Mosaic plasmids and mosaic replicons: evolutionary lessons from the analysis of genetic diversity in IncFII-related replicons

A. Mark Osborna,1, Fernanda M. da Silva Tatley2,4, Lafras M. Steyn3, Roger W. Pickup5 and Jon R. Saunders1

School of Biological Sciences, Life Science Building, The University of Liverpool, PO Box 147, Liverpool L69 7ZB, UK1
Department of Medical Biochemistry2 and Department of Medical Microbiology3, Medical School, University of Cape Town, Observatory 7925, Cape Town, South Africa
Department of Biochemistry, University of Otago, Dunedin, New Zealand4
Institute of Freshwater Ecology, Windermere Laboratories, The Ferry House, Ambleside, Cumbria LA22 0LP, UK5

Author for correspondence: A. Mark Osborn. Tel: +44 1206 87 3763. Fax: +44 1206 87 2592. e-mail: osborn{at}essex.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The alpha replicons of the multi-replicon plasmids pGSH500 and pLV1402 have been characterized by DNA sequence analysis. Analysis of the DNA sequence of a 3672 bp HindIII fragment from pFDT100, which contains the pGSH500 alpha replicon, revealed similarity to a number of replicons belonging to, or related to, those of the IncFII family. The replicon region contains copA, tapA, repA and oriR, and replication initiation and termination sites are related to those from the IncFII replicon of R1. A copB gene was found to lie upstream of the HindIII site in the parental plasmid pGSH500. Downstream of oriR, a 707 bp region shows 72·6% identity to a region of the Escherichia coli chromosome at 43·3', suggesting this region of pGSH500 may have been incorporated into the plasmid during a past chromosomal recombination event. Oligonucleotide primers homologous to consensus regions in the copB and repA genes, and the oriR regions from a number of IncFII-related replicons were used to amplify replication regions from pLV1402. Analysis of the amplified regions has shown the presence of copB, copA, tapA and repA genes. Phylogenetic analysis of Rep protein sequences from the RepFIIA family of antisense-control-regulated replicons revealed the presence of three distinct subgroups of Rep proteins. Comparative analysis of DNA and protein sequences from members of the RepFIIA family provides evidence supporting the roles of both non-selective divergence in co-integrate (multi-replicon) plasmids and Chi-mediated-recombination in replicon evolution, and in particular, that such processes may have been widespread in the evolution of the RepFIIA family.

Keywords: plasmid, incompatibility, evolution, replicon, pGSH500

The EMBL accession numbers for the sequences reported in this paper are AJ009980 (pGSH500 alpha replicon) and AJ009981 (pLV1402 alpha replicon).

a Present address: Department of Biological Sciences, University of Essex, Wivenhoe Park, Colchester CO4 3SQ, UK.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Plasmid incompatibility is defined as the failure of two co-resident plasmids to be stably inherited in the absence of external selection (Novick et al., 1976 ) and forms the basis of a system to classify bacterial plasmids, especially amongst those from the Enterobacteriaceae (Datta & Hedges, 1972 ). The subsequent development of a series of probes specific to the replicons from a number of incompatibility (Inc) groups has allowed the rapid assignment of a replicon to a particular Inc group (Couturier et al., 1988 ).

The replicons from plasmids of many of the Inc groups have been extensively characterized, in terms both of DNA and protein structure, and of the mechanism of their replication and replication control. The replicons from plasmids belonging to IncB, IncFII, IncFIC, IncI1, IncK, IncL/M and IncZ have been shown to share a common replication control mechanism, namely antisense control, and to share some homology between both DNA and protein sequences (Athanasopoulos et al., 1995 ; Hama et al., 1990 ; Praszkier et al., 1991 ; Saadi et al., 1987 ). In the IncFII replicon of R1 an antisense RNA molecule (CopA) inhibits synthesis of the replication protein (RepA) by binding to the leader region of the repA mRNA (CopT). RepA synthesis is dependent upon translation of a short leader peptide (TapA) that is not expressed when CopA binds to CopT, thereby preventing translation of RepA and consequently preventing replication of the plasmid (Blomberg et al., 1992 ). Replication of R1 is unidirectional and is initiated by the binding of RepA to the minimal replication origin (oriR) (Diaz & Staudenbauer, 1982 ) with leading strand DNA synthesis starting about 380 bp downstream of oriR (Bernander et al., 1992 ; Masai & Arai, 1989 ).

A number of plasmids possess more than one replicon, e.g. the F plasmid (Lane & Gardner, 1979 ; Saadi et al., 1984 ), the Streptococcus plasmid pAM{alpha}1 (Perkins & Youngman, 1983 ) and the Bacillus plasmid pTB19 (Imanaka et al., 1984 ). More recently, the moderately promiscuous plasmid pGSH500 from a South African clinical isolate of Klebsiella pneumoniae has been found to carry two replicons: alpha (cloned as pFDT100) and beta (cloned as pFDT200). The latter displays a mosaic character and is related to, but compatible with, both the IncN replicon of pCU1 and the ori-2 of F (da Silva Tatley & Steyn, 1993 ). Two mercury-resistant environmental isolates, AH14 and AH16, from freshwater sediments of Windermere, Cumbria, UK, have subsequently been identified that both carry a plasmid (pLV1402 and pLV1403, respectively) with sequences homologous to both alpha and beta replicons. Analysis of partial sequences from the beta replicons of these plasmids showed them to be virtually identical (99·2% identity across 910 bp and 98·6% identity across 888 bp, respectively) to the beta replicon from pGSH500 (Osborn et al., 2000 ).

In this paper we describe the nucleotide sequences of the alpha replicons from pGSH500 and pLV1402, their relationship to each other and their relationship to a group of replicons belonging to, or related to, those of the IncFII family. In addition, we describe the relationship of these IncFII-related replicons to those from other antisense-control-regulated replicons and discuss the mechanisms involved in the evolution of this broader family.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, plasmids and oligonucleotides.
Escherichia coli strains used were XL-1 Blue (supE44 hsdR17 recA1 endA1 gyrA96 thi-1 relA1 lac {F'[proAB+ lacIq lacZ{Delta}M15 Tn10 (tetR)]}; Stratagene) and GW125a (recA thr-1 leu-6 thi-1 lacY1 galK2 ara-14 xyl-5 mtl1-proA2 his5 argE3 str-31 polA1) (Dorrington & Rawlings, 1989 ). XL-1 Blue was used for all DNA manipulations. GW125a was used to assess function of replicon subclones. The plasmid-carrying environmental isolate AH14 (Osborn et al., 2000 ) carries resistance to HgCl2. Plasmids (Table 1) were propagated in the E. coli strains above. Oligonucleotides used in this study were as follows: M13 reverse primer (RP), 5'-AAC AGC TAT GAC CAT G-3'; M13 -20 primer, 5'-GTA AAA CGA CGG CCA GT-3'; CA1, 5'-ATG TCG CA[C/G] A[C/T][A/C/T] GAA AAT GC-3' (position 442); RA1, 5'-CAA AAC G[A/G]G A[A/C/G/T]G TGA A[G/C]C C-3' (position 1129); RA2, 5'-GG[C/G] TTC AC[A/C/G/T] TC[C/T] CGT TTT GA-3' (position 1111); OR1, 5'-CCT TGC AGT T[A/T][A/T] [A/C/T]TG TG[A/G] [A/G]TA A-3' (position 2048). Nucleotide positions are according to those of Ryder et al. (1982) for the sequence of R1 (RSC13) (accession no. V00351).


View this table:
[in this window]
[in a new window]
 
Table 1. Plasmids and constructs

 
Cloning and sequencing of the pFDT100 replicon.
DNA fragments for sequencing were subcloned into pBluescript KS(+) (Table 1 using standard protocols (Sambrook et al., 1989 ). Template DNA was prepared using Qiagen plasmid midi kits. Subclones were sequenced on both strands with the M13 -20 and RP primers together with a number of custom-synthesized primers.

PCR amplification and determination of DNA sequence of IncFII-like sequences.
Primers were designed to consensus sequences in the copB and repA genes, and the oriR region of replicons from plasmids pFM82139 (Rodríguez-Peña et al., 1997 ), pYVe439-80 (Vanooteghem & Cornelis, 1990 ), pSU316 (Lopez et al., 1989 ), NR1 (Womble et al., 1985 ), R100 and R1 (Ryder et al., 1982 ). Template DNA was prepared as described previously (Osborn et al., 1993 ). PCR products were amplified using Taq DNA polymerase (2·5 U) with the primer pairs: CA1 and RA1, RA2 and OR1, and CA1 and OR1. Reactions were performed for 30 cycles [94 °C for 1 min, 53 °C (CA1/RA1) or 54 °C (RA2/OR1 and CA1/OR1) for 1 min, 72 °C for 2 min] followed by 1 cycle of 72 °C for 10 min. Primer and dNTP concentrations were 30 pmol and 50 µM, respectively. PCR products were visualized following electrophoresis on 0·8% agarose TAE gels containing ethidium bromide (1 µg ml-1). PCR products were sequenced as described above using custom synthesized primers.

Computer analysis.
Sequences were analysed using the GCG set of programs (version 8, August 1994, Genetics Computer Group, Madison, Wisconsin, USA). The structures of the major hairpin loops of RNA molecules were predicted by the method of Zuker (1989) , with free energy values calculated according to the energy parameters of Walter et al. (1994) . RepA protein sequences were aligned using CLUSTAL W (Thompson et al., 1994 ). A neighbour-joining tree (Saitou & Nei, 1987 ) was constructed showing bootstrap values (1000 replicates) (Felsenstein, 1985 ).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The pGSH500 and pLV1402 alpha replicons are related to the IncFII replicon family
Analysis of the DNA sequence of the 3672 bp HindIII fragment from pFDT100 (containing the alpha replicon of pGSH500) identified similarity to a number of plasmid replicons belonging to, or related to, those of the IncFII family. On the basis of similarity to these replicons, the following genetic features were identified: copA, tapA, repA and oriR (Fig. 1). The clone pFDT100 does not contain a copB gene. However, PCR amplification from the parental plasmid pGSH500, using primers CA1 and RA1, showed the presence of a copB gene starting upstream of the HindIII site which defines the 5' end of pFDT100 (Fig. 1).



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 1. Map of the alpha replicons of plasmids pGSH500 and pLV1402, and of the pGSH500 subclones pFDT100, pFDT101 and pFDT102. Restriction endonuclease sites are as follows: H, HindIII, S, SstI and K, KpnI. Boxes A–C show alignment with the corresponding regions of the IncFII replicon of R1. Map coordinates of R1 are according to Masai & Arai (1989) . Box A: start of the oriR region showing DnaA box (boxed) and imperfect palindrome (bold) involved in RepA-binding. Box B: similarity between Ter sites. Identical bases are shown in bold. Box C: proposed in vivo (bold) and in vitro (underlined) start sites for leading strand DNA synthesis in R1.

 
Plasmid pLV1402 from the mercury-resistant strain AH14 has been shown to carry sequences homologous to the pFDT100 replicon by hybridization analysis (Osborn et al., 2000 ). Oligonucleotide primers CA1, RA1, RA2 and OR1, based on consensus sequences in the copB, repA and oriR regions in a number of IncFII and IncFII-related replicons, were used to amplify replicon regions from AH14(pLV1402). Analysis of the sequence of the amplified region from pLV1402 has shown the presence of copB, copA, tapA and repA sequences (Fig. 1).

An extended family of RepA proteins from IncFII-related replicons
The predicted sequence of the 286 aa RepA protein from pFDT100 was most closely related to those from pLV1402 (86·4% identity), the Salmonella enteritidis plasmid pFM82139 (85·8% identity; Rodríguez-Peña et al., 1997 ) and the virulence plasmids pYVe439-80 from Yersinia enterocolitica (Vanooteghem & Cornelis, 1990 ) and pCD1 (Hu et al., 1998 ) from Yersinia pestis (82·5% identity). TFASTA searches using the pFDT100 RepA sequence identified further similarity with a number of RepA proteins from both IncFII replicons and those from more distantly related replicons, such as the IncZ replicon on pIE545 (Praszkier et al., 1991 ) and the RepA proteins from a number of plasmids isolated from Buchnera sp. (Bracho et al., 1995 ; van Ham et al., 1997 ; Silva et al., 1998 ). TFASTA searches using RepA proteins from these Buchnera plasmids further extended this family of RepA proteins to include those from replicons of the IncB, FIC, I1, K and L/M groups. A neighbour-joining tree (Saitou & Nei, 1987 ) of RepA sequences was constructed (Fig. 2) and revealed the presence of three distinct subgroups of RepA proteins from the extended family of RepFIIA antisense-control-regulated replicons (Saadi et al., 1987 ). On this basis, we now propose that the extended RepFIIA family should describe and include the replicons of the I complex (IncI1, IncB/O, IncK and IncL/M).



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 2. Neighbour-joining dendrogram derived from the deduced protein sequences encoded by repA genes from IncFII and IncFII-related replicons. Bootstrap values (1000 replicates) are shown adjacent to the node being considered.

 
Antisense control of pGSH500 and pLV1402 alpha replicons
Replication of IncFII replicons is under the control of a small antisense RNA encoded by copA (Wagner & Simons, 1994 ). CopA forms a stem–loop structure with a stretch of 6 bases (CGCCAA) forming the loop that binds to a region of the RepA mRNA called CopT, to form a duplex that prevents translation of the TapA reading frame. This permits the formation of a stem–loop structure that sequesters the repA ribosome-binding site and blocks translation of the RepA reading frame (Blomberg et al., 1992 ). Comparison of the predicted secondary structures of CopA molecules from pFDT100 and pLV1402 with those from the closely related pFM82139 and pYVe439-80 replicons, and from the archetypal IncFII replicon R1 (Fig. 3) showed conservation of a 6 base sequence, CGCCAA composing the loop of CopA. Furthermore, a short ORF of 25 aa, upstream of repA in both pFDT100 and pLV1402, bears high similarity to the tapA gene proposed for R1 (Blomberg et al., 1992 ) and for pFM82139 (Rodríguez-Peña et al., 1997 ). This suggests that in both replicons, RepA synthesis is similarly regulated by CopA through inhibition of leader peptide translation. Mutations in both stem and loop regions have been shown to result in changes in incompatibility in both IncFII (Brady et al., 1983 ) and the IncI complex plasmids (Nikoletti et al., 1988 ) and variation in these regions is believed to be the basis for incompatibility in antisense-control-regulated replicons. Comparison of the predicted stem sequences of the five CopA molecules clearly shows variation in this region (Fig. 3).



View larger version (24K):
[in this window]
[in a new window]
 
Fig. 3. Predicted structures of the major hairpins formed by CopA from pFDT100, pFM82139, pLV1402, pYVe439-80 and R1, as predicted by the method of Zuker (1989) . Free energy was calculated according to the energy parameters of Walter et al. (1994) . Bases proposed to interact with CopT are shown in bold.

 
Replication origin, initiation and termination sites in pFDT100
Alignment of the oriR regions of pFDT100 and R1 (Fig. 1, box A) reveals strong conservation in the DnaA box and the imperfect palindrome that defines the 5' end of the RepA-binding site in R1 (Giraldo & Diaz, 1992 ). A second imperfect palindrome in pFDT100 (5'-ctTTTAAAca-3') is identical to that which defines the 3' end of the R1 RepA-binding site and overlaps the first of three A+T-rich repeat sequences believed to be involved in DNA melting at oriR. Additional sequence conservation is seen between the TerR sites of R1 and pFDT100 (Fig. 1, box B), suggesting that in pFDT100, replication termination and the prevention of multimerization is controlled in a similar manner to that of R1 (Krabbe et al., 1997 ).

The site of leading strand synthesis in plasmid R1 has been located at a point about 380 bp downstream of the minimal oriR in in vivo (Bernander et al., 1992 ) and in vitro (Masai & Arai, 1989 ) studies. Comparison of the pFDT100 sequence with that of R1 in this region reveals a single base pair difference between the proposed in vitro start sites, whilst the proposed in vivo start sites are identical (Fig. 1, box C).

To further define the limit of the minimal alpha replicon of pGSH500, the subclones pFDT101 and pFDT102 (Fig. 1) were transformed into the E. coli polA1 mutant strain GW125a (Dorrington & Rawlings, 1989 ). Replication from the DNA polymerase I-dependent ColE1 replicon from the vector regions of pFDT101 and pFDT102 is prevented in this strain. Both constructs were able to replicate in GW125a, showing that both subclones of pFDT100 still carry functional replicons. The SstI site that defines the 3' end of pFDT101 lies upstream of both the TerR sequences and the proposed start site for leading strand synthesis in pFDT100, showing that these regions are not required for replication. Similarly, sequences equivalent to the R1 and R100 initiation sites (Bernander et al., 1992 ) are located outside of the minimal origin. Interestingly, in pFDT101, a region of vector sequence (3'-agCCCCgACaCcC-5') which lies 336 bp downstream of oriR bears some similarity to the pFDT100 start site (see Fig. 1, box C). This region may function as the start site for leading strand synthesis in pFDT101, and perhaps significantly, the CCCC motif, which has been shown to be the in vivo start site in R1, is conserved in this region.

pGSH500 carries sequences homologous to the E. coli chromosome
FASTA searches of the region downstream of the alpha replicon revealed a region of 707 bp which showed 72·6% identity to a region on the E. coli chromosome (43·3'). Raha et al. (1993) had previously proposed that this region is non-coding in both the E. coli and Salmonella typhimurium chromosomes. However, translation of the pFDT100 DNA sequence in this region revealed two putative ORFs, ORF1 (138 aa) and ORF2 (171 aa). TFASTA searches revealed homology (80·3% identity across 132 aa) between ORF1 and the 217 aa gene product of the hypothetical E. coli gene yedG (Blattner et al., 1997 ). Further homology to yedG was found upstream of ORF1, suggesting ORF1 is a truncated version of yedG. In the E. coli chromosome the initiation codon of yedG is found at a position that would lie upstream of the E. coli-homologous region of pLV1402.

Role of Chi sites in the formation of mosaic structures in antisense-control-regulated replicons
The replicons from the Inc groups B, FII, FIC, I1, K, L/M and Z encode related replication proteins (Fig. 2) and their replication is controlled by a common mechanism involving antisense control by a small RNA molecule. Comparison of the DNA sequences of replicons from these Inc groups reveals that a number of these replicons are mosaic in structure (Fig. 4). Alignment of these replicons suggests that the IncFIC replicon consists of a rep gene that is closely related to those from the IncB, K and I1 replicons, whilst the upstream inc region containing copA is closely related to those from IncFII replicons. Conversely, the IncZ replicon is a mosaic structure consisting of a rep gene closely related to that of IncFII replicons, whilst its inc region is closely related to that of the IncB, K and I1 replicons. The breakpoint of these mosaic structures is found where tapA overlaps with the rep gene.



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 4. Mosaic structures of the family of antisense-control-regulated replicons related to IncFII replicons and sequences of their potential Chi sites. Relationships between replicons are shown by shading as follows: black, IncB/I/K-like sequences; grey, IncFII-like sequences; white, IncL/M sequences. The positions of genes involved in replication in R1 and IncFII plasmids are shown above. In the other replicons a small ORF encoding a leader peptide, whose expression is similarly required for Rep protein expression, is found at a position similar to tapA. In ColIb-P9 the gene encoding the replication protein is called repZ. Chi sites are shown in bold with variation from the Chi sequence shown in lower case. The start codons of the genes encoding the replication protein are boxed.

 
Increased levels of recombination have been found to be associated with the presence of Chi sites, which consist of the nucleotide octamer 5'-GCTGGTGG-3' (Smith et al., 1981 ). Chi sites are recognized by the E. coli enzyme RecBCD which promotes homologous recombination via the RecBCD pathway (Smith, 1987 ; Taylor & Smith, 1995 ). Examination of the replicons from these Inc groups reveals the presence of Chi-like sequences at the start of the genes encoding replication proteins from all six replicons (Fig. 4), suggesting that the mosaic IncFIC and IncZ replicons are the product of recombination events mediated by the presence of Chi sites. In the IncFII-related replicons from plasmids pGSH500, pLV1402, pFM82139 and pYVe439-80 only sequences with very limited identity to Chi were found at the start of repA. In contrast, Chi-related sequences could be identified in the replicons of pSU316 (IncFIII; Lopez et al., 1989 ), pSU212 (IncFVI; Lopez et al., 1991 ) and ColV2-K94 (Weber & Palchaudhuri et al., 1986 ) in this region.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The alpha replicons from plasmids pGSH500 and pLV1402 have been identified as members of a broad family related to the IncFII replicons of R100 and R1. The predicted sequences of the Rep proteins from pGSH500 and pLV1402 showed these to be most closely related to those from plasmids pFM82139 and pYVe439-80. Significant sequence conservation is found between pFDT100 and R1 in the replication origin and the replication initiation and termination sites (Fig. 1). Whilst providing additional evidence for replication proceeding by a similar mechanism to R1, the sequence alignments also reveal a number of conserved bases in all three regions. In particular, the conservation of the CCCC motif, which is the in vivo replication start site in R1, suggests this sequence may have an equivalent role in pFDT100.

It is clear from DNA and protein sequence comparisons that the pYVe439-80, pCD1 and pFM82139 replicons and the alpha replicons from pGSH500 and pLV1402 are related to IncFII plasmids and their close relatives, the IncFIII and IncFVI replicons. However, it is also apparent that they are outliers of the IncFII family and it is likely that many other replicons exist in plasmids from members of the Enterobacteriaceae that will be related to this plasmid family, yet similarly divergent. The divergence of the replicons from pYVe439-80 and pCD1, pFM82139 and pGSH500 may also be a consequence of these plasmids being found in hosts other than E. coli (i.e. species of Yersinia, Salmonella and Klebsiella, respectively) and may represent adaptation of such plasmids to their hosts.

Compatibility between R100 and pYVe439-80 results from sequence variation in the stem region of CopA (Vanooteghem & Cornelis, 1990 ). Similarly, compatibility between members of the extended RepFIIA family is also a consequence of sequence variation in both loop and upper stem sequences in CopA (Athanasopoulos et al., 1995 ). Given the variation between stem sequences of CopA, from the other IncFII-related replicons shown in Fig. 3, including those of the pGSH500 and pLV1402 alpha replicons and from pFM82139, it is likely that some or all of these replicons may be compatible with each other. The sequence CGCCAA that forms the proposed loop in CopA is conserved across the majority of replicons from the extended RepFIIA family. Furthermore, some conservation of this loop sequence is seen across a range of other antisense-controlled replicons. This is most notable in ColE2 replicons (CCGCCAA), but also to some extent in the RNA I molecules of ColE1 (CUACCAA) and pT181 (UGACCAA) (Hiraga et al., 1994 ), suggesting that this sequence and consequently the replication control regions themselves may have a common evolutionary origin.

The mosaic nature of plasmids is broadly recognized, with a range of mobile genetic elements, e.g. transposons, integrons, gene cassettes and IS elements, commonly found inserted into the plasmid backbone (Hall & Vockler, 1987 ; Pansegrau et al., 1994 ). In pGSH500, the presence of an E. coli-homologous region, immediately downstream of the alpha replicon in pGSH500, provides evidence for the acquisition of chromosomal regions by plasmids, possibly during recombination events.

The beta replicons of pGSH500 and pLV1402 are very closely related (99·2% identity at the DNA level; Osborn et al., 2000 ). Determination of plasmid copy number for pGSH500 and its derived minireplicons, alpha (pFDT100) and beta (pFDT200), indicates that replication of pGSH500 proceeds from the beta replicon (da Silva Tatley & Steyn, 1993 ). Given the divergence between the alpha replicons of pGSH500 and pLV1402, it is attractive to propose that these replicons have been free to diverge, perhaps during recombination events with other IncFII-related replicons. Such a hypothesis is in agreement with the model of Sykora (1992) for non-selective divergence of plasmid replicons in co-integrate (i.e. multi-replicon) plasmids, where one replicon is strongly conserved due to selective pressure (i.e. the beta replicon), whilst the second replicon (the alpha replicon) is free to diverge.

Comparison of the IncFIC replicon P307 with those of the IncFII replicons of R1 and R100 has shown the presence of a number of homologous and non-homologous regions, which has been proposed to result from recombination events (Saadi et al., 1987 ). Our comparison of the DNA sequences of members of the extended RepFIIA family of antisense-controlled replicons has revealed the presence of mosaic replicons. The presence of Chi-like sequences at the junction of these mosaics suggests that Chi-related recombination events have played an important role in the evolution of this replicon family. Smith et al. (1981) have demonstrated that a number of single base pair mutations within the Chi sequence can abolish its activity. However, three sequences in wild-type {lambda} phage that differ at the first (G to A), second (C to T) and fourth (G to A) positions have partial Chi activity (6, 11 and 38%, respectively; Cheng & Smith, 1984 ). The imperfect Chi sequences at the start of the repA genes may similarly have partial Chi activity and have been involved in recombination events between replicons. Alternatively, the Chi-like sequences seen in these replicons may have since diverged from the archetypal E. coli Chi sequence. The presence of Chi-related sequences in replicons from other Inc groups related to IncFII, e.g. IncFIII and IncFVI, would suggest that there is great potential for the evolution of new replicons by Chi-mediated recombination events. Chi-related sequences have been similarly implicated in recombination events in the formation of mosaic RepFIB-related repA genes in plasmids from natural populations of E. coli (Boyd et al., 1996 ). Mosaic regions have also been identified in the ColE2 family of antisense-control-regulated replicons, with one breakpoint between segments again found between the inc and rep regions (Hiraga et al., 1994 ). However, no Chi sites could be identified in this region, excluding the possibility of Chi-mediated homologous recombination in mosaic formation for the ColE2 replicons.

The existence of mosaic replicons causes additional complications for the classification of bacterial plasmids and for attempts to assess the evolutionary relationships between plasmids and their replicons. Plasmids are often classified on the basis of their incompatibility relationships and a series of replicon (rep) probes have been developed for a number of incompatibility groups to facilitate the classification of new plasmids by hybridization analysis. However, the individual rep probes of the RepFIIA family of antisense-controlled replicons show considerable cross-hybridization (Couturier et al., 1988 ), which limits their effective use in demonstrating membership of this broad family. Classification of plasmids of this group is further complicated as incompatibility relationships between these replicons can result from single base pair changes. Consequently, any attempt to characterize new members of this family should be made with these considerations in mind. Moreover, descriptions of replicon evolution should in some cases consider inc and rep regions as separate evolutionary units. The problems of plasmid classification are further complicated by the occurrence of multi-replicon elements where replication may proceed from different replicons on the plasmid in different hosts. We suggest that replicons from such plasmids are themselves treated and classified as separately evolving systems.

In conclusion, comparative analysis of members of the extended RepFIIA family has provided evidence supporting the roles both of non-selective divergence in multi-replicon plasmids and of Chi-mediated homologous recombination in the evolution of plasmid replicons. Finally, this work underlines the considerable value of such systematic analyses of the wealth of hitherto under-utilized sequence data that is available to further our understanding of the evolution of bacterial plasmids.


   ACKNOWLEDGEMENTS
 
This work was funded by the Natural Environmental Research Council, UK, and by the Medical Research Council of South Africa. This work benefited from the use of the SEQNET facility at Daresbury, UK. The authors would also like to thank Angela Rosin for automated DNA sequencing.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Athanasopoulos, V., Praszkier, J. & Pittard, A. J. (1995). The replication of an IncL/M plasmid is subject to antisense control. J Bacteriol 177, 4730-4741.[Abstract]

Bernander, R., Krabbe, M. & Nordström, K. (1992). Mapping of the in vivo start site for leading strand DNA synthesis in plasmid R1. EMBO J 11, 4481-4487.[Abstract]

Blattner, F. R., Plunkett, G., Bloch, C. A. & 14 other authors (1997). The complete genome sequence of Escherichia coli K-12. Science 277, 1453–1577.[Abstract/Free Full Text]

Blomberg, P., Nordström, K. & Wagner, E. G. H. (1992). Replication control of plasmid R1: RepA synthesis is regulated by CopA RNA through inhibition of leader peptide translation. EMBO J 11, 2675-2683.[Abstract]

Boyd, E. F., Hill, C. W., Rich, S. M. & Hartl, D. L. (1996). Mosaic structure of plasmids from natural populations of Escherichia coli. Genetics 143, 1091-1100.[Abstract/Free Full Text]

Bracho, A. M., Martínez-Torres, D., Moya, A. & Latorre, A. (1995). Discovery and molecular characterization of a plasmid localized in Buchnera sp. bacterial endosymbiont of the aphid Rhopalosiphum padi. J Mol Evol 41, 67-73.[Medline]

Brady, G., Frey, J., Danbara, H. & Timmis, K. N. (1983). Replication control mechanisms of plasmid R6-5 and their effects on interactions of the RNA-I control element with its target. J Bacteriol 154, 429-436.[Medline]

Cheng, K. C. & Smith, K. R. (1984). Recombinational hotspot activity of Chi-like sequences. J Mol Biol 180, 371-377.[Medline]

Couturier, M., Bex, F., Bergquist, P. L. & Maas, W. K. (1988). Identification and classification of bacterial plasmids. Microbiol Rev 52, 375-395.

Datta, N. & Hedges, R. W. (1972). R factors identified in Paris, some conferring gentamicin resistance, constitute a new compatibility group. Ann Inst Pasteur (Paris) 123, 222-223.

Diaz, R. & Staudenbauer, W. L. (1982). Origin and direction of mini-R1 plasmid DNA replication in cell extracts of Escherichia coli. J Bacteriol 150, 1077-1084.[Medline]

Dorrington, R. A. & Rawlings, D. E. (1989). Identification and sequence of the basic replication region of a broad-host-range plasmid isolated from Thiobacillus ferrooxidans. J Bacteriol 171, 2735-2739.[Medline]

Felsenstein, J. (1985). Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 783-791.

Giraldo, R. & Diaz, R. (1992). Differential binding of wild-type and a mutant RepA protein to oriR sequence suggests a model for the initiation of plasmid R1 replication. J Mol Biol 228, 787-802.[Medline]

Hall, R. M. & Vockler, C. (1987). The region of the IncN plasmid R46 coding for resistance to ß-lactam antibiotics, streptomycin/spectinomycin and sulphonamides is closely related to antibiotic resistance segments found in IncW plasmids and in Tn21-like transposons. Nucleic Acids Res 15, 7491-7501.[Abstract]

van Ham, R. C. H. J., Moya, A. & Latorre, A. (1997). Putative evolutionary origin of plasmids carrying the genes involved in leucine biosynthesis in Buchnera aphidicola (endosymbiont of aphids). J Bacteriol 179, 4768-4777.[Abstract]

Hama, C., Takizawa, T., Morikawa, H., Urasaki, Y. & Mizobuchi, K. (1990). Organization of the replication control region of plasmid ColIb-P9. J Bacteriol 172, 1983-1991.[Medline]

Hiraga, S. I., Sugiyama, T. & Itoh, T. (1994). Comparative analysis of the replicon regions of eleven ColE2-related plasmids. J Bacteriol 176, 7233-7243.[Abstract]

Hu, P., Elliott, J., McCready, P., Skowronski, E., Garnes, J., Kobayashi, A., Brubaker, R. R. & Garcia, E. (1998). Structural organization of virulence-associated plasmids of Yersinia pestis. J Bacteriol 180, 5192-5202.[Abstract/Free Full Text]

Imanaka, T., Ano, T., Fujii, M. & Aiba, S. (1984). Two replication determinants of an antibiotic-resistance plasmid, pTB19 from a thermophilic Bacillus. J Gen Microbiol 130, 1399-1408.[Medline]

Krabbe, M., Zabielski, J., Bernander, R. & Nordström, K. (1997). Inactivation of the replication-termination system affects the replication mode and causes unstable maintenance of plasmid R1. Mol Microbiol 24, 723-735.[Medline]

Lane, D. & Gardner, R. C. (1979). Second EcoRI fragment of F capable of self-replication. J Bacteriol 139, 141-151.[Medline]

Lopez, J., Crespo, P., Rodriguez, J. C., Andres, I. & Ortiz, J. M. (1989). Analysis of IncF plasmids evolution: nucleotide sequence of an IncFIII replication region. Gene 78, 183-187.[Medline]

Lopez, J., Delgado, D., Andrés, I., Ortiz, J. M. & Rodríguez, J. C. (1991). Isolation and evolutionary analysis of a RepFVIB replicon of the plasmid pSU212. J Gen Microbiol 137, 1093-1099.[Medline]

Masai, H. & Arai, K. I. (1989). Leading strand synthesis of R1 plasmid replication in vitro is primed by primase alone at a specific site downstream of oriR.J Biol Chem 264, 8082-8090.[Abstract/Free Full Text]

Nikoletti, S., Bird, P., Praszkier, J. & Pittard, J. (1988). Analysis of the incompatibility determinants of I-complex plasmids. J Bacteriol 170, 1311-1318.[Medline]

Novick, R. P., Clowes, R. C., Cohen, S. N., Curtiss, R.III, Datta, N. & Falkow, S. (1976). Uniform nomenclature for bacterial plasmids: a proposal. Bacteriol Rev 40, 168-189.[Medline]

Osborn, A. M., Bruce, K. D., Strike, P. & Ritchie, D. A. (1993). Polymerase chain reaction–restriction fragment length polymorphism analysis shows divergence among mer determinants from Gram-negative soil bacteria indistinguishable by DNA–DNA hybridisation.Appl Environ Microbiol 59, 4024-4030.[Abstract]

Osborn, A. M., Pickup, R. W. & Saunders, J. R. (2000). Development and application of molecular tools in the study of IncN-related plasmids from lakewater sediments. FEMS Microb Lett 186, 203-208.[Medline]

Pansegrau, W., Lanka, E., Barth, P. T. & 7 other authors (1994). Complete nucleotide sequence of Birmingham IncP{alpha} plasmids. J Mol Biol 239, 623–663.[Medline]

Perkins, J. B. & Youngman, P. (1983). Streptococcus plasmid pAM{alpha}1 is a composite of two separable replicons, one of which is closely related to Bacillus plasmid pBC16. J Bacteriol 155, 607-615.[Medline]

Praszkier, J., Wei, T., Siemering, K. & Pittard, J. (1991). Comparative analysis of the replication regions of IncB, IncK and IncZ plasmids. J Bacteriol 173, 2393-2397.[Medline]

Raha, M., Kihara, M., Kawagishi, I. & MacNab, R. M. (1993). Organization of the Escherichia coli and Salmonella typhimurium chromosomes between flagellar regions IIIa and IIIb, including a large non-coding region. J Gen Microbiol 139, 1401-1407.[Medline]

Rodríguez-Peña, J. M., Buisán, M., Ibáñez, M. & Rotger, R. (1997). Genetic map of the virulence plasmid of Salmonella enteritidis and nucleotide sequence of its replicons. Gene 188, 53-61.[Medline]

Ryder, T. B., Davison, D. B., Rosen, J. I., Ohtsubo, E. & Ohtsubo, H. (1982). Analysis of plasmid genome evolution based on nucleotide-sequence comparison of two related plasmids of Escherichia coli. Gene 17, 299-310.[Medline]

Saadi, S., Maas, W. K. & Bergquist, P. L. (1984). RepFIC, a basic replicon of IncFI plasmids that has homology with a basic replicon of IncFII plasmids. Plasmid 12, 61-64.[Medline]

Saadi, S., Maas, W. K., Hill, D. F. & Bergquist, P. L. (1987). Nucleotide sequence analysis of RepFIC, a basic replicon present in IncFI plasmids P307 and F, and its relation to the RepA replicon of IncFII plasmids. J Bacteriol 169, 1836-1846.[Medline]

Saitou, N. & Nei, M. (1987). The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4, 406-425.[Abstract]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Silva, F. J., van Ham, R. C. H. J., Sabater, B. & Latorre, A. (1998). Structure and evolution of the leucine plasmids carried by the endosymbiont (Buchnera aphidicola) from aphids of the family Aphididae. FEMS Microb Lett 168, 43-49.[Medline]

da Silva-Tatley, F. M. & Steyn, L. M. (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.[Medline]

Smith, G. R. (1987). Mechanism and control of homologous recombination in Escherichia coli.Annu Rev Genet 21, 179-201.[Medline]

Smith, G. R., Kunes, S. M., Schultz, D. W., Taylor, A. & Triman, K. L. (1981). Structure of Chi hotspots of generalized recombination. Cell 24, 429-436.[Medline]

Sykora, P. (1992). Macroevolution of plasmids: a model for plasmid speciation. J Theor Biol 159, 53-65.[Medline]

Taylor, A. F. & Smith, G. R. (1995). Strand specificity of nicking of DNA at Chi sites by RecBCD enzyme. J Biol Chem 270, 24459-24467.[Abstract/Free Full Text]

Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994). CLUSTAL-W – Improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice.Nucleic Acids Res 22, 4673-4680.[Abstract]

Vanooteghem, J. C. & Cornelis, G. R. (1990). Structural and functional similarities between the replication region of the Yersinia virulence plasmid and the RepFIIA replicons. J Bacteriol 172, 3600-3608.[Medline]

Wagner, E. G. H. & Simons, R. W. (1994). Antisense RNA control in bacteria, phages and plasmids. Annu Rev Microbiol 48, 713-742.[Medline]

Walter, A. E., Turner, D. H., Kim, J., Lyttle, M. H., Muller, P., Matthews, D. H. & Zuker, M. (1994). Coaxial stacking of helixes enhances binding of oligoribonucleotides and improves predictions of RNA folding. Proc Natl Acad Sci USA 91, 9218-9222.[Abstract/Free Full Text]

Weber, P. C. & Palchaudhuri, S. (1986). Incompatibility repressor in a RepA-like replicon of the IncFI plasmid ColV2-K94. J Bacteriol 166, 1106-1112.[Medline]

Womble, D. D., Sampathkumar, P., Easton, A. M., Luckow, V. A. & Rownd, R. H. (1985). Transcription of the replication control region of the IncFII R-plasmid NR1 in vitro and in vivo. J Mol Biol 181, 395-410.[Medline]

Zuker, M. (1989). On finding all suboptimal foldings of an RNA molecule. Science 244, 48-52.[Medline]

Received 15 March 2000; revised 13 June 2000; accepted 19 June 2000.