INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Clinical Staphylococcus aureus strains often harbor multiple plasmids, ranging from small rolling-circle (RC) replicating plasmids that are cryptic or encode only a single resistance determinant to larger multiresistance and conjugative plasmids (12, 32, 36). Three groups of multiresistance plasmids have been recognized in staphylococci. Isolates from the 1960s and 1970s were commonly found to carry multiresistance plasmids conferring resistance to penicillin and heavy metals or other inorganic ions (48). Such -lactamase-heavy-metal resistance plasmids characteristically contain the -lactamase-encoding transposon Tn552 or a derivative and operons mediating resistance to arsenical, cadmium, and/or mercuric ions (36). Some -lactamase-heavy-metal resistance plasmids also contain Tn551, conferring resistance to macrolide-lincosamide-streptogramin type B antibiotics (33); an IS256-bounded composite aminoglycoside resistance transposon, Tn4001 (30); and/or a qacA or qacB antiseptic and disinfectant multidrug resistance determinant (30).

pSK41-like conjugative multiresistance plasmids were first detected in strains isolated in the mid 1970s. Such plasmids commonly contain a Tn4001 hybrid structure (6) and IS257-flanked cointegrated copies of small plasmids, such as the aminoglycoside resistance plasmid pUB110 (7). Other determinants encoded by cointegrated episomes carried by pSK41 family plasmids include smr, which confers multidrug resistance to antiseptics and disinfectants, and dfrA, which mediates trimethoprim resistance (4). Mupirocin resistance plasmids related to pSK41 have also been identified (31).

Clinical S. aureus strains isolated in Australia and the United Kingdom in the 1980s typically contained a multiresistance plasmid related to pSK1 (28, 30, 50, 55). pSK1 family plasmids commonly carry a qacA gene (40, 53), Tn4001 (39), and an IS257-containing structure termed Tn4003, which encodes trimethoprim resistance (11, 41); some pSK1 family plasmids also possess a Tn552-like -lactamase resistance transposon (17).

Although the replication of staphylococcal RC plasmids has been studied in considerable detail, comparatively little attention has been paid to the replication systems of larger plasmids from this genus, which are presumed to replicate via the theta mode (21, 49). To date, the replication initiation region of only a single staphylococcal multiresistance plasmid has been characterized, that of the -lactamase-heavy-metal resistance plasmid pSX267 from Staphylococcus xylosus (16, 19). To broaden our understanding of the replicative mechanisms of staphylococcal multiresistance plasmids and the evolutionary relationships among them, we have characterized the replication regions from the prototypes of the pSK1 and pSK41 plasmid families and a -lactamase-heavy-metal resistance plasmid, pI9789::Tn552.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Bacterial strains, plasmids, and culture conditions. The Escherichia coli host strain used was DH5 (F endA hsdR17 supE44 thi-1 recA1 gyrA96 relA1 80dlacZM15; Bethesda Research Laboratories), while the S. aureus host strains used were RN4220 (23) and the rifampin- and novobiocin-resistant strain SK982 (29). The S. aureus strain RN1965 (43) was used as a source of a reference set of plasmids for plasmid incompatibility tests (see below). pUC18 (58) and derivatives were used as cloning vectors. The general culture conditions were 37°C on Luria-Bertani (LB) agar or in LB medium (45). Where appropriate, antimicrobial agents were used at the following concentrations: ampicillin, 100 µg/ml; cadmium chloride, 50 µM; erythromycin, 20 µg/ml; gentamicin, 20 µg/ml; neomycin, 10 µg/ml; novobiocin, 2 µg/ml; rifampin, 20 µg/ml; tetracycline, 10 µg/ml; trimethoprim, 100 µg/ml.

DNA manipulations. Plasmid DNA was isolated from E. coli using a Quantum Prep plasmid miniprep kit (Bio-Rad), while DNA was isolated from S. aureus as described by Lyon et al. (28). Protoplast transformation of S. aureus was performed as described by Götz et al. (18). Electroporation of S. aureus with a Bio-Rad gene pulser was performed as described previously (25). DNA cloning and transformation of E. coli was performed using standard techniques (45). Restriction endonucleases and T4 DNA ligase (Promega) were used in accordance with the manufacturer's instructions. PCRs were carried out with Pfu DNA polymerase (Stratagene) according to the manufacturer's recommendations. Oligonucleotide primers for PCR and nucleotide sequencing were made with a Beckman Oligo 1000 synthesizer.

Nucleotide sequence determination and data analysis. Dideoxy sequencing (46) was performed with SequiTherm cycle-sequencing kits (Epicentre Technologies) or Sequenase version 2 sequencing kits (United States Biochemical) according to the manufacturers' recommendations. All restriction sites were crossed, and all sequence was determined on both DNA strands; derivatives of pBR322 (5) or pUC119 (56) containing pSK1 or pI9789::Tn552 restriction fragments were utilized as double-stranded sequencing substrates. Sequences were stored and assembled with the program SEQUENCHER (Gene Codes Corporation) and analyzed with the Genetics Computer Group (9) and PHYLIP (10) packages maintained by the Australian National Genomic Information Service, University of Sydney. Sequence similarities were assessed with pairwise alignments generated with the program GAP (9) using the Dayhoff 250 PAM matrix (8). Statistical significance (Z) was calculated using the formula Z = (a  m)/, where a is the alignment score, m is the mean of 100 alignment scores where one of the sequences has been randomly shuffled, and  is the standard deviation of m (37). Z scores greater than 3, 6, or 10 were taken to be indicative of "possible," "probable," or "highly probable" evolutionary relatedness, respectively (26). Phylogenetic trees were drawn by using the program TREEVIEW (34).

Assay of plasmid segregational stability. An overnight culture of the strain(s) to be assayed, grown in medium selective for the plasmid, was diluted in saline, and a viable count was performed using nonselective LB agar plates. Additionally, 100 µl of the 10² dilution was used to inoculate 10 ml of LB medium. After overnight growth without selection, this culture was diluted, counted, and subcultured as before; this process was repeated until approximately 100 generations in the nonselective medium were achieved. One hundred colonies from each day's viable-count plates were patched onto media with and without selection for the plasmid so that the proportion of the population retaining the resistance phenotype conferred by the plasmid could be quantitated. DNA was isolated from selected colonies to confirm the absence or presence of the relevant plasmid.

Plasmid incompatibility tests. S. aureus strain RN1965 (43) contains RC plasmids representing incompatibility groups (Inc) 3 to 5 and larger plasmids defining Inc1, Inc2, and Inc6. The six plasmids carried by this strain are pRN3025 (Inc1), pRN2003 (Inc2), pT127 (Inc3), pC221 (Inc4), pS177 (Inc5), and pK545 (Inc6), which confer resistance to erythromycin, cadmium, tetracycline, chloramphenicol, streptomycin, and neomycin, respectively. Plasmid incompatibility was determined by using mixed-culture transfer experiments (29) employing RN1965 as the donor and SK982 containing the plasmid under test as the recipient. Transferrants were selected on LB agar containing rifampin, novobiocin, and selection for the incoming plasmid only, and were subsequently patched onto media selective for both incoming and resident plasmids to determine coexistence and hence compatibility. This was confirmed by inoculating the cultures into 10 ml of LB medium selective for only the incoming plasmid, plating them with the same selection for single colonies, and patching them onto LB agar selective for both incoming and resident plasmids; maintenance of the resident plasmid by more than 90% of the cells carrying the incoming plasmid was taken to indicate compatibility. DNA isolations were subsequently performed on representative isolates to confirm the presence of both plasmids.

Nucleotide sequence accession numbers. The nucleotide sequences of the pSK1 and pI9789::Tn552 replication regions are available under the GenBank accession numbers AF203376 and AF203377, respectively.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Replication regions of the multiresistance plasmids pSK1 and pI9789::Tn552. The putative replication regions from the prototype of the pSK1 family of plasmids (12) and the -lactamase-heavy-metal resistance plasmid pI9789::Tn552 (42, 48) were identified based on similarity to the predicted rep region of the conjugative S. aureus multiresistance plasmid, pSK41 (4), and the demonstrated rep region of the S. xylosus -lactamase-heavy-metal resistance plasmid, pSX267, which is believed to replicate via the theta mode (16). The functionality of the pSK1 and pI9789::Tn552 rep regions was subsequently demonstrated in S. aureus (see below). The extent of nucleotide sequence similarity among the replication regions of pSK1, pI9789::Tn552, and pSK41, which represent the three groups of large staphylococcal multiresistance plasmids identified to date (12), is illustrated in the multiple sequence alignment shown in Fig. 1; due to the high degree of identity (89%) between the pI9789::Tn552 and pSX267 replication regions, for clarity, only the longer pI9789::Tn552 sequence is shown. The replication initiation genes of these staphylococcal plasmids encode products similar (17  Z  43) to those of plasmids from other gram-positive genera, including pLS32 from Bacillus natto (52), pLJ1 (51), pLH1 (54), and pSAK1 (GenBank entry Z50862) from Lactobacillus species, and the enterococcal plasmids pAD1 (57), pCF10 (44), and pPD1 (14) (Fig. 2A), a number of which have been proposed to utilize the theta mode of replication (see also reference 21). Phylogenetic analysis (Fig. 2B) revealed that the Rep proteins fall into generic clusters, providing no evidence of horizontal transmission of the rep genes among these genera.

View larger version (10988K): [in this window] [in a new window]   FIG. 1.   Nucleotide sequence alignment of the replication regions of pSK1, pI9789::Tn552, and pSK41. The pSK41 sequence shown corresponds to nt 12991 to 14216 of GenBank entry AF051917. Nucleotides present in two or more sequences at any one position are shaded, and insertions or deletions are denoted by dashes. The nucleotide sequences are numbered on the right. Translation start and stop codons of the indicated genes (outlined) are boxed. Directly repeated sequences are indicated by arrows, whereas inverted repeats are shown by half arrows.

View larger version (80K): [in this window] [in a new window]   FIG. 2.   (A) Multiple sequence alignment of the deduced pSK1 and pI9789::Tn552 rep products and related proteins. The amino acid sequences of the replication initiation proteins from the indicated plasmids were obtained from the following GenBank entries: pSX267, X92404; pSK41, AF051916; pCF10, L14285; pPD1, D78016; pAD1, L01794; pLH1, AJ222725; pLJ1, J04240; pSAK1, Z50862; and pLS32, D49467. Residue numbering is shown on the right. Amino acids common to five or more sequences at any position are shaded, globally conserved residues are indicated in the consensus (cons) line, and insertions or deletions are denoted by dashes. (B) Phylogenetic analysis of the Rep proteins shown in panel A. The unrooted tree was constructed by using the programs PROTDIST and NEIGHBOR (10) from the multiple alignment generated by PILEUP (9) shown in panel A. Branch lengths in arbitrary units are indicated. All nodes had bootstrap values of greater than 80%, and an equivalent tree topology was obtained with the program PROTPARS (10).

Functional analysis of the pSK1 replication region. To confirm the functionality of the pSK1 replication region identified above and to investigate the role, if any, of orf245 in the process, two minireplicons, pSK4829 and pSK4833, were constructed in E. coli using the plasmid pWE180. This plasmid is a derivative of pUC18 (59) containing an erythromycin resistance determinant, ermC. Although ermC confers resistance in both S. aureus and E. coli, the ColE1-derived replicon of pWE180 is functional only in the latter. The two pSK1 DNA segments amplified and separately cloned into appropriately restricted pWE180 corresponded to the rep gene and upstream intergenic region in the case of pSK4833 (nt 842 to 2287 [Fig. 1]) and a larger fragment also encompassing orf245 in pSK4829 (nt 3 to 2287 [Fig. 1]). pSK4829 and pSK4833 plasmid DNA isolated from E. coli was able to transform S. aureus RN4220 cells to erythromycin resistance via protoplast transformation. Plasmid isolation and agarose gel electrophoresis demonstrated the presence of pSK4829 and pSK4833 DNA in the RN4220 transformants, confirming that both of these plasmids contain a replication origin that functions in S. aureus. It should be noted that a recombinant plasmid equivalent to pSK4833, but containing the pI9789::Tn552 rep region rather than that from pSK1, was similarly shown to replicate in S. aureus RN4220.

View larger version (22K): [in this window] [in a new window]   FIG. 3.   Segregational stabilities of pSK1, pSK41, and derivatives in S. aureus strain RN4220. The values indicate the proportion of cells in a population, cultured in the absence of plasmid selection for the indicated number of generations, subsequently able to grow when plated on medium containing selection for the plasmid. Open and solid shapes distinguish data for each of two independent experiments. (A) Values for pSK1, pSK4829, and pSK4833, tested in S. aureus strain RN4220, are denoted by squares, diamonds and circles, respectively. The presence of pSK1 was indicated by growth in the presence of 100 µg of trimethoprim/ml, whereas retention of pSK4829 or pSK4833 was indicated by growth in the presence of 20 µg of erythromycin/ml. (B) Values for pSK41 and pSK5413, tested in S. aureus strains SK982 and RN4220, respectively, are denoted by squares and triangles, respectively. The presence of pSK41 was indicated by growth in the presence of 20 µg of gentamicin/ml, whereas retention of pSK5413 was indicated by growth in the presence of 10 µg of tetracycline/ml.

Functional analysis of the replication region from the conjugative multiresistance plasmid pSK41. The results discussed above implicating orf245 in the stable maintenance of pSK1 prompted us to investigate the replication system of pSK41, since this plasmid possesses a related rep gene but lacks an orf245 homolog. However, in an equivalent position and orientation, pSK41 possesses a gene, orf86, whose deduced product contains a putative helix-turn-helix DNA-binding domain (4), a feature shared by Orf245 and homologs (3, 47). We therefore intended to construct two pSK41 minireplicons, one containing pSK41 rep and the upstream intervening sequence and the other containing orf86 through rep, to determine if orf86 plays a role equivalent to that of orf245 on pSK1; however, despite numerous attempts we have not been able to obtain a recombinant plasmid containing orf86, either in the presence of rep or in isolation. pSK5413 was constructed by ligating an amplified fragment corresponding to nt 12795 to 14727 of pSK41 (GenBank entry AF051917), encoding the rep gene and the upstream intergenic region, into pUC18 and the subsequent insertion of the staphylococcal tetracycline resistance determinant, tetA(K) (20), as a marker. pSK41 was maintained in a population with almost absolute fidelity after 100 generations (Fig. 3B). After electroporation into S. aureus RN4220, segregational stability assays indicated that pSK5413 was maintained almost as efficiently as pSK41 (Fig. 3B), suggesting that orf86 does not contribute to the segregational stability of pSK41 and that the replication machinery of pSK41 requires no accessory gene equivalent to pSK1 orf245 for maximal efficiency.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

The nucleotide sequence identity evident among pSK1, pI9789::Tn552, pSX267, and pSK41 is consistent with incompatibility classifications determined here and previously (19, 27, 48). In particular, the sequence repeats present within the rep genes, which are likely to represent the replication origin of each plasmid, are largely conserved on the Inc1 plasmids, pSK1, pI9789::Tn552, and pSX267, whereas the repeats present within the rep gene of pSK41, which does not belong to Inc1, differ considerably (Fig. 1). As is the case for pSX267, database entries for three other staphylococcal plasmids contain incomplete sequences of probable orf245-like genes. It is therefore likely that the multiresistance plasmids, pSK156 (35), pIP630 (1), and pIP1156 (2), all belong to the same plasmid replication complex as pSK1 and pI9789::Tn552.

The staphylococcal multiresistance plasmids described here represent the third type of theta-mode replicon to possess an orf245-like gene, with pUCL22- and pUCL287-like plasmids representing the other two. Comparative sequence analyses reveal that genes encoding products sharing statistically significant amino acid sequence similarity to pSK1 Orf245 are represented on plasmids possessing yet more classes of replication initiation genes. Across the entire family of these proteins, sequence conservation is restricted to two linear segments, as shown in Fig. 4; viz., an N-terminal region confined to the first half of a predicted helix-turn-helix DNA-binding domain present in each protein and a segment located within the C-terminal halves of the proteins. Another theta-replicating plasmid, pIP404 from Clostridium perfringens (15), which encodes a replication initiation gene related to those of the broad-host-range plasmids pAM1 and pIP501, also possesses a gene encoding an Orf245 homolog (Z = 4.4). As on pUCL22 and pUCL287, the pIP404 orf245-like gene is encoded downstream of the rep gene, but rather than having an overlapping organization, in this case the two genes are separated by 757 nt of noncoding sequence (Fig. 4) which includes the probable origin of replication. Notably, our analysis has also identified orf245-like genes on members of two distinct classes of plasmids that replicate via an RC mechanism. The streptococcal plasmid pSSU1 and the Lactobacillus plasmid pLH2 (38) belong to the pE194/pMV158 family of RC plasmids. On pSSU1, the rep and orf245-like genes are separated by a small open reading frame that slightly overlaps them (Fig. 4), whereas on pLH2 these genes are separated by almost 2 kb. Finally, another streptococcal plasmid, pVA380-1, which belongs to the pC194/pUB110 RC plasmid family, contains an orf245 homolog immediately downstream of its replication initiation gene (Fig. 4); interestingly, this gene has been assigned a role in the mobilization of pVA380-1 (24).

View larger version (58K): [in this window] [in a new window]   FIG. 4.   (A) Multiple amino acid sequence alignment of segments conserved in Orf245 and related proteins. Residues present in at least 75% of the sequences at any one position are shaded, and amino acids present in at least 90% are indicated in a consensus line at the bottom. The dots denote sequences between the conserved regions that have been omitted for clarity. The amino acid sequences were obtained from the following GenBank entries: pUCL22, X60454; pJW563, X85168; pJW565, Y12736; pSRQ900, AF001314; pMD136, AF069302; pND324, U44843; pND861, AF034786; pIL7, Z25475; pIL105, AF116286; pIL2614, U90222; pSL2, X56550; pCIS3, AF153414; pNZ4000 (C1 to C4), AF036485; pUCL287, X75607; pSBO1, AB021464; pLrham (unnamed plasmid from Lactobacillus rhamnosus), AF037091; pIP404, M32882; pSSU1, AB019522; pLH2, X81981; and pVA380-1, L23803. For each segment, the amino acid number of the rightmost amino acid of a sequence is shown. Proteins from plasmids possessing related replication initiation genes are bracketed. The Orf245-like protein from pSL2, and two of the four homologs carried by pNZ4000, appear to be truncated and therefore do not possess the second conserved region. (B) For the top plasmid of each replicon class shown in panel A, the genetic organization of the gene encoding the Orf245-like protein, with respect to its associated rep gene, is represented diagrammatically. Distinct rep genes are represented by different patterned boxes, whereas orf245 homologs are denoted by solid boxes, and the genes' directions of transcription are indicated by arrowheads on the boxes; note that genes are not drawn to scale.

Despite the prevalence of rep gene-associated orf245 homologs, only the plasmid maintenance assays of the pSK1 minireplicons described here and the findings of Benachour et al. (3) concerning the stabilities and copy numbers of pUCL287 derivatives have implied a role for these genes in plasmid replication. The association of orf245 homologs with a range of seemingly distinct types of plasmid replication initiation genes is both intriguing and perplexing. Such a distribution implies a significant role for this gene, a notion supported experimentally for pUCL287 and now pSK1. However, for each of the six replicon types represented in Fig. 4, there are examples of naturally occurring plasmids which lack an orf245-like gene. Such a "patchy" distribution is consistent with plasmids variously acquiring and losing a gene of this type, possibly indicating that orf245-like genes play a functionally and evolutionarily significant role only under certain conditions. For example, it is conceivable that such a gene is required for plasmid survival in some hosts and/or environments but not in others. The precise role of orf245 and related genes has yet to be elucidated, although the effect on plasmid copy number observed when repB287 was removed from pUCL287 (3) suggests that at least in this case the gene influences replication directly rather than by enhancing plasmid maintenance as a stand-alone segregation mechanism. We are pursuing the working hypothesis that orf245 exerts a regulatory effect on rep transcription. It is hoped that such studies will ultimately provide insights into the interaction among plasmids, bacterial hosts, and their environment.

The sequence similarity detected between the pSK1 replication region and those of the -lactamase-heavy-metal resistance plasmids and pSK41-like conjugative plasmids establishes that all three recognized groups of large staphylococcal multiresistance plasmids utilize evolutionarily related theta-mode replication systems. It is sobering to consider the impact of this single replicon type on the worldwide development of antimicrobial-resistant staphylococcal strains.