|Home | About | Journals | Submit | Contact Us | Français|
Vibrio anguillarum is a fish pathogen that causes vibriosis, a serious hemorrhagic septicemia, in wild and cultured fish. Many serotype O1 strains of this bacterium harbor the 65kb plasmid pJM1 carrying the majority of genes encoding the siderophore anguibactin iron transport system that is one of the most important virulence factors of this bacterium. We previously identified a replication region of the pJM1 plasmid named ori1. In this work we determined that ori1 can replicate in E. coli and that the chromosome-encoded proteins DnaB, DnaC and DnaG are essential for its replication whereas PolI, IHF and DnaA are not required. The copy number of the pJM1 plasmid is 1–2, albeit cloned smaller fragments of the ori1 region replicate with higher copy numbers in V. anguillarum while in E. coli we did not observe an obvious difference of the copy numbers of these constructs which were all high. Furthermore, we were able to delete the ori1 region from the pJM1 plasmid and identified a second replication region in pJM1 that we named ori2. This second replication region is located on ORF25 that is within the trans-acting factor (TAFr) region, and showed that it can only replicate in V. anguillarum.
Vibrio anguillarum, commonly found in the aquatic environment, is the cause of vibriosis, a serious hemorrhagic septicemia of both wild and cultured fish (Actis et al., 2011; Naka and Crosa, 2011b). 23 serotypes have been reported in V. anguillarum, and serotypes O1 and O2 are the major causative agents of this disease (Sorensen and Larsen, 1986; Toranzo and Barja, 1990; Larsen et al., 1994). Many of the O1 serotype strains carry the 65kb plasmid pJM1, harboring the majority of genes involved in the siderophore anguibactin transport system, the most important virulence factor of this bacterium (Crosa, 1980; Crosa and Walsh, 2002; Di Lorenzo et al., 2003; Actis et al., 2011). In the pJM1 plasmid, there are several characteristic regions surrounded by transposon or insertion sequences (Fig. 1), and we proposed that pJM1 was acquired by V. anguillarum during evolution to gain the ability to compete for iron in environmental and/or host conditions where the amount of free iron is extremely limited (Tolmasky and Crosa, 1995; Di Lorenzo et al., 2003; Naka et al., 2008). The pJM1 iron transport operon (ITBO) carries the genes encoding enzymes for the biosynthesis of anguibactin (angR and angT) and transport (fatABCD), and also encodes antisense RNAs that influence the expression of the ITBO (Actis et al., 1985, 1988; Salinas et al., 1989; Farrell et al., 1990; Salinas et al., 1993; Tolmasky et al., 1993; Waldbeser et al., 1993; Actis et al., 1995; Salinas and Crosa, 1995; Waldbeser et al., 1995; Chen and Crosa, 1996; Chen et al., 1996; Wertheimer et al., 1999; Lopez et al., 2007; Stork et al., 2007; Naka et al., 2010) (Fig. 1). There is another region encoding the trans-acting factor (TAF) that is subdivided in two subregions TAFr and TAFb. TAFr is involved in the activation of the transcription of the ITBO operon via mechanisms that are still unknown, whereas TAFb carries the anguibactin biosynthesis genes, angBCDE (Tolmasky et al., 1988; Welch et al., 2000; Di Lorenzo et al., 2003) (Fig. 1). We identified a replication origin, named ori1 between open reading frames ORF49 and 50 (Chen, 1995; Di Lorenzo et al., 2003) (Fig. 1), and in this study, we discuss results of an in depth analysis of this replication origin. Furthermore, we identified that homologues of replication related genes potentially encoding, DNA helicase, ParA and ParB. The latter two are located within TAFr and thus far from ori1 (Di Lorenzo et al., 2003) (Fig. 1). In this work we also demonstrate that within TAFr there is a second origin of replication, ori2, that is able to replicate independently of ori1 in V. anguillarum and that is inactive in E. coli.
Strains, plasmids and primers used in this study are listed in Tables 1 and and2.2. Vibrio anguillarum strains were grown at 25°C in trypticase soy broth supplemented with 1% NaCl (TSBS) or on TSBS with 1.5% agar (TSAS). In some experiments, V. anguillarum was incubated in CM9 minimal medium (Naka and Crosa, 2011a). E. coli strains were grown at 37 °C in LB broth or M9 minimal medium (Sambrook and Russell, 2001), or on LB agar. When needed, antibiotics were added at the following concentrations, for E. coli, ampicillin (Amp) 100μg/ml, chloramphenicol (Cm) 30μg/ml and kanamycin (Km) 50μg/ml; for V. anguillarum, Cm 10μg/ml, Km 50μg/ml and rifampicin (Rif) 100μg/ml.
The pJM1 plasmid was digested with Sau3AI EcoRI, XhoI or BamHI. The obtained fragments containing ori1 were cloned into the appropriate restriction enzyme sites located at both ends of the 3kb Km resistance fragments. The plasmid copy number was determined by the method described by De Graaff et al. (1976). Briefly, 3–5 ml of exponential phase cultures of V. anguillarum H775-3 and E. coli strains harboring plasmids were labeled with [3H] thymine. Cell lysates were obtained by the modified method of Young and Sinsheimer (1967) by replacing sodium lauryl sulfate with 1.2% Sarkosyl. After adding cesium chloride and ethidium bromide in the cleared lysate, the refractive index was adjusted with the TES buffer to 1.3955. The samples were centrifuged in a Beckman type 50 rotor for longer than 40 hours, and the gradients were collected on Whatman filters. The DNA was precipitated by trichloroactic acid, and the radioactivity was measured in a liquid scintillation counter. Copy numbers were calculated by the following equation: copy number = [cpm of plasmid peak/cpm of chromosome peak] × [MWc (molecular weight of chromosome)/MWp (molecular weight of plasmid)].
To test whether any proteins synthesized from plasmid and/or chromosome were involved in the replication of Rep181, which was assessed by culturing E. coli HB101 harboring Rep181 with or without adding Cm, a protein synthesis inhibitor. pBR322 was used as a control since this plasmid can replicate exponentially in the presence of Cm (Clewell, 1972). E. coli HB101 carrying Rep181 or pBR322 were incubated overnight in M9 minimal medium with appropriate antibiotics. The overnight culture was diluted (50 times) into fresh M9 minimal medium, separated into two portions, and to one of the portions 170 μg/ml Cm were added. Samples were harvested at different time points, and plasmid DNA was extracted. Extracted DNA was linearized by digesting with PstI and examined using a 0.8% agarose gel. The intensity of DNA bands on the gel was compared, and the data was normalized to the cell density (OD600).
E. coli C2110 that is a DNA polymerase I (PolI) deficient strain was used to transform Rep181 to check the requirement of PolI for ori1. To test whether ori1 require DnaA and integration host factor (IHF), Rep181 was transformed into the following strains: EH3896 (dnaA+) and EH3894 (dnaA−) for DnaA, and N99 (IHF+) and IHF mutants such as HN545 (hip157) and HN678 (himA) for IHF, and analyzed the existence of colonies after overnight incubation at 37°C. The necessity of DnaB, DnaC and DnaG for ori1 replication was determined using the temperature sensitive E. coli strain mutants E107 (dnaB ts), PC1 (dnaC ts) and PC3 (dnaG ts), respectively. Rep181 was transformed into these strains, and incubated at 25°C. Obtained colonies were cultured into M9 minimal medium at 25°C until exponential phase and separated into two portions, one portion was incubated at 25°C and the other portion was incubated at 37°C. Plasmids were extracted from cells harvested at different time points and their banding profile and concentration were determined.
A Km resistance gene was amplified by PCR with primers, Km-SmaI F and Km-SmaI R using pKD4 as a template, and cloned into pBluescript generating pBlue-Km-SmaI. The DNA fragment containing angA was PCR amplified by using primers, angAp Km SpeI and angAp Km SacI, cloned into pCR-bluntII-TOPO (Invitrogen, Carlsbad, CA), and then subcloned into pBluescript. The plasmid was digested with StuI, and the Km cassette (SmaI sites in both sides) from pBlue-Km-SmaI was ligated in the StuI site. angA::Km thus obtained was moved into the SpeI and SacI site of the pDM4 plasmid generating pDM4-angA::Km. This plasmid was conjugated into V. anguillarum 775(pJM1), and first recombinants were selected by Cm and Km resistance, and grew on TSAS supplemented with 15% sucrose and Km to obtain second recombinants which were confirmed by Cm sensitivity, pJM1 plasmid digestion pattern and PCR.
Δori1 has been constructed by the allelic exchange method. DNA fragments that locate upstream and downstream of ori1 were amplified by PCR using primer sets, ori1-mut-upSacI and ori1-mut-upEcoRV, and ori1-mut-dnEcoRV and ori1-mut-dnXhoI, respectively. The amplified fragments were separately cloned into pCR-BluntII-TOPO (Invitrogen), and subsequently cloned together in pBluescript. The inserted DNA digested with SacI and XhoI was purified, and ligated into pDM4 generating pDM4-Δori1. This plasmid was conjugated into V. anguillarum by the method described before, and first recombinants were selected by antibiotics resistance to 100 μg/ml Rif and 10 μg/ml Cm. Second recombinants were obtained as described before.
The replication origin of the pJM1 plasmid of V. anguillarum 775 was first identified by our group within a stretch of 1.8 kbs (Chen, 1995). Wu et al. (2004) later, isolated the replication origin from pEIB1 that is highly related the pJM1 plasmid. The nucleotide sequence of the replication region is identical. We designated the replication origin as ori1 to avoid confusion with another replication origin mentioned later. We have made a series of constructs, Rep181, Rep182 and Rep183, encompassing the ori1 region (Fig. 1) that we cloned to the kanamycin (Km) resistance gene, and determined their copy number in V. anguillarum H775-3, a pJM1 cured strain, and E. coli HB101. Our results (Table 3) indicated that the copy number of the pJM1 plasmid in V. anguillarum is 1 to 2. Except for the shortest construct containing a 1.8 kb fragment encompassing ori1 that showed a high copy number, 65, the rest of the derivatives showed medium copy numbers, 7 to 16 indicating that some factors located outside of the 1.8kb fragment control the copy number of the pJM1 plasmid. However, we did not observe this effect when we determined the copy number of the pJM1 derivatives in E. coli (all constructs showed similar copy number: 50–75) indicating that this copy number control specifically occurs in V. anguillarum.
Two types of mechanisms for the initiation of plasmid replication have been reported. One mechanism found in ColE1 related plasmids, involves opening of the helix through an RNA transcript. The other mechanism found in other plasmids is that a plasmid-encoded Rep protein and/or chromosome-encoded factors bind to a specific region in an ori, and facilitate opening of the DNA helix (Kues and Stahl, 1989).
To test whether ori1 requires protein synthesis from chromosome or the plasmid for its replication, we tested the replication of Rep181 in the presence or absence of chloramphenicol (Cm) that is an inhibitor of protein synthesis. As shown in Fig. 2 in the absence of Cm the copy number of Rep181 increases along with the incubation time while it remains low even after 16 hours incubation (panel A and C). The pBR322 plasmid in which replication can be processed without proteins was used as a control, and as we expected, the copy number of this plasmid increased in the presence of Cm (panel B and C). From these results, we can conclude that proteins from the plasmid or the chromosome are essential for the replication of the pJM1 replication origin ori1.
It has been reported that many plasmids require proteins originated from the chromosome-encoded genes (Scott, 1984; Tolmasky et al., 2010). The DnaA protein plays a key role on the initiation of replication of the bacterial chromosome as well as several plasmids in combination with plasmid-encoded Rep proteins by unwinding of the DNA in the ori region and helicase loading (Bramhill and Kornberg, 1988; Kaguni, 1997; Messer et al., 2001). The integration host factor (IHF) is a DNA-binding protein that enhances the bending of DNA in the ori region through its interaction with DNA to promote initiation of replication (Friedman, 1988; Rice, 1997), and IHF requires for the replication of plasmids such as pSC101 and R6K (Biek and Cohen, 1989; Dellis and Filutowicz, 1991). DnaB (helicase), DnaC (helicase accessory protein) and DnaG (primase) also involved in the replication of some plasmids such as RK2 and ColE1 plasmids (Pinkney et al., 1988; Konieczny and Helinski, 1997; Wang et al., 2004). In this study, we determined which chromosome-encoded factors are essential for the replication of ori1 in E. coli. Our results shown in Table 4 indicate that ori1 replication requires DnaB, DnaC and DnaG while PolI, IHF and DnaA were not indispensable
Some plasmids carry several replication origins thus we determined whether ori1 is essential for the replication of the pJM1 plasmid. First, we constructed pJM1-Km in which a Km resistance gene from pKD4 inserted into the angA gene that has been shown dysfunctional due to the lack of the N-terminal region, and then deleted the ori1 replication region by the allelic exchange method. This experiment generated a pJM1 derivative in which the ori1 region from the pJM1-Km plasmid was deleted demonstrating that there must be another replication origin in the pJM1 plasmid.
In the pJM1 plasmid there is a region that carries genes for replication and partition related proteins such as parA, parB and the DNA helicase gene homologues (Di Lorenzo et al., 2003), thus we suspected that the second replication origin could be located around this region. We PCR amplified several fragments around this region, and cloned into the suicide vector pDM4 that carries the R6K origin (Milton et al., 1996). The pDM4 derivatives thus obtained were conjugated into V. anguillarum H775-3, the pJM1-cured strain. Table 5 shows the constructs that were successfully conjugated and that replicated in this strain indicating that the second replication origin is indeed located within these regions. The common overlapping region in all of the constructs corresponds approximately to ORF25. We then cloned the ORF25 gene in pDM4 generating pDM4-ORF25. When pDM4-ORF25 was transformed into E. coli with or without λpir that provides the π protein that R6K origin requires for the replication. Transformants were only obtained when E. coli carries λpir demonstrating that ORF25 does not replicate in E. coli (Table 5). Furthermore, we were only able to transform pJM1-Km but not pJM1-KmΔori1 into E. coli (Table 5). Next, we conjugated pDM4-ORF25 into V. anguillarum H775-3, and successfully obtained colonies while no colonies were obtained when empty pDM4 was attempted to be conjugated into H775-3 (Table 5). These results indicate that the second replication origin of the pJM1 plasmid is included within ORF25, and it replicates only in V. anguillarum. We designated this replication origin as ori2 (Fig. 1). Furthermore, the stability of pDM4-ORF25 was tested, and the results indicated that the pDM4-ORF25 was very stable in V. anguillarum H775-3 as compared with pDM4-ori1 that has been shown unstable in V. anguillarum (data not shown).
Many V. anguillarum O1 serotype strains carry the 65 kb plasmid pJM1 harboring genes encoding the siderophore anguibactin iron uptake system that is the most important virulence factor for this bacterium (Crosa, 1980; Crosa and Walsh, 2002; Di Lorenzo et al., 2003; Actis et al., 2011). We have identified two replication regions in the pJM1 plasmid. One is located between ORFs 49 and 50 (Chen, 1995), in this study we identified a second replication region on ORF25 that is located within the TAFr region involved in up-regulation of the iron transport and biosynthesis operon (ITBO) (Tolmasky et al., 1988; Welch et al., 2000; Di Lorenzo et al., 2003). ORF25 is located close to genes potentially encoding replication and partition related proteins such as DNA helicase (ORF19), ParB (ORF27) and ParA (ORF28). We previously reported that the region including ORF19-ORF45 in the pJM1 plasmid is flanked by two identical ISVme-like sequences (Di Lorenzo et al., 2003) (Fig. 1). This region includes ori2 and replication related proteins, and many unknown or hypothetical proteins indicating that the ori2 region might be acquired by a pJM1 ancestor plasmid carrying ori1 through transposition events and horizontal transfer. The Rep181 plasmid including the ori1 region was not stable in V. anguillarum while the ori2 plasmid, pDM4-ORF25 was stable in this bacterium suggesting that these replication and partition related proteins might be important for the stability of the ori1 region in V. anguillarum.
The ori1 region of the pJM1 plasmid can replicate in both V. anguillarum and E. coli and is similar to the replication origin found in pEBI1 (Wu et al., 2004), while the ori2 region can only replicate in V. anguillarum indicating that some V. anguillarum chromosome-encoded proteins are required for the ori2 replication in E. coli. We showed that the ori1 replication origin of the pJM1 plasmid requires DnaB, DnaC and DnaG but not PolI, IHF and DnaA. Since ori2 cannot replicate in E. coli, we were not able to test its requirement of chromosome-encoded proteins.
The copy number of the pJM1 plasmid was determined as 1–2 while shorter fragments carrying the ori1 region exhibited higher copy numbers as compared to the whole pJM1 plasmid. The pJHC9-8 plasmid carrying ori1, ori2 and possible replication proteins showed slightly higher copy number than the whole pJM1 plasmid and lower copy number than shorter constructs Rep181-184. Furthermore, the shortest constructs Rep181 showed much higher copy number as compared with rest of constructs. From these results we could speculate that some factors that negatively control copy number might exist in the region in Rep182 and Rep183 but absent from Rep181 increasing the copy number of Rep181. Also, ori2 and/or possible replication proteins found in the pJM1 plasmid could affect the copy number because Rep181-4 in which ori2 and possible replication proteins are missing showed higher copy number than pJHC9-8 that carries all ori1, ori2 and possible replication protein genes. These copy number controls occurred only in V. anguillarum because all the constructs show similar copy number in E. coli. Experiments are being carried out to assess the contribution of these two origins and/or the putative partition and replication proteins to pJM1 replication and maintenance.
This work was supported by Grant AI19018 from the National Institutes of Health to J.H.C.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.