Specificity of the interactions between the Rep proteins and the origins of replication of Staphylococcus aureus plasmids pT181 and pC221

Summary pT181 and pC221 are closely relatedStaphylococcus aureus plasmids with the same genome organization, which is characterized by the overlapping of the origin of replication with the sequence encoding a protein, Rep, essential for plasmid replication. Former results have shown the lack of in vivo cross-complementation between these two plasmids, while in vitro studies have revealed the ability of both Rep proteins to act on either origin. One possible explanation for this difference was based on a previous analysis of the incompatibility expressed by the origin of replication of these plasmids, showing that the origin embedded in therep gene competes for Rep utilization with the origin of a test plasmid and that changes in the sequence of the origin reduce its ability to compete. To avoid this problem, in the present work special hybrids were constructed in which the origin of replication overlapping therep gene was mutationally inactivated, without changing the amino acid sequence of the encoded protein. The level of Rep expression by these hybrids could be varied by taking advantage of what is presently known about the control of Rep synthesis in plasmid pT181. The results of complenentation studies conducted using these hybrids have shown that: (i) at the usual level of expression for a wild-type plasmid each Rep protein can initiate replication strictly from its corresponding origin; (ii) when overproduced, the pT181 RepC protein could also act efficiently on the pC221 origin; a functional pT181 origin present in the same host completely prevented this complementation; (iii) in excess, the RepD protein encoded by pC221 could replicate a plasmid carrying the pT181 origin but could not ensure the hereditary stability of such a plasmid in the absence of another active replication system; (iv) when overproduced both RepC and RepD could act on the origin of replication of three other related plasmids pS194, pC223 and pUB112.     

Replication termination for staphylococcal plasmids: plasmids pT181 and pC221 cross-react in the termination process.

We present data which indicate that (i) the origin of replication of plasmids pT181 and pC221 can also function as termination signals; (ii) termination of replication occurs when a round of replication initiated either by RepC at the pT181 origin or by RepD at the pC221 origin reaches either of these origins, proving that the two plasmids cross-react for termination of replication; and (iii) the replication initiated at the origin of another staphylococcal plasmid, pE194, does not terminate at the origin of pT181 or pC221, indicating the existence of a specific relationship between the initiation and termination of a replication event.     

Specificity of RepC protein in plasmid pT181 DNA replication

The plasmid pT181 of Staphylococcus aureus consists of 4437 base pairs and encodes resistance to tetracycline. Initiation of pT181 DNA replication specifically requires the plasmid-encoded initiator protein, RepC. The initiator protein binds specifically to a 32-base pair sequence within the pT181 origin of replication. RepC protein also has a nicking-closing activity that is specific for the pT181 origin. Replication of pT181 initiates by covalent extension of the nick and proceeds by a rolling circle mechanism. Two other small, multicopy plasmids pC221 and pS194 belong to the pT181 family and have common structural organization and replication properties. The replication proteins and replication origins of these plasmids have extensive sequence homologies, although they belong to different incompatibility groups. In spite of this homology, the replication proteins and replication origins of these three plasmids do not show any cross-reactivity in vivo. We have carried out a series of in vitro experiments to determine the specificity of pT181-encoded initiator protein, RepC. DNA binding experiments showed that although the binding of RepC to the pT181 origin was very efficient, little or no binding was seen with pC221 and pS194 origins. The nicking-closing activity of RepC was found to be equally efficient with the pC221 and pS194 plasmids. The plasmids pC221 and pS194 replicated efficiently in a RepC-dependent in vitro system. However, replication of these plasmids was greatly reduced in the presence of a competing pT181 origin. The results presented here suggest that nicking-closing by RepC at the origin is not sufficient for maximal replication and that tight binding of RepC to the origin plays an important role in the initiation of DNA replication.     

Plasmid replication initiator protein RepD increases the processivity of PcrA DNA helicase

The replication initiator protein RepD encoded by the Staphylococcus chloramphenicol resistance plasmid pC221 stimulates the helicase activity of the Bacillus stearothermophilus PcrA DNA helicase in vitro. This stimulatory effect seems to be specific for PcrA and differs from the stimulatory effect of the Escherichia coli ribosomal protein L3. Whereas L3 stimulates the PcrA helicase activity by promoting co-operative PcrA binding onto its DNA substrate, RepD stimulates the PcrA helicase activity by increasing the processivity of the enzyme and enables PcrA to displace DNA from a nicked substrate. The implication of these results is that PcrA is the helicase recruited into the replisome by RepD during rolling circle replication of plasmids of the pT181 family.     

Plasmid pT181 replication is decreased at high levels of RepC per plasmid copy

The replication of staphylococcal plasmid pT181 is indirectly controlled at the level of the synthesis of its replication initiator, RepC. As a result, high levels of RepC synthesis per plasmid copy were expected to lead to autocatalytic plasmid replication, which secondarily would affect host physiology. Surprisingly, RepC overexpression was found to lead to a rapid decrease in pT181 copy number and replication rate. These effects depended on the ratio of RepC lo the PT181 replication origin rather than on the absolute amount of RepC in the cell. In a wild-type host, the increase in RepC/plasmid copy also inhibited chromosome replication and cell division. The changes in host physiology did not play any role in the decrease in pT181 replication caused by RepC overexpression since pT181 replication responded in the same way in a host mutant insensitive to the effects of RepC induction. These results suggest that pT181, the prototype of an entire class of plasmids from Gram-positive bacteria, responds to overexpression of its replication initiator by a decrease in plasmid replication.     

Plasmids of the pT181 family show replication-specific initiator protein modification.

The rolling circle plasmids of Staphylococcus aureus regulate their replication by controlling initiator (Rep) protein synthesis. It was demonstrated recently that the pT181 initiator protein RepC is inactivated during pT181 replication by the addition of an oligodeoxynucleotide, giving rise to a new form, RepC* (A. Rasooly and R. P. Novick, Science, 262:1048-1050). We establish here that this initiator modification occurs with four other members of the pT181 family and that it occurs in Bacillus subtilis as well as S. aureus. These results suggest that Rep conversion to Rep* is probably universal among plasmids of the pT181 family and is not host dependent.     

Comparative analysis of five related staphylococcal plasmids

The genomic organization of five small multicopy staphylococcal plasmids comprising the pT181 family has been analyzed. In addition to pT181, the family presently includes the streptomycin resistance plasmid pS194 and the chloramphenicol resistance plasmids pC221, pC223, and pUB112. Although they belong to five different incompatibility groups, the five plasmids have similar basic replicons, use the same basic copy control mechanism, and have a common structural organization. It has been demonstrated previously that pT181 and pC221 encode trans-active replication proteins (RepC and RepD, respectively) which specifically recognize the respective plasmid's origin of replication in both cases is initiated by site-specific nicking and 3' extension. The other three plasmids in this family encode similar replication proteins; 63% of the predicted amino acid residues are identical for all five and the least similar pair shows 75% identity at the amino acid level. However, despite this homology, the replication proteins and origins of replication of different members in this family did not show cross complementation in vivo. Outside of the basic replicon, which comprises about one-third of each plasmid's genome, functional organization is also conserved. The resistance determinants are all located in the same position, immediately downstream of the replication protein coding sequence, and all are transcribed in the same direction. The three chloramphenicol resistance determinants encode highly homologous chloramphenicol transacetylases which are unrelated to the tet and str gene products. Three of the five plasmids form relaxation complexes and the involved genome segments are closely related. The other two are not homologous to these three in the corresponding region, but are homologous to each other and encode a site-specific recombinase, Pre. It is suggested that the replication, resistance, and relaxation complex regions of these plasmids can be regarded as conserved segments ("cassettes") assembled in various combinations, but always with the same spatial arrangement.     

An 18-base-pair sequence is sufficient for termination of rolling-circle replication of plasmid pT181.

pT181 and related plasmids of gram-positive bacteria replicate by a rolling-circle mechanism. The replication initiator protein of pT181, RepC, has origin-specific nicking-closing activities. Replication of the plasmid pT181 leading strand initiates by covalent extension of the RepC-generated nick, and the origin of replication contains signals for both initiation and termination of DNA replication. We have investigated the sequence requirements for the initiation and termination steps by using plasmids containing two pT181 origins. In vitro replication experiments showed that 18- and 24-bp synthetic oligonucleotides containing the RepC nick site were active in the termination of replication. However, initiation of replication required a larger region which also includes the RepC binding site. Plasmids containing the 18- and 24-bp region were also found to be nicked by the RepC protein. Our results demonstrate that sequence requirements for initiation and termination of pT181 replication overlap, but while the RepC binding site is required for initiation, it is dispensable for termination.     

Biochemical characterization of the Staphylococcus aureus PcrA helicase and its role in plasmid rolling circle replication

Previous genetic studies have suggested that a putative chromosome-encoded helicase, PcrA, is required for the rolling circle replication of plasmid pT181 in Staphylococcus aureus. We have overexpressed and purified the staphylococcal PcrA protein and studied its biochemical properties in vitro. Purified PcrA helicase supported the in vitro replication of plasmid pT181. It had ATPase activity that was stimulated in the presence of single-stranded DNA. Unlike many replicative helicases, PcrA was highly active as a 5' --> 3' helicase and had a weaker 3' --> 5' helicase activity. The RepC initiator protein encoded by pT181 nicks at the origin of replication and becomes covalently attached to the 5' end of the DNA. The 3' OH end at the nick then serves as a primer for displacement synthesis. PcrA helicase showed an origin-specific unwinding activity with supercoiled plasmid pT181 DNA that had been nicked at the origin by RepC. We also provide direct evidence for a protein-protein interaction between PcrA and RepC proteins. Our results are consistent with a model in which the PcrA helicase is targeted to the pT181 origin through a protein-protein interaction with RepC and facilitates the movement of the replisome by initiating unwinding from the RepC-generated nick.     

Bacillus anthracis and Bacillus cereus PcrA helicases can support DNA unwinding and in vitro rolling-circle replication of plasmid pT181 of Staphylococcus aureus

Replication of rolling-circle replicating (RCR) plasmids in gram-positive bacteria requires the unwinding of initiator protein-nicked plasmid DNA by the PcrA helicase. In this report, we demonstrate that heterologous PcrA helicases from Bacillus anthracis and Bacillus cereus are capable of unwinding Staphylococcus aureus plasmid pT181 from the initiator-generated nick and promoting in vitro replication of the plasmid. These helicases also physically interact with the RepC initiator protein of pT181. The ability of PcrA helicases to unwind noncognate RCR plasmids may contribute to the broad-host-range replication and dissemination of RCR plasmids in gram-positive bacteria.     

Role of individual monomers of a dimeric initiator protein in the initiation and termination of plasmid rolling circle replication

Plasmids of the pT181 family encode initiator proteins that act as dimers during plasmid rolling circle (RC) replication. These initiator proteins bind to the origin of replication through a sequence-specific interaction and generate a nick at the origin that acts as the primer for RC replication. Previous studies have demonstrated that the initiator proteins contain separate DNA binding and nicking-closing domains, both of which are required for plasmid replication. The tyrosine residue at position 191 of the initiator RepC protein of pT181 is known to be involved in nicking at the origin. We have generated heterodimers of RepC that consist of different combinations of wild type, DNA binding, and nicking mutant monomers to identify the role of each of the two monomers in RC replication. One monomer with DNA binding activity was sufficient for the targeting of the initiator to the origin, and the presence of Tyr-191 in one monomer was sufficient for the initiation of replication. On the other hand, a dimer consisting of one monomer defective in DNA binding and the other defective in origin nicking failed to initiate replication. Our results demonstrate that the monomer that promotes sequence-specific binding to the origin must also nick the DNA to initiate replication. Interestingly, whereas Tyr-191 of the initiator was required for nicking at the origin to initiate replication, it was dispensable for termination, suggesting that alternate amino acids in the initiator may promote termination but not initiation.     

The inactivated plasmid inititator protein RepC/RepC* may have a regulatory role.

During replication of the plasmid pT181, the initiator protein RepC is modified by the addition of an oligodeoxynucleotide, giving rise to a new form, RepC*. Here we show that during in vitro replication, RepC* is radioactively labeled, suggesting that the source of the RepC* oligodeoxynucleotide is the newly synthesized pT181 DNA. The RepC/RepC* heterodimer retains its ability to bind the pT181 double-strand origin and, therefore, it may act as a competitive inhibitor of the RepC homodimer during replication.     

Functional organization of the plasmid pT181 replication origin

Replication of the staphylococcal plasmid pT181 is initiated at the origin (ori) with the introduction of a site-specific nick by the plasmid-encoded initiator protein RepC. Deletion analysis showed that a sequence of about 70 base-pairs is required for full ori function, including the ability to compete with a co-resident wild-type origin for the trans-acting RepC protein. A shorter sequence of 43 base-pairs is sufficient for origin function in the absence of competition. Single and double point mutations within these 43 base-pairs were used to determine the sequence requirement for replication within the minimal origin. Deletion mutants and point mutants were tested in replication and competition assays in vivo and in vitro, and in a RepC-mediated nicking assay.     

Kinetic mechanism of initiation by RepD as a part of asymmetric, rolling circle plasmid unwinding

Some bacterial plasmids carry antibiotic resistance genes and replicate by an asymmetric, rolling circle mechanism, in which replication of the two strands is not concurrent. Initiation of this replication occurs via an initiator protein that nicks one DNA strand at the double-stranded origin of replication. In this work, RepD protein from the staphylococcal plasmid pC221 carries this function and allows PcrA helicase to bind and begin unwinding the plasmid DNA. This work uses whole plasmid constructs as well as oligonucleotide-based mimics of parts of the origin to examine the initiation reaction. It investigates the phenomenon that nicking, although required to open a single-stranded region at the origin and so allow PcrA to bind, is not required for another function of RepD, namely to increase the processivity of PcrA, allowing it to unwind plasmid lengths of DNA. A kinetic mechanism of RepD initiation is presented, showing rapid binding of the origin DNA. The rate of nicking varies with the structure of the DNA but can occur with a rate constant of >25 s(-1) at 30 °C. The equilibrium constant of the nicking reaction, which involves a transesterification to form a phosphotyrosine bond within the RepD active site, is close to unity.     

Molecular structure of the replication origin of a Bacillus subtilis (natto) plasmid, pUH1.

The structure of a 2.0-kb BstEII DNA sequence necessary and sufficient for the replication of a 5.7-kb Natto plasmid, pUH1, which is responsible for gamma-polyglutamate production by Bacillus subtilis (natto), has been characterized by using a trimethoprim resistance gene derived from B. subtilis chromosomal DNA as a selective marker. The 2.0-kb DNA sequence contains an open reading frame, rep, stretching for 999 bp; a promoter region for rep expression; and a possible replication origin for the plasmid upstream of the promotor. The predicted Rep protein has highly homologous amino acid sequences with rep14 of pFTB14 in B. amyloliquefaciens, RepB of pUB110, and protein A, which is necessary for pC194 replication in staphylococci throughout the protein molecule, but is not homologous with RepC of staphylococcal plasmid pT181.     

Molecular structure of the replication origin of a Bacillus subtilis (natto) plasmid, pUH1.

The structure of a 2.0-kb BstEII DNA sequence necessary and sufficient for the replication of a 5.7-kb Natto plasmid, pUH1, which is responsible for gamma-polyglutamate production by Bacillus subtilis (natto), has been characterized by using a trimethoprim resistance gene derived from B. subtilis chromosomal DNA as a selective marker. The 2.0-kb DNA sequence contains an open reading frame, rep, stretching for 999 bp; a promoter region for rep expression; and a possible replication origin for the plasmid upstream of the promotor. The predicted Rep protein has highly homologous amino acid sequences with rep14 of pFTB14 in B. amyloliquefaciens, RepB of pUB110, and protein A, which is necessary for pC194 replication in staphylococci throughout the protein molecule, but is not homologous with RepC of staphylococcal plasmid pT181.     

RepC is rate limiting for pT181 plasmid replication

The effect on pT181 plasmid replication of the concentration of the plasmid-coded initiator protein, RepC, has been analyzed. In one type of experiment, plasmid replication was found to stop immediately after the addition of an inhibitory concentration of chloramphenicol (Cm) to growing cultures. Chromosomal replication showed the slow turnoff that is usual for Cm inhibition. Because plasmid replication rate is determined autogenously, no host factor can be rate limiting, suggesting that the specific factor affected is Rep C. In another type of experiment, we constructed a translational fusion between the repC coding sequence and a translationally inducible Cm-acetylase gene, cat-86, using pUB110 as the carrier replicon. The fusion plasmid showed an eightfold amplification of its own copy number and a similar amplification of a co-resident pT181 plasmid upon Cm induction. The amplified plasmids did not show autocatalytic runaway replication but rather established stable elevated copy numbers, indicating the existence of a secondary level of regulation. These results suggest that RepC is rate limiting for pT181 replication and support the hypothesis that pT181 replication is regulated at the level of RepC synthesis. The nature of the secondary regulation is unknown.     

The PcrA3 mutant binds DNA and interacts with the RepC initiator protein of plasmid pT181 but is defective in its DNA helicase and unwinding activities

Plasmid rolling-circle replication initiates by covalent extension of a nick generated at the plasmid double-strand origin (dso) by the initiator protein. The RepC initiator protein binds to the plasmid pT181 dso in a sequence-specific manner and recruits the PcrA helicase through a protein-protein interaction. Subsequently, PcrA unwinds DNA at the nick site followed by replication by DNA polymerase III. The pcrA3 mutant of Staphylococcus aureus has previously been shown to be defective in plasmid pT181 replication. Suppressor mutations in the repC initiator gene have been isolated that allow pT181 replication in the pcrA3 mutant. One such suppressor mutant contains a D57Y change in the RepC protein. To identify the nature of the defect in PcrA3, we have purified this mutant protein and studied its biochemical activities. Our results show that while PcrA3 retains its DNA binding activity, it is defective in its helicase and RepC-dependent pT181 DNA unwinding activities. We have also purified the RepC D57Y mutant and shown that it is similar in its biochemical activities to wild-type RepC. RepC D57Y supported plasmid pT181 replication in cell-free extracts made from wild-type S. aureus but not from the pcrA3 mutant. We also demonstrate that both wild-type RepC and its D57Y mutant are capable of a direct physical interaction with both wild-type PcrA and the PcrA3 mutant. Our results suggest that the inability of PcrA3 to support pT181 replication is unlikely to be due to its inability to interact with RepC. Rather, it is likely that a defect in its helicase activity is responsible for its inability to replicate the pT181 plasmid.