Phylogeny of the replication regions of pPT23A-like plasmids from Pseudomonas syringae

Ane Sesma1, George W. Sundin2 and Jesús Murillo1

Instituto de Agrobiotecnología y Recursos Naturales, CSIC-UPNA, and Laboratorio de Patología Vegetal, Departamento de Producción Agraria, Universidad Pública de Navarra, 31006 Pamplona, Spain1
Department of Plant Pathology and Microbiology, Texas A&M University, College Station, TX 77843-2132, USA2

Author for correspondence: Jesús Murillo. Tel: +34 948 169133. Fax: +34 948 169732. e-mail: jesus{at}unavarra.es


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
It was previously shown that most Pseudomonas syringae strains contain one or more plasmids with cross-hybridizing replication regions and other areas of homology, and these plasmids were designated the pPT23A-like family. The majority of these plasmids encode genes conferring epiphytic fitness or resistance to antibacterial compounds and those investigated in this study are essential for pathogenicity or increased virulence. The phylogeny of 14 pPT23A-like plasmids from five P. syringae pathovars was studied by comparing a fragment of the sequence of their repA genes (encoding a replicase essential for replication). In the phylogenetic tree obtained, four groups (<=88·8% identity between their members) could be identified. The first group contained the plasmids from three P. syringae pv. tomato strains, a P. syringae pv. apii strain and five out of the seven P. syringae pv. syringae strains, with identity ranging between 88·8 and 100%. The clustering of the pv. syringae strains did not reflect host specialization or previously reported phylogenetic relationships. The second group contained the plasmids from two strains of pv. glycinea and pv. tomato (95·5% identity), and it also included the previously sequenced replicon of a pathogenicity plasmid from P. syringae pv. phaseolicola. The plasmids from the remaining two pv. syringae strains were distantly related to the other plasmid sequences. Hybridization experiments using different genes or transposable elements previously described as plasmid-borne in P. syringae, showed that the gene content of highly related plasmids could be dissimilar, suggesting the occurrence of major plasmid reorganizations. Additionally, the phylogeny of the different native plasmids did not always correlate with the phylogeny of their harbouring strains, as determined by the analysis of extragenic repetitive consensus (ERIC) and arbitrarily primed PCR (AP-PCR) products. Collectively, these results suggest that pPT23A-like plasmids were, in most cases, acquired early during evolution.

Keywords: pathogenicity and virulence plasmids, avirulence gene avrD, repeated DNA, rulAB, ultraviolet light resistance genes

Abbreviations: AP-PCR, arbitrarily primed PCR; ERIC, extragenic repetitive consensus

The EMBL accession numbers for the sequences reported in this paper are AJ276998–AJ277021.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The phytopathogenic bacterial species Pseudomonas syringae is genetically diverse and has traditionally been subdivided into groups of pathovars according to host range (Young et al., 1996 ). A recent analysis of DNA reassociation rates has allowed the separation of P. syringae into nine genomospecies which could contain one to several pathovars with non-overlapping host ranges (Gardan et al., 1999 ). Likewise, the results of other studies have indicated that the pathovar distinction based on host-range data does not always correlate well with other molecular, biochemical and physiological typing methods (Young et al., 1992 ; Saunier et al., 1996 ; Sawada et al., 1999 ).

Little is known about the specific genetic determinants that affect host range in P. syringae or other necro genic bacteria, although it is becoming clear that host specificity is defined by the coordinated action of ‘positive’ factors and the antagonistic action of ‘negative’ factors. The negative factors, or avirulence (avr) genes, are the best known and act by restricting the plant cultivars or species that can be infected by a given isolate (Vivian et al., 1997 ). The positive factors would include determinants that govern the ability to produce disease or increase virulence in a given plant host. In P. syringae, phytotoxins and growth regulators generally increase the aggressiveness of the bacterium towards a plant host. However, there is growing evidence to suggest that avr genes could also be the main positive factors determining virulence and/or pathogenicity (Kearney & Staskawicz, 1990 ; Swarup et al., 1991 ; Lorang et al., 1994 ; Ritter & Dangl, 1995 ; Yang et al., 1996 ; Jackson et al., 1999 ). For example, the ability of P. syringae pv. phaseolicola 1449B to infect beans and soybean is conferred by a plasmid-encoded pathogenicity island of about 30 kb which contains three avr genes and four other ORFs with the characteristics of avr genes (Jackson et al., 1999 ). One of the ORFs, designated virPphA, confers the ability to infect beans and to cause a hypersensitive response in certain soybean cultivars (Jackson et al., 1999 ). Avirulence gene avrPphF, also included within this pathogenicity island, determines the capacity to infect soybean and causes a specific gene-for-gene hypersensitive response on the bean cultivar Red Mexican (Tsiamis et al., 2000 ).

Many of the determinants involved in virulence and pathogenicity of P. syringae, including several that clearly influence host range, are encoded on native plasmids. Outstanding examples are the genes involved in the biosynthesis of the phytotoxin coronatine (Sato, 1988 ; Bender et al., 1999 ), auxins (Comai & Kosuge, 1980 ; Glickmann et al., 1998 ) and avr genes (Vivian et al., 1997 ). In some cases, related genes or gene clusters are conserved in unrelated pathovars; examples of this include the coronatine biosynthetic cluster which is present in five pathovars (Mitchell, 1982 ; Wiebe & Campbell, 1993 ; Cuppels & Ainsworth, 1995 ) and avrD sequences which have been detected within a wide pathovar range (Yucel et al., 1994 ). However, the relationships among different native plasmids from P. syringae have not been studied extensively and authors have reported both similarity and dissimilarity among plasmids from a given pathovar (Curiale & Mills, 1983 ; Denny, 1988 ; King, 1989 ; Sundin et al., 1994 ). The possibility of plasmid transfer within natural populations via conjugation must also be considered in terms of the introduction of novel genes affecting host interactions into unrelated P. syringae pathovar strains.

The majority of the native plasmids identified in P. syringae belong to the recently described pPT23A-like family; these plasmids share replication sequences and, in most cases, additional areas of homology (Murillo & Keen, 1994 ; Sundin & Bender, 1996 ; Glickmann et al., 1998 ; Sesma et al., 1998 ; Gibbon et al., 1999 ). Furthermore, many P. syringae strains contain two to six coexisting pPT23A-like plasmids (Murillo & Keen, 1994 ; Sesma et al., 1998 ), suggesting their potential role in the bacterial life cycle and underlying their capacity to overcome incompatibility. In the best characterized case, that of pPT23A and pPT23B, both from P. syringae pv. tomato PT23, approximately 74% of these plasmids consist of repeated sequences (Murillo & Keen, 1994 ). We are interested in the evolution of the pPT23A-like plasmid family, in particular the molecular evolution of the plasmids and of the sequences they encode with emphasis on the host-selective forces resulting in the compartmentalization of specific plasmid genes within a limited pathovar range. The replication regions from two pPT23A-like plasmids contain a determinant, repA, that is highly homologous to the major replication gene of ColE2 replicons, a plasmid group widespread in Escherichia coli and found in other members of the {gamma}-Proteobacteria, and to a plasmid from Thiobacillus intermedius (Gibbon et al., 1999 ). In this study, we report an analysis of repA sequences as a strategy to examine the phylogeny of native P. syringae plasmids; the repA determinant is essential for replication of the pPT23A-like plasmids and thus is an appropriate locus for comparative study.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, plasmids and growth conditions.
Escherichia coli DH10B was used for cloning purposes. The plasmidless Pseudomonas syringae pv. syringae FF5 (Sundin & Bender, 1993 ) was used to test the replication ability of selected clones. Other P. syringae strains are listed in Table 1. E. coli was grown in LB medium at 37 °C and P. syringae was cultivated in King’s medium B (Hispanlab; King et al., 1954 ) or in LB at 28 °C. When necessary, media were supplemented with ampicillin or kanamycin at a final concentration of 100 or 25 µg ml-1, respectively.


View this table:
[in this window]
[in a new window]
 
Table 1. Origin and characteristics of the pPT23A-like plasmids used in this study

 
Genetic and molecular biology techniques.
Standard molecular biology techniques were followed (Sambrook et al., 1989 ). Plasmid DNA minipreparations were performed using 1·5 ml of an overnight culture in King’s medium B following a modified alkaline lysis procedure (Zhou et al., 1990 ), and intact plasmids were separated by electrophoresis on 0·6% agarose (Hispanlab) gels in 1x TAE buffer (40 mM Tris/acetate, 1 mM EDTA, pH 8·0) at 3·2 V cm-1 for 5–6 h at room temperature or for about 16 h at 4 °C (Murillo et al., 1994 ). Cloning of DNA fragments was done essentially as described by Crouse et al. (1983) . DNA was introduced into Pseudomonas by electroporation (Keen et al., 1992 ).

Fragments to be used as probes were either amplified by PCR or separately cloned in pBluescript (Stratagene) or pK184 (Jobling & Holmes, 1990 ); in this case, they were excised from the vector and separated in low-melting-point agarose before labelling. Labelling of DNA with digoxigenin and hybridization of uncut plasmid DNA separated by electrophoresis and transferred to nylon membranes (Hybond N+; Amersham) were performed following the manufacturer’s instructions (Boehringer Mannheim).

Cloning of origins of replication.
The possession of pPT23A-like plasmids among the strains included in this study either was already reported (Murillo & Keen, 1994 ; Sundin & Bender, 1996 ; Sesma et al., 1998 ; Sundin & Murillo, 1999 ) or was examined by hybridization using the 0·8 kb EcoRI fragment from the replication region of pPT23A (Gibbon et al., 1999 ) as probe. repA genes were amplified using primers 532 (5'-GAACGGTGGACTTATGG-3') and 1588 (5'-CTCCAGCTTGCGGCCCC-3') which flank a fragment of 1399 bp containing 1279 bp of the repA coding region plus 120 bp upstream of the putative start codon (Gibbon et al., 1999 ). Total plasmid DNA was used as template for strains containing a single native plasmid and for P. syringae pv. tomato strain OK-1. The origins of replication from pPT23A, pR6C and pAV505 (see Table 1 for details) were cloned in plasmids pAKC (Murillo & Keen, 1994 ), pORI601 (Sesma et al., 1998 ) and pPPY50 (Gibbon et al., 1999 ), respectively, which were also used as templates for amplification. Amplifications were carried out in a total volume of 25 µl using 1 µl DNA as template under the following conditions: 4 mM MgCl2, 0·75 mM each dNTPs, 1 pmol each primer µl-1 and 1·0 U Taq polymerase. PCR was performed in a Perkin Elmer 480 thermocycler with one cycle of 94 °C for 5 min, 55 °C for 5 min and 72 °C for 90 s, followed by 32 cycles at 94 °C for 1 min, 55 °C for 1 min, 72 °C for 90 s and a final extension step of 10 min at 72 °C. Ten microlitres of the PCR mix was loaded on a 1% agarose gel and bands of the correct size were purified using a gel extraction kit (Qiaquick; Qiagen) and ligated to vector pCR2.1 (Invitrogen). The repA gene from pAV505 was not cloned because its complete sequence was already available (accession no. AJ222648; Gibbon et al., 1999 ). The amplicon cloned from total plasmid DNA from strain OK-1 was confirmed to be derived from plasmid pOK1B, by comparison of the restriction profile of repA amplicons from purified DNA of each of the two native plasmids of OK-1 with that of the cloned amplicon. For restriction analysis, DNA was amplified with primers 532 and 1588 as above and PCR products digested with HaeIII or Sau3AI were separated on 2·5% high-resolution agarose (MS8; Hispanlab) gels. As templates, we used purified plasmid DNA for the cloned amplicons and for P. syringae strains containing a single native plasmid. For strains containing two or more pPT23A-like plasmids, these were separated by electrophoresis as indicated above and template DNA was purified from appropriate bands excised from the gels using 0·2 µm Nanosep MF columns (Pall Filtron) as described previously (Sesma et al., 1998 ).

Analysis of nucleotide sequences.
Plasmid DNA was purified using Qiagen columns. Nucleotide sequencing was done using the Big Dye kit (ABI) following the instructions of the manufacturer; sequence reactions were run at the Genetic Technologies Center, Texas A&M University. Both ends of the cloned amplicons were sequenced using universal and custom-synthesized primers based on the sequence of the minimal replication regions of plasmids pPT23A and pAV505 from P. syringae (Gibbon et al., 1999 ). Sequence comparison, alignment and construction of phylogenetic trees using the neighbour-joining method was done using the programs CLUSTAL W and MULTALIN at EBI, Cambridge, UK, or IBCP, Lyon, France. The strength of the tree topology was assessed by the bootstrap method using the CLUSTALX software package (Thompson et al., 1997 ). Phylogenetic trees were viewed using the NJPLOT software (Saitou & Nei, 1987 ).

ERIC, arbitrarily primed PCR and data analysis.
The genetic relationships among P. syringae strains was examined by PCR using primers for extragenic repetitive consensus (ERIC) and the arbitrarily primed PCR (AP-PCR) techniques. For ERIC analysis, primers ERIC1R and ERIC2 were used for amplification as described by McManus & Jones (1995) . AP-PCR was carried out with a single 20 bp oligonucleotide primer complementary to the IS50 portion of Tn5 as described previously (Sundin & Murillo, 1999 ).

Amplification bands or bands resulting from digestion of amplification products from the origins of replication were scored as 1 (present) or 0 (absent), and a similarity matrix was computed using Dice’s coefficient with the program NTSYS-PC (Applied Biostatistics) as described previously (Sundin & Murillo, 1999 ).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Conservation of DNA sequences among pPT23A-like plasmids
pPT23A-like plasmids are those that contain sequences that hybridize to the origin of replication of plasmid pPT23A from Pseudomonas syringae pv. tomato PT23 (Gibbon et al., 1999 ). Although they can share extensive regions of homology (Murillo & Keen, 1994 ; Sundin et al., 1994 ; Sesma et al., 1998 ; Sundin & Murillo, 1999 ), little is known about particular genes that might be broadly conserved among them. To study the conservation of DNA sequences among 14 pPT23A-like plasmids originating from P. syringae pathovars apii, glycinea, phaseolicola, syringae and tomato (Table 1), we performed a series of Southern hybridizations using seven probes for sequences that have been previously detected on pPT23A or in the highly related plasmid pPT23B, also from strain PT23 (Table 2). None of the probes hybridized to all the tested plasmids. Five of the probes (avirulence gene avrD, genes involved in the synthesis of coronatine, plasmid stability determinants and the insertion sequence IS1240) hybridized only to 2–4 of the 14 analysed plasmids, indicating their poor conservation. On the other hand, the rulAB genes, which confer resistance to UV light (Sundin et al., 1996 ; Sundin & Murillo, 1999 ), hybridized to the native plasmids of three pv. tomato and the seven pv. syringae strains, corroborating its widespread distribution (Sundin et al., 1996 ; Sesma et al., 1998 ; Sundin & Murillo, 1999 ). The insertion sequence IS801, which was originally detected in a P. syringae pv. phaseolicola native plasmid (Romantschuk et al., 1991 ), was not present in the native plasmids from any of the P. syringae pv. syringae strains, while it hybridized to the plasmids from the remaining seven strains (Table 2). Only in a few cases (rulAB genes, IS1240, IS801 and stbCBAD genes) was hybridization to the probes in more than one pPT23A-like plasmid from the same strain observed (Table 3). For example, five of the pPT23A-like plasmids of P. syringae pv. glycinea race 6 hybridized to IS801, while three of them showed hybridization to the stbCBAD genes.


View this table:
[in this window]
[in a new window]
 
Table 2. Conservation of relevant genetic determinants in different pPT23A-like plasmids as determined by DNA hybridization

 

View this table:
[in this window]
[in a new window]
 
Table 3. Hybridization of different coexisting pPT23A-like plasmids to selected DNA probes

 
Phylogeny of the repA sequence of pPT23A-like plasmids
With the availability of the complete sequence of several plasmids from diverse bacterial species, it is apparent that plasmid genomes are mosaics of sequences that can be acquired from varied sources (for example, see Tauch et al., 2000 ). It is likely that plasmids of the pPT23A-like family are also structural mosaics; for example, PAI recently characterized from P. syringae pv. phaseolicola is located on a pPT23A-like family plasmid (Jackson et al., 1999 ). Nevertheless, the absolute conservation of repA among pPT23A-like family plasmids and the ability of the cloned repA sequence to evict a number of pPT23A-like family plasmids in incompatibility experiments (Sesma et al., 1998 ) suggests that analysis of repA is a good starting point in determining phylogenetic relationships among this plasmid group. However, since we do not have knowledge of the complete sequence of the plasmids under study, we will conservatively refer to our results in terms of gaining an understanding of the phylogeny of the repA sequence from pPT23A-like family plasmids.

The minimal fragment able to support the replication of pPT23A was defined as an approximately 1·6 kb fragment that spans gene repA (Fig. 1) (Gibbon et al., 1999 ), which encodes a 437 aa putative replicase that is essential for plasmid replication. A 1399 bp fragment containing 1279 bp of the 1311 bp repA coding region plus 120 bp upstream of the putative start codon (Fig. 1; Gibbon et al., 1999 ) was amplified from the pPT23A-like plasmids and separately cloned in vector pCR2.1 which, in our hands, did not replicate in P. syringae. All the cloned amplicons, except those originating from p5D425A and pB76A, were able to replicate in the plasmidless strain P. syringae pv. syringae FF5 (data not shown), suggesting that they were functional in their parental plasmids. Since in all cases the repA gene was in the opposite orientation with respect to the vector lac promoter, as determined by nucleotide sequencing (data not shown), this suggests that the 120 bp preceding the putative start codon could be sufficient to initiate transcription and for supporting autonomous replication in P. syringae.



View larger version (6K):
[in this window]
[in a new window]
 
Fig. 1. Replication region from pPT23A. A box shows the position and extent of the replicase (repA) gene; vertical black bars inside the box denote nucleotide changes that result in residues that are not conserved between the replicases of pPT23A and pAV505. The putative start and stop codons are shown. Numbers refer to nucleotide positions in the pPT23A sequence (Gibbon et al., 1999 ). The extent of the fragment amplified and cloned is shown below the replication region; the position of the primers used for amplification (532 and 1588) is shown and the fragments sequenced are denoted as black boxes. E, EcoRI; H, HindIII; V, EcoRV.

 
The digestion profile of the cloned amplicons with HaeIII and Sau3AI was identical to the profile of the amplicons obtained from the corresponding native plasmids with primers 532 and 1588 (Fig. 2 and data not shown), indicating that no major changes or reorganizations had occurred during the cloning procedure. Most of the amplicons had a characteristic restriction profile with both enzymes (Fig. 2). DNA bands were used to calculate genetic distance and to construct the tree shown in Fig. 3. In general, plasmids from all the strains of the same pathovar did not cluster together, suggesting their diverse origin.



View larger version (76K):
[in this window]
[in a new window]
 
Fig. 2. Restriction analysis of cloned amplicons from different replication regions. DNA was amplified, digested with HaeIII (a) or Sau3AI (b) and separated by gel electrophoresis. DNA amplified from the respective native plasmids showed the same restriction pattern as the cloned amplicons (not shown).

 


View larger version (19K):
[in this window]
[in a new window]
 
Fig. 3. Dendrogram derived from restriction data of the amplicons. Cluster analysis was performed using the Dice similarity coefficient as described (Sundin & Murillo, 1999 ). The scale at the top indicates the degree of genetic relatedness.

 
The RepA proteins from plasmids pPT23A, from P. syringae pv. tomato PT23, and pAV505, from P. syringae pv. phaseolicola HRI1302A, are nearly identical from residues 1 to 373 at their N-terminal ends (96·5% identity), but poorly conserved from positions 374 to 437 in their C termini (50% identity) (Fig. 1; Gibbon et al., 1999 ). We therefore sequenced both ends of the cloned amplicons to study the phylogenetic relationships of their corresponding native plasmids. The sequence corresponding to the 3' end of the repA genes contained 290 nt for all the cloned amplicons, while the 5' sequence comprised 291 or 279 nt (see below), of which 176 nt corresponded to the putative start of the repA gene and the rest to the upstream sequences. In all cases, the sequence of the primers used to amplify the repA genes were excluded. The sequences obtained from the cloned amplicon originating from pPT23A were identical to the previously obtained repA sequence (Gibbon et al., 1999 ). Three of the amplicons, corresponding to plasmids pPT23A, pOK1B and pPSR14, contained a deletion of 12 nt located 11 (pPT23A and pOK1B) or 19 nt (pPSR14) before the ATG start (Fig. 4). In some of the amplicons, a direct repeat of 9 nt was found bordering the deletion area; since the deleted DNA spanned one of the repeats plus the intervening DNA, it is possible that the deletions were generated by replicative slippage (Hancock, 1995 ). To avoid overestimating sequence divergence, the 20 nt defined by the ends of the deletions (see Fig. 4) were excluded from all the comparison analyses.



View larger version (53K):
[in this window]
[in a new window]
 
Fig. 4. Alignment of a portion of the 5' sequences from the cloned amplicons showing the deletion found upstream of the repA start codon. Numbers on the top refer to nucleotide positions in the pPT23A sequence (Gibbon et al., 1999 ), while numbers on the right indicate nucleotide positions of the sequenced amplicons. The 20 nt included in the box were excluded from all the phylogenetic analyses. The putative start codon for repA is indicated in bold. Arrows indicate the position of the direct repeats found in the sequences. Alignment was done using the program MULTALIN at IBCP. Dashes indicate gaps introduced to maximize the alignment.

 
The 3' and 5' sequences of each amplicon were treated as a single continuous sequence to obtain the consensus tree shown in Fig. 5. The overall genetic analysis resulted in the clustering of the plasmid amplicons into four distinct groups (identity >=88·8%; bootstrap values >=95%; designated A, B, C and D in Fig. 5). Plasmid amplicons from three pv. tomato strains, pv. apii and five out of the seven pv. syringae strains formed a well defined group. Among the pv. syringae strains, no obvious correlation was found between host of isolation and phylogenetic proximity. For instance, the sequences of the amplicons from pBBS325A and pB8617A, isolated from beans in Colorado and New York, respectively, were identical and they showed only 13 nt differences with the amplicon from p5D425A, which was isolated from apricot in California. The second group included the plasmid amplicons from the pv. glycinea and pv. phaseolicola strains and from the remaining pv. tomato strain. Nucleotide sequence identities among the groups were >92·9% for the pv. tomato and pv. apii amplicons from Group A, >91·3% for the pv. syringae plasmids from Group A and >90·0% for the amplicons from Group B. Additionally, the amplicons from plasmids pPSR14 and pCG131 from P. syringae pv. syringae formed two separate groups that showed only 77·1–87·7% identity with the rest of the sequences. pCG131 was the most divergent of the P. syringae pv. syringae plasmids, perhaps owing to its geographical (New Zealand) or plant (millet) origin, although the closest relatives were the plasmid amplicons from P. syringae pv. phaseolicola (86·5% identity) and pv. syringae 7B12 (85·7% identity).



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 5. Phylogenetic tree generated with the 3' and 5' end sequences of each amplicon treated as a single continuous sequence. The tree was constructed by the neighbour-joining method using the CLUSTAL W Multiple Sequence Alignment interactive program available at EBI and displayed using NJPLOT (Saitou & Nei, 1987 ). The four plasmid groups are indicated on the right (sequence identity >=88·8%). Only bootstrap percentages that were lower than 95% are shown (numbers in italics).

 
When analysed separately (data not shown), the 5' sequences did not allow for a clear separation of the different plasmid groups. However, the 3' sequences were very discriminative and allowed a similar grouping to that obtained with the complete sequence, except that it was not possible to allocate pAV505 to any of the groups. Also, the 3' sequences were more conserved than the 5' sequences among any given plasmid group. For instance, the sequences of plasmid pPT23A showed 92·1/93·8% identity (5'/3' sequences) to the DC3000 plasmid, which is highly related to pPT23A, while they were 92·8/77·2% identical to the sequences of plasmid from B120, which clustered in a separate plasmid group.

The DNA fingerprint of the strains belonging to pv. syringae and pv. tomato was determined by using ERIC-PCR and AP-PCR. The position of the strains in the resulting dendrograms and their genetic relatedness (Fig. 6) agreed in general with results obtained by other authors using combinations of these strains (Hendson et al., 1992 ; Legard et al., 1993 ; Sundin & Murillo, 1999 ). In many cases, the relatedness between two strains did not reflect the relatedness between their native plasmids (Figs 5 and 6). For instance, strains 8B48 and B76 showed a low relative genetic similarity, although the plasmids they harboured, pPSR11 and pB76A, respectively, were 98% identical, which is the highest identity value among the native plasmids if we exclude pB8617A and pBBS325A (100% identical).



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 6. Dendrogram of genetic relatedness among P. syringae pv. syringae and pv. tomato strains. Data from the ERIC and AP-PCR fingerprint patterns generated from each strain were combined and cluster analysis was performed using the Dice similarity coefficient as described (Sundin & Murillo, 1999 ). The scales at the top indicates the degree of genetic relatedness between the strains tested.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
The pPT23A-like plasmid family encompasses a large majority of P. syringae plasmids of known importance to host–pathogen interactions. This plasmid group is widely distributed among P. syringae pathovars; however, inter- and intrapathovar plasmid differentiation is readily apparent regarding the occurrence of specific genes encoded by individual pPT23A-like plasmids. In this study, we observed a limited interpathovar distribution of genes affecting host specificity and virulence (avrD and cfa, respectively), while the rulAB determinant, which is important for survival and population maintenance in the phyllosphere (Sundin & Murillo, 1999 ), was widely distributed among pathovars. These results corroborate previous examples of the limited distribution of host specificity determinants among pathovars; indeed, host range has always been an important determinator of pathovar identification in general.

Knowledge of individual plasmid genotypes may provide important clues in the evolution of the pPT23A-like plasmid family. A determinant such as rulAB may have been acquired early in the evolution of this plasmid family and may have been fixed within the plasmid genome because of its importance to phyllosphere fitness, a trait which is universally important to the success of P. syringae strains (Beattie & Lindow, 1995 ). The acquisition of loci such as avr genes might have stimulated the emergence of new pathovars and, concomitant with that, the opportunity for continued plasmid evolution within the context of a new host–pathogen interaction. Further analysis will determine if certain plasmid lineages are defined by the presence of additional determinants of importance to host–pathogen dynamics. Likewise, the limited distribution of determinants of importance to plasmid maintenance (IncC and stbCBAD) could be illustrative of the requirement of such features only in certain host backgrounds, or may reflect a possibly large diversity of maintenance systems among pPT23A-like plasmids. The distribution of IS elements such as IS801 and IS1240 may also be interesting in terms of the timing of evolutionary acquisition. IS801-like sequences, and those of other transposable elements, have been found to flank avr sequences and sequence regions suspected to be pathogenicity islands in several P. syringae pathovars (Kim et al., 1998 ; Jackson et al., 1999 ). These observations imply that the mobility of these types of sequences among strains is important, possibly from the standpoint of a pathogen adding to its host range or combating a new plant resistance gene. It should be noted that IS801 and avr genes were not detected in pv. syringae in this study or a previous study (Sundin & Bender, 1996 ), indicating that the mobility of these sequences may be limited by factors which are currently unknown.

We utilized the repA gene as our starting point for large-scale phylogenetic analysis of the pPT23A-like plasmid family. repA was appropriate because of its requirement for plasmid replication in P. syringae and because it is the only gene currently known to be distributed among all pPT23A-like plasmids. Dendrograms generated from analyses of restriction digest patterns or 5' and 3' sequence data of repA lacked congruence. This could indicate that the restriction digest data were not substantial enough to resolve relationships among the plasmids, or that additional diversity was present within portions of repA that were not sequenced. Nevertheless, similar sequencing strategies have been utilized to examine the diversity present among replication genes of closely related plasmids (Burgos et al., 1996 ; Turner et al., 1996 ) and imply that the relationships generated from our analysis of the 5' and 3' sequence data (Fig. 5) would be the most valid. These results indicate that the repA sequences from plasmids isolated from different P. syringae pathovars were not always clearly distinguished. From these observations, we can infer that ecological factors such as plasmid transfer or similar selection pressures faced by different pathovars could result in the current distribution of repA sequences among P. syringae pathovars. There have been several examinations of plasmid transfer in planta (Bender & Cooksey, 1986 ; Burr et al., 1988 ; Sundin et al., 1989 ; Björklöf et al., 1995 ). However, although these studies indicate that plasmid transfer can occur, they do not provide evidence for the natural occurrence of plasmid transfer or if plasmids are mobile among more distantly related strains. In one retrospective analysis of plasmid and host genotypes examining a population of P. syringae pv. syringae under bactericide (copper and streptomycin) selection pressure, recent plasmid transfer events were inferred, but limited to closely related host strains (Sundin et al., 1994 ). The issue of transfer of pPT23A-like plasmids into P. syringae strains that already contain an indigenous pPT23A-like plasmid(s) must also be addressed. In one study utilizing the P. syringae pv. syringae strains 4A39 and FF3, each of which contained a single pPT23A-like plasmid, conjugation experiments were done resulting in the transfer of the copper resistance (CuR) plasmid pPSR4 into strain FF3 which contained the streptomycin resistance (SmR) transposon Tn5393 on plasmid pPSR5 (Sundin & Bender, 1996 ). Further experiments showed that, in the absence of selection, the plasmids were incompatible; however, selection for the CuR and SmR markers enabled the FF3 strain to harbour both pPSR4 and pPSR5 for 32±8 generations, after which pPSR5 was lost, but Tn5393 transposed into pPSR4 (Sundin & Bender, 1996 ). This experiment showed that surface exclusion or restriction modification systems did not preclude the transfer of a pPT23A-like plasmid into a strain harbouring an incompatible plasmid, at least for these P. syringae pv. syringae hosts. Plasmid incompatibility may prevent the establishment of a plasmid in a new P. syringae background; however, the extensive homology present on pPT23A-like plasmids may contribute to genetic rearrangements resulting in the acquisition of new sequences by these plasmids. Whether such barriers to plasmid transfer exist in other P. syringae pathovars is currently unknown. The results of other studies of Agrobacterium and Rhizobium spp. have also indicated the long-term stability of plasmid and host genotypes (Young & Wexler, 1988 ; Otten et al., 1996 ; Wernegreen et al., 1997 ). Our results suggest that most of the PT23A-like plasmids studied have been maintained for long periods within their respective hosts. Our current analysis of strains HS191, 7B12 and B120, however, does not preclude the possible occurrence of plasmid transfer in nature.


   ACKNOWLEDGEMENTS
 
We thank the researchers listed in Table 1 for the gifts of P. syringae strains. We are also grateful to C. Manceau for helpful suggestions and to J. L. Jacobs for excellent technical assistance. G.W.S. acknowledges support from the US Department of Agriculture (NRICGP 9702832) and the Texas Agricultural Experiment Station. J.M. gratefully acknowledges support from the Spanish CICYT (BIO94-0442, BIO97-0598) and from Caja Rural de Navarra and Universidad Pública de Navarra for a short stay in Texas A&M University.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Beattie, G. A. & Lindow, S. E. (1995). The secret life of foliar bacterial pathogens on leaves. Annu Rev Phytopathol 33, 145-172.

Bender, C. & Cooksey, D. (1986). Indigenous plasmids in Pseudomonas syringae pv. tomato: conjugative transfer and role in copper resistance. J Bacteriol 165, 534-541.[Medline]

Bender, C. L., Alarcón-Chaidez, F. & Gross, D. C. (1999). Pseudomonas syringae phytotoxins: mode of action, regulation, and biosynthesis by peptide and polyketide synthetases. Microbiol Mol Biol Rev 63, 266-292.[Abstract/Free Full Text]

Björklöf, K., Suoniemi, A., Haahtela, K. & Romantschuk, M. (1995). High frequency of conjugation versus plasmid segregation of RP1 in epiphytic Pseudomonas syringae populations. Microbiology 141, 2719-2727.[Abstract]

Burgos, P. A., Velazquez, E. & Toro, N. (1996). Identification and distribution of plasmid-type A replicator region in rhizobia. Mol Plant–Microbe Interact 9, 843-849.[Medline]

Burr, T. J., Norelli, J. L., Katz, B., Wilcox, W. F. & Hoying, S. A. (1988). Streptomycin resistance of Pseudomonas syringae pv. papulans in apple orchards and its association with a conjugative plasmid. Phytopathology 78, 410-413.

Comai, L. & Kosuge, T. (1980). Involvement of plasmid deoxyribonucleic acid in indoleacetic acid synthesis in Pseudomonas savastanoi. J Bacteriol 143, 950-957.[Medline]

Crouse, G. F., Frischauf, A. & Lehrach, H. (1983). An integrated and simplified approach for cloning into plasmids and single-stranded phages. Methods Enzymol 101, 78-89.[Medline]

Cuppels, D. A. (1986). Generation and characterization of Tn5 insertion mutations in Pseudomonas syringae pv. tomato. Appl Environ Microbiol 51, 323-327.

Cuppels, D. A. & Ainsworth, T. (1995). Molecular and physiological characterization of Pseudomonas syringae pv. tomato and Pseudomonas syringae pv. maculicola strains that produce the phytotoxin coronatine. Appl Environ Microbiol 61, 3530-3536.[Abstract/Free Full Text]

Curiale, M. S. & Mills, D. (1983). Molecular relatedness among cryptic plasmids in Pseudomonas syringae pv. glycinea. Phytopathology 73, 1440-1444.

Denny, T. P. (1988). Phenotypic diversity in Pseudomonas syringae pv. tomato. J Gen Microbiol 134, 1939-1948.

Gardan, L., Shafik, H., Belouin, S., Broch, R., Grimont, F. & Grimont, P. A. D. (1999). DNA relatedness among the pathovars of Pseudomonas syringae and description of Pseudomonas tremae sp. nov. and Pseudomonas cannabina sp. nov. (ex Sutic and Dowson 1959). Int J Syst Bacteriol 49, 469-478.[Abstract]

Gibbon, M. J., Sesma, A., Canal, A., Wood, J. R., Hidalgo, E., Brown, J., Vivian, A. & Murillo, J. (1999). Replication regions from plant-pathogenic Pseudomonas syringae plasmids are similar to ColE2-related replicons. Microbiology 145, 325-334.[Abstract]

Glickmann, E., Gardan, L., Jacquet, S., Hussain, S., Elasri, M., Petit, A. & Dessaux, Y. (1998). Auxin production is a common feature of most pathovars of Pseudomonas syringae. Mol Plant–Microbe Interact 11, 156-162.[Medline]

Hancock, J. M. (1995). The contribution of DNA slippage to eucaryotic nuclear 18S rRNA evolution. J Mol Evol 40, 629-639.[Medline]

Hanekamp, T., Kobayashi, D., Hayes, S. & Stayton, M. M. (1997). Avirulence gene D of Pseudomonas syringae pv. tomato may have undergone horizontal gene transfer. FEBS Lett 415, 40-44.[Medline]

Hendson, M., Hildebrand, D. C. & Schroth, M. N. (1992). Relatedness of Pseudomonas syringae pv. tomato, Pseudomonas syringae pv. maculicola and Pseudomonas syringae pv. anthirrini. J Appl Bacteriol 73, 455-464.

Jackson, R. W., Athanassopoulos, E., Tsiamis, G., Mansfield, J. W., Sesma, A., Arnold, D. L., Gibbon, M. J., Murillo, J., Taylor, J. D. & Vivian, A. (1999). Identification of a pathogenicity island, which contains genes for virulence and avirulence, on a large native plasmid in the bean pathogen Pseudomonas syringae pathovar phaseolicola. Proc Natl Acad Sci U S A 96, 10875-10880.[Abstract/Free Full Text]

Jobling, M. G. & Holmes, R. K. (1990). Construction of vectors with the p15a replicon, kanamycin resistance, inducible lacZ{alpha} and pUC18 or pUC19 multiple cloning sites. Nucleic Acids Res 18, 5315-5316.[Medline]

Kearney, B. & Staskawicz, B. J. (1990). Widespread distribution and fitness contribution of Xanthomonas campestris avirulence gene avrBs2. Nature 346, 385-386.[Medline]

Keen, N. T., Shen, H. & Cooksey, D. A. (1992). Introduction of DNA into plant pathogenic bacteria. In Molecular Plant Pathology: A Practical Approach , pp. 45-50. Edited by S. J. Gurr, M. J. McPherson & D. J. Bowles. Oxford:IRL Press.

Kim, J. F., Charkowski, A. O., Alfano, J. R., Collmer, A. & Beer, S. V. (1998). Transposable elements and bacteriophage sequences flanking Pseudomonas syringae avirulence genes. Mol Plant–Microbe Interact 11, 1247-1252.

King, E. O., Ward, N. K. & Raney, D. E. (1954). Two simple media for the demonstration of pyocyanin and fluorescein. J Lab Clin Med 44, 301-307.

King, G. J. (1989). Plasmid analysis and variation in Pseudomonas syringae. J Appl Bacteriol 67, 489-496.

Kobayashi, D. Y., Tamaki, S. J. & Keen, N. T. (1990). Molecular characterization of avirulence gene D from Pseudomonas syringae pv. tomato. Mol Plant–Microbe Interact 3, 94-102.[Medline]

Legard, D. E., Aquadro, C. F. & Hunter, J. E. (1993). DNA sequence variation and phylogenetic relationships among strains of Pseudomonas syringae pv. syringae inferred from restriction site maps and restriction fragment length polymorphism. Appl Environ Microbiol 59, 4180-4188.[Abstract]

Lorang, J. M., Shen, H., Kobayashi, D., Cooksey, D. & Keen, N. T. (1994). avrA and avrE in Pseudomonas syringae pv. tomato PT23 play a role in virulence on tomato plants. Mol Plant–Microbe Interact 7, 508-515.

McManus, P. S. & Jones, A. L. (1995). Genetic fingerprinting of Erwinia amylovora strains isolated from tree-fruit crops and Rubus spp. Phytopathology 85, 1547-1553.

Mitchell, R. E. (1982). Coronatine production by some phytopathogenic pseudomonads. Physiol Plant Pathol 20, 83-89.

Murillo, J. & Keen, N. T. (1994). Two native plasmids of Pseudomonas syringae pathovar tomato strain PT23 share a large amount of repeated DNA, including replication sequences. Mol Microbiol 12, 941-950.[Medline]

Murillo, J., Shen, H., Gerhold, D., Sharma, A. K., Cooksey, D. A. & Keen, N. T. (1994). Characterization of pPT23B, the plasmid involved in syringolide production by Pseudomonas syringae pv. tomato. Plasmid 31, 275-287.[Medline]

Otten, L., de Ruffray, P., Momol, E. A., Momol, M. T. & Burr, T. J. (1996). Phylogenetic relationships between Agrobacterium vitis isolates and their Ti plasmids. Mol Plant–Microbe Interact 9, 782-786.

Rangaswamy, V., Mitchell, R., Ullrich, M. & Bender, C. (1998). Analysis of genes involved in biosynthesis of coronafacic acid, the polyketide component of the phytotoxin coronatine. J Bacteriol 180, 3330-3338.[Abstract/Free Full Text]

Ritter, C. & Dangl, J. L. (1995). The avrRpm1 gene of Pseudomonas syringae pv. maculicola is required for virulence on Arabidopsis. Mol Plant–Microbe Interact 8, 444-453.[Medline]

Romantschuk, M., Richter, G. Y., Mukhopadhyay, P. & Mills, D. (1991). IS801, an insertion sequence element isolated from Pseudomonas syringae pathovar phaseolicola. Mol Microbiol 5, 617-622.[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.

Sato, M. (1988). In planta transfer of the gene(s) for virulence between isolates of Pseudomonas syringae pv. atropurpurea. Ann Phytopathol Soc Jpn 54, 20-24.

Saunier, M., Malandrin, L. & Samson, R. (1996). Distribution of Pseudomonas syringae pathovars into twenty-three O serogroups. Appl Environ Microbiol 62, 2360-2374.[Abstract]

Sawada, H., Suzuki, F., Matsuda, I. & Saitou, N. (1999). Phylogenetic analysis of Pseudomonas syringae pathovars suggests the horizontal gene transfer of argK and the evolutionary stability of hrp gene cluster. J Mol Evol 49, 627-644.[Medline]

Sesma, A. (2000). Characterization of native plasmids and virulence genes in the phytopathogen Pseudomonas syringae. PhD thesis, Universidad Pública de Navarra, Pamplona, Spain.

Sesma, A., Sundin, G. & Murillo, J. (1998). Closely related replicons coexisting in the phytopathogen Pseudomonas syringae show a mosaic organization of the replication region and altered incompatibility behavior. Appl Environ Microbiol 64, 3948-3953.[Abstract/Free Full Text]

Sundin, G. W. & Bender, C. L. (1993). Ecological and genetic analysis of copper and streptomycin resistance in Pseudomonas syringae pv. syringae. Appl Environ Microbiol 59, 1018-1024.[Abstract]

Sundin, G. W. & Bender, C. L. (1996). Molecular analysis of closely related copper- and streptomycin-resistance plasmids in Pseudomonas syringae pv. syringae. Plasmid 35, 98-107.[Medline]

Sundin, G. W. & Murillo, J. (1999). Functional analysis of the Pseudomonas syringae rulAB determinant in tolerance to ultraviolet B (290–320 nm) radiation and distribution of rulAB among P. syringae pathovars. Environ Microbiol 1, 75-88.[Medline]

Sundin, G. W., Jones, A. L. & Fulbright, D. W. (1989). Copper resistance in Pseudomonas syringae pv. syringae from cherry orchards and its associated transfer in vitro and in planta with a plasmid. Phytopathology 79, 861-865.

Sundin, G. W., Demezas, D. H. & Bender, C. L. (1994). Genetic and plasmid diversity within natural populations of Pseudomonas syringae with various exposures to copper and streptomycin bactericides. Appl Environ Microbiol 60, 4421-4431.[Abstract]

Sundin, G. W., Kidambi, S. P., Ullrich, M. & Bender, C. L. (1996). Resistance to ultraviolet light in Pseudomonas syringae: sequence and functional analysis of the plasmid-encoded rulAB genes. Gene 177, 77-81.[Medline]

Swarup, S., De Feyter, R., Brlansky, R. H. & Gabriel, D. W. (1991). A pathogenicity locus from Xanthomonas citri enables strains from several pathovars of X. campestris to elicit cankerlike lesions on citrus. Phytopathology 81, 802-809.

Tauch, A., Krieft, S., Kalinowski, J. & Pühler, A. (2000). The 51,409-bp R-plasmid pTP10 from the multiresistant clinical isolate Corynebacterium striatum M82B is composed of DNA segments initially identified in soil bacteria and in plant, animal, and human pathogens. Mol Gen Genet 263, 1-11.[Medline]

Thompson, J. D., Gibson, T. J., Plewniak, F., Jeanmougin, F. & Higgins, D. G. (1997). The CLUSTALX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res 25, 4876-4882.[Abstract/Free Full Text]

Tsiamis, G., Mansfield, J. W., Hockenhull, R. & 8 other authors (2000). Cultivar-specific avirulence and virulence functions assigned to avrPphF in Pseudomonas syringae pv. phaseolicola, the cause of bean halo-blight disease. EMBO J19, 3204–3214.[Abstract/Free Full Text]

Turner, S. L., Rigottier-Gois, L., Power, R. S., Amarger, N. & Young, J. P. (1996). Diversity of repC plasmid-replication sequences in Rhizobium leguminosarum. Microbiology 142, 1705-1713.[Abstract]

Vivian, A., Gibbon, M. J. & Murillo, J. (1997). The molecular genetics of specificity determinants in plant pathogenic bacteria. In The Gene-for-Gene Relationship in Plant Parasite Interactions , pp. 293-328. Edited by I. R. Crute, E. B. Holub & J. J. Burdon. Wallingford:CABI.

Wernegreen, J. J., Harding, E. E. & Riley, M. A. (1997). Rhizobium gone native: unexpected plasmid stability of indigenous Rhizobium leguminosarum. Proc Natl Acad Sci U S A 94, 5483-5488.[Abstract/Free Full Text]

Wiebe, W. L. & Campbell, R. N. (1993). Characterization of Pseudomonas syringae pv. maculicola and comparison with P. s. pv. tomato. Plant Dis 77, 414-419.

Yang, Y., Yuan, Q. & Gabriel, D. W. (1996). Watersoaking function(s) of XcmH1005 are redundantly encoded by members of the Xanthomonas avr/pth gene family. Mol Plant–Microbe Interact 8, 627-631.

Young, J. & Wexler, M. (1988). Sym plasmid and chromosomal genotypes are correlated in field populations of Rhizobium leguminosarum. J Gen Microbiol 134, 2731-2739.

Young, J. M., Takikawa, Y., Gardan, L. & Stead, D. E. (1992). Changing concepts in the taxonomy of plant pathogenic bacteria. Annu Rev Phytopathol 30, 67-105.

Young, J. M., Saddler, G. S., Takikawa, Y., De Boer, S. H., Vauterin, L., Gvozdyak, R. I. & Stead, D. E. (1996). Names of plant pathogenic bacteria 1864–1995. Rev Plant Pathol 75, 721-763.

Yucel, I., Boyd, C., Debnam, Q. & Keen, N. T. (1994). Two different classes of avrD alleles occur in pathovars of Pseudomonas syringae. Mol Plant–Microbe Interact 7, 131-139.[Medline]

Zhou, C., Yang, Y. & Jong, A. Y. (1990). Miniprep in ten minutes. BioTechniques 8, 172-173.[Medline]

Received 23 March 2000; revised 30 June 2000; accepted 11 July 2000.