RP4 plasmid in which only traE has been deleted. The replication and transfer properties of this plasmid will be examined in isogenic strains in which topB remains intact or has been disrupted.
In addition to the plasmid RP4 gene product, a search of the NCBI nonredundant data base using the BlastP program (39) revealed that there are 12 additional open reading frames that show extensive protein sequence similarity to topB (Table II). Interestingly, seven of the eight homologues that show the greatest protein sequence similarity to topB have been found on conjugative plasmids (perhaps eight of nine if one considers that the Streptococcus agalactiae homologue is a partial sequence; Table II) that have been isolated from both Gramnegative and Gram-positive bacteria.
One could envision that the plasmid-encoded topB homologue could be involved in two distinct processes. Many of the conjugative plasmids are large, low copy number genomes. To ensure that these large plasmids are faithfully segregated, these plasmids often encode their own partitioning systems that actively segregate the newly synthesized plasmid DNA molecules into the daughter cells. A component of such a system may be a potent decatenase. Presumably, this activity would serve as a "fail-safe" mechanism to ensure that the replicated chromosomes are always fully decatenated prior to partitioning of the newly synthesized plasmid molecules. In addition, in the case of broad host range, conjugative plasmids, the presence of a plasmid-encoded topoisomerase/decatenase would always guarantee the presence of a functional and compatible topoisomerase activity, regardless of the host organism. Alternatively, one could envision a role for these enzymes during conjugative transfer. The topoisomerase could serve as a swivel to facilitate the unwinding of the donor strand that is transferred to the recipient host. In either case, the presence of multiple examples of Topo III homologues contained on conjugative plasmids is compelling and suggests that the presence of topB-like genes may define a distinct subclass of these types of plasmids. The NCBI nonredundant data base was analyzed with the BlastP program (39) using the BLOSUM62 matrix. p values are given in scientific notation. In general, p values Ͻ 10 Ϫ6 (e Ϫ6 ) are considered significant (39). The gene symbol and references for each homologue are given. Protein sequence alignments of many these polypeptides have been published previously (40). hyp.prt., hypothetical protein.