Analysis of the Replication Elements of the pMJ101 Plasmid from the Fish Pathogen Vibrio ordalii

Plasmid 42, 20 –30 (1999) Article ID plas.1999.1406, available online at http://www.idealibrary.com on

Analysis of the Replication Elements of the pMJ101 Plasmid from the Fish Pathogen Vibrio ordalii Carla Bidinost,* Paula J. Wilderman,† ,1 Caleb W. Dorsey,† and Luis A. Actis† ,2 *Departamento de Bioquı´mica Clı´nica, Facultad de Ciencias Quı´micas, Universidad Nacional de Co´rdoba, 5016 Co´rdoba, Argentina; and †Department of Microbiology, Miami University, Oxford, Ohio 45056 Received February 11, 1999; revised April 6, 1999 Vibrio ordalii is a major cause of vibriosis in wild and cultured marine salmonids and carries pMJ101, a 30-kb cryptic plasmid that replicates in the absence of DNA polymerase I without producing single-stranded intermediates. A recombinant derivative harboring the pMJ101 replication region proved to be compatible with pJM1, a plasmid containing the iron acquisition system required for the virulence of V. anguillarum 775, another important pathogen that causes vibriosis. Sequence analysis of a 1.56-kb fragment harboring the pMJ101 replication region revealed the presence of typical features found in DNA origins including an AT-rich region, 11 dam-methylation sites of which 5 are within the putative ori region, and five copies of the 9-bp consensus sequence for DnaA binding. Gel retardation assays demonstrated that the latter replication element indeed binds DnaA purified from Escherichia coli. A potential open reading frame encoding a hydrophilic protein with a predicted pI of 10.3 and an M r of 33,826 was found adjacent to the ori region. Although these properties are typical of DNA-binding proteins, no significant homology was found between this predicted protein, named RepM, and other previously characterized proteins. Reverse transcriptase– polymerase chain reaction analysis of total RNA demonstrated the presence of repM mRNA in V. ordalii. The major initiation site of this mRNA was located 187 nucleotides upstream of the GTG initiation codon as determined by nuclease S1 protection assays. This transcription initiation site is preceded by putative 210 and 235 promoter sequences that control the expression of the repM replication gene. These results demonstrate that the replication region of pMJ101 shares some structural and sequence similarities with other DNA replication regions, which include DnaA binding and methylation sites and an open reading frame encoding a distinct protein required for its replication. © 1999 Academic Press

Vibrio ordalii is one of the major causes of vibriosis in wild and cultured marine salmonids in Japan and the Pacific Northwest of the United States (Actis et al., 1998). This bacterium has been isolated from natural infections in coho salmon, as well as from experimental infections in chum salmon and spring chinook salmon (Ransom et al., 1984). V. ordalii is phenotypically and genetically distinct from V. anguillarum 775 (Schiewe and Crosa, 1981), which also causes disseminated infections in salmonids (Schiewe et al., 1981).

The molecular characterization of different isolates of V. ordalii revealed the presence of a high-copy-number plasmid in all strains examined to date (Schiewe and Crosa, 1981). This plasmid, which was designated pMJ101, is a 30-kb extrachromosomal element that has no DNA sequence homology with the pJM1 virulence plasmid present in V. anguillarum 775 (Crosa et al., 1980). Over the past two decades, the genetic and molecular analyses of pJM1 have been invaluable in determining its impact in virulence. This plasmid is essential for the virulence of V. anguillarum and encodes a highaffinity siderophore-mediated iron acquisition system that allows this fish pathogen to acquire this essential micronutrient from the infected host (Crosa, 1989, 1997). However, not much is known about the functions encoded by pMJ101

1 Current address: Department of Microbiology, Campus Box B175, University of Colorado Health Sciences Center, 4200 East Ninth Avenue, Denver, CO 80262. 2 To whom correspondence should be addressed at Department of Microbiology, Miami University, 40 Pearson Hall, Oxford, OH 45056. Fax: (513) 529-2431. E-mail: [email protected].

0147-619X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.

20

REPLICATION ELEMENTS OF pMJ101 PLASMID FROM Vibrio ordalii

and the role of this cryptic extrachromosomal element in the infections caused by V. ordalii. Recent work initiated the characterization of pMJ101 at the molecular and genetic level (Bidinost et al., 1994). A restriction map was obtained, and cloning and transformation experiments allowed the isolation of the replication region of pMJ101. The essential replication functions of this plasmid were localized to a 2.4-kb EcoRV–HindIII restriction fragment. Furthermore, recombinant clones carrying this fragment were able to replicate in Escherichia coli deficient in DNA polymerase I (PolI, PolA) or integration host factor, although DnaA was required for the stable maintenance of pMJ101. The ability of pMJ101 derivatives to replicate in the absence of PolI suggested that a pMJ101encoded protein is required for the initiation of its replication. This hypothesis was supported by the electrophoretic analysis of radiolabeled plasmid-mediated proteins that identified a 36kDa polypeptide encoded within the replication region of pMJ101. Transposon insertion within the DNA region encoding this protein abolished its expression as well as replication in the absence of PolI. It was also shown in replicon typing experiments that pMJ101 does not share detectable DNA sequence homology with other plasmids belonging to the 18 plasmid incompatibility groups already defined, strongly suggesting that it represents a new plasmid group. In this work, we describe the analysis of DNA sequences and gene products involved in the replication of pMJ101. In addition, the presence of pMJ101 single-stranded intermediates and the incompatibility between pMJ101 and pJM1 were examined by Southern blot DNA hybridization and conjugation experiments, respectively. MATERIALS AND METHODS Bacterial Strains, Culture Conditions, and Plasmids V. ordalii 45-S was used as the source of the plasmid pMJ101 (Schiewe et al., 1977). A nalidixic acid-resistant (Nal r) 3 derivative of V. an3

Abbreviations used: Amp, ampicillin; Nal R, nalidixic

21

guillarum 775, which harbors the pJM1 virulence plasmid (Crosa et al., 1977), was used in conjugation experiments to determine plasmid incompatibility between pMJ101 and pJM1. This Nal r derivative was obtained as described previously (Miller, 1972). E. coli strains JM109 (Yanisch-Perron et al., 1985), HB101 (Boyer and Roulland-Doussoix, 1969), and DH5a (Sambrook et al., 1989) were used as transformation hosts for molecular cloning and DNA sequencing experiments and were grown at 37°C in LB medium (Sambrook et al., 1989). V. ordalii was grown at 30°C in brain– heart medium supplemented with 1% (w/v) NaCl and 0.02% (w/v) MgCl 2 (Schiewe and Crosa, 1981). V. anguillarum was cultured at 30°C on trypticase soy broth supplemented with 1% NaCl (TN) or TN agar containing the appropriate antibiotics. All antibiotics were added to culture media to a final concentration of 20 mg/ml (unless otherwise stated). The plasmid pBluescript (Stratagene) was used as cloning vector to determine the nucleotide sequence of the pMJ101 replication region. Plasmid Conjugation and Incompatibility Experiments The recombinant plasmid pCL2, which is a pBR322 derivative that harbors the pMJ101 replication region and encodes resistance to ampicillin (Amp) (Bidinost et al., 1994), was introduced into a Nal r derivative of the wild-type strain of V. anguillarum 775 (pMJ1) by conjugation from E. coli HB101 using the helper plasmid pRK2073 as previously described (Actis et al., 1985). Transconjugants were selected on TN agar containing 50 mg/ml Nal and 500 mg/ml Amp after overnight incubation at 30°C. The incompatibility between the pMJ101 and the pJM1 replicons was tested by agarose gel electrophoresis and Southern blot analysis of plasmid DNA isolated from V. anguillarum 775 transconjugants harboring pMJ101 and pCL2. These transconjugants were repeatedly cultured at 30°C in TN broth and agar containing 200 acid resistant; ORF, open reading frame; RT-PCR, reverse transcription-polymerase chain reaction.

22

BIDINOST ET AL.

and 500 mg/ml Amp, respectively, before plasmid isolation. General Nucleic Acid Procedures Plasmid DNA was isolated from the Vibrio and E. coli strains by the method of Birnboim and Doly (1979) and purified by ultracentrifugation in CsCl– ethidium bromide density gradients (Sambrook et al., 1989). DNA restriction and modification enzymes (New England Biolabs) were used as suggested by the manufacturer. DNA agarose gel electrophoresis and Southern blot hybridization assays were done using standard protocols (Sambrook et al., 1989). DNA probes were amplified using the polymerase chain reaction and labeled with [a- 32P]dCTP using the random oligolabeling system (Feinberg and Vogelstein, 1983). DNA amplification reactions were done as described previously (Barancin et al., 1998) using pMJ101 as a template and the oligonucleotide primers shown in Fig. 1. The PCR products were purified using the GeneClean II Kit (Bio 101 Inc.) according to the manufacturer’s instructions. The oligonucleotide primers were synthesized using a Beckman Oligo 1000M DNA synthesizer. The DNA sequence of this region was determined by the dideoxynucleotide chain-termination method (Sanger et al., 1979) using the T7 Sequenase sequencing kit (Amersham Life Science). DNA and protein sequences were analyzed using the Blast network service at the National Center for Biotechnology Information and the University of Wisconsin Genetics Computer (GCG) software package. RNA Isolation and Reverse Transcription Total RNA was isolated with the Trizol reagent (Gibco BRL). Residual DNA was removed by the addition of RNase-free DNase I (Boehringer Mannheim) and incubation at 37°C for 30 min. First-strand synthesis of cDNA for reverse transcription-PCR (RT-PCR) was catalyzed using SuperScript II reverse transcriptase (Gibco BRL) under the experimental conditions described previously (Barancin et al., 1998). The repM cDNA was amplified using PCR with

primers 112 (59-CCAAGGGCAATAAGAAATTAACCC-39) and 113 (59-GGTAGTCTTTACGCTTGCGAG-39). The location of these primers within the replication region of pMJ101 is indicated in Fig. 1. Determination of the 59 End of the repM mRNA by S1 Mapping The repM transcription initiation site(s) was determined by S1 mapping as described previously (Wu and Janssen, 1997). The DNA probe used was prepared by PCR using primers 164 (59-GGTCTACGCTTCCCGCAAGCCG39) and 165 (59-GGGTTAATTTCTTATTGCCCTTGG-39). The location of these primers within the replication region of pMJ101 is indicated in Fig. 1. Primer 165 was end-labeled with [g- 32P]ATP using T4 polynucleotide kinase and used together with unlabeled primer 164 to PCR amplify the repM promoter region using pMJ101 as a template. This 474-bp fragment was purified using low-melting-point agarose gel electrophoresis and 10 6 cpm was coprecipitated with 60 – 80 mg of total RNA that was isolated from V. ordalii 45-S as described previously (Wu and Janssen, 1997). The precipitate was dissolved in hybridization buffer and incubated at 85°C for 15 min. To allow annealing, the temperature of the water bath was lowered to 49°C over a period of 3 h and continued overnight at 49°C. The samples were digested with nuclease S1 (New England Biolabs) and the double-stranded hybrids were analyzed by electrophoretic comparison under denaturing conditions to a ladder obtained by dideoxynucleotide sequencing using primer 165 and the plasmid pCL2 as a template. This plasmid is a pBR322 derivative that contains the complete pMJ101 replication region (Bidinost et al., 1994). The gel was dried and exposed to Kodak X-ray film using intensifying screens at 280°C. DNA Mobility Shift and Competition Assays A 474-bp DNA fragment from the pMJ101 replication region containing the five DnaA boxes was amplified by PCR with oligonucleotide primers 164 and 165. The fragment was

REPLICATION ELEMENTS OF pMJ101 PLASMID FROM Vibrio ordalii

purified using GeneClean and 59 end-labeled with [g- 32P]ATP using T4 polynucleotide kinase. Reactions (20 ml) containing 25 fmol (5 ng) of the 59 end-labeled fragment and various amounts (5– 40 ng) of DnaA were incubated for 10 min at room temperature in a buffer containing 20 mM Hepes–KOH (pH 8.0), 2.5 mM magnesium acetate, 15% glycerol, 0.4% Triton X-100, 2 mM EDTA, 4 mM dithiothreitol, 5 mg/ml bovine serum albumin, and 0.5 mM ATP. For competition assays, various amounts (5–200 ng) of unlabeled 474-bp fragment were added to a reaction containing 25 fmol of endlabeled fragment and 20 ng of DnaA. Samples were loaded onto a low-ionic-strength 6% polyacrylamide gel (19 parts acrylamide:1 part bisacrylamide; 1.5 mm thick; 15 cm long) and subjected to electrophoresis for 3– 4 h at 125 V in 0.53 TBE buffer (Sambrook et al., 1989). The gel was dried and exposed to Kodak X-ray film using intensifying screens at 280°C. Detection of Single-Stranded Plasmid Intermediates The presence of single-stranded pMJ101 DNA was tested by Southern blot hybridization of plasmid DNA transferred to nitrocellulose with and without prior alkaline denaturation (Chaussee and Hill, 1998; te Riele et al., 1986). Image Processing The images of ethidium bromide-stained agarose gels and autoradiograms were either scanned using the Agfa StudioScan IIsi scanner and the Fotolook 2.08 software program or digitally acquired using the Gel Print 2000i system from BioPhotonics. The images were then processed with Adobe Photoshop 4.0 and Canvas 3.5.5 computer programs. RESULTS AND DISCUSSION Cloning and Nucleotide Sequencing of the pMJ101 Origin of Replication We have previously shown that the pMJ101 replication functions are located within an EcoRV–HindIII restriction fragment (Bidinost et al., 1994). To analyze these replication func-

23

tions at the molecular level, this fragment was subcloned into pBluescript and its nucleotide sequence was determined by the dideoxynucleotide chain-termination method (Sanger et al., 1979) (GenBank Accession No. U68169). Figure 1 shows the nucleotide sequence of the pMJ101 replication region that has an overall G1C content of 40%. This is somewhat in contrast to the G1C content (44%) of either pMJ101 or V. ordalii chromosomal DNA (Schiewe and Crosa, 1981). This bias is most likely due to the large A1T-rich region, typical of DNA origins of replication (Bramhill and Kornberg, 1988), with a G1C content of 25% that maps between nucleotides 150 and 290 (Fig. 1). Two additional signatures of replication origins in prokaryotes were found in the pMJ101 replicon: (i) five copies of the 9-bp consensus binding sequence of the DnaA protein (Fuller et al., 1984) and (ii) 11 dam-methylation sites (GATC), 5 of which are located within the putative ori region (Fig. 1). A potential open reading frame (ORF), which starts with the GTG alternative initiation codon (Stormo et al., 1982) at position 598 and a stop at position 1477 with the termination codon TAA, was adjacent to the predicted DnaA boxes and DNA methylation sites. Computer analysis showed that this ORF has the potential to encode a 293-amino-acid hydrophilic polypeptide with a predicted isoelectric point and molecular size of 10.3 and 33,826 Da, respectively. The size of this virtual protein is very close to the 36-kDa radiolabeled polypeptide previously detected using an in vitro transcription–translation system and the plasmid DNA templates pCL2, pCL2.1, pCL2.3, and pCL13 (Bidinost et al., 1994). All of these are pBR322 derivatives that contain the complete pMJ101 replication region and replicate in the C2110 polI mutant strain of E. coli (Bidinost et al., 1994). This radiolabeled protein was not detected when the plasmids pCL44 and pCL46 were used as templates. These two plasmids do not replicate in the C2110 strain and carry Tn5 insertions mapping within the predicted ORF shown in Fig. 1 (Bidinost et al., 1994). Accordingly, we have designated this ORF as repM, the structural gene for the pMJ101 replication protein.

24

BIDINOST ET AL.

REPLICATION ELEMENTS OF pMJ101 PLASMID FROM Vibrio ordalii

The difference between the predicted size and the size obtained experimentally for RepM may be attributed to the high pI deduced for this protein, which most likely affects its electrophoretic mobility. A similar behavior was observed during the analysis of the RepI protein encoded within the REPI replication region of the plasmid pColV-K30 (Perez-Casal et al., 1989). This protein was estimated to have a pI of 10.69 and a molecular size of 35 kDa. However, RepI migrated as a 39-kDa protein upon sodium dodecyl sulfate–polyacrylamide gel electrophoresis. Although the hydrophilic and isoelectric properties of RepM are typical of DNA-binding proteins, no significant homology to other proteins was found in the GenBank database. Interestingly, it was previously shown that pMJ101 belongs to a new incompatibility group with replication and maintenance functions unrelated to already characterized plasmids (Bidinost et al., 1994; Schiewe and Crosa, 1981). Accordingly, we propose that repM encodes a distinct plasmid DNA replication protein with no apparent homology to known bacterial proteins. Our contention that the pMJ101 plasmid present in the marine pathogen V. ordalii belongs to a unique incompatibility group is in further agreement with recent reports (Dahlberg et al., 1997; Sobecky et al., 1997, 1998), which showed that plasmids from marine bacteria show no detectable homology with plasmids from clinical bacterial isolates. These results together with our data indicate that bacterial isolates obtained from infected fish and marine sediments are a large source of plasmids that belong to new incompatibility groups that remain to be characterized.

25

FIG. 2. Agarose gel electrophoresis of RT, RT-PCR, and PCR products and total RNA. Total RNA was isolated from V. ordalii and cDNA was made using reverse transcriptase and random hexamers as primers. The repM cDNA was detected by PCR amplification using primers 112 and 113. PCR amplification of total RNA and pMJ101 with the same set of primers was used as negative and positive controls, respectively. Lanes 1, HindIII-digested l DNA; 2, cDNA; 3, amplification of the cDNA shown in lane 2; 4, amplification of pMJ101 DNA; 5, amplification of total RNA; and 6, total RNA. The size in base pairs for some of the l HindIII fragments and the PCR fragments amplified from cDNA and pMJ101 is shown (in bp) on the left and right sides of the figure, respectively. RT-PCR was conducted under RNase-free conditions while PCR and agarose gel electrophoresis were done using standard conditions without any precaution to avoid RNase contamination.

Expression of the repM Gene The transcription of repM in V. ordalii was tested by RT-PCR using as a template total RNA treated with RNase-free DNase I. Lane 2 in Fig. 2 shows that cDNA was generated by the reverse transcription of RNA isolated from the wild-type strain. When this cDNA sample was used as a template for PCR using primers 112 and 113, which are located within the repM ORF (Fig. 1), a 788-bp fragment was detected by agarose gel electrophoresis (Fig. 2, lane 3). A DNA fragment displaying the same size was obtained by PCR amplification of pMJ101 DNA using the same set of primers (lane 4).

FIG. 1. Nucleotide sequence of the pMJ101 replication region and predicted amino acid sequence of the RepM polypeptide. Putative Shine-Dalgarno (SD) and 210 and 235 promoter regions are indicated by the horizontal bars. Putative Dam-methylation sites (GATC) are in italics and the conserved 7-bp repeats containing them are shown within the open boxes. The horizontal arrows indicate the position and orientation of DnaA boxes. The position of the two BglII sites located within the replication region are indicated by the black boxes. PCR primers used for various DNA amplifications are indicated by their numbers within the black rectangles. The repM transcription initiation sites determined by S1 mapping are indicated by the arrows pointing upward, with the larger arrow identifying the most prominent fragment protected from nuclease S1 digestion.

26

BIDINOST ET AL.

The PCR experiment was repeated using total RNA instead of cDNA as the template. The electrophoretic analysis of this reaction showed the absence of the 788-bp fragment and the presence of a faster migrating band that comigrated with the total RNA sample (compare lanes 5 and 6). The RNA nature of this faster migrating band present in lanes 5 and 6 was confirmed by its sensitivity to boiled DNasefree RNaseA (data not shown). These data confirmed that the RNA specimen was not contaminated with pMJ101 DNA and demonstrated that the presence of the 788-bp fragment indeed is the result of the transcription of repM in V. ordalii. These results, together with those obtained previously using Tn5 insertional mutagenesis and plasmid replication in E. coli C2110 (Bidinost et al., 1994), demonstrate that RepM is encoded within the replication region of pMJ101 and it is required for its replication in V. ordalii as well as in E. coli in the absence of DNA polymerase I. Location of the repM Transcription Initiation Site The proposed repM ORF is preceded by a putative GCAAG Shine–Dalgarno (1974) sequence and upstream promoter-like elements (Lisser and Margalit, 1993) (Fig. 1). The transcriptional initiation site was determined by nuclease S1 mapping using a 474-bp PCR fragment encompassing the putative repM promoter region as a probe. Three bands 1 base apart were detected when V. ordalii total RNA was hybridized with this 39-end radiolabeled probe and digested with nuclease S1 (Fig. 3, lane 1). These bands map the repM mRNA initiation site to the three consecutive Ts located at positions 410 through 412; however, the most intense band corresponds to position 411 (Fig. 1). No bands were detected when the radiolabeled probe was hybridized with E. coli tRNA (Fig. 3, lane 2), an unrelated RNA used to determine the specificity of the hybridization reaction. Furthermore, these three bands are smaller than the probe used to protect the 59 end of the repM mRNA, which was located in the top portion of the sequencing gel (data not shown). These re-

FIG. 3. Nuclease S1 mapping of repM mRNA. The 59 end-labeled probe, which was obtained by PCR amplification of pMJ101 with primers 164 and 165, was hybridized to either V. ordalii 45S total RNA (lane 1) or E. coli tRNA (lane 2). The hybridization reactions were treated with nuclease S1 and the fragments were separated by electrophoresis on a 6% polyacrylamide sequencing gel. The DNA sequencing ladders represent the sense strand; the arrows indicate the transcriptional initiation sites, with the major site indicated by the largest arrow. The position of a potential 210 promoter region is indicated to the right of the sequence.

sults indicate that the transcription of repM initiates at three contiguous sites, although preferential initiation occurs at position 411, which is located 187 nucleotides upstream from the GTG translation initiation codon. The nucleotide sequence analysis of the region upstream of the repM mRNA initiation site showed the presence of potential promoter elements such as the 235 and 210 regions indicated in Fig. 1, with the predicted 210 sequence of TAAAAT that maps 6 nucleotides upstream of the major transcription initiation site. This potential promoter element differs in only one nucleotide from the TATAAT consensus sequence reported previously (Lisser and Margalit, 1993). Taken together, all these results provide strong evidence that the predicted 210 and 235 promoter regions located upstream of repM transcription initiation site control the expression of this plasmid replication gene. Further-

REPLICATION ELEMENTS OF pMJ101 PLASMID FROM Vibrio ordalii

27

FIG. 4. Analysis of DnaA binding to the pMJ101 replication region by gel mobility shift assays. (A) Binding assay. The 474-bp fragment from pMJ101 containing the five DnaA boxes was PCR amplified using primers 164 and 165 and end-labeled with [ 32P]ATP and T4 polynucleotide kinase. Five nanograms of the end-labeled fragment (lanes 1–3) was incubated for 10 min at room temperature with 0 ng (lane 1), 5 ng (lane 2), and 40 ng (lane 3) of purified DnaA. (B) Competition assay. Five nanograms of the end-labeled 474-bp fragment (lanes 1– 6) was incubated for 10 min at room temperature with either 0 ng (lane 1) or 20 ng of DnaA (lanes 2– 6) together with 0 ng (lane 2), 5 ng (lane 3), 50 ng (lane 4), 100 ng (lane 5), and 200 ng (lane 6) of unlabeled 474-bp fragment. DNA–protein complexes were separated on a low-ionic-strength 6% polyacrylamide gel and visualized using autoradiography.

more, the location of the repM promoter region determined by this approach agrees with the previous observation that deletions within the BglII sites located within the putative promoter region abolish the pMJ101 replication functions (Bidinost et al., 1994). DnaA Binding to the pMJ101 Replication Region The analysis of the nucleotide sequence of the pMJ101 replication region revealed the presence of five DnaA boxes located 59 of the repM-coding region (Fig. 1). To determine whether these DnaA boxes are functional, we analyzed the binding of purified E. coli DnaA protein to a 474-bp DnaA box-containing fragment from pMJ101 using gel mobility shift assays. This fragment was amplified by PCR using primers 164 and 165 (Fig. 1) and endlabeled with [g- 32P]ATP using T4 polynucleotide kinase. Figure 4A shows that DNA–protein complexes were readily formed

when 5 ng of DnaA was added to the reaction mixture, causing a significant shift in the electrophoretic mobility of the labeled fragment (compare lanes 1 and 2). With the addition of 40 ng of DnaA, essentially all of the labeled DNA fragment was retarded (lane 3). To corroborate the specificity of the binding of DnaA to the 474-bp DNA probe, we examined the effect of adding different amounts of unlabeled 474-bp fragment as competitor DNA. With fixed amounts of DnaA protein (20 ng) and labeled DNA fragment (5 ng), the addition of 5 ng unlabeled DNA fragment increased slightly the electrophoretic mobility of the 474-bp labeled fragment (compare lanes 2 and 3 of Fig. 4B). Furthermore, the addition of 50, 100, and 200 ng of unlabeled competitor abolished completely the retardation of the labeled DNA fragment (lanes 4 – 6). These findings demonstrate that the DnaA boxes located upstream of the repM gene are active binding sites and provide further support for the requirement of this

28

BIDINOST ET AL.

DNA-binding protein, which is encoded by the chromosome of the host cell, for the replication of pMJ101 (Bidinost et al., 1994). Presence of Single-Stranded DNA Most plasmids in gram-negative bacteria replicate by theta replication without involving a single-stranded DNA intermediate. However, a number of gram-negative plasmids replicate via a rolling circle mode (del Solar et al., 1993), which is typical of gram-positive bacteria and includes the presence of single-stranded DNA intermediates. The lack of homology of the pMJ101 replicon to well-characterized DNA replication regions led us to examine whether replication of pMJ101 involves a singlestranded DNA molecule. For this purpose, total DNA from V. ordalii was analyzed by Southern blot hybridization using nondenatured and alkaline-denatured DNA, as it was described for the detection of single-stranded DNA during plasmid DNA replication in Bacillus subtilis and Staphylococcus aureus (te Riele et al., 1986) and DNA transformation of Neisseria gonorrhoeae (Chaussee and Hill, 1998). Although not shown, these experiments demonstrated that radiolabeled pMJ101 hybridized with V. ordalii total DNA and purified pMJ101 that were subjected to agarose gel electrophoresis and NaOH treatment prior to transfer to nitrocellulose. However, no signal was detected when the same two samples were transferred to nitrocellulose without prior alkaline treatment and hybridized with the same probe. These results indicate that single-stranded intermediates are not produced during pMJ101 replication, which suggests that this plasmid replicates via a theta replication mechanism as is normally found in most gramnegative plasmids. Incompatibility Analysis of pMJ101 Although DNA hybridization experiments showed that the V. ordalii pMJ101 and V. anguillarum pJM1 plasmids do not share detectable DNA sequence homology, even under lowstringency conditions (Schiewe and Crosa, 1981), their incompatibility was never tested functionally. In this work, we examined the

ability of pCL2 to coexist with pJM1 in a Nal r derivative of V. anguillarum 775 after triparental conjugation. The pCL2 plasmid is a pBR322 derivative that harbors the pMJ101 replication region and replicates in the absence of PolI (Bidinost et al., 1994). Numerous transconjugant colonies grew on TN agar plates containing 50 mg/ml Nal and 500 mg/ml Amp after incubation at 30°C. The plasmid content of eight transconjugants was analyzed after three passages on the same solid medium and culture in TN broth containing 200 mg/ml Amp. For simplicity, we show the results obtained with one of the V. anguillarum 775 transconjugants since all of them displayed the same behavior. Agarose gel electrophoresis and Southern blot hybridization of EcoRI-digested plasmid DNA with radiolabeled pMJ101 revealed the presence of pCL2, which displayed the same restriction and hybridization pattern as that of purified pCL2 digested with EcoRI (Figs. 5A and 5B). The presence of pJM1 in this transconjugant was tested by restriction analysis and Southern blot hybridization of SalI-digested plasmid DNA probed with radiolabeled pJM1 plasmid. Figures 5C and 5D show that the transconjugant contained a pJM1 plasmid that displayed the same hybridization profile as that detected with pJM1 isolated from the wild-type V. anguillarum 775 strain. These results show that selection for a recombinant plasmid harboring the replication region of pMJ101 did not cause the curing and/or rearrangement of the wild-type pJM1 plasmid when conjugated into V. anguillarum. These data are in agreement with the results published previously by Schiewe and Crosa (1981), which showed the absence of detectable homology between pMJ101 and pJM1, and demonstrate that the replication systems of these two plasmids are both structurally and functionally unrelated. In summary, this work shows some of the basic genetic and molecular elements involved in the replication of pMJ101, which are unique to this plasmid found in all V. ordalii strains examined so far. Despite this work, pMJ101 remains a largely uncharacterized extrachromosomal element whose role in the life cycle and/or virulence of V. ordalii is still unknown.

REPLICATION ELEMENTS OF pMJ101 PLASMID FROM Vibrio ordalii

29

obtain a V. ordalii 45-S plasmidless derivative. This derivative will be an invaluable tool to assess the functions encoded by this V. ordalii universal plasmid and its potential role in the pathogenesis of vibriosis. ACKNOWLEDGMENTS We thank to A. Rosa and L. C. Patrito (Universidad Nacional de Co´rdoba, Argentina) for suggestions and helpful discussions as well as encouragement and assistance during this project and J. H. Crosa (Oregon Health Sciences University) and M. E. Tolmasky (California State University Fullerton) for critically reading the manuscript. We also thank J. Kaguni (Michigan State University) for kindly providing us with purified DnaA and G. R. Janssen (Miami University) for helping us with S1 mapping. This work was supported by grants from Consejo Nacional de Investigaciones Cientı´ficas y Te´cnicas (CONICET) of Argentina, Consejo de Investigaciones Cientı´ficas y Tecnolo´gicas de Co´rdoba (CONICOR) to L.A.A. and L. C. Patrito, and research funds from Miami University to L.A.A. Contributions to this work were made by students taking the Microbial and Molecular Genetics Laboratory course, Department of Microbiology, Miami University. FIG. 5. Plasmid analysis of a V. anguillarum 775 transconjugant harboring pJM1 and pCL2. Plasmid DNA isolated from a transconjugant colony (lanes 1) was digested with either EcoRI (A and B) or SalI (C and D) and the restriction fragments were size-fractionated by electrophoresis in 0.8% agarose containing ethidium bromide (A and C). The bands shown in A and C were transferred to nitrocellulose and probed with pMJ101 (B) and pJM1 (D). Plasmids pCL2 (A and B, lane 2) and pJM1 (C and D, lane 2) purified by ultracentrifugation in a CsCl gradient were used as positive controls. HindIII-digested l DNA (A–D, lane 3) was used as a molecular weight marker.

A classical and simple genetic approach used to determine the role of extrachromosomal elements in bacteria is the isolation of a plasmidless isogenic derivative using different chemical and/or physical agents (Crosa et al., 1980; Tolmasky et al., 1993). Although not shown in this report, we failed to cure pMJ101 using different chemical agents and growth conditions, a phenomenon that is most likely due to the relative high copy number and stability of this plasmid within its natural bacterial host. Therefore, the analysis of the replication system of pMJ101 is important not only to understand how this relatively large plasmid replicates, but it is also essential to develop a more rational approach to

REFERENCES Actis, L. A., Potter, S., and Crosa, J. H. (1985). Ironregulated outer membrane protein OM2 of Vibrio anguillarum is encoded by virulence plasmid pJM1. J. Bacteriol. 161, 736 –742. Actis, L. A., Tolmasky, M. E., and Crosa, J. H. (1998). Vibriosis. In “Fish Diseases and Disorders, Vol. 3, Viral, Bacterial and Fungal Infections” (P. T. K. Woo and D. W. Bruno, Eds.), pp. 523–557. CAB International Wallingford, Oxon/New York. Barancin, C. E., Smoot, J. C., Findlay, R. H., and Actis, L. A. (1998). Plasmid-mediated histamine biosynthesis in the bacterial fish pathogen Vibrio anguillarum. Plasmid 39, 235–244. Bidinost, C., Crosa, J. H., and Actis, L. A. (1994). Localization of the replication region of the pMJ101 plasmid from Vibrio ordalii. Plasmid 31, 242–250. Birnboim, H. C., and Doly, J. (1979). A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7, 1513–1523. Boyer, H. W., and Roulland-Doussoix, D. (1969). A complementation analysis of the restriction and modification of DNA in Escherichia coli. J. Mol. Biol. 41, 459 – 472. Bramhill, D., and Kornberg, A. (1988). A model for initiation at origins of DNA replication. Cell 54, 915–918. Chaussee, M. S., and Hill, S. A. (1998). Formation of single-stranded DNA during DNA transformation of Neisseria gonorrhoeae. J. Bacteriol. 180, 5117–5122. Crosa, J. H. (1989). Genetics and molecular biology of

30

BIDINOST ET AL.

siderophore-mediated iron transport in bacteria. Microbiol. Rev. 53, 517–530. Crosa, J. H. (1997). Signal transduction and transcriptional and posttranscriptional control of iron-regulated genes in bacteria. Microbiol. Mol. Biol. Rev. 61, 319 –336. Crosa, J. H., Hodges, L., and Schiewe, M. (1980). Curing of a plasmid is correlated with an attenuation of virulence in the marine fish pathogen Vibrio anguillarum. Infect. Immun. 27, 897–902. Crosa, J. H., Schiewe, M., and Falkow, S. (1977). Evidence for plasmid contribution to the virulence of the fish pathogen Vibrio anguillarum. Infect. Immun. 18, 509 –513. Dahlberg, C., Linberg, C., Torsvik, V. L., and Hermansson, M. (1997). Conjugative plasmids from bacteria in marine environments show various degrees of homology to each other and are not related to well-characterized plasmids. Appl. Environ. Microbiol. 63, 4692– 4697. del Solar, G., Moscoso, M., and Espinosa, M. (1993). Rolling circle replicating plasmids from gram-positive and gram-negative bacteria: A wall falls. Mol. Microbiol. 8, 789 –796. Feinberg, A. P., and Vogelstein, B. (1983). A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132, 6 –13. Fuller, R. S., Funnell, B. E., and Kornberg, A. (1984). The dnaA protein complex with the E. coli chromosomal replication origin (oriC) and other DNA sites. Cell 38, 889 –900. Lisser, S., and Margalit, H. (1993). Compilation of E. coli mRNA promoter sequences. Nucleic Acids Res. 21, 1507–1516. Miller, J. (1972). “Experiments in Molecular Genetics” Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Perez-Casal, J. F., Gammie, A. E., and Crosa, J. H. (1989). Nucleotide sequence and expression of the minimum REPI replication region and incompatibility determinants of pColV-K30. J. Bacteriol. 171, 2195–2201. Ransom, D. P., Lannan, C. N., Rohovec, J. S., and Fryer, J. L. (1984). Comparison of histopathology caused by Vibrio anguillarum and Vibrio ordalii in three species of Pacific salmon. J. Fish Dis. 7, 107–115. Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989). “Molecular Cloning: A Laboratory Manual,” 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Sanger, F., Nicklen, S., and Coulsen, A. (1979). DNA sequencing with chain-termination inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463–5467. Schiewe, M. H., Crosa, J. H., and Ordal, E. J. (1977). Deoxyribonucleic acid relationships among marine vibrios pathogenic to fish. Can. J. Microbiol. 23, 954 – 950. Schiewe, M. H., and Crosa, J. H. (1981). Molecular characterization of Vibrio anguillarum biotype 2. Can. J. Microbiol. 27, 1011–1018. Schiewe, M. H., Trust, T. J., and Crosa, J. H. (1981). Vibrio ordalii sp. nov.: A causative agent of vibriosis in fish. Curr. Microbiol. 6, 343–348. Shine, J., and Dalgarno, L. (1974). The 39-terminal sequence of Escherichia coli 16S ribosomal RNA: Complementary to nonsense triplets and ribosome binding sites. Proc. Natl. Acad. Sci. USA 71, 1342–1346. Sobecky, P. A., Mincer, T. J., Chang, M. C., and Helinski, D. R. (1997). Plasmids isolated from marine microbial communities contain replication and incompatibility regions unrelated to those of known plasmid groups. Appl. Environ. Microbiol. 63, 888 – 895. Sobecky, P. A., Mincer, T. J., Chang, M. C., Toukdarian, A., and Helinski, D. R. (1998). Isolation of broad-hostrange replicons from marine sediment bacteria. Appl. Environ. Microbiol. 64, 2822–2830. Stormo, G. D., Schneider, T. D., and Gold, L. (1982). Characterization of translational initiation sites in E. coli. Nucleic Acids Res. 10, 2971–2996. te Riele, H., Michel, B., and Ehrlich, S. D. (1986). Are single-stranded circle intermediates in plasmid DNA replication? EMBO J. 5, 631– 637. Tolmasky, M. E., Actis, L. A., and Crosa, J. H. (1993). Virulence plasmids. In “Plasmids: A Practical Approach” (K. G. Hardy, Ed.), pp. 95–118. IRL Press, NY. Wu, C.-J., and Janssen, G. R. (1997). Expression of a streptomycete leaderless mRNA encoding chloramphenicol acetyltransferase in Escherichia coli. J. Bacteriol. 179, 6824 – 6830. Yanisch-Perron, C., Vieira, J., and Messing, J. (1985). Improved M13phage cloning vectors and host strains: Nucleotide sequence of the M13mp18 and pUC19 vectors. Gene 33, 103–109. Communicated by D. R. Helinski