Replication Control and Other Stable Maintenance Mechanisms of Plasmids
Chapter
122
DONALD R. HELINSKI, ARESA E. TOUKDARIAN, andRICHARD P. NOVICK
The majority of gram-negative and gram-positive bacteria contain plasmids which are distinguished by their physical separation from the chromosome of the host cell as well as by their stable maintenance in this extrachromosomal state. As the result of a wide range of studies of plasmids in diverse bacteria, it is clear that these extrachromosomal elements determine a remarkable array of phenotypic traits in bacteria of medical, agricultural, environmental, and commercial importance. Thus, to fully understand the genetic and phenotypic properties of a bacterium, one must include an understanding of the one or more plasmid elements stably maintained in that organism. Because of both the similarities and differences in the mechanisms of initiation and regulation of replication of plasmids of gram-negative and gram-positive bacteria, a true understanding of the basic biology of plasmids is greatly enhanced by an examination of properties of plasmids from both major classes of bacteria. To keep within the space confines of this chapter, we have restricted ourselves to the initiation and regulation of plasmid replication and other mechanisms responsible for the stable maintenance of plasmids.
The early studies on plasmids, for the most part carried out in gram-negative bacteria, established two types of plasmids on the basis of overall genetic content. One type, designated nonconjugative or nontransmissible, has genes that are required for the initiation and regulation of autonomous replication but do not possess a functional set of genes that are required for conjugal transfer. If additional genes that provide a distinctive phenotype for the host are present, the plasmid is often given a specific designation. For example, a plasmid of Escherichia coli that carries a set of genes involved in colicin production is designated a Col plasmid. If the additional genes determine resistance to one or more antibiotics, then the element is usually designated an R plasmid. Col plasmids and R plasmids are broadly distributed, both geographically and with respect to their presence in a wide variety of nonpathogenic and pathogenic bacteria. The second type of plasmids is the conjugative or self-transmissible elements that are also designated sex factors. Plasmids of this class possess a functional set of genes that are involved in conjugal mating and DNA transfer as well as those required for autonomous replication. These two categories of plasmids are found in both gram-negative and gram-positive bacteria.
The classic early studies with the F sex factor of E. coli, in addition to establishing sexuality in bacteria, provided the basis for the development of fundamental concepts of molecular biology, including the role of repressors in the control of gene expression and the concept of an operon. Early studies with R plasmids underscored the importance of these extrachromosomal elements in the worldwide dissemination of antibiotic resistance in a wide range of bacteria, as well as the increased incidence of plasmid-specified multiple antibiotic resistance in response to increased use of antibiotics in human and veterinary medicine and in agriculture. Later work on R plasmids was key to the development of the concept of a transposable element and a mobile insertion sequence. There is also little question that R plasmids, along with several other plasmid elements including Col plasmids, provided one of the key factors in the development of gene cloning techniques.
Plasmid elements generally are in the form of double-stranded circular DNA molecules that are covalently intact and are isolated in the form of supercoiled DNA. As more attention has been given in the past few years to bacteria other than E. coli, linear double-stranded DNA plasmids also have been discovered and characterized. Most plasmid elements of gram-negative bacteria are of the narrow host range type in that they are stably maintained in their natural host or closely related bacteria but fail to replicate or regulate their copy number in distantly related bacteria. Several well-characterized plasmids in gram-negative bacteria, however, have been found to have an extended host range, e.g., plasmid RK2 and other members of the IncP1 family and plasmid RSF1010. While these plasmid elements are capable of replicating in most gram-negative bacteria, their host range for controlled replication does not extend to the gram-positive bacteria. Many naturally occurring plasmids in gram-positive bacteria exhibit an extended host range within the gram-positive bacteria. With rare exceptions, a naturally occurring plasmid is incapable of controlled replication in both gram-negative and gram-positive bacteria.
Originally, plasmids were classified based on their incompatibility group (Inc), which is a property of the mechanisms by which they control replication initiation and stable inheritance. Plasmids of the same incompatibility group share these functions and as a consequence they cannot be stably coinherited. This is caused by two features of bacterial plasmid maintenance: random selection for replication and random assortment for partitioning. Classification of plasmids based on incompatibility is useful for practical purposes; e.g., in strain construction only compatible replicons can be stably introduced into the host. This chapter, however, will consider plasmids according to their structural features as well as their modes of replication. It will also include, where relevant, the plasmid form of the E. coli phages P1 and λ (λdv).
While plasmids vary considerably in size and copy number, in a particular bacterial host growing under specific conditions a given plasmid is maintained at a defined copy number. The key factor in the maintenance of a plasmid in the extrachromosomal state is the ability of the plasmid to control the frequency of replication initiation such that there is on the average one replication event per plasmid copy per generation. Studies of the regulation of plasmid replication involving a number of different replicons clearly show that plasmids employ different strategies to assure their maintenance in a host cell at a characteristic copy number, with similar mechanisms being used by both gram-negative and gram-positive bacteria. These different regulatory mechanisms have in common the ability to correct the stochastic fluctuations in plasmid copy number that occur during cell growth and division.
The responsiveness of a particular regulatory mechanism may on occasion fail in a particular cell to correct for a lower than normal copy number, with the net result that plasmidless daughter cells are produced through random segregation of the plasmid copies during cell division. To correct for this inefficiency in the regulation of plasmid copy number and for the relatively high probability of generating plasmidless cells by random segregation, particularly in the case of a low-copy-number plasmid, many plasmids encode genetic systems to assure stable retention in a growing bacterial population. For some plasmids a cis-acting partition mechanism ensures that each daughter cell receives at least one copy of a plasmid molecule after cell division. Advances in our understanding of the genetic elements that code for a plasmid partition mechanism have come largely from the analyses of plasmids of gram-negative bacteria; however, it is likely that similar mechanisms are specified by plasmids of gram-positive bacteria. Additionally, several plasmid-based toxin/antitoxin systems that cause the inhibition or death of plasmid-free segregants have been well characterized. These killing systems have been shown to involve the production of a protein toxin that acts intracellularly to cause the death of a plasmidless daughter cell. The mechanisms of plasmid-specified copy-number control, partitioning, and postsegregational cell killing that contribute to stable maintenance of plasmid elements in gram-negative and gram-positive bacteria are the major themes of this chapter.
Since it is not possible to adequately treat all aspects of plasmid maintenance in bacteria, we refer the reader to several recent reviews on various aspects of plasmid biology that will provide additional details and in some cases different perspectives on this subject (31, 49, 81, 86, 99, 114, 125, 155, 166, , 187, 195, 224, 228, 229, , 268, 296, 320, 338; G. del Solar, M. J. Ruiz-Echevarria, M. Espinosa, and R. Diaz, in press).
Regardless of plasmid size, the basic replicon of a plasmid generally consists of a contiguous set of information that includes a definable origin where DNA replication initiates and one or more adjoining controlling elements. This information is often contained within a segment that is 3 kb or less in size. The compactness of most plasmid replicons has certainly facilitated their isolation and characterization. The origin of replication (ori) has been delineated for most plasmids to within a several-hundred-base-pair segment. It contains recognition sites for those plasmid and host proteins involved in replication initiation. In addition, most replicons encode a structural gene, often designated rep, for a plasmid-specific protein that, in those cases studied, is required for the initiation of replication. Generally the Rep protein is capable of acting in trans; however, there are well-studied systems (e.g., plasmid R1) where this protein preferentially acts at an origin in cis to the rep gene. For most plasmids of gram-negative bacteria that have been characterized, replication proceeds from the origin in a unidirectional or bidirectional mode via a theta replication mechanism. However, a few plasmids of gram-negative bacteria have been shown to replicate via a rolling-circle (RC) mechanism (e.g., references 156 and 337). This is unlike the situation with gram-positive bacteria, where the plasmids characterized to date predominately replicate via an RC mechanism that involves the formation of a single-stranded circular DNA intermediate.
Most plasmids have a single origin of replication. However, in the well-studied case of the theta replicating plasmid R6K, three origins of replication have been defined under the control of a single replication initiation protein (54, 131). All three origins and the initiation protein gene are contained within a 4-kb segment within this 38-kb plasmid. In addition, as a consequence of more detailed studies of relatively large plasmids, an increasing number of plasmids of gram-negative bacteria have been shown to contain more than one replicon (53). For example, the various members of the IncFI family, including the classic F plasmid, contain either fully functional systems or remnants of three independent replication systems (22). The best-studied replicon in the F plasmid, designated FIA, has two origins of replication, with both origins controlled by the RepE initiation protein. An FIA-related replicon is also found on some IncH1 plasmids, while the ori-2 origin of the FIA replicon is found on a naturally occurring plasmid of Klebsiella pneumoniae (58). This Klebsiella plasmid, pGSH500, also contains a pCU1-related replicon (58). A further demonstration of the variety and plasticity of plasmids is seen in the IncN plasmid, pCU1, which has a broad host range and contains three origins of replication clustered within a 1.2-kb region (15, 152).
A variety of plasmid replicons isolated from gram-negative and gram-positive bacteria have been characterized with respect to the key elements controlling the initiation of replication. In each case, a negative feedback control system has been identified which is responsible for maintenance of the plasmid at a steady-state copy number that is characteristic of the particular plasmid in a given host bacterium. In the case of broad-host-range plasmids this copy number may vary depending on the bacterial species. Any plasmid copy-control system must respond to upward or downward fluctuations in copy number during exponential growth of the bacterium and also allow for a relatively rapid rate of initiation of replication of a particular plasmid element after its entry into a cell via transformation, transduction, or conjugation. Despite the variety of replicons, two basic schemes for the regulation of replication of plasmid copy number have been elucidated. In one scheme the plasmid replicon encodes a small, diffusible RNA molecule that acts as an antisense transcript that negatively regulates a key step in the replication of the plasmid. The second basic mechanism is found in plasmids, for the most part in gram-negative bacteria, that contain a series of direct repeats, designated iterons, that are located within the functional sequence of nucleotides that make up the origin of replication and are essential for both the initiation of replication and its control. These direct repeats and the plasmid-specified replication initiation protein which binds to these iterons make up the essential elements for replication control of iteron-containing plasmids.
The ColE1-type and the IncFII-type plasmids of E. coli are the best understood replicons with respect to the regulation of initiation of replication (47, 74, 155, 166, 210, 228, 245, 329; del Solar et al., in press). In both cases the regulatory element is a small RNA molecule; in the ColE1-type plasmids this molecule regulates the functionality of an RNA primer that is necessary for replication initiation, while for the IncFII-type plasmids the regulatory RNA molecule controls the synthesis of the essential cis-acting replication initiation protein. In addition to the extensively studied plasmid ColE1, the ColE1-type family includes the naturally occurring plasmids p15A, RSF1030, pMB1, and CloDF13 as well as the cloning vectors pACYC184, pBR322, and related plasmids (166, 229). Unlike most plasmids in gram-negative bacteria, the ColE1 family of plasmids do not encode a replication initiation protein, and they require the host DNA polymerase I (Pol I) for early synthesis of the leading strand. Replication initiation involves the synthesis of a 700-base preprimer RNA, designated RNA II, whose transcription begins 555 bases upstream of the origin (defined as the transition point between primer RNA and DNA) (Fig. 1) (304, 305). The 3' end of the RNA II molecule "couples" with the DNA molecule at the origin of replication by hybrid formation. The formation of a functional coupled complex is dependent on the generation of a specific secondary structure at the 5' end of the RNA II molecule that in turn allows the formation of a stem-loop at the 3' end of the RNA II-coupled complex that is recognized by RNase H (304, 305, 306). Cleavage of the RNA-DNA hybrid then occurs at the replication origin, releasing a 3'-hydroxyl group that serves as the primer for DNA synthesis catalyzed by Pol I. Replication then proceeds unidirectionally with the initiation of lagging-strand synthesis at specific ColE1 sequences. It has been further shown for ColE1 that there are two alternative modes of replication that allow for replication in the absence of either RNase H or both RNase H and Pol I (56). However, regardless of the replication mode, RNA II must be produced in the correct structural configuration to act as the primer for DNA replication.
Copy-number control of ColE1 is mainly dependent on the 108-bp RNA I transcript. RNA I binds to nascent RNA II by complementary base-pairing and thereby prevents the formation of the hybrid between RNA II and the DNA template that is recognized by RNase H. While the ColE1 plasmid does not encode a replication initiation protein, it does specify a protein (Rop) that lowers the copy number of the plasmid by stabilizing the RNA I-RNA II complex by binding to both RNA molecules (42, 121, 303, 306, 313). In addition to the role of the Rop protein in the regulation of ColE1 copy number, the rate of decay of RNA I clearly is a major factor in copy-number determination. It has been shown recently that the decay rate of RNA I is determined to a major extent by polyadenylation of the transcript (333) and by RNase E cleavage (179).
The requirement of Pol I for DNA replication, while infrequently found among plasmids, is not restricted to the ColE1 family. Plasmid pCU1 exhibits a requirement for Pol I for one (oriV) of the three functional origins of this plasmid; replication originating at the other two origins apparently is Pol I independent (152). Colicinogenic plasmids ColE2 and ColE3 also require this polymerase for replication (286). Interestingly, these relatively small plasmids, with the smallest defined origin sequence discovered to date (less than 50 bp), also specify a replication protein that is essential for replication (126, 223, 336). Finally, the naturally occurring gram-positive bacterial plasmid pAMβ1 also shows a requirement for Pol I (see below).
IncFII plasmids, including R1, NR1, and R6-5, are another well-studied group in which the key element in the regulatory apparatus for the initiation of replication is a small antisense RNA molecule (28, 227, 228, 261, 329). Unlike the ColE1 family, however, replication is controlled by the rate of synthesis of a plasmid-encoded replication protein (RepA). Plasmid R1 is maintained at a copy number of 1 to 2 per chromosome. The low copy number of this family of plasmids makes it especially important to have a regulatory control circuit that rapidly responds to changing copy number. As shown in Fig. 1, the minimal replicon (approximately 2.5 kb within the 100-kb plasmid) contains two promoters (copBp and copAp) that regulate transcripts of the repA gene initiating at a third promoter, repAp. The copB promoter is constitutively active, while repAp is repressed by the small protein product of the copB gene. RepA protein production is also controlled by the 80-base antisense RNA product of copA. The homology between the RepA mRNA leader region, referred to as CopT, and CopA allows the two RNA molecules to form a specific RNA duplex. The formation of this duplex results in posttranscriptional control of RepA protein synthesis. Thus copy number of the R1 group of plasmids is determined by the rate of synthesis and degradation of RepA mRNA, the repressor activity of the CopB protein at the repA promoter, the concentration of the CopA RNA (which is determined by the rates of synthesis and decay of CopA RNA), the rate of formation of the RNA duplex of CopA and CopT, and the efficiency of translation of the RepA mRNA (228). Since it is estimated that the half-life of CopA is of the order of 1 to 2 min, the concentration of CopA is proportional to plasmid copy number. The interaction of these various regulatory elements results in the sensing of plasmid copy number during cell growth and provides for rapid adjustments in replication initiation frequency to reestablish the steady-state copy number. It is of interest that plasmid R1 is a case where a low copy number is corrected by an active mechanism through action of the copB promoter. This results in an increase in mRNA for RepA and a decrease in CopA RNA production because of convergent transcription. A detailed kinetic analysis of the interaction of these various elements in the control of replication of the R1 replicon has recently been carried out (228).
The majority of plasmids in gram-negative bacteria that have been isolated and characterized to date contain one or more clusters of direct repeats within the replication origin region. These direct repeats, designated iterons, have been shown to provide specific multiple binding sites for the plasmid replication initiation protein. Additional direct and/or indirect repeats are also often found in the origin region of iteron-regulated plasmids (Fig. 2). These replicons invariably also encode a trans-acting replication initiation protein (often designated Rep), an A+T-rich region adjacent to the iterons, and generally one or more sites related to the consensus DnaA-binding site. The host-encoded DnaA protein is essential for chromosomal as well as plasmid replication. In E. coli it serves as the initiation protein for oriC by binding to the DnaA boxes in the origin (30). The iteron cluster in the minimal origin region consists of direct repeats generally of a size of 17 to 22 bp with a spacer nucleotide sequence between the smaller iteron sequences to provide a distance of approximately 22 bp between identical iteron sequence positions. This arrangement allows binding of the plasmid initiation protein to the iterons in the same phase of the DNA helix. The number of iterons in a fully functional origin varies depending on the plasmid element; e.g., plasmid pSC101 contains three iterons (52) while plasmid R6K contains seven (280). Deletion of two of the seven iterons within the R6K origin results in a significant loss of origin activity (160).
A number of iteron-containing plasmids contain a second cluster of direct repeats that is nonessential for replication but plays an important role in copy number control; i.e., deletion of the auxiliary cluster of repeats has been shown for several plasmid systems to result in a higher plasmid copy number and, in some cases, destabilization of the plasmid (5, 145, 219, 234, 297). The orientation of the repeats and the spacing between the repeats are variable in the auxiliary cluster, unlike the iterons within the origin.
The plasmid Rep protein gene is usually located adjacent to the replication origin (83). The Rep protein itself is generally of the size of 30 to 40 kDa and is usually found to be dimeric in solution (2, 45, 85, 176, 186, 190). As discussed below, however, for several plasmid initiation proteins the active form of the Rep protein for binding to the iterons has been shown to be the monomeric form (136, 186, 326). In the case of some plasmids, but not for all, the expression of the rep gene is autoregulated (51, 80, 259, 290, 318, 334). The binding site in the promoter is usually indirect repeats of sequences related to the plasmid iteron sequences. Although the Rep protein is required in a positive sense for the initiation of replication, in the case of plasmids R6K (84), P1 (239), and pSC101 (128), substantial overexpression of the Rep protein can result in the inhibition of replication. Mutants of the Rep proteins for plasmids R6K (130, 193, 279), RK2 (71, 76, 119), F (23, 122, 136, 270), P1 (277), Rts1 (339), R1 (209, 235a), ColV-K30 (94), and pSC101 (332) have been isolated which result in the plasmid being maintained at an increased copy number from severalfold to 20-fold or more. In several cases the copy-up phenotype has been shown to be suppressed by the presence of low levels of the wild-type Rep protein (71, 122, 192, 279). These observations have led to the proposal that the Rep proteins of iteron-containing plasmids exhibit both positive and negative activities. A model to account for these activities of the plasmid initiation protein is discussed below.
A general model for the initiation of DNA replication that is applicable to plasmid systems where replication occurs by the theta mode involves the following sequence of events (Fig. 3 and 4). The replicon-specific initiation protein recognizes and binds to specific sequences (e.g., iterons) at the replication origin and effects structural changes such as bending and/or localized melting of an adjacent A+T-rich region, unwinding of the duplex DNA, or both. The initiation protein, perhaps in concert with the host DnaA protein, guides host replication proteins into the open complex. This prepriming complex establishes the starting point for replication to proceed. The prepriming complex, designated as a specialized nucleoprotein complex, consists then of both plasmid-encoded and host-chromosomally encoded proteins in association with origin DNA sequences.
It is clear that all plasmid systems depend substantially or almost entirely on host DNA replication proteins (163). In most cases the plasmid specifies only the replication initiation protein. An extreme case is the ColE1-type plasmids, where replication initiation is totally dependent on host proteins (47, 74). In the case of plasmid RSF1010 and the related plasmid R1162, however, two proteins, in addition to the initiation protein RepC, are encoded by the plasmid and are required for replication (88, 118, 166). The RSF1010-specified proteins RepA and RepB exhibit helicase and primase activities, respectively. The encoding of these two additional enzymatic activities by the RSF1010 plasmid undoubtedly is of importance to the broad-host-range properties of this plasmid. Plasmid RK2, on the other hand, encodes two replication initiation proteins (TrfA-44 and TrfA-33) by using two translational start sites in the same reading frame within the trfA gene (162, 273, 275). While the 33-kDa TrfA protein appears to be sufficient for the initiation of replication of this plasmid in most gram-negative bacteria, the 44-kDa protein is essential for replication of this plasmid in Pseudomonas aeruginosa (69, 273). Thus, in this case the plasmid utilizes two forms of the initiation protein in order to extend its host range. It is remarkable that the 33-kDa TrfA protein alone is capable of initiating replication of this plasmid in a wide range of gram-negative bacteria, indicating an ability of this protein to adapt to the host replication machinery of a variety of distantly related bacteria.
Figures 3 and 4 illustrate several of the key steps in the initiation of replication of iteron-containing plasmids. While the major form of purified plasmid initiation proteins is a dimer in solution, it is now documented for plasmids P1 (326), F (136), and pSC101 (186) and indicated in the case of plasmid pPS10 (220) that the active form of the replication protein that binds to the iterons at the origin of replication is a monomer. E. coli heat shock proteins have been shown to promote the disassembly of the dimeric form of the RepA protein of plasmid P1 (326) in addition to activating the monomeric form for binding to the origin iterons (49, 57, 218). Evidence has been obtained for a similar role of the heat shock proteins in the case of the RepE initiation protein of the F plasmid. It has been suggested that the leucine-zipper motif of the RepA protein of pPS10 is involved in the dimerization of this protein (del Solar et al., in press). For several plasmids in which the expression of the initiation protein is autoregulated, it has been found that the dimer is the active form for binding to the promoter-operator region of the initiation protein gene (81, 136, 186, 220).
Binding of the plasmid Rep protein to the cluster of direct repeats at the origin has been readily demonstrated for a variety of iteron-containing plasmids by utilizing gel mobility shift electrophoresis or DNA footprinting. In the cases examined, the binding of Rep proteins to the iterons appears to be noncooperative. Interestingly, in the case of the replication initiation protein π of plasmid R6K, a single amino acid substitution changes the binding from a noncooperative to a cooperative mode (84a). There is no general pattern, however, with respect to an iteron-binding domain among the various Rep proteins. A mutational analysis of the RK2 TrfA protein resulted in the isolation of a number of DNA-binding mutants that had a single amino acid substitution distributed over a 20% region of the protein towards the N-terminal end of the protein (46, 176). In these same studies, however, it was found that the deletion of the two C-terminal amino acids of the TrfA protein resulted in a complete loss of DNA-binding activity.
Analyses of several plasmid systems clearly indicate the formation of a specialized nucleoprotein structure as a result of binding of the replication protein to the iterons, in some cases involving several host-encoded proteins. A structure, designated the O-some, consisting of eight λ O protein monomers wrapped by the origin DNA sequence, has been visualized by electron microscopy (68). This interaction results in bending of the DNA at the origin and helix destabilization (267). A similar complex has been shown to form in the case of the origin of replication of the pSC101 plasmid, but involving in addition to the RepA protein the host-encoded proteins DnaA and IHF (integration host factor) (283, 284). A key role for the IHF and DnaA proteins in the formation of a specialized nucleoprotein structure involving origin DNA and a plasmid Rep protein has also been demonstrated for the R6K plasmid (81, 82). While the DnaA protein plays a central role in the initiation of replication of the E. coli chromosomal origin of replication by forming an open replication initiation complex in the A+T-rich region of the origin (30), and also has been shown to be required for replication of a variety of plasmid systems (31, 163), it must work in concert with a plasmid Rep protein to activate a plasmid origin. We are only at the beginning stages in our understanding of the interplay between the DnaA and plasmid Rep proteins in activating a plasmid origin. Evidence has been obtained for RepA protein facilitation of DnaA protein binding to the origin of the R1 plasmid (189). In this system, binding of the R1 RepA protein also results in melting of the A+T-rich region at the origin (189, 237). The RepA-DnaA complex may play a role in guiding the DnaB/DnaC complex to a loading site within the open complex at the R1 origin. Recently, an interaction between the DnaA and DnaB proteins has been demonstrated in vitro (188). Unlike the situation with the λ (267), R1 (189, 237), and R6K (216) plasmid systems, where the plasmid replication initiation protein brings about strand opening at the origin, the P1 RepA protein alone does not melt the P1 origin region (218). With the P1 system the DnaA protein, binding to the DnaA boxes flanking the cluster of iterons, carries out this essential strand-opening step. The P1 RepA protein shows some stimulation of this activity of the DnaA protein. The P1 RepA protein, however, plays a key role in binding to the iterons and forming a nucleoprotein complex involving the wrapping of the DNA around the RepA protein monomers (217). Recent studies with plasmid RK2 have shown that both the DnaA protein and the replication initiation protein TrfA bind independently of each other to their specific binding sequences at the origin of replication (A. Blasina and D. R. Helinski, submitted for publication). Thus, while the various iteron-containing plasmids require the interaction of the Rep protein and specific host proteins at the origin of replication, the precise role of these various proteins in the activation of the replication origin varies depending on the plasmid system.
In every case examined at the in vitro level, plasmids of gram-negative bacteria show a requirement for DNA Pol III for elongation of the replication fork. In the case of plasmid R6K, a specific region of the plasmid carries a replication termination signal (ter) that is not specific for the R6K replicon (17, 159). The ter sequence has been characterized and shown to interact specifically with an E. coli host protein designated Ter or Tus to form a functional termination complex (183).
Completion of the replication process for a theta-type plasmid further requires resolution of the catenated form of the two circular daughter DNA molecules that is generated by the elongation/termination process. The important role of the type 2 topoisomerase IV in the resolution of supercoiled catenates generated by duplication of plasmid molecules in E. coli has been demonstrated using an E. coli strain carrying a mutation in the topoisomerase IV gene (6).
Direct repeats within the functional origin or as part of the auxiliary clusters adjacent to the origin, when inserted into a plasmid normally compatible with the iteron-containing plasmid, show strong incompatibility against the iteron plasmid (5, 145, 177, 194, 244, 299, 310, 335). In addition, the insertion of additional homologous iterons into an iteron-containing plasmid results in a reduction of copy number of the plasmid (192, 239). It was proposed initially that the incompatibility resulting from the insertion of iterons into a heterologous replicon was due to titration of a rate-limiting initiation protein (310). The fact that the initiation protein binds to the iterons and that the severity of incompatibility or copy-number reduction is proportional to the number of iterons inserted is consistent with this proposal (192, 239, 299). Furthermore, in the case of plasmid R1162 it was shown that increasing the concentration of the RepC initiation protein increased plasmid copy number and could also overcome iteron-induced incompatibility, again consistent with a titration model for the regulation of plasmid copy number (153). According to this model, replication initiates when sufficient replication protein is present in the cell to fully load the iteron-binding sites at the origin. This results in a duplication of iterons at the origin, with the consequence that there is an inadequate ratio of initiation protein to iterons for further initiation of replication until additional Rep protein is synthesized during the cell cycle.
Extensive studies carried out on the regulation of initiation of replication of the iteron-containing plasmids R6K (192, 207), P1 (3, 239), and RK2 (70, 154) have revealed properties of these plasmid systems that argue against a simple titration model for the regulation of plasmid copy number. For instance, in the case of many of these plasmid systems the expression of the rep gene is autoregulated (e.g., plasmids R6K [80], P1 [51], pSC101 [318, 334], and F [23, 190, 319]), while for plasmid RK2, regulatory elements outside the basic replicon region regulate the expression of the trfA initiation gene, thus dampening oscillations in the concentration of unbound replication protein during the cell cycle (79, 294, 295). Furthermore, increasing the concentration in vivo for the replication initiation protein for plasmids R6K, RK2, P1, pSC101, or Rts1 does not result in a proportional increase in plasmid copy number (70, 84, 128, 239, 291). For plasmid RK2, a 170-fold increase in the concentration of the TrfA protein results in only a 30% increase in copy number of a minimal RK2 replicon (70). Similarly, decreasing the level of the R6K π initiation protein to 1% of the normal level in E. coli (10,000 π dimers per cell) does not result in a decrease in the copy number of this plasmid (84). In response to these and other observations that are inconsistent with a simple titration model, several modifications of this model have been proposed. For example, the autoregulation paradox gave rise to the proposal that the Rep protein has a different affinity for iterons in the promoter-operator region of the rep gene than for the iterons at the origin, with the latter set of iterons possessing a lower affinity for the Rep protein (50). The higher affinity for the iterons controlling rep gene expression would result in a slower response of the rep promoter to decreasing concentrations of the Rep protein.
Trawick and Kline have proposed a two-form model for replication control of F (307). The autorepressor form of the Rep protein would only bind to the control locus. The form that is active in replication initiation would be generated by an irreversible conversion of a small amount of the autorepressor form to the initiation form and would bind to both the control locus and the origin. While this model has commanded some interest, it does not provide for a plasmid-specified element that controls the conversion of the autorepressor form to the initiator form in order to control the frequency of initiation of replication at the plasmid origin. As stated above, several Rep proteins have now been shown to exist as two forms, i.e., dimers and monomers. In the case of two Rep proteins, specified by the autoregulated plasmid systems F (136) and pSC101 (186), the dimer form functions as the autorepressor and the monomer form functions as the initiator protein. In these cases, however, the conversion of one form to the other is reversible.
In response to the data that have been obtained for several systems that are inconsistent with the original or a modified form of the titration model, an alternative regulatory mechanism has been proposed that involves a reversible pairing of plasmid molecules at their origin of replication, mediated specifically by the plasmid Rep protein and resulting in an inactivation of the origin. This model, called handcuffing or coupling, was first proposed for the iteron-containing plasmids R6K (192) and P1 (239). According to this model (Fig. 3), the Rep protein binds to the iteron cluster at the replication origin, and when the iterons are fully loaded, one of two events occurs, depending on the concentration of Rep-bound iterons. If the concentration of the protein-iteron complex is low, as for the case of a lower than normal plasmid copy number, replication is initiated at the origin by the plasmid Rep protein in concert with host replication proteins. However, when the copy number of the plasmid is at or above the normal level, the concentration of Rep protein-bound iterons is sufficiently high to reversibly couple all of the plasmid molecules to each other at the origin region. When the plasmid (iteron) concentration becomes sufficiently low, as in the case of cell growth without any initiation of plasmid replication, the coupled complexes dissociate into active origins. This mechanism of plasmid copy-number sensing is dependent on the iterons and the Rep protein and involves direct communication between plasmid molecules at their origin.
Both genetic and biochemical data have been obtained to support this model for the regulation of replication of plasmids R6K, P1, and RK2. The fact that increasing the concentration of Rep protein for each of these plasmid systems does not result in an increase in plasmid copy number, once the plasmid is established in the cell, is consistent with this model since it is the concentration of the iteron-Rep protein complex that is the determinant of copy number and not the Rep protein concentration alone. In addition, this model leads to the prediction that the Rep protein contains a function that is responsible for Rep-Rep association when complexed to the iterons and that this activity may be separable from the iteron-binding activity of the protein. In fact, copy-up mutants of the Rep proteins of plasmids R6K (85), P1 (218), and RK2 (46) have been obtained that are not altered in their affinity for the iterons at the replication origin, and at least in the case of the RK2 Rep protein, these mutants do not show a substantial difference in the ratio of monomer to dimer forms but are capable of maintaining the plasmid at a substantially higher copy number in E. coli (J. Cereghino and D. R. Helinski, unpublished data). In the case of R6K and RK2, providing a small amount of wild-type protein in trans suppresses this high-copy-number phenotype (71, 85, 279). Biochemical evidence in support of the handcuffing model includes the demonstration of the ability of the R6K, RK2, and P1 Rep proteins to couple plasmid molecules at their specific origins of replication (192, 215, 218; A. Blasina, B. L. Kittell, A. E. Toukdarian, and D. R. Helinski, submitted for publication). It has been further shown that some but not all copy-up mutants of the R6K Rep protein (206, 207) and all three copy-up mutants of the P1 RepA examined (218) were defective in this coupling activity in vitro. Finally, using an in vitro replication system, it was found for plasmids RK2 (154) and P1 (3) that the addition of iterons on a supercoiled heterologous plasmid would inhibit replication even in the presence of excess Rep protein. A copy-up replication initiation protein of plasmid RK2 showed substantially less sensitivity to iteron addition to the in vitro replication system, consistent with decreased activity in coupling iteron-containing molecules (154). A role for the auxiliary iterons of plasmid P1 which control plasmid copy number in the formation of coupled complexes has been demonstrated by electron microscopy (50, 239) and by iteron inhibition of plasmid replication in vitro (3). Examination of the P1 RepA protein-DNA complexes by electron microscopy showed looping between the auxiliary iteron region (incA) and the origin iterons. It was proposed that, like intermolecular coupling, the RepA-mediated looping can cause steric hindrance to the origin function by a similar mechanism (239).
While both genetic and biochemical data have been obtained in support of the coupling model of replication control, more direct proof for the formation of these complexes in vivo and a more consistent correlation between the copy-up phenotype and reduced activity of the replication initiation protein in the formation of paired complexes in vitro for the R6K (π) and RK2 (TrfA) Rep proteins must be obtained before this mechanism is established as the regulatory strategy for the three plasmid systems. Furthermore, it is likely that not all iteron-containing plasmids will satisfy the basic predictions of this model. A copy-up mutation in the RepE protein of plasmid F has been shown to be altered in the proportion of monomer to dimer forms of this protein (136). Also, the concentration of the RepC initiation protein of plasmid RSF1010/R1162 in E. coli is limiting, and therefore, in this bacterium, plasmid copy number is increased in response to an increase in RepC protein (153). Even in the case of the three plasmid systems (R6K, RK2, and P1) where the data are most consistent with the coupling model, there is evidence supporting a role for other control elements in determining plasmid copy number (1, 81, 207, 223, 295, 331). Evidence has been obtained for a role of methylation (1) and the E. coli SeqA protein in the regulation of P1 replication (35a). Data also have been obtained for a role of membranes in the replication of plasmid RK2 (87, 198). It is clear that much remains to be learned before there is a clear understanding of all of the major factors controlling plasmid copy number and the mechanism of their interactions.
Plasmids have been identified in most species of gram-positive bacteria, although in some important pathogens such as Listeria monocytogenes and Streptococus pneumoniae plasmids are quite uncommon. The vast majority of plasmids in gram-positive eubacteria are circular supercoiled DNA molecules ranging in size from 1.2 kb (149) to 100 kb (18, 41, 165). Linear plasmids have been identified thus far among the gram-positive bacteria only in Streptomyces and Borrelia spp.
Plasmids from gram-positive bacteria are classified primarily according to their structure and mode of replication and secondarily according to sequence homologies. Five replication types have been described. As noted, the first and commonest is the asymmetric RC (rolling-circle) type (110). Less common are plasmids known to use the theta mechanism, and these are of two very different types: the "conventional" theta, presumably using a short RNA primer, and a novel type using a long RNA primer and initiated by Pol I. The fourth and fifth classes are linear plasmids of two types, found in Borrelia sp. and Streptomyces sp., respectively (125). Although both of these genera contain circular as well as linear plasmids, the chromosomes of both are linear (178, 262). The linear plasmids of Borrelia have hairpin ends and are thought to replicate by a modified RC mechanism. The linear Streptomyces plasmids have covalently bound 5'-terminal proteins. It has been suggested that the chromosomes and linear plasmids of Streptomyces replicate by the same mechanism, and that this is also true for Borrelia sp. If these suggestions are correct, they have important implications for the long-standing question of whether plasmids primarily coevolved with the host chromosome or primarily evolved through horizontal spread from common ancestral sources.
The well-established principle that genes essential for plasmid replication and maintenance are clustered, forming a minimal replicon, has enabled the isolation and characterization of essential replication genes from a wide variety of plasmids. Sequence analysis of these has revealed homologies that can be used to define families of functionally related replicons. A compilation of these is presented in Table 1. One generalization that is immediately apparent is that RC plasmids are never very large (≤10 kb), presumably because of the structural instabilities inherent in large RC molecules (see below), whereas theta and linear plasmids have a very wide range of sizes, nearly up to that of the bacterial chromosome. Incompatibility groupings are of less utility in classification since they tend to identify only plasmids with virtually identical replicons, especially in the case of countertranscript regulation where a single base change can generate a new incompatibility type (304).
Table 1Homologies among plasmids |
RC plasmids are widely dispersed among the gram-positive bacteria, and many families have members in very distant species, suggesting the occurrence of horizontal transfer. Additionally, many plasmids appear to have been assembled from cassettes, and these cassettes seem sometimes to have been reassorted across species lines. Thus a cassette containing a single-strand origin (SSO) that is functional in one set of species may be associated with a cassette containing a double-strand origin (DSO) (plus-strand origin) and rep gene that have a different host range.
The theta mode of replication has been formally demonstrated for plasmids from gram-positive bacteria only recently (38). However, since the standard RC mode does not appear to be used for plasmids of >10 kb in size, it is very likely that all of the larger plasmids in gram-positive bacteria, as well as those in gram-negative bacteria, use the theta mode of replication. The linear Borrelia plasmids could represent an exception to this rule which will become apparent once the mode of replication of these plasmids has been detailed. It is interesting that conjugative plasmids in eubacteria replicate by the theta mechanism even though conjugative transfer proceeds by a modified RC mechanism. That is, the RC process used for vegetative replication does not seem to be adaptable to conjugation, and vice versa. Indeed, there are several RC plasmids that encode RC mobilization functions that are separate and distinct from the RC functions used for vegetative replication. Indeed, the two sets of functions are unrelated and therefore of different ancestry, implying that they have been assembled by recombination. Three members of the pT181 family, namely pC221, pC223, and pS194 (230), exemplify this type of genome organization.
All linear plasmids thus far identified in gram-positive bacteria belong to a group of replicons known as invertrons, owing to their long terminal inverted repeat sequences (266). The majority of these, including the linear plasmids of Streptomyces, but not those of Borrelia, contain proteins covalently bound to their 5' termini.
Leading-Strand Initiation: the DSO.
RC replication in plasmids follows in essential details the pattern and strategy shown by the single-stranded coliphages (see the review by Baas and Jansz [13]). These plasmids always encode an initiator protein which introduces a site-specific nick in the leading- or plus-strand origin (now referred to as the DSO, or double-strand origin) (158), generating a 3'-OH priming terminus. In the pT181 family, the initiator protein becomes covalently attached to the 5' site of the nick by a phosphotyrosine bond (293) and is very likely to remain attached throughout replication, as is the case with φX174. With pLS1 (a deletion derivative of pMV158), the initiator protein does not become covalently bound to the DNA (213), as is the case with the M13/fd group of single-stranded phages. Although it is widely assumed that all of the RC plasmid initiator proteins use a tyrosine-OH to form a permanent or transient protein-DNA bond during the nicking event, the only plasmids for which there is direct evidence for a tyrosine-DNA bond is the pT181 family, in which this linkage has been demonstrated biochemically (293). Although one or more conserved tyrosines are always found in predicted RC initiator proteins (161), it is probably unwise, at this juncture, to conclude that all RC replicons use a tyrosine hydroxyl for initiation. For the RC plasmids, aside from those of the pT181 family, the putative active tyrosine has been identified on the basis of motifs conserved in φX194 and in pC194. This identification is supported by site-directed mutagenesis experiments in which replacement of the putative active tyrosine of pC194 (Tyr-214) caused a 95% reduction in in vivo replication activity (222). However, replacement of either of two glutamate residues, one neighboring the putative active-site tyrosine and the other some distance away, had an even more profound effect on the activity of the protein. Noirot-Gros et al. (222) point out that carboxyl residues are "invariably present in the active sites of enzymes which cleave DNA" and suggest that Glu-210 (near Tyr-214) may be involved in the termination of replication (see below). This suggestion is supported by the presence of a glutamate residue at the corresponding position in the initiator protein of each of the 22 plasmids belonging to the pC194/pUB110 family. This conserved glutamate, however, is also present in the φX174 initiator protein, where any role in termination is more difficult to envision, given the well-established participation of a second active-site tyrosine in φX174 termination (117, 314a).
Two RC initiators (those of the pT181 and pMV158 families) have thus far been studied biochemically. Since they differ profoundly in several ways, their similarities may represent general properties of the class. The pT181 initiator proteins are dimeric (252), whereas those of pMV158 are hexameric (213). Primary sequences are unrelated. The pT181 initiators form a covalent tyrosine-DNA bond (293); the pMV158 initiator protein does not (213). Nevertheless, both types have site-specific nicking and topoisomerase I-like activities (158, 213). Both nick single-stranded DNA containing their recognition sequence and catalyze strand exchange (see Fig. 5) (213; C. Thomas, personal communication). The strand-exchange reaction is also characteristic of proteins that initiate RC conjugative DNA transfer (240). With two exceptions, all of these proteins in gram-positive bacteria, gram-negative bacteria, and RC viruses of eukaryotes as well as prokaryotes have recently been shown to constitute a vast and widespread single class characterized by certain shared amino acid motifs. The most striking of these are HUHUUU (where H = histidine and U = bulky hydrophobic residue) and YXX[KH](Y), where [KH] is lysine or histidine, X is any residue, and (Y) represents those few initiator proteins that have a second active tyrosine (161). The conserved tyrosine is assumed to be the residue that is joined to DNA during the nicking or topoisomerase I-like activities, and the HUH motif, which is also widely conserved among certain metalloenzymes, is likely to be involved in binding the divalent cations that are universally required for the function of these proteins. The two exceptions are the pT181 family and the pSN2 family (230). The pSN2 family initiator proteins have neither motif and are quite different from the others, in being only 16 kDa in size. The pT181 initiator proteins have a variant of the active tyrosine motif, YxxH, seen also with plasmids of the pE194 family, but lack the HUHUUU motif (127).
The topoisomerase I-like activity of RC initiator proteins is a side reaction since it not only decreases superhelix density, but also fails to generate a replicatable substrate. It is likely that it is an inevitable consequence of the site-specific nicking and strand-transfer (closing) activities necessary for the replication functions of these proteins. Indeed, RC replication may be viewed as a topoisomerase I reaction in which the nicking and closing components are separated by a round of replication. Thus, the well-established requirement for supercoiling for RC replication could simply represent the need to provide a single-stranded region which would allow the nicking reaction but prevent completion of the topoisomerase I-like reaction (closing) until after a round of replication has occurred. Several families of RC plasmids have an inverted repeat surrounding the nicking site (Fig. 6), and in one case this has been shown to form a cruciform in vivo, stimulated by binding of the initiator protein (221). In DSOs lacking a strong potential hairpin, one would presume that the single-stranded configuration is maintained during initiation by some other means.
In the pT181 family, the initiator proteins recognize the DSO using a recognition element consisting of six amino acid residues that is located some 80 residues C-terminal to the active tyrosine (230, 322). The corresponding regions of the DSO, however, are adjacent, indicating that the protein is folded so as to juxtapose the two regions. A similar organization has recently been described for the pMV158 initiator protein (63). The recognition region of the DSO represents the site of tight binding of the initiator protein, whereas the region surrounding the nick site is not affected by the initiator protein, at least by gel mobility shift analysis. Thus one may envision initiation as involving first a tight binding to the DSO recognition site, followed by a bending of the DNA (157) and possibly a deformation of the protein which brings the nicking site closer to the active tyrosine and promotes melting of the DNA. Binding efficiency is enhanced by a cis-acting sequence element, cmp (98), located about 1 kb from the DSO, which binds a specific 60-kDa cellular protein that is probably involved in its function (M. Gennaro, personal communication). The properties of this replication enhancer are quite similar to those of the pSC101 par locus (129, 204), which enhances replication by influencing the local supercoiling of the plasmid DNA and may also facilitate partitioning by enabling decatenation of plasmid molecules. Melting is facilitated by an A+T-rich region distal to the DSO hairpin (P.-Z. Wang and R. P. Novick, unpublished data), resulting in extrusion of the cruciform, which enables the active-site tyrosine of the initiator protein to attack the nicking target. The net result is an initiation complex in which the DSO is in an open configuration with the leading strand nicked and the initiator protein bound at its recognition site, on the downstream side of the nick (see Fig. 7).
It is interesting that the pMV158 and pT181 DSOs are organized similarly despite the lack of significant sequence homology between them (Fig. 6). Whether this represents a general plan for DSOs of RC replicons remains to be seen.
Following establishment of an initiation complex, the next step would have to be realignment of the sequences so that the 3'-OH priming site would once again pair with the template strand (Fig. 7) and the DNA configuration would serve as a suitable substrate for the replicase. The replicase is presumed to be Pol III, though the only direct evidence is inhibition by 5-hydroxy-azo-uracil (R. Novick, unpublished data; S. Khan, personal communication) and lack of any requirement for Pol I (in Bacillus subtilis) (S. Projan and R. Novick, unpublished data). Staphylococcus aureus pcrA mutants, defective in a helicase activity similar to the Rep helicase of E. coli, accumulate the pT181 initiation complex (133), suggesting that the helicase is a key factor in converting this complex to a form that can be used by Pol III. Suppression of the defective replication seen in this helicase mutant occurs by mutations localized to an N-terminal subdomain of the initiator protein, presumably delineating the site of interaction between the initiator protein and the helicase (134).
Mechanism of Termination.
It is widely accepted that termination of RC replication is effected by a strand-transfer reaction catalyzed by the initiator protein (Fig. 5). Unlike φX174, whose initiator protein has two closely placed active tyrosines that alternate from cycle to cycle (117, 314a), most of the plasmid initiator proteins have only a single active tyrosine. With pT181, the initiator protein is a dimer and it is supposed that the hydroxyl of the active tyrosine of the second subunit would serve as the attacking group during termination (251). Alternatively, as proposed by Noirot-Gros et al. (222), the carboxyl group of a conserved neighboring glutamate could serve as the attacking group.
Direct evidence for a role of the initiator protein in termination consists of initiator protein mutations that cause the accumulation of multimers. Such mutations have been isolated for plasmids pIM13 (248), pT48 (43), and pUB110 (24). It is notable that as termination of RC plasmid replication is extremely efficient (135), multimers are very uncommon, and to date there is no evidence for any multimer resolution function among these plasmids (97). Thus, the accumulation of multimers is reasonable evidence for defective termination. As with the filamentous single-stranded coliphages, RC plasmids with two tandem origins show initiation at one and termination at the other. Constructs of this type have been used to show that sequence requirements for termination are less stringent than for initiation. Thus, the pUB110 DSO can be used by the pC194 initiator protein for termination but not for initiation (199). The central hairpin of the pT181 DSO, containing the nick site (see Fig. 6), is sufficient for termination but not for initiation (135). An explanation for this difference may be that since the initiator protein is already bound to the DNA, specific recognition need not be required for termination; the nicking site and flanking sequences involved in any required structure are evidently sufficient.
RC plasmids have been widely used as cloning vectors in gram-positive bacteria, and it has been repeatedly observed that increasing their size above 10 kb causes structural and hereditary instability. Structural instability is discussed below; in the present context, the accumulation of high-molecular-weight (HMW) multimers of plasmids containing large inserts has been repeatedly observed (112, 151, 316) and is thought to be a consequence of a failure to terminate replication (112). Perhaps, owing to the longer time required for polymerization, the initiator protein may dissociate from the displaced leading strand, a possibility that would seem to exist only where the initiator protein is not covalently bound. Alternatively, the 5' end of the leading strand may not be held in juxtaposition to the DSO if the displaced leading strand exceeds a certain length. In some studies (112) but not others (151), the base composition of the insert was important for the formation of HMW DNA. The presence of HMW DNA would be expected, by virtue of the copy control system, to reduce the monomer copy number proportionately and thus cause instability. This is sometimes but not always true (151). In fact, in some cases, the HMW DNA amounts to several times the normal unit copy number of the plasmid, indicating that any copy control expressed by this material is abnormal. Additionally, the accumulation of HMW DNA is inhibitory for cell growth and thus causes counterselection against the plasmid.
Minus-Strand Replication: the SSO.
RC plasmid replication is asymmetric, as the leading strand is displaced as a single-stranded monomeric circle. Synthesis of the lagging strand on this single-stranded template generally begins at a specific, often palindromic site, the SSO. A variety of SSOs have been identified by deletion analysis on the basis of reduced double-strand copy number and the accumulation of circular single-stranded DNA corresponding to the displaced leading strand (65, 113, 184, 271). Four different classes have thus far been identified, SSOA, SSOU, SSOT (114), and SSOW (269; C. Bruand, S. D. Ehrlich, and L. Janniere, submitted for publication), which are distinguished on the basis of sequence, host range, and requirement for RNA polymerase (Rpo; see Table 2).
Table 2SSO classes |
Several plasmids have been found to contain two or more SSOs, often with very different host specificities (314). Thus the SSOs may be important determinants of host range and host specificity among the RC plasmids. Cassette exchange involving SSOs would seem to have occurred frequently, since very different SSOs are found within families of closely related plasmids; in some cases, these "foreign" SSOs are closely related to SSOs from other plasmid families (113, 247, 271).
Complementary-strand synthesis is accomplished by a mechanism that relies wholly on host functions. With the exception of the SSO itself, all plasmid functions are dispensable and SSO-dependent complementary-strand synthesis can be carried out by cell-free extracts prepared from a plasmid-free strain (26). Interestingly, the efficiency of lagging-strand replication for a particular SSO is heavily host dependent. Thus some SSOs are quite host specific (113, 271), while others function effectively in several different species (29, 271). Furthermore, plasmids lacking any defined SSO can be constructed, and these replicate, often rather poorly, so long as the DSO is functional. In other words, single-stranded circular plasmid DNA can be used as a template for complementary-strand synthesis, with very variable efficiency, in the absence of any specific signaling or initiation sequence. This situation resembles that seen with φX174, which requires a full-scale primosome for lagging-strand initiation. Very possibly, a similar mechanism is used for lagging-strand synthesis by plasmids lacking a specific SSO.
It is assumed that lagging-strand synthesis cannot begin until the SSO has been displaced by leading-strand extension and has become single stranded, allowing the formation of a hairpin (or other required secondary structure). Lagging-strand synthesis initiates within this structure (26) and most commonly requires the host Rpo, which is assumed to lay down an RNA primer, though this has yet to be demonstrated directly. As the SSOs generally lack consensus promoter sequences and as their secondary structure is imperfectly double stranded at best, RNA synthesis using such a template would have to involve a highly novel Rpo mechanism. Although single-stranded bubbles are preferential sites for the initiation of transcription by Rpo (7), the underlying substrate is basically double stranded, and fully single-stranded molecules are not used as templates. Certain SSOs function in the presence of rifampin, indicating a lack of Rpo involvement (269). These SSOs could utilize primase in a manner analogous to that used by the single-stranded phage G4, or could utilize a full-scale primosome as does φX174. Sometimes a plasmid will have both an Rpo-dependent and an Rpo-independent SSO (269).
Two major families replicating by the theta mode have been characterized, the pAMβ1 and pWV02 families (see Table 1). Many additional plasmids using the theta mode have been identified but not yet characterized. Unlike the RC plasmids, which differ in only minor ways from one another with respect to replication mechanisms, the two known theta classes are profoundly different, and their differences may parallel those between plasmids of the ColE1 and iteron families of the gram-negative bacteria.
The replicon structure for pWV02, the prototype of the pWV02 class, is shown in Fig. 8. The major open reading frame, rep, is required for replication and presumably encodes an initiator protein. 5' to the rep coding sequence is a highly conserved region of about 300 bp which is also essential for replication. Contained in this region is a 36-bp A+T-rich (∼90% A+T) sequence, which is almost totally conserved among the nine plasmids studied, and a set of 3 1/2 22-bp repeats that could represent iterons comparable to those of the plasmids of gram-negative bacteria. Rather puzzling is the presence in pWV02 of a set of complicated and precisely conserved repetitive sequences 3' to rep, covering about 1.8 kb and completely dispensable. It seems likely that novel principles of plasmid DNA organization and function will be revealed by study of this fascinating set of plasmids.
Plasmids of the pAMβ1 family replicate by a unidirectional theta mechanism apparently quite similar to that used by ColE2 and ColE3 of E. coli (137). The organization of a typical replicon, that of pAMβ1, is diagrammed in Fig. 8. Initiation of replication utilizes a long primer, probably generated by Rpo from the Prep promoter (see Fig. 8). This transcript also covers the initiator protein gene (rep), whose role is thought to be processing of the transcript to generate a primer terminus at ori (39). Unidirectional theta replication is initiated by Pol I and later switches to Pol III, possibly at or near a lagging-strand initiation site (SSI) that resembles the n '-primosome assembly sites (pas) of E. coli plasmids (Bruand et al., submitted).
Downstream of the leading-strand origin are located genes encoding a site-specific recombinase of the resolvase-invertase family and a replicon-specific topoisomerase I. The recombinase is utilized for multimer resolution (see below), and both enzymes are apparently involved in the switch from Pol I to Pol III (see Fig. 9). The Pol I replication complex is paused or halted by the bound resolvase, and this pause is a signal for the switch to Pol III. In the absence of the topoisomerase, long D-loops are then generated (L. Jannière, personal communication). It is not clear precisely how the topoisomerase prevents this.
As noted above, there are a large number of plasmids in gram-positive bacteria that probably utilize the theta mode of replication. For several of these, the functional organization of the replicon has been analyzed (see, for example, reference 324); however, details of the mode of replication and of the roles of the replication genes have not yet been determined.
Thus far, the only true gram-positive bacteria in which linear plasmids have been identified are Streptomyces sp. and the related Rhodococcus spp. (reviewed in references 125 and 263). These plasmids, ranging in size from 9 to over 600 kb, belong to a large class of genetic elements known as invertrons, owing to their long terminal inverted repetitions (263), to which are covalently attached proteins that serve as primers for DNA synthesis. In the now classical mode of replication of these elements, de novo DNA synthesis is initiated at one or both ends of the linear molecule, primed by the terminal proteins rather than by RNA or DNA termini (see review [264]). Rather surprisingly, however, the two linear streptomyces plasmids that have been analyzed thus far, pSCL (272) and pSLA2 (48), have both been found to replicate bidirectionally from an internal origin, as shown in Fig. 10A, rather than from the ends. It has been proposed by Chang and Cohen (48) that the role of the terminal proteins in this mechanism is to prime the DNA synthesis required to fill in the 5'-terminal gap left on the lagging strand by the inability of the standard lagging-strand replication mechanism (RNA-primed Okazaki fragments) to complete a linear molecule. As suggested (48), this mechanism may be widespread in Streptomyces and may also be utilized for replication of the chromosome, which is linear in at least a few species (178). Spirochetes such as Borrelia sp. are considered here as they have a number of features in common with the gram-positive bacteria. As noted above, they have linear chromosomes and plasmids (125); the latter are characterized by hairpin telomeres in which the two strands are connected by a terminal loop. It is thought that this terminal loop is the site of initiation of replication and that this replication proceeds by a modified RC mechanism (a speculative model is shown in Fig. 10B).
All known plasmids of gram-positive bacteria use a plasmid-encoded protein, the initiator protein, for the initiation of replication at the DSO and control their initiation rate (copy number) primarily by controlling initiator protein synthesis. To maintain stable copy numbers, plasmids must thus also control the fate of the initiator protein following its use. There is, however, no evidence for any control of lagging-strand replication, nor is there evidence for any control of the function of the initiator protein, as is the case with the iteron plasmids of gram-negative bacteria.
Regulatory mechanisms have thus far been analyzed in only a few cases, namely the pT181 family, the pUB110 family, the pMV158 family, and the pAMβ1 family, and an immediate generalization is that RC and theta plasmids share strategies for regulating initiator protein synthesis. Thus several families of RC plasmids use antisense RNAs (countertranscripts or countertranscript RNAs) to regulate initiator protein synthesis, and at least one family of theta plasmids, the pAMβ1 family, does so also. For the RC prototype, pT181, countertranscript RNAs act by causing translational attenuation of the initiator protein mRNA (233). A nearly identical mechanism has recently been described for two members of the pAMβ1 family, pIP501 (33) and pAMβ1 (172), and a similar mechanism is used by an RC plasmid from Lactobacillus pentosus (247). A schematic diagram of these parallel systems is presented in Fig. 11. In other families, such as the pMV158 and pUB110 families, countertranscript RNAs are also used (8, 64, 167), but their mechanism of action has yet to be determined. Many plasmids from gram-positive bacteria exhibit dual copy-number control, in which there is a repressor protein as well as a countertranscript RNA. Typical examples are the pMV158 family (64) and the pAMβ1 family (172). In both cases, the repressor gene is 5' to the initiator protein gene and the repressor acts on the initiator protein gene promoter. In one case, that of pE194, this protein binds to a site upstream of the repressor gene so that the repressor gene is autoregulated (167). Although this type of autoregulatory circuit was proposed many years ago by Sompayrac and Maaløe (276), its effectiveness as a copy control mechanism is questionable. As mentioned above, since an autorepressor will maintain its own concentration at a constant level regardless of its gene dosage, it will not respond appropriately to fluctuations in plasmid copy number. The result will be a very broad distribution of copy numbers in individual cells and a consequent segregational instability of the plasmid. This has been seen experimentally and documented theoretically with λdv (330). The coupling of an autorepressor with a countertranscript RNA evidently obviates the disadvantages of the autorepressor, since these plasmids are perfectly stable. Nevertheless, it remains unclear whether the autorepressor confers any regulatory advantage.
Repeated nucleotide sequences are frequently found in or near the minimal replicon in all types of plasmids from gram-positive bacteria as well as in those from gram-negative bacteria. In the pWV02 family, as noted above, a set of 3 1/2 22-bp tandem direct repeats 5' to the initiator protein gene is essential for replication, and it has been suggested that these may be iterons functionally similar to those of the gram-negative bacteria (269). No regulatory role has been ascribed to these sequences, however, and their remarkably high degree of conservation between pairs of compatible plasmids makes it unlikely that they function as plasmid-specific regulatory elements. For example, pWV04 and pWV05 are compatible, yet their iteron sets differ at only two positions. Additionally, and also unlike the iterons of plasmids from gram-negative bacteria, with one exception, pSL2A, all the copies in any one set are identical or differ at not more than one position. In contrast, the iterons of plasmids from gram-negative bacteria, which are the major determinants of incompatibility among these plasmids, differ considerably from plasmid to plasmid and among the members of any one plasmid’s set. Since these iterons function as sequence-specific initiator protein-binding sites, and since such sites can often tolerate a substitution or two without a major change in specificity, if the pWV02 repeats are determinants of plasmid specificity (incompatibility) they must function by a novel mechanism. It is suggested, incidentally, that the term "iteron" be reserved for directly repeated sequences that have a well-defined role in the regulation of replication, as is the case for these sequence elements in the classical iteron plasmids of gram-negative bacteria.
Fate of the Initiator Protein.
Most RC plasmids regulate their replication by controlling initiator protein synthesis in a hyperbolic manner. This means that synthesis of the initiator protein is not directly coupled to replication but is free to vary in response to the concentration of an inhibitor such as antisense RNA. This arrangement permits a rapid response to fluctuations, especially down fluctuations in copy number (124). Although such initiator proteins are maintained at a low, rate-limiting level, the maintenance of a stable copy number requires that they not be recycled, as is the initiator protein of φX174. This requirement has been confirmed by the demonstration of an inactive form of the initiator protein for the pT181 family plasmids that is generated during replication (251). The inactivation is associated with the covalent linkage to one subunit of the initiator protein dimer of an oligonucleotide corresponding to the 10 to 12 nucleotides immediately 3' to the DSO nick site (see Fig. 6). This oligonucleotide is probably attached to the active-site tyrosine and is probably generated during the termination step by a mechanism such as that shown in Fig. 12, in which replication proceeds a short distance past the DSO nick site, generating the DSO hairpin, which is cleaved by the free active tyrosine of the second subunit of the initiator protein. This would initiate a concerted strand transfer cascade, generating a single-stranded monomer corresponding to the displaced leading strand, circularizing the new leading strand, and releasing the initiator protein as an inactive heterodimer. Note that the replication-termination cycle as depicted here involves two nickings, one at initiation and the other at termination, and that termination involves two closings, one for the new and one for the old leading strand. The overall model, however, remains speculative.
Replication Control in Plasmid Cointegrates.
Many bacterial plasmids and chromosomes contain one or more secondary replicons in addition to the one that normally drives replication. These secondary replicons have to be suppressed under normal circumstances. A current theory of negative regulation of replication predicts that if two replicons are joined, the one with the higher intrinsic copy number will drive the cointegrate and the one with the lower copy number will be suppressed because its negative regulator will be maintained at an elevated level. In practice, this prediction has been confirmed in only one case, that of F::ColE1 cointegrates (311). With some artificial cointegrates, the replicons are both used, at frequencies related to their intrinsic copy numbers. It is not known whether this ever occurs in naturally occurring plasmids. In other cointegrates, one replicon shuts off the other in a manner that has nothing to do with the negative regulatory function or the intrinsic copy number of either.
The classical case occurs where replication of the low-copy-number plasmid is controlled by an all-or-none switch. Function of both replicons is seen when the negative regulation is hyperbolic (initiation frequency is inversely proportional to the concentration of the regulatory molecule), and this, in retrospect, is quite predictable. The third situation, known as cis inhibition, has been seen in both prokaryotes and eukaryotes. In well-studied cases this inhibition turns out to be unrelated to any replication function of the inhibitory replicon but is rather caused by "poison" sequences carried by the latter (see, for example, references 78 and 182). A case in point is the inhibition of pT181 replication by pE194 sequences in cis (249). Here deletion of the pE194 SSO or of the pT181 countertranscript RNA promoter relieves the cis inhibition. The countertranscript RNA promoter is strong and reads away from the nearby pT181 DSO, which would increase negative supercoiling at the DSO (180). This, one might think, would promote rather than inhibit the initiation of replication, as it would favor DNA melting in the DSO region. A possible clue to this seeming paradox comes from the observation that the cis-inhibited origin cycles the pT181 initiator protein, converting it to the above-mentioned inactive form containing an oligonucleotide covalently attached to one subunit (A. Rasooly and R. Novick, unpublished data). This implies that replication is initiated at the cis-inhibited pT181 DSO, but is then aborted. One possible explanation is that the nicked initiation complex remains rotationally constrained and that the negative supercoiling stimulated by transcription from the countertranscript promoter forces the DSO cruciform to reextrude prematurely, immediately following the initiation of replication. This structure would then be recognized and cleaved by the second subunit, aborting the replication cycle. The role of the pE194 SSO would then have to be in prevention of the cancellation of these negative supercoils by the positives generated in front of the countertranscript. A naturally occurring example of this has recently been described by Pujol (C. Pujol, Ph.D. thesis, Université Paris-Sud, U.F.R. Scientifique d’Orsay, Paris, France, 1995) and coworkers, who have identified a cis-inhibited pT181-like plasmid within the pAMβ1 genome. As the initiator protein of this cis inhibited plasmid reacted with anti-RepC antibodies, Western immunoblotting showed that the modified form was produced (Rasooly and Novick, unpublished), suggesting that this type of cis inhibition may be common.
Many other natural plasmid counterparts have been described in both gram-negative and gram-positive bacteria (238, 243), and the bacterial chromosome is the repository of a large variety of integrated plasmids, all of which are unable to replicate in situ. Prophage-like repressors are of course possible; indeed, the integrating plasmids of Streptomyces spp. encode functions (imp) that specifically repress plasmid replication, quite like prophage repressors (108). Alternatively, integration may interrupt an essential replication function, such as the initiator protein gene (40, 191, 238). cis inhibition more or less like that described above may also be common.
Our basic understanding of the major features of plasmid incompatibility has not changed significantly over the past decade (229). It is clear that the sharing of any function required for the regulation of plasmid replication by two plasmid elements is likely to result in incompatibility. This is true for plasmids of gram-negative and gram-positive bacteria. Random selection for replication causes disparities in the copy pools of two coresident isogenic plasmids, which cannot be corrected since a copy-number control system does not recognize the two as different (19, 115, 260). In addition, as a consequence of random assortment for partitioning, plasmids that use a common partitioning system show weak segregational incompatibility due to random selection for partitioning (11, 227, 232). This has been shown at least for plasmids of gram-negative bacteria. Analogous partitioning systems for plasmids of gram-positive bacteria have not been characterized to date.
Unidirectional or vectorial incompatibility is shown by pairs of coresident plasmids where one member contains a replicon function used by the other but not by itself. Thus, incompatibility has been seen with cloned copy control elements, a cloned DSO or origin iterons, and cloned partitioning elements. In addition, a disparity in the functionality of a replicon element between otherwise isogenic plasmids can result in asymmetric segregational incompatibility since the plasmid with the more efficient version of the element will outcompete the other. For example, this has been seen with pairs of pT181 derivatives differing in the replication enhancer, also known as the competitivity (cmp) locus. This locus effects the ability to bind and use the plasmid-specific initiation protein (98). Another interesting case is found in Streptomyces, where the imp replication repressor is an important cause of vectorial plasmid incompatibility (108).
Much of our understanding of the mechanistic details of incompatibility involving replication control elements has come from studies on the ColE1 plasmid (47, 74). Because RNA I of ColE1 is a diffusible negative replication regulator, it is the primary incompatibility determinant of the ColE1 plasmid. Consequently, a normally compatible coresident plasmid with a DNA insert that produces RNA I cannot coexist with a ColE1 replicon dependent on the identical RNA I for copy-number control (289). It also follows that single base changes in RNA I, with the complementary change in the overlapping primer (RNA II), can alter the specificity of the regulatory RNA I molecule and result in the creation of a new incompatibility group (168). The Rop protein of ColE1, although involved in copy-number control, is not an incompatibility determinant because it is not essential for replication even though it enhances the incompatibility exerted by RNA I (235, 306). Similarly, the ColE1 replication origin is not an incompatibility determinant because it does not titrate a diffusible inhibitor or activator molecule.
A similar set of observations have been made for the IncFII plasmids which, unlike the ColE1-type plasmids, encode a replication initiation protein. The CopA RNA, like RNA I, is the major incompatibility determinant of the IncFII plasmids (210, 261). As shown for ColE1, changes in CopA nucleotides that are critical for CopA-CopT interactions should change the incompatibility grouping of the plasmid. A divergence in the sequence of CopA is most likely responsible for the compatibility observed between the IncFII plasmids and several naturally occurring plasmids that apparently have a similar replicon structure and regulatory mechanism for replication (329). Because the RepA protein of the IncFII plasmids is cis-acting, the replication origin of these plasmids is not an incompatibility determinant (298).
It has been well established that the direct repeats at the origin and in the copy-control region of the iteron-containing plasmids are incompatibility determinants. This fact was initially established for the F plasmid, where it was shown that inserting a cluster of five 22-bp iterons into a compatible coresident plasmid resulted in incompatibility with the F replicon (299). As described above, this finding has been extended to a number of iteron-containing plasmids. In some cases it has been shown that this incompatibility, the severity of which is correlated directly with the number of iterons inserted, is not relieved in vivo by providing additional amounts of the plasmid-specific initiation protein. This latter observation has been used in support of the coupling model for plasmid replication control.
The major determinants of incompatibility in plasmids of gram-positive bacteria are the small RNA counter-transcript regulatory molecules and the origin of replication. It has been shown that the antisense RNAs of plasmid pT181, when cloned, exert strong incompatibility against pT181 replicons (231). Unlike the ColE1 and IncFII plasmids, however, the pT181 origin is also an incompatibility determinant (132, 231). This activity of the origin is consistent with its ability to bind the plasmid initiation protein and, therefore, to effectively titrate the rate-limiting, trans-acting RepC initiation protein. Interestingly, a comparative analysis of four naturally occurring plasmids of Staphylococcus that are related to plasmid pT181 showed that while each of the plasmids exhibited considerable structural similarity of their replicons, including considerable similarity in the sequences of their antisense RNAs and origin regions, there was sufficient nucleotide sequence divergence to place each of the plasmids in a separate incompatibility group (249).
It is likely that the phenomenon of incompatibility is operative in natural environments in placing limitations on plasmid combinations stably maintained in gram-negative and gram-positive bacteria. Plasmids that possess more than one functional replicon are likely to be less restricted in this regard.
Plasmid stability and instability can be divided into structural and hereditary types. Natural plasmids in their native hosts are hereditarily highly stable, and for the most part are also structurally stable, by virtue of having survived the forces of natural selection. Instabilities of both types arise primarily as a consequence of transfer to a non-native host or of in vitro manipulation.
The streptococcal plasmid pAMβ1 is an example of host-specific structural instability in a natural plasmid. When transferred to B. subtilis, pAMβ1 undergoes site-specific excision of a 4.4-kb segment very closely related to the Bacillus plasmid pBC16, which replicates well in B. subtilis but not in Streptococcus faecalis (243). Here, the excision utilizes a pair of directly repeated sequences which are likely to have been used for formation of the cointegrate in the first place.
It has been observed frequently for artificial constructs that plasmids containing foreign DNA undergo deletions and other rearrangements. These rearrangements are often site specific and they usually cross a splice junction. In a number of cases, rearrangements can be determined to result from recombination between short direct repeats ranging from 3 to 20 bp in length (37, 75). Sometimes, however, repeated sequences are not found at the endpoints. Most of the studies detailing this have been done with B. subtilis (75). Of particular interest is the observation that plasmids of gram-positive bacteria that use the theta mode of replication are structurally much more stable than the RC plasmids (151). This difference has been attributed to the fact that single-stranded DNA is much more prone to replication errors than double-stranded DNA. For example, direct repeats in single-stranded DNA (i.e., in RC plasmids) are prone to slipped mispairing during lagging-strand replication, which causes deletions at least 1,000-fold more frequently than the same repeats in double-stranded DNA (i.e., for theta plasmids). In the latter case, for deletions caused by classical homologous recombination, slipped mispairing is also possible, but evidently occurs much less frequently in double-stranded DNA. The increasing probability of short direct repeats with increasing size is probably one of the reasons that RC plasmids of >10 kb are rare in nature and may account for the difficulty of constructing and stably maintaining large plasmids that replicate via the RC mechanism. These observations have led to the proposal that theta-replicating plasmids are much more useful as cloning vectors than RC plasmids (141, 151).
A variety of changes, including transfer to a new host, mutations, deletions, rearrangements, and insertions of foreign DNA, may upset the delicate balance between plasmid and host that is required for hereditary stability, with a concomitant loss of the plasmid from the culture at some detectable frequency. It is often quite difficult to determine precisely the cause of this destabilization. However, one can discern at least three well-defined possibilities: (i) replication defects; (ii) partitioning defects; (iii) growth disadvantages for plasmid-containing cells. Genetic determinants involved in replication are not properly classed as stability determinants, though for practical reasons it may sometimes be useful to do so. For example, with respect to plasmids of gram-positive bacteria, defects in conversion from single stranded to double stranded, owing to SSO defects or poor SSO function in foreign hosts, usually cause plasmid instability, and SSOs have been referred to, probably incorrectly, as stability determinants.
Plasmids evidently ensure their hereditary stability either by high copy numbers or by specific genetic mechanisms. Among the latter are three: multimer resolution, equipartitioning, and postsegregational killing. Of these mechanisms, equipartitioning is generally used by low- or intermediate-copy-number plasmids. High-copy-number plasmids seem to rely primarily on random partitioning; multimer resolution systems are necessary if multimer formation occurs with significant frequency. In general, RC plasmids have extremely efficient replication termination mechanisms and consequently do not generally encode multimer resolution systems. Postsegregational killing is used by low- and intermediate-copy-number plasmids as a fail-safe mechanism. These specific genetic mechanisms responsible for plasmid hereditary stability have been the subject of a number of reviews (11, 12, 103, 208, 225, 229; del Solar et al., in press).
The occurrence of plasmid multimers is often observed, with the individual units in the multimers generally oriented head-to-tail. The multimers likely originate by defects in the termination of replication or by homologous recombination of monomers. The formation of oligomers decreases the number of individual molecules for distribution to daughter cells at cell division and, thus, has an effect similar to that of a reduction of copy number. This can have serious consequences, particularly in the case of a normally low-copy-number plasmid. A number of different plasmids in gram-negative and gram-positive bacteria possess a multimer resolution system (mrs) that is site specific and will resolve dimers and higher multiple forms to monomers. In many cases the plasmid mrs system includes both a cis-acting site and an encoded recombinase that is specific for this site. This is the case in gram-negative bacteria for the mrs of plasmids P1 (285), F (170), RK2/RP4 (73, 106, 109, 253), and R46 (67) and a virulence plasmid of Salmonella (164). For ColE1 and related plasmids, however, multimer resolution involves a cis-acting site (cer) on the plasmid and four E. coli chromosomally encoded proteins, XerA(ArgR), XerB, XerC, and XerD (27). In addition to the plasmid-specific multimer resolution systems found for a variety of plasmids, many plasmids also carry functional transposons that also encode an mrs (e.g., the TnpR resolvases of the Tn3 family). The role that these transposons play in reducing the number of plasmid multimers has not been systematically examined.
A multimer resolution system has been identified in the pAMβ1 family of plasmids from gram-positive bacteria. All of the well-studied members, pAMβ1 (250), pIP50 (250), and pSM19035 (257, 258), possess this system, which likely serves as a stability function. These plasmids encode proteins belonging to the resolvase-invertase family but differing from the classical Tn3 resolvase in binding to only two rather than three separate sites in the target region. They mediate both resolution and inversion, but only in the presence of an unknown host factor (257). Interestingly, as noted above, they are also involved in replication. It has been proposed that the Res protein of plasmid pIP404 of Clostridium perfringens also is involved in the resolution of multimers of this naturally occurring plasmid (96).
Unlike the partitioning of chromosomes of eukaryotic cells during cell division, the process of segregation of bacterial chromosomal or plasmid DNA to daughter cells is poorly understood. It is clear that for certain plasmid elements one can identify genes that act in trans on an in-cis site and these elements provide better than random segregational stability without affecting plasmid copy number or the growth properties of the host. On the basis of these properties, shared by several well-studied plasmid elements, it has been proposed that the stabilization observed is due to an active partitioning process and for this reason the region is designated par (for extensive reviews on plasmid partitioning systems see references 11, 12, 225, 327, and 329, and del Solar et al., in press). Most of our information comes from the work on plasmids F, P1 and R1/NR1. In the case of all three of these plasmid systems, genetic information includes, as a contiguous set, genes that encode two trans-acting proteins and an in-cis site that has been considered to be an analog of the centromere region of eukaryotic chromosomes. For those plasmids examined, the par region is not replicon specific but is effective when present on a heterologous replicon. In addition to these three extensively studied plasmids, stabilization elements on plasmids RK2/RP4 (73, 106, 253, 254) and pTar (93) and a Salmonella virulence plasmid (164) exhibit properties that suggest their functioning also as a partition system.
Partitioning defects are, broadly speaking, any defects that cause an increase in the probability of generating cells which lack the relevant plasmid (plasmid negatives) despite normal replication and copy number. The hallmark of a partitioning (or segregation) defect is a linear semilog survival curve. In practice, this is observed rather infrequently. Much more frequently, semilog survival curves for unstable plasmids show a downward convexity (acceleration), which is a likely indication that the plasmid negatives have a growth advantage, often too small to be detected in side-by-side measurements of growth rate (226). Misinterpretation of plasmid loss kinetics following the deletion of a region of a plasmid has on occasion resulted in the faulty assignment of a plasmid segment as a partitioning function. True segregation defects have two known causes, defects in partitioning and defects in self-correction of the plasmid copy pool.
The most extensively studied par systems are those of the plasmids F and P1. They also show a number of similarities in structure and function. Each system contains a cis-acting site and two trans-acting proteins: SopA (43.7 kDa) and SopB (35.4 kDa) for F (211), and ParA (44 kDa) and ParB (38 kDa) for P1 (4, 10). The ParA and ParB proteins of P1 show extensive homology to the corresponding SopA and SopB proteins of F (327). Despite this similarity in trans-acting proteins, the in-cis sites for P1 and F are quite different. This site for F (SopC) contains 12 43-bp direct repeats downstream of the sopB gene (123, 171, 211), while the corresponding centromere-like site for P1 contains a 20-bp inverted repeat (4). Further delineation of the in-cis site of F revealed that only one of the 43-bp repeats is necessary for par activity (25, 171). The SopA protein binds to and autoregulates its promoter (212); this repression is enhanced by the SopB protein (212). Similarly, the ParA protein regulates expression of the parAB operon by binding to its promoter (60); this binding is stimulated by the ParB protein (89). The amino acid sequence of both ParA and SopA contains an ATP-binding motif (214). ATPase activity has been demonstrated in ParA, and this activity is stimulated by ParB and DNA (60). The finding of this enzymatic activity for ParA has led to the speculation that ParA-catalyzed ATP hydrolysis may provide energy for a key step in the plasmid partitioning process. Both SopB (212, 323) and ParB (59) bind specifically to their corresponding in-cis sites, and in ParB this binding is enhanced by the IHF protein of E. coli (59, 90). IHF has also been shown to promote bending of the in-cis par region (91, 120). The properties of the F and P1 par systems have given rise to two different basic models for plasmid partitioning. Both models, which are an outgrowth of the original model of Jacob, Brenner, and Cuzin (139), invoke a role for the trans-acting proteins in translocating a partition complex consisting of a plasmid-DNA specialized nucleoprotein structure to a site on both sides of the cellular division plane during the segregation process. The basic difference in the two models is the proposal in one case that two plasmid molecules are paired at their specific in-cis sites before attachment to the host receptor site (10, 12), while in the alternative model freely diffusible plasmid molecules associate individually with the specific receptors in the host (11, 139). Obviously, much remains to be done before we have an understanding of the mechanistic details of the plasmid partitioning process in bacteria.
The par systems of F and P1 are the prototypes for plasmid partitioning systems and bear a number of similarities to each other and the IncFII plasmids R1/NR1/R100 (55, 142, 329). Given the likelihood of partitioning systems in a number of other plasmids, it will be interesting to determine how much variation there is among different par systems. More recently, the stabilization region of the broad-host-range plasmid RK2 (designated par) has received considerable attention (106, 253, 266). This region consists of two operons within a 3.2-kb segment of DNA (106, 254). One of the operons (parCBA) encodes a site-specific resolution system that is capable of resolving multimers (73, 106, 109, 253), while the adjacent operon (parDE) encodes a toxin/antitoxin system that stabilizes by killing plasmidless cells (73, 109, 255, 256). The two adjacent operons are divergent in their arrangement and are autoregulated (61, 72). The parA gene encodes a 24-kDa resolvase that functions at a specific site that spans the promoter of the parCBA operon (73, 106, 253). The ParA protein belongs to the Tn3 family of resolvases (106). The functions of the ParB or ParC proteins are largely unknown; however, the ParB protein displays nuclease activity (E. Grohmann and H. Schwab, submitted for publication). It has been speculated that the parCBA region specifies a nucleoprotein complex that contains both site-specific recombination and partitioning activities (73, 106, 254). Much work remains to be done before the mechanism of stabilization by the parCBA region is understood, particularly since it appears to function as a broad-host-range stabilization system. Another region of plasmid RK2 consisting of the korA incC korB korF and the kfrA operons has been proposed to stabilize by partitioning, on the basis of DNA sequence homology to known par systems and gene inactivation studies (214, 327). While this system alone is dispensable for stabilization of RK2 in E. coli (274), it may function as a partitioning system in certain other gram-negative bacteria.
Assuming random diffusion of plasmids at cell division, one can readily predict the frequency of appearance of plasmidless cells per generation for a plasmid maintained at a given copy number. Studies with plasmid pSC101 have raised the possibility that in some cases plasmid molecules are not freely diffusible but are compartmentalized, possibly in the form of plasmid aggregates. A region designated par in pSC101 substantially decreases the loss rate of this plasmid (196). In contrast to the F, P1, and IncFII plasmids, the pSC101 stabilization region does not encode a protein but consists of two direct repeats bordering an inverted repeat (196, 205, 312). One of the functions of this stabilization region may be to increase the random diffusion of the plasmid, perhaps by disaggregating plasmid molecules. The pSC101 par region possesses a binding site for E. coli DNA gyrase (321), and it has been shown that the loss of plasmid stability by deletion of the par region can be compensated for to a large extent by a topA mutation of E. coli, suggesting an important role of superhelicity in the stable maintenance of pSC101 and most likely other plasmid systems (203). Studies are continuing to determine whether there are multiple roles for this par region in the stable maintenance of pSC101.
For the gram-positive plasmid pADI, deletion of a region outside the minimal replicon region causes plasmid instability. It has been proposed that this region is a par locus and that the functional element consists of two small noncoding RNAs (325). Available data to date are insufficient to support this remarkable possibility convincingly: first, mutants with defects in this region show downwardly accelerating semilog survival curves, and one particular deletion is lethal, suggesting kil/kor or postsegregational lethality rather than a par system; second, plausible coding regions exist within the two RNAs and their translation has not been ruled out. Further experimentation is required to clarify the mechanism of stabilization by this system.
Finally, a gene outside the minimal replicon, encoding a 15-kDa protein, is required for incorporation of the plasmid pTH1030 into B. subtilis spores (175). Though this gene does not seem to affect hereditary stability in growing cells, it clearly qualifies as a plasmid stability determinant.
A number of plasmid-encoded systems have evolved that, by virtue of their ability to kill cells which have have lost a specific plasmid element, stabilize the plasmid in a growing population of cells. These systems fall into two types, depending on whether the toxin responsible for the killing is excreted and works extracellularly or whether the toxin functions intracellularly on the cell that has lost the plasmid. Examples of the former type include both bacteriocins (e.g., colicins produced by colicinogenic plasmids of E. coli [241]) and microcins (16), both of which are antibiotically active polypeptides. In both cases, cells carrying plasmids that produce either a bacteriocin or a microcin are immune to the antibiotic that they produce, while cells in the population that have lost the plasmid lose their immunity to the toxin and are killed. Two types of stabilization systems that kill plasmidless cells from within have been described to date. In one case, both toxin and antitoxin proteins are specified by the plasmid but the toxin protein is more stable, or less susceptible to degradation, than is the antitoxin protein, while in the other case the synthesis of the toxin protein is prevented by inhibition of translation by a plasmid-specified antisense RNA. Both of these toxin/antidote systems have been described in some detail for plasmid systems (103, 208, 265).
Colicinogenic plasmids of E. coli encode protein toxins that are transported outside the host bacterium and act on other E. coli or closely related bacteria that do not carry that particular Col plasmid. While this mechanism of cell killing undoubtedly contributes to the stable maintenance of a Col plasmid, the activity of the toxin is dependent on close proximity of plasmid-containing and plasmid-free cells, unlike the toxin/antitoxin systems, where the toxin acts intracellularly. The first clear demonstration of stabilization by a toxin/antitoxin system came from studies with the ccd system of the F plasmid. It was observed that the loss of the F plasmid from an E. coli cell resulted in filament formation and ultimately cell death (140, 146, 202). The ccd operon, located near the oriS replication origin of the IncFIA replicon, consists of two genes (ccdA and ccdB) that encode a protein toxin (CcdB) and a protein (CcdA) that inhibits the activity of this toxin by association with the toxin protein (140, 146, 200). A third gene (repD) which specifies a resolvase that may play a role in site-specific recombination is also part of this operon (170). The CcdB protein autoregulates expression of the operon; the CcdA protein enhances the binding of the CcdB protein to its promoter (62, 287, 288). Stabilization of a plasmid carrying the ccd system is thought to be the result of greater stability of the toxin protein than the antitoxin and, therefore, the toxic activity of the CcdB protein is eventually released in cells that have lost the plasmid (315). In an elegant study, the target of the toxin activity of the CcdB protein was shown to be the GyrA protein (21, 201). The CcdA protein interacts with the CcdB-GyrA complex, and the resulting inactivation of CcdB by binding to CcdA releases GyrA activity. Studies with the F plasmid underscore the functioning of at least two mechanisms to assure stable maintenance of this low-copy-number plasmid: the F partitioning system and the backup ccd system. This is not unusual in that several plasmids have now been shown to carry multiple systems to stabilize their maintenance in a growing population of bacteria. In those plasmids examined, these systems function independent of the replicon that carries the system.
Plasmid R1 is another good example of a plasmid that carries more than one stabilization system. There are actually two toxin-antitoxin systems encoded by plasmid R1. The parD system of Rl shares the basic features of the F ccd system in that it consists of two genes in an operon that encode a toxin protein (Kid) and its antitoxin (Kis) (34, 35). The Kid protein has been shown in vitro to inhibit DNA replication, and this inhibition is prevented by the presence of the Kis protein, which forms a strong complex with the Kid protein (del Solar et al., in press). Interestingly, the Kis and CcdA proteins exhibit considerable homology to each other (35). The same system is found in the closely related plasmid R100. In this case the genes are designated pemK and pemI for the toxin and antitoxin, respectively (308). In addition to the parD system, plasmid R1 carries the parB operon that includes a toxin protein gene (hok) and an antitoxin sequence (sok) that specifies a relatively unstable antisense RNA (100, 104). Translation of the hok messenger is coupled to the synthesis of another structural gene in the parB operon, designated mok, and the expression of the mok messenger is inhibited by the sok RNA (102, 105, 292). Thus, the production of the Hok protein is indirectly inhibited by translational inhibition of the Mok protein. The relatively rapid degradation of the sok mRNA in plasmid-free cells is thought to result in release of inhibition of the Mok protein and, consequently, the synthesis of the Hok protein toxin which is a membrane protein. Expression of the Hok protein results in the formation of "ghost cells" and a loss of cell viability (104). A system analogous to the parB hok/sok system, designated flm or stm, is also found on the F plasmid. Finally, genes with functional and structural homology to the hok/sok system have been identified on the E. coli chromosome (101, 103, 208, 246). The significance of these chromosomally located toxin/antidote systems, designated gef and relF, is not known.
It is now clear that plasmids often carry both a set of genes that have the properties of a partitioning system and a set of toxin/antitoxin genes, with the latter system correcting for errors in the regulation of replication that result in only a single copy of the plasmid in a dividing cell or a failure of the partitioning system. Recently, the well-studied P1 plasmid was shown to encode a toxin/antitoxin system (designated phd/doc) to complement the P1 partitioning system (174). In addition, the broad-host-range plasmid RK2, as described above, carries the parDE operon, which specifies antitoxin and toxin proteins and functions to kill plasmidless cells of E. coli and other gram-negative bacteria (254, 255, 256). As with the ccd, parD, and parB killing systems, the protein products of the RK2 parDE system are relatively small; their sizes are approximately 11 kDa (ParD) and 9 kDa (ParE) (254). Also, as with the ccd and parD systems, the ParD antitoxin protein forms a tight complex with the ParE toxin protein (E. P. Johnson, A. R. Strom, and D. R. Helinski, submitted for publication). In addition to the parCBA and parDE systems in RK2, evidence has been presented for the presence of a postsegregational arrest system (pas) that appears to inhibit the proliferation of plasmid-free cells (143). As with other plasmids carrying multiple systems for stabilization, much work remains to clarify the relative contributions of the individual systems to plasmid stability under different environmental conditions. Recent studies with RK2 have provided insight into the relative contributions of the parCBA and parDE systems toward stabilizing an intact RK2 plasmid in a variety of gram-negative bacteria (274; C. Easter, P. Sobecky, and D. R. Helinski, manuscript in preparation). The results of these studies suggest that the two systems differ in their contribution to stable maintenance depending on the host and the culture conditions. Finally, these studies further indicate a role for conjugal transfer in the stable maintenance of a conjugative plasmid in a growing population of bacteria.
It is clear from detailed analyses of a variety of plasmid systems in gram-negative and gram-positive bacteria that highly ordered structures involving RNA (repressor) and RNA (target) molecules, or specialized nucleoprotein structures involving plasmid-specified initiation proteins and a specific replication origin sequence that often includes direct nucleotide sequence repeats, are key elements in the regulation of plasmid copy number. In every case it is the concentration of at least one of the components of these highly ordered complexes that determines the frequency of initiation of replication. Particularly for low- or intermediate-copy-number plasmids, mechanisms other than copy-number control are required to limit the number of plasmidless cells that are generated on the basis of simple random distribution of plasmids to daughter cells during cell division. One of these mechanisms appears to be active partitioning of plasmid molecules to assure that each daughter cell receives at least one copy of the plasmid. It is now clear that there are also a variety of plasmid-encoded toxin-antidote systems that act as a backup process to partitioning/regulatory failures. These systems kill those cells that fail to receive a specific plasmid molecule on cell division. Frequently, a particular plasmid will contain, in addition to a partitioning system, one or more toxin-antitoxin genes to provide for stable maintenance. The variety of replication control mechanisms and other stable maintenance systems underscores the ability of plasmids to evolve diverse strategies to assure their hereditary stability.
References
1. Abeles, A., T. Brendler, and S. Austin. 1993. Evidence for two levels of control of P1 oriR and host oriC replication origins by DNA adenine methylation. J. Bacteriol. 175:7801–7807.
2. Abeles, A. L. 1986. P1 plasmid replication: purification and DNA binding activity of the replication protein RepA. J. Biol. Chem. 261:3548–3555.
3. Abeles, A. L., and S. J. Austin. 1991. Antiparallel plasmid-plasmid pairing may control P1 plasmid replication. Proc. Natl. Acad. Sci. USA 88:9011–9015.
4. Abeles, A. L., S. A. Friedman, and S. J. Austin. 1985. Partition of unit-copy miniplasmids to daughter cells. III. The DNA sequence and functional organization of the P1 partition region. J. Mol. Biol. 185:261–272.
5. Abeles, A. L., K. M. Snyder, and D. K. Chattoraj. 1984. P1 plasmid replication: replicon structure. J. Mol. Biol. 173:307–324.
6. Adams, E. A., E. M. Shekhtman, E. L. Zechiedrich, M. B. Schmid, and N. R. Cozzarelli. 1992. The role of topoisomerase IV in partitioning bacterial replicons and the structure of catenated intermediates in DNA replication. Cell 71:277–288.
7. Aiyar, S. E., J. D. Helmann, and P. L. de Hasets. 1994. A mismatch bubble in double-stranded DNA suffices to direct precise transcription initiation by Escherichia coli RNA. J. Biol. Chem. 269:13179–13184.
8. Alonso, J. C., and R. H. Tailor. 1987. Initiation of plasmid pC194 replication and its control in Bacillus subtilis. Mol. Gen. Genet. 210:476–484.
9. Armstrong, K. A., R. Acosta, E. Ledner, Y. Machida, R. Pancotto, M. McCormick, H. Ohtsubo, and E. Ohtsubo. 1984. A 37 × 103 molecular weight plasmid-encoded protein is required for replication and copy number control in the plasmid pSC101 and its temperature-sensitive derivative pHS1. J. Mol. Biol. 174:331–347.
10. Austin, S., and A. Abeles. 1983. Partition of unit-copy miniplasmids to daughter cells. II. The partition region of miniplasmid P1 encodes an essential protein and a centromere-like site at which it acts. J. Mol. Biol. 169:373–387.
11. Austin, S., and K. Nordstrom. 1990. Partition-mediated incompatibility of bacterial plasmids. Cell 60:351–354.
12. Austin, S. J. 1988. Plasmid partition. Plasmid 20:1–9.
13. Baas, P., and H. Jansz. 1988. Single-stranded DNA phage origins. Curr. Top. Microbiol. Immunol. 136:31–70.
14. Baker, T. A., and S. H. Wickner. 1992. Genetics and enzymology of DNA replication in Escherichia coli. Annu. Rev. Genet. 26:447–477.
15. Banerjee, S. K., B. T. Luck, H.-Y. Kim, and V. N. Iyer. 1992. Three clustered origins of replication in a promiscuous-plasmid replicon and their differential use in a PolA+ strain and a ΔPolA strain of Escherichia coli K-12. J. Bacteriol. 174:8139–8143.
16. Baquero, F., and F. Moreno. 1984. The microcins. FEMS Microbiol. Lett. 23:117–175.
17. Bastia, D., J. Germino, J. H. Crosa, and J. Ram. 1981. The nucleotide sequence surrounding the replication terminus of R6K. Proc. Natl. Acad. Sci. USA 78:2095–2099.
18. Baum, J. A., and J. Gonzalez, Jr. 1992. Mode of replication, size and distribution of naturally occurring plasmids in Bacillus thuringiensis. FEMS Microbiol. Lett. 75:143–148.
19. Bazaral, M., and D. R. Helinski. 1970. Replication of a bacterial plasmid and an episome in Escherichia coli. Biochemistry 9:399–406.
20. Behnke, D., P. Tomich, and D. Clewell. 1980. Electron microscopic mapping of deletions on a streptococcal plasmid carrying extraordinarily long inverted repeats. Plasmid 4:139–147.
21. Bernard, P., and M. Couturier. 1992. Cell killing by the F plasmid CcdB protein involves poisoning of DNA-topoisomerase II complexes. J. Mol. Biol. 226:735–745.
22. Berquist, P. L., S. Saadi, and W. K. Maas. 1986. Distribution of basic replicons having homology with RepFIA, RepFIB and RepFIC among incF group plasmids. Plasmid 15:19–34.
23. Bex, F., P. Pierard, A. Desmyter, P. Dreze, M. Colet, and M. Couturier. 1986. Mini-F E protein: the carboxy-terminal end is essential for E gene repression and mini-F copy number control. J. Mol. Biol. 189:293–303.
24. Bidnenko, V. E., A. Gruss, and S. D. Ehrlich. 1993. Mutation in the plasmid pUB110 Rep protein affects termination of rolling-circle replication. J. Bacteriol. 175:5611–5616.
25. Biek, D. P., and J. Shi. 1994. A single 43-bp sopC repeat of plasmid mini-F is sufficient to allow assembly of a functional nucleoprotein partition complex. Proc. Natl. Acad. Sci. USA 91:8027–8031.
26. Birch, P., and S. A. Khan. 1992. Replication of single-stranded plasmid pT181 DNA in vitro. Proc. Natl. Acad. Sci. USA 89:290–294.
27. Blakely, G., G. May, R. McCulloch, L. K. Arciszewska, M. Burke, S. T. Lovett, and D. J. Sherratt. 1993. Two related recombinases are required for site-specific recombination at dif and cer in E. coli K12. Cell 75:351–361.
28. Blomberg, P., H. M. Engdahl, C. Malmgren, P. Romby, and E. G. Wagner. 1994. Replication control of plasmid R1: disruption of an inhibitory RNA structure that sequesters the repA ribosome-binding site permits tap-independent RepA synthesis. Mol. Microbiol. 12:49–60.
29. Boe, L., M.-F. Gros, H. te Riele, S. D. Ehrlich, and A. Gruss. 1989. Replication origins of single-stranded-DNA plasmid pUB110. J. Bacteriol. 171:3366–3372.
30. Bramhill, D., and A. Kornberg. 1988. Duplex opening by dnaA protein at novel sequences in initiation of replication at the origin of the E. coli chromosome. Cell 52:743–755.
31. Bramhill, D., and A. Kornberg. 1988. A model for initiation at origins of DNA replication. Cell 54:915–918.
32. Brantl, S., D. Behnke, and J. Alonso. 1990. Molecular analysis of the replication region of the conjugative Streptococcus agalactiae plasmid pIPa501 in Bacillus subtilis. Comparison with plasmids pAM beta 1 and pSM19035. Nucleic Acids Res. 18:4783–4790.
33. Brantl, S., E. Birch-Hirschfeld, and D. Behnke. 1993. RepR protein expression on plasmid pIP501 is controlled by an antisense RNA-mediated transcription attenuation mechanism. J. Bacteriol. 175:4052–4061.
34. Bravo, A., G. de Torrontegui, and R. Diaz. 1987. Identification of components of a new stability system of plasmid R1, ParD, that is close to the origin of replication of this plasmid. Mol. Gen. Genet. 210:101–110.
35. Bravo, A., S. Ortega, G. de Torrontegui, and R. Diaz. 1988. Killing of Escherichia coli cells modulated by components of the stability system ParD of plasmid R1. Mol. Gen. Genet. 215:146–151.
35a. Brendler, T., A. Abeles, and S. Austin. 1995. A protein that binds to the P1 origin core and the oriC 13mer region in a methylation-specific fashion is the product of the host seqA gene. EMBO J. 14:4083–4089.
36. Brennan, R. G., and B. W. Matthews. 1989. The helix-turn-helix DNA binding motif. J. Biol. Chem. 264:1903–1906.
37. Bron, S., S. Holsappel, G. Venema, and B. P. Peeters. 1991. Plasmid deletion formation between short direct repeats in Bacillus subtilis is stimulated by single-stranded rolling-circle replication intermediates. Mol. Gen. Genet. 226:88–96.
38. Bruand, C., S. D. Ehrlich, and L. Janniere. 1991. Unidirectional theta replication of the structurally stable Enterococcus faecalis plasmid pAMtβ1. EMBO J. 10:2171–2177.
39. Bruand, C., E. Le Chatelier, S. D. Ehrlich, and L. Janniere. 1993. A fourth class of theta-replicating plasmids: the pAMβ1 family from gram-positive bacteria. Proc. Natl. Acad. Sci. USA 90:11668–11672.
40. Byrne, M. E., M. T. Gillespie, and R. A. Skurray. 1991. 4',4″-Adenyltransferase activity on conjugative plasmids isolated from Staphylococcus aureus is encoded on an integrated copy of pUB110. Plasmid 25:70–75.
41. Carlton, B. C., and J. M. Gonzalez, Jr. 1985. Plasmids and delta-endotoxin production in different subspecies of Bacillus thuringiensis, p. 246–252. In J. A. Hoch and P. Setlow (ed.), Molecular Biology of Microbial Differentiation. American Society for Microbiology, Washington, D.C.
42. Castagnoli, L., M. Scarpa, M. Kokkinidis, D. W. Banner, D. Tsernoglou, and G. Cesareni. 1989. Genetic and structural analysis of the ColEI Rop (Rom) protein. EMBO J. 8:621–629.
43. Catchpole, I., and K. G. Dyke. 1991. Replication mutants of Staphylococcus aureus macrolide-lincosamide-streptogramin B resistance plasmid pT48. Mol. Microbiol. 5:959–968.
44. Ceglowski, P., R. Lurz, and J. Alonso. 1993. Functional analysis of pSM19035-derived replicons in Bacillus subtilis. FEMS Microbiol. Lett. 109:145–150.
45. Cereghino, J. L., and D. R. Helinski. 1993. Essentiality of the three carboxyl-terminal amino acids of the plasmid RK2 replication initiation protein TrfA for DNA binding and replication activity in gram-negative bacteria. J. Biol. Chem. 268:24926–24932.
46. Cereghino, J. L., D. R. Helinski, and A. E. Toukdarian. 1994. Isolation and characterization of DNA-binding mutants of a plasmid replication initiation protein utilizing an in vivo binding assay. Plasmid 31:89–99.
47. Cesareni, G., C. M. Helmer, and L. Castagnoli. 1991. Control of ColE1 plasmid replication by antisense RNA. Trends Genet. 7:230–235.
48. Chang, P. C., and S. N. Cohen. 1994. Bidirectional replication from an internal origin in a linear streptomyces plasmid. Science 265:952–954.
49. Chattoraj, D. K. 1995. Role of molecular chaperones in the initiation of plasmid DNA replication, p. 81–98. In J. K. Setlow (ed.), Genetic Engineering. Plenum Press, New York.
50. Chattoraj, D. K., R. J. Mason, and S. H. Wickner. 1988. Mini-P1 plasmid replication: the autoregulation-sequestration paradox. Cell 52:551–557.
51. Chattoraj, D. K., K. M. Snyder, and A. L. Abeles. 1985. P1 plasmid replication: multiple functions of RepA protein at the origin. Proc. Natl. Acad. Sci. USA 82:2588–2592.
52. Churchward, G., P. Linder, and L. Caro. 1983. The nucleotide sequence of replication and maintenance functions encoded by plasmid pSC101. Nucleic Acids Res. 11:5645–5659.
53. Couturier, M., F. Bex, P. L. Bergquist, and W. K. Maas. 1988. Identification and classification of bacterial plasmids. Microbiol. Rev. 52:375–395.
54. Crosa, J. H. 1980. Three origins of replication are active in vivo in the R-plasmid RSF1040. J. Biol. Chem. 255:11075–11077.
55. Dam, M., and K. Gerdes. 1994. Partitioning of plasmid R1. Ten direct repeats flanking the parA promoter constitute a centromere-like partition site, parC, that expresses incompatibility. J. Mol. Biol. 236:1289–1298.
56. DasGupta, S., H. Masukata, and J. Tomizawa. 1987. Multiple mechanisms for initiation of ColE1 DNA replication: DNA synthesis in the presence and absence of ribonuclease H. Cell 51:1113–1122.
57. DasGupta, S., G. Mukhopadhyay, P. P. Papp, M. S. Lewis, and D. K. Chattoraj. 1993. Activation of DNA binding by the monomeric form of the P1 replication initiator RepA by heat shock proteins DnaJ and DnaK. J. Mol. Biol. 232:23–34.
58. da Silva-Tatley, F. M., and L. M. Steyn. 1993. Characterization of a replicon of the moderately promiscuous plasmid, pGSH500, with features of both the mini-replicon of pCU1 and the ori-2 of F. Mol. Microbiol. 7:805–823.
59. Davis, M. A., and S. J. Austin. 1988. Recognition of the P1 plasmid centromere analog involves binding of the ParB protein and is modified by a specific host factor. EMBO J. 7:1881–1888.
60. Davis, M. A., K. A. Martin, and S. J. Austin. 1992. Biochemical activities of the parA partition protein of the P1 plasmid. Mol. Microbiol. 6:1141–1147.
61. Davis, T. L., D. R. Helinski, and R. C. Roberts. 1992. Transcription and autoregulation of the stabilizing functions of broad-host-range plasmid RK2 in Escherichia coli, Agrobacterium tumefaciens and Pseudomonas aeruginosa. Mol. Microbiol. 6:1981–1994.
62. de Feyter, R., C. Wallace, and D. Lane. 1989. Autoregulation of the ccd operon in the F plasmid. Mol. Gen. Genet. 218:481–486.
63. de la Campa, A. G., G. H. del Solar, and M. Espinosa. 1990. Initiation of replication of plasmid pLS1. The initiator protein RepB acts on two distant DNA regions. J. Mol. Biol. 213:247–262.
64. del Solar, G., and M. Espinosa. 1992. The copy number of plasmid pLS1 is regulated by two trans-acting plasmid products: the antisense RNA II and the repressor protein, RepA. Mol. Microbiol. 6:83–94.
65. Deng, Z., T. Kieser, and D. A. Hopwood. 1988. "Strong incompatibility" between derivatives of the Streptomyces multi-copy plasmid pIJ101. Mol. Gen. Genet. 214:286–294.
66. de Vos, W. 1987. Gene cloning and expression in lactic streptococci. FEMS Microbiol Rev. 46:281–295.
67. Dodd, H. M., and P. M. Bennett. 1987. The R46 site-specific recombination system is a homologue of the Tn3 and gamma delta (Tn1000) cointegrate resolution system. J. Gen. Microbiol. 133:2031–2039.
68. Dodson, M., J. Roberts, R. McMacken, and H. Echols. 1985. Specialized nucleoprotein structure at the origin of replication of bacteriophage lambda: complexes with lambda O protein and with lambda O, lambda P, and Escherichia coli DnaB proteins. Proc. Natl. Acad. Sci. USA 82:4678–4682.
69. Durland, R. H., and D. R. Helinski. 1987. The sequence encoding the 43-kilodalton TrfA protein is required for efficient replication or maintenance of minimal RK2 replicons in Pseudomonas aeruginosa. Plasmid 18:164–169.
70. Durland, R. H., and D. R. Helinski. 1990. Replication of the broad-host-range plasmid RK2: direct measurement of intracellular concentrations of the essential TrfA replication protein and their effect on plasmid copy number. J. Bacteriol. 172:3849–3858.
71. Durland, R. H., A. Toukdarian, F. Fang, and D. R. Helinski. 1990. Mutations in the trfA replication gene of the broad-host-range plasmid RK2 result in elevated plasmid copy numbers. J. Bacteriol. 172:3859–3867.
72. Eberl, L., M. Givskov, and H. Schwab. 1992. The divergent promoters mediating transcription of the par locus of plasmid RP4 are subject to autoregulation. Mol. Microbiol. 6:1969–1979.
73. Eberl, L., C. S. Kristensen, M. Givskov, E. Grohmann, M. Gerlitz, and H. Schwab. 1994. Analysis of the multimer resolution system encoded by the parCBA operon of broad-host-range plasmid RP4. Mol. Microbiol. 12:131–41.
74. Eguchi, Y., T. Itoh, and J. Tomizawa. 1991. Antisense RNA. Annu. Rev. Biochem. 60:631–652.
75. Ehrlich, S. D., H. Bierne, E. d’Alencon, D. Vilette, M. Petranovic, P. Noirot, and B. Michel. 1993. Mechanisms of illegitimate recombination. Gene 135:161–166.
76. Fang, F., R. H. Durland, and D. R. Helinski. 1993. Mutations in the gene encoding the replication-initiation protein of plasmid RK2 produce elevated copy numbers of RK2 derivatives in Escherichia coli and distantly related bacteria. Gene 133:1–8.
77. Fernandes-Gonzalez, C., R. Cadenas, M. Noirot-Gros, J. Martin, and J. Gil. 1994. Characterization of a region of plasmid pBL1 of Brevibacterium lactofermentum involved in replication via the rolling-circle model. J. Bacteriol. 176:3154–3161.
78. Figurski, D. H., R. J. Meyer, and D. R. Helinski. 1979. Suppression of ColE1 replication properties by the IncP-1 plasmid RK2 in hybrid plasmids constructed in vitro. J. Mol. Biol. 133:295–318.
79. Figurski, D. H., C. Young, H. C. Schreiner, R. F. Pohlman, D. H. Bechhofer, A. S. Prince, and T. F. D’Amico. 1985. Genetic interactions of broad host range plasmid RK2: evidence for a complex replication regulon, p. 227–241. In D. R. Helinski, S. N. Cohen, D. B. Clewell, D. A. Jackson, and A. Hollander (ed.), Plasmids in Bacteria. Plenum Publishing Corp., New York.
80. Filutowicz, M., G. Davis, A. Greener, and D. R. Helinski. 1985. Autorepressor properties of the π-initiation protein encoded by plasmid R6K. Nucleic Acids Res. 13:103–114.
81. Filutowicz, M., S. Dellis, I. Levchenko, M. Urh, F. Wu, and D. York. 1994. Regulation of replication of an iteron-containing DNA molecule. Prog. Nucleic Acid Res. Mol. Biol. 48:239–273.
82. Filutowicz, M., and R. Inman. 1991. A compact nucleoprotein structure is produced by binding of Escherichia coli integration host factor (IHF) to the replication origin of plasmid R6K. J. Biol. Chem. 266:24077–24083.
83. Filutowicz, M., M. McEachern, A. Greener, P. Mukhopadyay, E. Uhlenhopp, R. Durland, and D. R. Helinski. 1985. Role of the π initiation protein and direct nucleotide sequence repeats in the regulation of plasmid R6K replication, p. 125–140. In D. R. Helinski, S. N. Cohen, D. B. Clewell, A. Jackson, and A. Hollaender (ed.), Plasmids in Bacteria. Plenum Publishing Corp., New York.
84. Filutowicz, M., M. J. McEachern, and D. R. Helinski. 1986. Positive and negative roles of an initiation protein at an origin of replication. Proc. Natl. Acad. Sci. USA 83:9645–9649.
84a. Filutowicz, M., D. York, and I. Levchenko. 1994. Cooperative binding of initiator protein to replication origin conferred by single amino acid substitution. Nucleic Acids Res. 22:4211–4215.
85. Filutowicz, M. E., E. Uhlenhopp, and D. R. Helinski. 1986. Binding of purified wild-type and mutant π initiation proteins to a replication origin region of plasmid R6K. J. Mol. Biol. 187:225–239.
86. Firshein, W. 1989. Role of the DNA/membrane complex in prokaryotic DNA replication. Annu. Rev. Microbiol. 43:89–120.
87. Escherichia coli. J. Bacteriol. 150:1234–1243.
88. Frey, J., and M. Bagdasarian. 1989. The molecular biology of IncQ plasmids, p. 79–94. In C. M. Thomas (ed.), Promiscuous Plasmids of Gram-Negative Bacteria. Academic Press, London.
89. Friedman, S. A., and S. J. Austin. 1988. The P1 plasmid-partition system synthesizes two essential proteins from an autoregulated operon. Plasmid 19:103–112.
90. Funnell, B. E. 1988. Participation of Escherichia coli integration host factor in the P1 plasmid partition system. Proc. Natl. Acad. Sci. USA 85:6657–6661.
91. Funnell, B. E., and L. Gagnier. 1993. The P1 plasmid partition complex at parS. II. Analysis of ParB protein binding activity and specificity. J. Biol. Chem. 268:3616–3624.
92. Gabant, P., A. O. Chahdi, and M. Couturier. 1994. Nucleotide sequence and replication characteristics of RepHI1B: a replicon specific to the IncHI1 plasmids. Plasmid 31:111–120.
93. Gallie, D. R., and C. I. Kado. 1987. Agrobacterium tumefaciens pTAR parA promoter region involved in autoregulation, incompatibility and plasmid partitioning. J. Mol. Biol. 193:465–478.
94. Gammie, A. E., M. E. Tolmasky, and J. H. Crosa. 1993. Functional characterization of a replication initiator protein. J. Bacteriol. 175:3563–3569.
95. Garnier, T., and S. T. Cole. 1988. Identification and molecular genetic analysis of replication functions of the bacteriocinogenic plasmid pIP404 from Clostridium perfringens. Plasmid 19:151–160.
96. Garnier, T., W. Saurin, and S. T. Cole. 1987. Molecular characterization of the resolvase gene, res, carried by a multicopy plasmid from Clostridium perfringens: common evolutionary origin for prokaryotic site-specific recombinases. Mol. Microbiol. 1:371–376.
97. Gennaro, M. L., J. Kornblum, and R. P. Novick. 1987. A site-specific recombination function in Staphylococcus aureus plasmids. J. Bacteriol. 169:2601–2610.
98. Gennaro, M. L., and R. P. Novick. 1988. An enhancer of DNA replication. J. Bacteriol. 170:5709–5717.
99. Georgopoulos, C. 1989. The E. coli dnaA initiation protein: a protein for all seasons. Trends Genet. 5:319–321.
100. Gerdes, K. 1988. The parB (hok/sok) locus of plasmid R1: a general purpose plasmid stabilization system. Bio/Technology 6:1402–1405.
101. Gerdes, K., F. W. Bech, S. T. Jorgensen, O. A. Lobner, P. B. Rasmussen, T. Atlung, L. Boe, O. Karlstrom, S. Molin, and K. von Meyenburg. 1986. Mechanism of postsegregational killing by the hok gene product of the parB system of plasmid R1 and its homology with the relF gene product of the E. coli relB operon. EMBO J. 5:2023–2029.
102. Gerdes, K., K. Helin, O. W. Christensen, and O. A. Lobner. 1988. Translational control and differential RNA decay are key elements regulating postsegregational expression of the killer protein encoded by the parB locus of plasmid R1. J. Mol. Biol. 203:119–129.
103. Gerdes, K., L. K. Poulsen, T. Thisted, A. K. Nielsen, J. Martinussen, and P. H. Andreasen. 1990. The hok killer gene family in gram-negative bacteria. New Biol. 2:946–956.
104. Gerdes, K., P. B. Rasmussen, and S. Molin. 1986. Unique type of plasmid maintenance function: postsegregational killing of plasmid-free cells. Proc. Natl. Acad. Sci. USA 83:3116–3120.
105. Gerdes, K., T. Thisted, and J. Martinussen. 1990. Mechanism of post-segregational killing by the hok/sok system of plasmid R1: sok antisense RNA regulates formation of a hok mRNA species correlated with killing of plasmid-free cells. Mol. Microbiol. 4:1807–1818.
106. Gerlitz, M., O. Hrabak, and H. Schwab. 1990. Partitioning of broad-host-range plasmid RP4 is a complex system involving site-specific recombination. J. Bacteriol. 172:6194–6203.
107. Gilbride, K. A., and J. L. Brunton. 1990. Identification and characterization of a new replication region in Neisseria gonorrhoeae β-lactamase plasmid pFA3. J. Bacteriol. 172:2439–2446.
108. Grant, S. R., S. C. Lee, K. Kendall, and S. N. Cohen. 1989. Identification and characterization of a locus inhibiting extrachromosomal maintenance of the Streptomyces plasmid SLPI. Mol. Gen. Genet. 217:324–331.
109. Grinter, N. J., G. Brewster, and P. T. Barth. 1989. Two mechanisms necessary for the stable inheritance of plasmid RP4. Plasmid 22:203–214.
110. Gros, M. F., H. te Riele, and S. D. Ehrlich. 1987. Rolling circle replication of single-stranded DNA plasmid pC194. EMBO J. 6:3863–3869.
111. Grosschedl, R., and G. Hobom. 1979. DNA sequences and structural homologies of the replication origins of lambdoid bacteriophages. Nature (London) 277:621–627.
112. Gruss, A., and S. D. Ehrlich. 1988. Insertion of foreign DNA into plasmids from gram-positive bacteria induces formation of high-molecular-weight plasmid multimers. J. Bacteriol. 170:1183–1190.
113. Gruss, A., H. F. Ross, and R. P. Novick. 1987. Functional analysis of a palindromic sequence required for normal replication of several staphylococcal plasmids. Proc. Natl. Acad. Sci. USA 84:2165–2169.
114. Gruss, A. D., and S. D. Ehrlich. 1989. The family of highly interrelated single-stranded deoxyribonucleic acid plasmids. Microbiol. Rev. 53:231–241.
115. Gustafsson, P., and K. Nordstrom. 1975. Random replication of the stringent plasmid R1 in Escherichia coli K-12. J. Bacteriol. 123:443–448.
116. Hagege, J., J. Pernodet, A. Friedmann, and M. Guerineau. 1993. Mode and origin of replication of pSAM2, a conjugative integrating element of Streptomyces ambofaciens. Mol. Microbiol. 10:799–812.
117. Hanai, R., and J. C. Wang. 1994. The mechanism of sequence-specific DNA cleavage and strand transfer by ϕX174 gene A* protein. J. Biol. Chem. 268:23830–23836.
118. Haring, V., and E. Scherzinger. 1989. Replication proteins of the IncQ plasmid RSF1010, p. 95–124. In C. M. Thomas (ed.), Promiscuous Plasmids of Gram-Negative Bacteria. Academic Press, London.
119. Haugan, K., P. Karunakan, J. M. Blatny, and S. Valla. 1992. The phenotypes of temperature-sensitive mini-RK2 replicons carrying mutations in the replication control gene trfA are suppressed nonspecifically by intragenic cop mutations. J. Bacteriol. 174:7026–7032.
120. Hayes, F., and S. J. Austin. 1993. Specificity determinants of the P1 and P7 plasmid centromere analogs. Proc. Natl. Acad. Sci. USA 90:9228–9232.
121. Helmer-Citterich, M., M. M. Anceschi, D. W. Banner, and G. Cesarini. 1988. Control of ColEI replication: low affinity specific binding of Rop (Rom) to RNA I and RNA II. EMBO J. 7:557–566.
122. Helsberg, M., J. Ebbers, and R. Eichenlaub. 1985. Mutations affecting replication and copy number control in plasmid mini-F both reside in the gene for the 29-kDa protein. Plasmid 14:53–63.
123. Helsberg, M., and R. Eichenlaub. 1986. Twelve 43-base-pair repeats map in a cis-acting region essential for partition of plasmid mini-F. J. Bacteriol. 165:1043–1045.
124. Highlander, S. K., and R. P. Novick. 1987. Plasmid repopulation kinetics in Staphylococcus aureus. Plasmid 17:210–221.
125. Hinnebusch, J., and K. Tilly. 1993. Linear plasmids and chromosomes in bacteria. Mol. Microbiol. 10:917–922.
126. Hiraga, S., T. Sugiyama, and T. Itoh. 1994. Comparative analysis of the replicon regions of eleven ColE2-related plasmids. J. Bacteriol. 176:7233–7243.
127. Ilyina, T. V., and E. V. Koonin. 1992. Conserved sequence motifs in the initiator proteins for rolling circle DNA replication encoded by diverse replicons from eubacteria, eucaryotes and archaebacteria. Nucleic Acids Res. 20:3279–3285.
128. Ingmer, H., and S. N. Cohen. 1993. Excess intracellular concentration of the pSC101 RepA protein interferes with both plasmid DNA replication and partitioning. J. Bacteriol. 175:7834–7841.
129. Ingmer, H., and S. N. Cohen. 1993. The pSC101 par locus alters protein-DNA interactions in vivo at the plasmid replication origin. J. Bacteriol. 175:6046–6048.
130. Inuzuka, M., and Y. Wada. 1985. A single amino acid alteration in the initiation protein is responsible for the DNA overproduction phenotype of copy number mutants of plasmid R6K. EMBO J. 4:2301–2307.
131. Inuzuka, N., M. Inuzuka, and D. R. Helinski. 1980. Activity in vitro of three replication origins of the antibiotic resistance plasmid RSF1040. J. Biol. Chem. 255:11041–11074.
132. Iordanescu, S. 1987. The Inc3B determinant of plasmid pT181. A mutational analysis. Mol. Gen. Genet. 207:60–67.
133. Iordanescu, S. 1991. The Staphylococcus aureus mutation pcrA3 leads to the accumulation of pT181 replication initiation complexes. J. Mol. Biol. 221:1183–1189.
134. Iordanescu, S. 1993. Plasmid pT181-linked suppressors of the Staphylococcus aureus pcrA3 chromosomal mutation. J. Bacteriol. 175:3916–3917.
135. Iordanescu, S., and S. J. Projan. 1988. Replication termination for staphylococcal plasmids: plasmids pT181 and pC221 cross-react in the termination process. J. Bacteriol. 170:3427–3434.
136. Ishiai, M., C. Wada, Y. Kawasaki, and T. Yura. 1994. Replication initiator protein RepE of mini-F plasmid: functional differentiation between monomers (initiator) and dimers (autogenous repressor). Proc. Natl. Acad. Sci. USA 91:3839–3843.
137. Itoh, T., and T. Horii. 1989. Replication of ColE2 and ColE3 plasmids: in vitro replication dependent on plasmid-coded proteins. Mol. Gen. Genet. 219:249–255.
138. Itoh, Y., Y. Kamio, and Y. Terawaki. 1987. Essential DNA sequence for the replication of Rts1. J. Bacteriol. 169:1153–1160.
139. Jacob, F., S. Brenner, and F. Cuzin. 1963. On the regulation of DNA replication in bacteria. Cold Spring Harbor Symp. Quant. Biol. 28:329–348.
140. Jaffe, A., T. Ogura, and S. Hiraga. 1985. Effects of the ccd function of the F plasmid on bacterial growth. J. Bacteriol. 163:841–849.
141. Janniere, L., C. Bruand, and S. D. Ehrlich. 1990. Structurally stable B. subtilis cloning vectors. Gene 87:53–61.
142. Jensen, R. B., M. Dam, and K. Gerdes. 1994. Partitioning of plasmid R1. The parA operon is autoregulated by ParR and its transcription is highly stimulated by a downstream activating element. J. Mol. Biol. 236:1299–1309.
143. Jovanovic, O. S., E. K. Ayers, and D. H. Figurski. 1994. Host-inhibitory functions encoded by promiscuous plasmids. Transient arrest of Escherichia coli segregants that fail to inherit plasmid RK2. J. Mol. Biol. 237:52–64.
144. Kamio, Y., A. Tabuchi, Y. Itoh, H. Katagiri, and Y. Terawaki. 1984. Complete nucleotide sequence of mini-Rts1 and its copy mutant. J. Bacteriol. 158:307–312.
145. Kamio, Y., and Y. Terawaki. 1983. Nucleotide sequence of an incompatibility region of mini-Rts1 that contains five direct repeats. J. Bacteriol. 155:1185–1191.
146. Karoui, H., F. Bex, P. Dreze, and M. Couturier. 1983. Ham22, a mini-F mutation which is lethal to host cell and promotes recA-dependent induction of lambdoid prophage. EMBO J. 2:1863–1868.
147. Kataoka, M., N. Kuno, T. Horiguchi, T. Seki, and T. Yoshida. 1994. Replication of the Streptomyces plasmid pSN22 through single-stranded intermediates. Mol. Gen. Genet. 242:130–136.
148. Kendall, K., and J. Cullum. 1986. Identification of a DNA sequence associated with plasmid integration in Streptomyces coelicolor. Mol. Gen. Genet. 202:243–245.
149. Khan, S. A., and R. P. Novick. 1982. Structural analysis of plasmid pSN2 in Staphylococcus aureus: no involvement in enterotoxin B production. J. Bacteriol. 149:642–649.
150. Kiewiet, R., S. Bron, K. de Jonge, G. Venema, and J. F. Seegers. 1993. Theta replication of the lactococcal plasmid pWV02. Mol. Microbiol. 10:319–327.
151. Kiewiet, R., J. Kok, J. F. M. L. Seegers, G. Venema, and S. Bron. 1993. The mode of replication is a major factor in segregational plasmid instability in Lactococcus lactis. Appl. Environ. Microbiol. 59:358–364.
152. Kim, H. Y., S. K. Banerjee, and V. N. Iyer. 1994. The incN plasmid replicon: two pathways of DNA polymerase I-independent replication. J. Bacteriol. 176:7735–7739.
153. Kim, K., and R. J. Meyer. 1985. Copy number of the broad host-range plasmid R1162 is determined by the amounts of essential plasmid-encoded proteins. J. Mol. Biol. 185:755–767.
154. Kittell, B. L., and D. R. Helinski. 1991. Iteron inhibition of plasmid RK2 replication in vitro: evidence for intermolecular coupling of replication origins as a mechanism for RK2 replication control. Proc. Natl. Acad. Sci. USA 88:1389–1393.
155. Kittell, B. L., and D. R. Helinski. 1993. Plasmid incompatibility and replication control, p. 223–242. In D. B. Clewell (ed.), Bacterial Conjugation. Plenum Press, New York.
156. Kleanthous, H., C. L. Clayton, and S. Tabaqchali. 1991. Characterization of a plasmid from Helicobacter pylori encoding a replication protein common to plasmids in Gram-positive bacteria. Mol. Microbiol. 5:2377–2389.
157. Koepsel, R. R., and S. A. Khan. 1986. Static and initiator protein-enhanced bending of DNA at a replication origin. Science 233:1316–1318.
158. Koepsel, R. R., R. W. Murray, W. D. Rosenblum, and S. A. Khan. 1985. The replication initiator protein of plasmid pT181 has sequence-specific endonuclease and topoisomerase-like activities. Proc. Natl. Acad. Sci. USA 82:6845–6849.
159. Kolter, R., and D. R. Helinski. 1978. Activity of the replication terminus of plasmid R6K in hybrid replicons in Escherichia coli. J. Mol. Biol. 124:425–441.
160. Kolter, R., and D. R. Helinski. 1982. Plasmid R6K DNA replication. II. Direct nucleotide sequence repeats are required for an active origin. J. Mol. Biol. 161:45–56.
161. Koonin, E. V., and T. V. Ilyina. 1993. Computer-assisted dissection of rolling circle DNA replication. Biosystems 30:241–268.
162. Kornacki, J. A., A. H. West, and W. Firshein. 1984. Protein encoded by the trans-acting replication and maintenance regions of broad host range plasmid RK2. Plasmid 11:48–57.
163. Kornberg, A., and T. Baker. 1992. DNA Replication, 2nd ed. Freeman, New York.
164. Krause, M., and D. G. Guiney. 1991. Identification of a multimer resolution system involved in stabilization of the Salmonella dublin virulence plasmid pSDL2. J. Bacteriol. 173:5754–5762.
165. Kronstad, J. W., H. E. Schnepf, and H. R. Whitely. 1983. Diversity of locations for Bacillus thuringiensis crystal protein genes. J. Bacteriol. 154:419–428.
166. Kues, U., and U. Stahl. 1989. Replication of plasmids in gram-negative bacteria. Microbiol. Rev. 53:491–516.
167. Kwak, J. H., and B. Weisblum. 1994. Regulation of plasmid pE194 replication: control of cop-repF operon transcription by Cop and of repF translation by countertranscript RNA. J. Bacteriol. 176:5044–5051.
168. Lacatena, R. M., and G. Cesareni. 1981. Base pairing of RNA I with its complementary sequence in the primer precursor inhibits ColE1 replication. Nature (London) 294:623–626.
169. Lacks, S. A., P. Lopez, B. Greenberg, and M. Espinosa. 1986. Identification and analysis of genes for tetracycline resistance and replication functions in the broad-host-range plasmid pLS1. J. Mol. Biol. 192:753–765.
170. Lane, D., R. de Feyter, M. Kennedy, S. H. Phua, and D. Semon. 1986. D protein of miniF plasmid acts as a repressor of transcription and as a site-specific resolvase. Nucleic Acids Res. 14:9713–9728.
171. Lane, D., R. Rothenbuehler, A. M. Merrillat, and C. Aiken. 1987. Analysis of the F plasmid centromere. Mol. Gen. Genet. 207:406–412.
172. Le Chatelier, E., S. D. Ehrlich, and L. Janniere. 1995. The pAMβ1 CopF repressor regulates plasmid copy number by controlling transription of the repE gene. Mol. Microbiol. 14:463–471.
173. Leenhouts, K. 1991. Nucleotide sequence and characterization of the broad-host-range lactococcal plasmid pWV01. Plasmid 26:55–66.
174. Lehnherr, H., and M. B. Yarmolinsky. 1995. Addiction protein Phd of plasmid prophage P1 is a substrate of the ClpXP serine protease of Escherichia coli. Proc. Natl. Acad. Sci. USA 92:3274–3277.
175. Lereclus, D., and O. Arantes. 1992. spbA locus ensures the segregational stability of pTH1030, a novel type of gram-positive replicon. Mol. Microbiol. 6:35–46.
176. Lin, J., and D. R. Helinski. 1992. Analysis of mutations in trfA, the replication initiation gene of the broad-host-range plasmid RK2. J. Bacteriol. 174:4110–4119.
177. Lin, L.-S., and R. J. Meyer. 1986. Directly repeated, 20-bp sequence of plasmid R1162 DNA is required for replication, expression of incompatibility, and copy number control. Plasmid 15:35–47.
178. Lin, Y. S., H. M. Kieser, D. A. Hopwood, and C. W. Chen. 1993. The chromosomal DNA of Streptomyces lividans 66 is linear. Mol. Microbiol. 10:923–933.
179. Lin-Chao, S., T. T. Wong, K. J. McDowall, and S. N. Cohen. 1994. Effects of nucleotide sequence on the specificity of rne-dependent and RNase E-mediated cleavages of RNA I encoded by the pBR322 plasmid. J. Biol. Chem. 269:10797–10803.
180. Liu, L. F., and J. C. Wang. 1987. Supercoiling of the DNA template during transcription. Proc. Natl. Acad. Sci. USA 84:7024–7027.
181. Llanes, C., P. Gabant, M. Couturier, and Y. Michel-Briand. 1994. Cloning and characterization of the Inc A/C plasmid RA1 replicon. J. Bacteriol. 176:3403–3407.
182. Lusky, M., and M. Botchan. 1981. Inhibition of SV40 replication in simian cells by specific pBR322 DNA sequences. Nature (London) 293:79–81.
183. MacAllister, T., G. S. Khatri, and D. Bastia. 1990. Sequence-specific and polarized replication termination in vitro: complementation of extracts of tus- Escherichia coli by purified Ter protein and analysis of termination intermediates. Proc. Natl. Acad. Sci. USA 87:2828–2832.
184. Madsen, S. M., L. Andrup, and L. Boe. 1993. Fine mapping and DNA sequence of replication functions of Bacillus thuringiensis plasmid pTX14–3. Plasmid 30:119–130.
185. Manen, D., and L. Caro. 1991. The replication of plasmid pSC101. Mol. Microbiol. 5:233–237.
186. Manen, D., G. L. Upegui, and L. Caro. 1992. Monomers and dimers of the RepA protein in plasmid pSC101 replication: domains in RepA. Proc. Natl. Acad. Sci. USA 89:8923–8927.
187. Marians, K. J. 1992. Prokaryotic DNA replication. Annu. Rev. Biochem. 61:673–719.
188. Marszalek, J., and J. M. Kaguni. 1994. DnaA protein directs the binding of DnaB protein in initiation of DNA replication in Escherichia coli. J. Biol. Chem. 269:4883–4890.
189. Masai, H., and K.-I. Arai. 1987. RepA and DnaA proteins are required for initiation of R1 plasmid replication in vitro and interact with oriR sequence. Proc. Natl. Acad. Sci. USA 84:4781–4785.
190. Masson, L., and D. S. Ray. 1988. Mechanism of autonomous control of the Escherichia coli F plasmid: purification and characterization of the repE gene product. Nucleic Acids Res. 16:413–424.
191. Matthews, P. R., K. C. Reed, and P. R. Stewart. 1987. The cloning of chromosomal DNA associated with methicillin and other resistances in Staphylococcus aureus. J. Gen. Microbiol. 133:1919–1929.
192. McEachern, M. J., M. A. Bott, P. A. Tooker, and D. R. Helinski. 1989. Negative control of plasmid R6K replication: possible role of intermolecular coupling of replication origins. Proc. Natl. Acad. Sci. USA 86:7942–7946.
193. McEachern, M. J., M. Filutowicz, and D. R. Helinski. 1985. Mutations in direct repeat sequences and in a conserved sequence adjacent to the repeats results in a defective replication origin in plasmid R6K. Proc. Natl. Acad. Sci. USA 82:1480–1484.
194. McEachern, M. J., M. Filutowicz, S. Yang, A. Greener, P. Mukhopadhyay, and D. R. Helinski. 1986. Elements involved in the copy number regulation of the antibiotic resistance plasmid R6K. Banbury Rep. 24:195–204.
195. McMacken, R., L. Silver, and C. Georgopoulos. 1987. DNA replication, p. 564–612. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. American Society for Microbiology, Washington, D.C.
196. Meacock, P. A., and S. N. Cohen. 1980. Partitioning of bacterial plasmids during cell division: a cis-acting locus that accomplishes stable plasmid inheritance. Cell 20:529–542.
197. Meyer, R. J., L.-S. Lin, K. Kim, and M. A. Brasch. 1985. Broad host-range plasmid R1162: replication, incompatibility, and copy-number control, p. 173–188. In D. R. Helinski, S. N. Cohen, D. B. Clewell, D. A. Jackson, and A. Hollaender (ed.), Plasmids in Bacteria. Plenum Publishing Corp., New York.
198. Michaels, K., J. Mei, and W. Firshein. 1994. TrfA-dependent, inner-membrane-associated plasmid RK2 DNA synthesis in Escherichia coli maxicells. Plasmid 32:19–31.
199. Michel, B., and S. D. Ehrlich. 1986. Illegitimate recombination occurs between the replication origin of the plasmid pC194 and a progressing replication fork. EMBO J. 5:3691–3696.
200. Miki, T., Z. T. Chang, and T. Horiuchi. 1984. Control of cell division by sex factor F in Escherichia coli. II. Identification of genes for inhibitor protein and trigger protein on the 42.84–43.6 F segment. J. Mol. Biol. 174:627–646.
201. Miki, T., J. A. Park, K. Nagao, N. Murayama, and T. Horiuchi. 1992. Control of segregation of chromosomal DNA by sex factor F in Escherichia coli. Mutants of DNA gyrase subunit A suppress letD (ccdB) product growth inhibition. J. Mol. Biol. 225:39–52.
202. Miki, T., K. Yoshioka, and T. Horiuchi. 1984. Control of cell division by sex factor F in Escherichia coli. I. The 42.84–43.6 F segment couples cell division of the host bacteria with replication of plasmid DNA. J. Mol. Biol. 174:605–625.
203. Miller, C. A., S. L. Beaucage, and S. N. Cohen. 1990. Role of DNA superhelicity in partitioning of the pSC101 plasmid. Cell 62:127–133.
204. Miller, C. A., and S. N. Cohen. 1993. The partition (par) locus of pSC101 is an enhancer of plasmid incompatibility. Mol. Microbiol. 9:695–702.
205. Miller, C. A., W. T. Tucker, P. A. Meacock, P. Gustafsson, and S. N. Cohen. 1983. Nucleotide sequence of the partition locus of Escherichia coli plasmid pSC101. Gene 24:309–315.
206. Miron, A., S. Mukherjee, and D. Bastia. 1992. Activation of distant replication origins in vivo by DNA looping as revealed by a novel mutant form of an initiator protein defective in cooperativity at a distance. EMBO J. 11:1205–1216.
207. Miron, A., I. Patel, and D. Bastia. 1994. Multiple pathways of copy control of gamma replicon of R6K: mechanisms both dependent on and independent of cooperativity of interaction of tau protein with DNA affect the copy number. Proc. Natl. Acad. Sci. USA 91:6438–6442.
208. Molin, S., L. Boe, L. B. Jensen, C. S. Kristensen, M. Givskov, J. L. Ramos, and A. K. Bej. 1993. Suicidal genetic elements and their use in biological containment of bacteria. Annu. Rev. Microbiol. 47:139–166.
209. Molin, S., R. Diaz, B. E. Uhlin, and K. Nordstrom. 1980. Runaway replication of plasmid R1 is not caused by loss of replication inhibitor activity of gene cop. J. Bacteriol. 143:1046–1048.
210. Molin, S., and K. Nordstrom. 1980. Control of plasmid R1 replication: functions involved in replication, copy number control, incompatibility, and switch-off of replication. J. Bacteriol. 141:111–120.
211. Mori, H., A. Kondo, A. Ohshima, T. Ogura, and S. Hiraga. 1986. Structure and function of the F plasmid genes essential for partitioning. J. Mol. Biol. 192:1–15.
212. Mori, H., Y. Mori, C. Ichinose, H. Niki, T. Ogura, A. Kato, and S. Hiraga. 1989. Purification and characterization of SopA and SopB proteins essential for F plasmid partitioning. J. Biol. Chem. 264:15535–15541.
213. Moscoso, M., G. del Solar, and M. Espinosa. 1995. Specific nicking-closing activity of the initiator of replication protein RepB of plasmid pMV158 on supercoiled or single-stranded DNA. J. Biol. Chem. 270:3772–3779.
214. Motallebi-Veshareh, M., D. A. Rouch, and C. M. Thomas. 1990. A family of ATPases involved in active partitioning of diverse bacterial plasmids. Mol. Microbiol. 4:1455–1463.
214a. Moyer, R. W., and R. L. Graves. 1981. The mechanism of cytoplasmic orthopoxvirus DNA replication. Cell 27:391–401.
215. Mukherjee, S., H. Erickson, and D. Bastia. 1988. Detection of DNA looping due to simultaneous interaction of a DNA-binding protein with two spatially separated binding sites on DNA. Proc. Natl. Acad. Sci. USA 85:6287–6291.
216. Mukherjee, S., I. Patel, and D. Bastia. 1985. Conformational changes in a replication origin induced by an initiator protein. Cell 43:189–197.
217. Mukhopadhyay, G., and D. K. Chattoraj. 1993. Conformation of the origin of P1 plasmid replication. Initiator protein induced wrapping and intrinsic unstacking. J. Mol. Biol. 231:19–28.
218. Mukhopadhyay, G., S. Sozhamannan, and D. K. Chattoraj. 1994. Relaxation of replication control in chaperone-independent initiator mutants of plasmid P1. EMBO J. 13:2089–2096.
219. Murotsu, T., K. Matsubara, H. Sugisaki, and M. Takanami. 1981. Nine unique repeating sequences in a region essential for replication and incompatibility of the mini-F plasmid. Gene 15:257–271.
220. Nieto, C., R. Giraldo, E. Fernandez-Tresguerres, and R. Diaz. 1992. Genetic and functional analysis of the basic replicon of pPS10, a plasmid specific for Pseudomonas isolated from Pseudomonas syringae patovar savastanoi. J. Mol. Biol. 223:415–426.
221. Noirot, P., J. Bargonetti, and R. P. Novick. 1990. Initiation of rolling circle replication of pT181 plasmid: initiator protein enhances cruciform extrusion at the origin. Proc. Natl. Acad. Sci. USA 87:8560–8564.
222. Noirot-Gros, M. F., V. Bidnenko, and S. D. Ehrlich. 1994. Active site of the replication protein of the rolling circle plasmid pC194. EMBO J. 13:4412–4420.
223. Nomura, N., H. Masai, M. Inuzuka, C. Miyazaki, E. Ohtsubo, T. Itoh, S. Sasamoto, M. Matsui, R. Ishizaki, and K. Arai. 1991. Identification of eleven single-strand initiation sequences (ssi) for priming of DNA replication in the F, R6K, R100 and ColE2 plasmids. Gene 108:15–22.
224. Nordstrom, K. 1990. Control of plasmid replication—how do DNA iterons set the replication frequency? Cell 63:1121–1124.
225. Nordstrom, K., and S. J. Austin. 1989. Mechanisms that contribute to the stable segregation of plasmids. Annu. Rev. Genet. 23:37–69.
226. Nordstrom, K., S. Molin, and H. Aagaard-Hansen. 1980. Partitioning of plasmid R1 in Escherichia coli. I. Kinetics of loss of plasmid derivatives deleted on the par region. Plasmid 4:215–227.
227. Nordstrom, K., S. Molin, and J. Light. 1984. Control of replication of bacterial plasmids: genetics, molecular biology and physiology of the plasmid R1 system. Plasmid 12:71–90.
228. Nordstrom, K., and E. G. Wagner. 1994. Kinetic aspects of control of plasmid replication by antisense RNA. Trends Biochem. Sci. 19:294–300.
229. Novick, R. P. 1987. Plasmid incompatibility. Microbiol. Rev. 51:381–395.
230. Novick, R. P. 1989. Staphylococcal plasmids and their replication. Annu. Rev. Microbiol. 43:537–565.
231. Novick, R. P., G. K. Adler, S. J. Projan, S. Carleton, S. K. Highlander, A. Gruss, S. A. Khan, and S. Iordanescu. 1984. Control of pT181 replication. I. The pT181 copy control function acts by inhibiting the synthesis of a replication protein. EMBO J. 3:2399–2405.
232. Novick, R. P., and F. C. Hoppensteadt. 1978. On plasmid incompatibility. Plasmid 1:421–434.
233. Novick, R. P., S. Iordanescu, S. J. Projan, J. Kornblum, and I. Edelman. 1989. pT181 plasmid replication is regulated by a countertranscript-driven transcriptional attenuator. Cell 59:395–404.
234. Nozue, H., K. Tsuchiya, and Y. Kamio. 1988. Nucleotide sequence and copy control function of the extension of the incI region (incI-b) of Rts1. Plasmid 19:46–56.
235. Nugent, M. E., T. J. Smith, and W. C. Tacon. 1986. Characterization and incompatibility properties of ROM- derivatives of pBR322-based plasmids. J. Gen. Microbiol. 132:1021–1026.
235a. Ohtsubo, H., T. B. Ryder, Y. Maeda, K. Armstrong, and E. Ohtsubo. 1986. DNA replication of the resistance plasmid R100 and its control. Adv. Biophys. 21:115–133.
236. Okumura, M. S., and C. I. Kado. 1992. The region essential for efficient autonomous replication of pSa in Escherichia coli. Mol. Gen. Genet. 235:55–63.
237. Ortega-Jimenez, S., R. Giraldo-Suarez, M. E. Fernandez-Tresguerres, A. Berzal-Herranz, and R. Diaz-Orejas. 1992. DnaA dependent replication of plasmid R1 occurs in the presence of point mutations that disrupt the dnaA box of oriR. Nucleic Acids Res. 20:2547–2551.
238. Oskam, L., D. J. Hillenga, G. Venema, and S. Bron. 1992. The integrated state of the rolling-circle plasmid pTB913 in the composite Bacillus plasmid pTB19. Mol. Gen. Genet. 233:462–468.
239. Pal, S. K., and D. K. Chattoraj. 1988. P1 plasmid replication: initiator sequestration is inadequate to explain control by initiator-binding sites. J. Bacteriol. 170:3554–3560.
240. Pansegrau, W., W. Schroder, and E. Lanka. 1993. Relaxase (TraI) of IncPα plasmid RP4 catalyzes a site-specific cleaving-joining reaction of single-stranded DNA. Proc. Natl. Acad. Sci. USA 90:2925–2929.
241. Pattus, F., D. Massotte, H. U. Wilmsen, J. Lakey, D. Tsernoglou, A. Tucker, and M. W. Parker. 1990. Colicins: prokaryotic killer-pores. Experientia 46:180–192.
242. Perez-Casal, J. F., A. E. Gammie, and J. H. Crosa. 1989. Nucleotide sequence analysis and expression of the minimum REP1 replication region and incompatibility determinants of pColV-K30. J. Bacteriol. 171:2195–2201.
243. Perkins, J. B., and P. Youngman. 1983. Streptococcus plasmid pAMα1 is a composite of two separable replicons, one of which is closely related to Bacillus plasmid pBC16. J. Bacteriol. 155:607–615.
244. Persson, C., and K. Nordstrom. 1986. Control of replication of the broad host range plasmid RSF1010: the incompatibility determinant consists of directly repeated DNA sequences. Mol. Gen. Genet. 203:189–192.
245. Polisky, B. 1988. ColEI replication control circuitry: sense from antisense. Cell 55:929–932.
246. Poulsen, L. K., N. W. Larsen, S. Molin, and P. Andersson. 1989. A family of genes encoding a cell-killing function may be conserved in all gram-negative bacteria. Mol. Microbiol. 3:1463–1472.
247. Pouwels, P. H., N. van Luijk, R. J. Leer, and M. Posno. 1994. Control of replication of the Lactobacillus pentosus plasmid p353–2: evidence for a mechanism involving transcriptional attenuation of the gene coding for the replication protein. Mol. Gen. Genet. 242:614–622.
248. Projan, S. J., M. Monod, C. S. Narayanan, and D. Dubnau. 1987. Replication properties of pIM13, a naturally occurring plasmid found in Bacillus subtilis, and of its close relative pE5, a plasmid native to Staphylococcus aureus. J. Bacteriol. 169:5131–5139.
249. Projan, S. J., and R. P. Novick. 1992. cis-Inhibitory elements in the pT181 replication system. Plasmid 27:81–92.
250. Pujol, C., S. D. Ehrlich, and L. Janniere. 1994. The promiscuous plasmids pIP501 and pAM beta 1 from gram-positive bacteria encode complementary resolution functions. Plasmid 31:100–105.
251. Rasooly, A., and R. Novick. 1993. Replication-specific inactivation of a rolling circle initiator protein. Science 262:1048–1050.
252. Rasooly, A., P. Z. Wang, and R. P. Novick. 1994. Replication-specific conversion of the Staphylococcus aureus pT181 initiator protein from an active homodimer to an inactive heterodimer. EMBO J. 13:5245–5251.
253. Roberts, R. C., R. Burioni, and D. R. Helinski. 1990. Genetic characterization of the stabilizing functions of a region of broad-host-range plasmid RK2. J. Bacteriol. 172:6204–6216.
254. Roberts, R. C., and D. R. Helinski. 1992. Definition of a minimal plasmid stabilization system from the broad-host-range plasmid RK2. J. Bacteriol. 174:8119–8132.
255. Roberts, R. C., C. Spangler, and D. R. Helinski. 1993. Characteristics and significance of DNA binding activity of plasmid stabilization protein ParD from the broad host-range plasmid RK2. J. Biol. Chem. 268:27109–27117.
256. Roberts, R. C., A. R. Strom, and D. R. Helinski. 1994. The parDE operon of the broad-host-range plasmid RK2 specifies growth inhibition associated with plasmid loss. J. Mol. Biol. 237:35–51.
257. Rojo, F., and J. C. Alonso. 1994. The beta recombinase from the streptococcal plasmid pSM19035 represses its own transcription by holding the RNA polymerase at the promoter region. Nucleic Acids Res. 22:1855–1860.
258. Rojo, F., F. Weise, and J. C. Alonso. 1993. Purification of the beta product encoded by the Streptococcus pyogenes plasmid pSM19035. A putative DNA recombinase required to resolve plasmid oligomers. FEBS Lett. 328:169–173.
259. Rokeach, L. A., L. Søgaard-Andersen, and S. Molin. 1985. Two functions of the E protein are key elements in the plasmid F replication control system. J. Bacteriol. 164:1262–1270.
260. Rownd, R. 1969. Replication of a bacterial episome under relaxed control. J. Mol. Biol. 44:387–402.
261. Rownd, R. H., D. D. Womble, X. Dong, V. A. Luckow, and R. P. Wu. 1985. Incompatibility and IncFII plasmid replication control, p. 335–354. In D. R. Helinski, S. N. Cohen, D. B. Clewell, D. A. Jackson, and A. Hollaender (ed.), Plasmids in Bacteria. Plenum Publishing Corp., New York.
262. Saint Girons, I., S. J. Norris, U. Gobel, J. Meyer, E. M. Walker, and R. Zuerner. 1992. Genome structure of spirochetes. Res. Microbiol. 143:615–621.
263. Sakaguchi, K. 1990. Invertrons, a class of structurally and functionally related genetic elements that includes linear DNA plasmids, transposable elements, and genomes of adeno-type viruses. Microbiol. Rev. 54:66–74.
264. Salas, M. 1991. Protein-priming of DNA replication. Annu. Rev. Biochem. 60:39–71.
265. Salmon, M. A., L. van Melderen, P. Bernard, and M. Couturier. 1994. The antidote and autoregulatory functions of the F plasmid CcdA protein: a genetic and biochemical survey. Mol. Gen. Genet. 244:530–538.
266. Saurugger, P. N., O. Hrabak, H. Schwab, and R. M. Lafferty. 1986. Mapping and cloning of the par-region of broad-host-range plasmid RP4. J. Biotechnol. 4:333–343.
267. Schnos, M., K. Zahn, R. B. Inman, and F. R. Blattner. 1988. Initiation protein induced helix destabilization at the lambda origin: a prepriming step in DNA replication. Cell 52:385–395.
268. Scott, J. R. 1984. Regulation of plasmid replication. Microbiol. Rev. 48:1–23.
269. Seegers, J. F., S. Bron, C. M. Franke, G. Venema, and R. Kiewiet. 1994. The majority of lactococcal plasmids carry a highly related replicon. Microbiology 140:1291–1300.
270. Seelke, R. W., B. C. Kline, J. D. Trawick, and G. D. Ritts. 1982. Genetic studies of F plasmid maintenance genes involved in copy number control, incompatibility, and partitioning. Plasmid 7:163–179.
271. Seery, L., and K. M. Devine. 1993. Analysis of features contributing to activity of the single-stranded origin of Bacillus plasmid pBAA1. J. Bacteriol. 175:1988–1994.
272. Shiffman, D., and S. N. Cohen. 1992. Reconstruction of a Streptomyces linear replicon from separately cloned DNA fragments: existence of a cryptic origin of circular replication within the linear plasmid. Proc. Natl. Acad. Sci. USA 89:6129–6133.
273. Shingler, V., and C. M. Thomas. 1984. Analysis of the trfA region of broad host-range plasmid RK2 by transposon mutagenesis and identification of polypeptide products. J. Mol. Biol. 175:229–249.
274. Sia, E. A., R. C. Roberts, C. Easter, D. R. Helinski, and D. H. Figurski. 1995. Different relative importance of the par operons and the effect of conjugal transfer on the maintenance of intact promiscuous plasmid RK2. J. Bacteriol. 177:2789–2797.
274a. Sloan, J., T. A. Warner, P. T. Scott, T. L. Bannam, D. I. Berryman, and J. I. Rood. 1992. Construction of a sequenced Clostridium perfringens-Escherichia coli shuttle plasmid. Plasmid 27:207–219.
275. Smith, C. A., and C. M. Thomas. 1984. Nucleotide sequence of the trfA gene of broad host-range plasmid RK2. J. Mol. Biol. 175:251–262.
276. Sompayrac, L., and O. Maaløe. 1973. Autorepressor model for control of DNA replication. Nature (London) New Biol. 241:133–135.
277. Sozhamannan, S., and D. K. Chattoraj. 1993. Heat shock proteins DnaJ, DnaK, and GrpE stimulate P1 plasmid replication by promoting initiator binding to the origin. J. Bacteriol. 175:3546–3555.
278. Sriprakash, K. S., and E. S. Macavoy. 1987. Characterization and sequence of a plasmid from the trachoma biovar of Chlamydia trachomatis. Plasmid 18:205–214.
279. Stalker, D. M., M. Filutowicz, and D. R. Helinski. 1983. Release of initiation control by a mutational alteration in the R6K pi protein required for plasmid DNA replication. Proc. Natl. Acad. Sci. USA 80:5500–5504.
280. Stalker, D. M., R. Kolter, and D. R. Helinski. 1979. Nucleotide sequence of the region of an origin of replication of the antibiotic resistance plasmid R6K. Proc. Natl. Acad. Sci. USA 76:1150–1154.
281. Stalker, D. M., R. Kolter, and D. R. Helinski. 1982. Plasmid R6K DNA replication. I. Complete nucleotide sequence of an autonomously replicating segment. J. Mol. Biol. 161:33–43.
282. Stalker, D. M., C. M. Thomas, and D. R. Helinski. 1981. Nucleotide sequence of the region of the origin of replication of the antibiotic resistance plasmid RK2. Mol. Gen. Genet. 181:8–12.
283. Stenzel, T. T., T. MacAllister, and D. Bastia. 1991. Cooperativity at distance promoted by the combined action of two replication initiator proteins and a DNA bending protein at the replication origin of pSC101. Genes Dev. 5:1453–1463.
284. Stenzel, T. T., P. Patel, and D. Bastia. 1987. The integration host factor of Escherichia coli binds to bent DNA at the origin of replication of the plasmid pSC101. Cell 49:709–717.
285. Sternberg, N., and D. Hamilton. 1981. Bacteriophage P1 site-specific recombination. I. Recombination between loxP sites. J. Mol. Biol. 150:467–486.
286. Tacon, W., and D. Sherratt. 1976. ColE plasmid replication in DNA polymerase I-deficient strains of Escherichia coli. Mol. Gen. Genet. 147:331–335.
287. Tam, J. E., and B. C. Kline. 1989. Control of the ccd operon in plasmid F. J. Bacteriol. 171:2353–2360.
288. Tam, J. E., and B. C. Kline. 1989. The F plasmid ccd autorepressor is a complex of CcdA and CcdB proteins. Mol. Gen. Genet. 219:26–32.
289. Tamm, J., and B. Polisky. 1983. Structural analysis of RNA molecules involved in plasmid copy number control. Nucleic Acids Res. 11:6381–6397.
290. Terawaki, Y., Z. Hong, Y. Itoh, and Y. Kamio. 1988. Importance of the C terminus of plasmid Rts1 RepA protein for replication and incompatibility of the plasmid. J. Bacteriol. 170:1261–1267.
291. Terawaki, Y., H. Nozue, H. Zeng, T. Hayashi, Y. Kamio, and Y. Itoh. 1990. Effects of mutations in the repA gene of plasmid Rts1 on plasmid replication and autorepressor function. J. Bacteriol. 172:786–792.
292. Thisted, T., and K. Gerdes. 1992. Mechanism of post-segregational killing by the hok/sok system of plasmid R1: sok antisense RNA regulates hok gene expression indirectly through the overlapping mok gene. J. Mol. Biol. 223:41–54.
293. Thomas, C. D., D. F. Balson, and W. V. Shaw. 1990. In vitro studies of the initiation of staphylococcal plasmid replication. J. Biol. Chem. 265:5519–5530.
294. Thomas, C. M. 1988. Recent studies on the control of plasmid replication. Biochim. Biophys. Acta 949:253–263.
295. Thomas, C. M., and G. Jagura-Burdzy. 1992. Replication and segregation: the replicon hypothesis revisited, p. 45–88. In S. Moran, C. Dow, and J. A. Cole (ed.), Prokaryotic Structure and Function. Cambridge University Press, Society for General Microbiology Ltd., Cambridge.
296. Thomas, C. M., and C. A. Smith. 1987. Incompatibility group P plasmids: genetics, evolution, and use in genetic manipulation. Annu. Rev. Microbiol. 41:77–101.
297. Thomas, C. M., D. S. Stalker, and D. R. Helinski. 1981. Replication and incompatibility properties of segments of the origin region of the broad host-range plasmid RK2. Mol. Gen. Genet. 181:1–7.
298. Timmis, K. N., L. Andres, and P. M. Slocombe. 1978. Plasmid incompatibility: cloning analysis of an Inc FII determinant of R6–5. Nature (London) 273:27–32.
299. Tolun, A., and D. R. Helinski. 1981. Direct repeats of the plasmid incC region express F incompatibility. Cell 24:687–694.
300. Tolun, A., and D. R. Helinski. 1982. Separation of the minimal replication region of the F plasmid into a replication origin segment and a trans-acting segment. Mol. Gen. Genet. 186:372–377.
301. Tomizawa, J. 1984. Control of ColEI plasmid replication: the process of binding of RNA I to the primer transcript. Cell 38:861–870.
302. Tomizawa, J. 1985. Control of ColEI plasmid replication: initial interaction of RNA I and the primer transcript is reversible. Cell 40:527–535.
303. Tomizawa, J. 1990. Control of ColEI plasmid replication. Intermediates in the binding of RNA I and RNA II. J. Mol. Biol. 212:683–694.
304. Tomizawa, J., and T. Itoh. 1981. Plasmid ColE1 incompatibility determined by interaction of RNA I with primer transcript. Proc. Natl. Acad. Sci. USA 78:6096–6100.
305. Tomizawa, J., T. Itoh, G. Selzer, and T. Som. 1981. Inhibition of ColE1 RNA primer formation by plasmid-specified small RNA. Proc. Natl. Acad. Sci. USA 78:1421–1425.
306. Tomizawa, J., and T. Som. 1984. Control of ColEI plasmid replication: enhancement of binding of RNA I to the primer transcript by the Rom protein. Cell 38:871–878.
307. Trawick, J. D., and B. C. Kline. 1985. A two-stage molecular model for control of mini-F replication. Plasmid 13:59–69.
308. Tsuchimoto, S., H. Ohtsubo, and E. Ohtsubo. 1988. Two genes, pemK and pemI, responsible for stable maintenance of resistance plasmid R100. J. Bacteriol. 170:1461–1466.
309. Tsurimoto, T., and K. Matsubara. 1981. Purified bacteriophage lambda O protein binds to four repeating sequences at the lambda replication origin. Nucleic Acids Res. 9:1789–1799.
310. Tsutsui, H., A. Fujiyama, T. Murotsu, and K. Matsubara. 1983. Role of nine repeating sequences of the mini-F genome for expression of F-specific incompatibility phenotype and copy number control. J. Bacteriol. 155:337–344.
311. Tsutsui, H., and K. Matsubara. 1981. Replication control and switch-off function as observed with a mini-F factor plasmid. J. Bacteriol. 147:509–516.
312. Tucker, W. T., C. A. Miller, and S. N. Cohen. 1984. Structural and functional analysis of the par region of the pSC101 plasmid. Cell 38:191–201.
313. Twigg, A. J., and D. Sherratt. 1980. Trans-complementable copy-number mutants of plasmid ColEI. Nature (London) 283:216–218.
314. van der Lelie, D., S. Bron, G. Venema, and L. Oskam. 1989. Similarity of minus origins of replication and flanking open reading frames of plasmids pUB110, pTB913 and pMV158. Nucleic Acids Res. 17:7283–7294.
314a. van Mansfeld, A. D. M., H. A. A. M. Teeffelen, P. D. Baas, and H. S. Jansz. 1986. Two juxtaposed tyrosyl-OH groups participate in phiX174 gene A protein catalysed cleavage and ligation of DNA. Nucleic Acids Res. 14:4229–4238.
315. Van Melderen, L., P. Bernard, and M. Couturier. 1994. Lon-dependent proteolysis of CcdA is the key control for activation of CcdB in plasmid-free segregant bacteria. Mol. Microbiol. 11:1151–1157.
316. Viret, J. F., and J. C. Alonso. 1987. Generation of linear multigenome-length plasmid molecules in Bacillus subtilis. Nucleic Acids Res. 15:6349–6367.
317. Vocke, C., and D. Bastia. 1983. Primary structure of the essential replicon of the plasmid pSC101. Proc. Natl. Acad. Sci. USA 80:6557–6561.
318. Vocke, C., and D. Bastia. 1983. The replication initiator protein of plasmid pSC101 is a transcriptional repressor of its own cistron. Proc. Natl. Acad. Sci. USA 82:2252–2256.
319. Wada, C., M. Imai, and T. Yura. 1987. Host control of plasmid replication: requirement for the sigma factor sigma32 in transcription of mini-F replication. Proc. Natl. Acad. Sci. USA 84:8849–8853.
320. Wagner, E. G., and R. W. Simons. 1994. Antisense RNA control in bacteria, phages, and plasmids. Annu. Rev. Microbiol. 48:713–742.
321. Wahle, E., and A. Kornberg. 1988. The partition locus of plasmid pSC101 is a specific binding site for DNA gyrase. EMBO J. 7:1889–1895.
322. Wang, P.-Z., S. J. Projan, V. Henriquez, and R. P. Novick. 1992. Specificity of origin recognition by replication initiator protein in plasmids of the pT181 family is determined by a six amino acid residue element. J. Mol. Biol. 223:145–158.
323. Watanabe, E., S. Inamoto, M. H. Lee, S. U. Kim, T. Ogua, H. Mori, S. Hiraga, M. Yamasaki, and K. Nagai. 1989. Purification and characterization of the sopB gene product which is responsible for stable maintenance of mini-F plasmid. Mol. Gen. Genet. 218:431–436.
324. Weaver, K. E., D. B. Clewell, and F. An. 1993. Identification, characterization, and nucleotide sequence of a region of Enterococcus faecalis pheromone-responsive plasmid pAD1 capable of autonomous replication. J. Bacteriol. 175:1900–1909.
325. Weaver, K. E., and D. J. Tritle. 1994. Identification and characterization of an Enterococcus faecalis plasmid pAD1-encoded stability determinant which produces two small RNA molecules necessary for its function. Plasmid 32:168–181.
326. Wickner, S., J. Hoskins, and K. McKenney. 1991. Monomerization of RepA dimers by heat shock proteins activates binding to DNA replication origin. Proc. Natl. Acad. Sci. USA 88:7903–7907.
327. Williams, D. R., and C. M. Thomas. 1992. Active partitioning of bacterial plasmids. J. Gen. Microbiol. 138:1–16.
328. Wirth, R., A. Friesenegger, and T. Horaud. 1992. Identification of new sex pheromone plasmids in Enterococcus faecalis. Mol. Gen. Genet. 233:157–160.
329. Womble, D., and R. Rownd. 1988. Genetic and physical map of plasmid NR1: comparison with other IncFII antibiotic resistance plasmids. Microbiol. Rev. 52:433–451.
330. Womble, D. D., and R. H. Rownd. 1986. Regulation of lambda dv plasmid DNA replication. A quantitative model for control of plasmid lambda dv replication in the bacterial cell division cycle. J. Mol. Biol. 191:367–382.
331. Wu, F., I. Levchenko, and M. Filutowicz. 1994. Binding of DnaA protein to a replication enhancer counteracts the inhibition of plasmid R6K γ origin replication mediated by elevated levels of R6K π protein. J. Bacteriol. 176:6795–801.
332. Xia, G., D. Manen, T. Goebel, P. Linder, G. Churchward, and L. Caro. 1991. A copy-number mutant of plasmid pSC101. Mol. Microbiol. 5:631–640.
333. Xu, F., S. Lin-Chao, and S. N. Cohen. 1993. The Escherichia coli pcnB gene promotes adenylylation of antisense RNAI of ColE1-type plasmids in vivo and degradation of RNAI decay intermediates. Proc. Natl. Acad. Sci. USA 90:6756–6760.
334. Yamaguchi, K., and Y. Masamune. 1985. Autogenous reguation of synthesis of the replication protein in plasmid pSC101. Mol. Gen. Genet. 200:362–367.
335. Yamaguchi, K., and M. Yamaguchi. 1984. The replication origin of pSC101: the nucleotide sequence and replication functions of the ori region. Gene 29:211–219.
336. Yasueda, H., T. Horii, and T. Itoh. 1989. Structural and functional organization of ColE2 and ColE3 replicons. Mol. Gen. Genet. 215:209–216.
337. Yasukawa, H., T. Hase, A. Sakai, and Y. Masamune. 1991. Rolling-circle replication of plasmid pKYM isolated from a Gram-negative bacterium. Proc. Natl. Acad. Sci. USA 88:10282–10286.
338. Yoshikawa, H., and N. Ogasawara. 1991. Structure and function of DnaA and the DnaA-box in eubacteria: evolutionary relationships of bacterial replication origins. Mol. Microbiol. 5:2589–2597.
339. Zeng, H., T. Hayashi, and Y. Terawaki. 1990. Site-directed mutations in the repA C-terminal region of plasmid Rts1: pleiotropic effects on the replication and autorepressor functions. J. Bacteriol. 172:2535–2540.
Chapter122
