Molecular evolution of Vibrio pathogenicity island-2 (VPI-2): mosaic structure among Vibrio cholerae and Vibrio mimicus natural isolates

William S. Jermyn and E. Fidelma Boyd

Department of Microbiology, UCC, National University of Ireland-Cork, Cork, Ireland

Correspondence
E. Fidelma Boyd
f.boyd{at}ucc.ie


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Vibrio cholerae is a Gram-negative rod that inhabits the aquatic environment and is the aetiological agent of cholera, a disease that is endemic in much of Southern Asia. The 57·3 kb Vibrio pathogenicity island-2 (VPI-2) is confined predominantly to toxigenic V. cholerae O1 and O139 serogroup isolates and encodes 52 ORFs (VC1758 to VC1809), which include homologues of an integrase (VC1758), a restriction modification system, a sialic acid metabolism gene cluster (VC1773–VC1783), a neuraminidase (VC1784) and a gene cluster that shows homology to Mu phage. In this study, a 14·1 kb region of VPI-2 comprising ORFs VC1773 to VC1787 was identified by PCR and Southern blot analyses in all 17 Vibrio mimicus isolates examined. The VPI-2 region in V. mimicus was inserted adjacent to a serine tRNA similar to VPI-2 in V. cholerae. In 11 of the 17 V. mimicus isolates examined, an additional 5·3 kb region encoding VC1758 and VC1804 to VC1809 was present adjacent to VC1787. The evolutionary history of VPI-2 was reconstructed by comparative analysis of the nanH (VC1784) gene tree with the species gene tree, deduced from the housekeeping gene malate dehydrogenase (mdh), among V. cholerae and V. mimicus isolates. Both gene trees showed an overall congruence; on both gene trees V. cholerae O1 and O139 serogroup isolates clustered together, whereas non-O1/non-O139 serogroup isolates formed separate divergent branches with similar clustering of strains within the branches. One exception was noted: on the mdh gene tree, V. mimicus sequences formed a distinct divergent lineage from V. cholerae sequences; however, on the nanH gene tree, V. mimicus clustered with V. cholerae non-O1/non-O139 isolates, suggesting horizontal transfer of this region between these species.


Abbreviations: CT, cholera toxin; TCP, toxin co-regulated pilus; VPI-2, Vibrio pathogenicity island-2


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Vibrio cholerae, the causative agent of cholera, is a major cause of morbidity and mortality in many parts of the world where clean water is not available; in some areas of Southern Asia, cholera is endemic. V. cholerae is a natural inhabitant of the aquatic ecosystem, and most V. cholerae isolates do not possess the ability to cause cholera. Among V. cholerae natural isolates, only O1 and O139 serogroups are associated with epidemic and pandemic cholera. The V. cholerae O139 serogroup first emerged in 1992 as a cause of epidemic cholera (Albert et al., 1993; Cholera Working Group, 1993). It is now believed that O139 serogroup isolates arose from an O1 strain that acquired the O139 antigen (Berche et al., 1994; Bik et al., 1995; Stroeher et al., 1997; Waldor & Mekalanos, 1994). All V. cholerae O1 and O139 serogroup isolates encode two main virulence factors: toxin co-regulated pilus (TCP) and cholera toxin (CT). TCP is encoded on a pathogenicity island, the Vibrio pathogenicity island (named VPI-1 in this paper), and CT is encoded by a filamentous phage, CTX{pi} (Karaolis et al., 1998; Kovach et al., 1996; Waldor & Mekalanos, 1996). Acquisition of TCP and CT by V. cholerae is sequential: first V. cholerae isolates acquired TCP, an essential intestinal colonization factor and the receptor for CTX{pi} (Taylor et al., 1987; Waldor & Mekalanos, 1996), and then they acquired CT via CTX{pi} (Waldor & Mekalanos, 1996). Several additional genomic regions have also been identified, predominantly among epidemic O1 and O139 serogroup isolates; these include RS1{pi}, Vibrio seventh pandemic island-I (VSP-I), VSP-II and VPI-2 (Faruque et al., 2002; Davis et al., 2002; Dziejman et al., 2002; Jermyn & Boyd, 2002; O'Shea et al., 2004a). RS1{pi} is associated with the CTX prophage in V. cholerae El Tor isolates and is required for the production of CTX{pi} (Davis & Waldor, 2000). VSP-I and VSP-II are genomic islands identified by microarray analysis among V. cholerae El Tor isolates (Dziejman et al., 2002). It was suggested by Dziejman et al. (2002) that the genes encoded on the VSP-I and VSP-II islands are likely to be responsible for the unique characteristics of the seventh pandemic (El Tor) strains.

VPI-2 is a 57·3 kb chromosomal region encoding ORFs VC1758 to VC1809 which has all the characteristics of a pathogenicity island (Jermyn & Boyd, 2002). All toxigenic V. cholerae O1 and O139 serogroup isolates contained VPI-2, whereas non-O1/nonO139 non-toxigenic isolates lacked the region. VPI-2 encodes several gene clusters: a type-1 restriction modification system, which may protect the bacteria from viral infection (Bickle & Kruger, 1993); a nannag gene cluster homologous to genes involved in sialic acid metabolism, which may act as a carbon and nitrogen source (Vimr et al., 2004); neuraminidase, which acts on higher-order gangliosides in the intestine converting them to GM1 gangliosides, with the subsequent release of sialic acids (Galen et al., 1992); and a region with homology to Mu phage. Among most V. cholerae O139 serogroup isolates, only a 20 kb region of VPI-2 was present, as ORFs VC1761 to VC1788 had been deleted, indicating the instability of the region (Jermyn & Boyd, 2002).

The species V. mimicus is closely related to V. cholerae; however, V. mimicus is phenotypically and genotypically distinct from V. cholerae and can be readily differentiated from V. cholerae (Boyd et al., 2000a; Byun et al., 1999; Davis et al., 1981; O'Shea et al., 2004a; Reen & Boyd, 2004). V. mimicus, unlike V. cholerae, is negative in sucrose, Voges–Proskauer, corn oil and Jordan tartrate reactions (Davis et al., 1981). Comparative sequence analysis of the housekeeping gene malate dehydrogenase (mdh) from V. mimicus isolates showed the mean pairwise divergence between V. cholerae and V. mimicus was approximately 10 %, which is equivalent to the divergence between Salmonella enterica LT2 and E. coli K-12 (Boyd et al., 1994, 2000a; Byun et al., 1999; O'Shea et al., 2004a). In addition, analysis of groE-I and groEL-II on chromosomes 1 and 2, respectively, of V. cholerae and V. mimicus also demonstrated the divergence of both species from one another (Reen & Boyd, 2004). The natural habitat of V. mimicus, similar to V. cholerae, is the aquatic ecosystem, where it has been found both as a free-living bacterium and in association with phytoplankton and crustaceans (Acuna et al., 1999; Campos et al., 1996). Consumption of V. mimicus-contaminated shellfish has been linked to the development of gastroenteritis (Acuna et al., 1999). The virulence determinants of V. mimicus have not been well characterized. Recently, it was demonstrated that some V. mimicus isolates also harbour VPI-1 (TCP) and CTX{pi} (CT) (Boyd et al., 2000b). Several V. mimicus strains isolated from clinical and environmental sources were shown to produce multiple toxins, including a haemolysin, zonula occludens toxin, a heat-stable enterotoxin, as well as a CT-like toxin (Campos et al., 1996; Chowdhury et al., 1994; Ramamurthy et al., 1994; Shi et al., 1998; Spira & Fedorka-Cray, 1983, 1984).

In this paper, we examined 17 V. mimicus clinical and environmental isolates of various serogroups for the presence of the 57·3 kb VPI-2 region. ORFs VC1773 to VC1787 were found to be present in all 17 V. mimicus stains examined. In addition, 11 V. mimicus isolates contained ORFs VC1758 (integrase) and VC1804 to VC1809. We reconstructed the phylogenetic relationship of the nanH (VC1784) gene among a range of V. cholerae and V. mimicus isolates and compared it to the species tree, as deduced from the housekeeping gene mdh. On the mdh gene tree, V. mimicus isolates formed a separate divergent lineage from V. cholerae. On the nanH gene tree, however, V. mimicus isolates clustered with V. cholerae non-O1/non-O139 isolates, indicating that the nanH gene from these species is closely related. Analysis of additional genes spanning VPI-2 among a subset of V. cholerae and V. mimicus were in agreement with the nanH data, suggesting that VPI-2 was horizontally transferred between these species.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains.
A total of 22 V. cholerae and 17 V. mimicus isolates were used in this study. These isolates were derived from both clinical and environmental sources with a wide geographic distribution (Table 1). V. cholerae and V. mimicus strains were grown in Luria–Bertani (LB) broth. All strains were stored at –70 °C in broth containing 20 % (v/v) glycerol.


View this table:
[in this window]
[in a new window]
 
Table 1. V. cholerae and V. mimicus strains used in PCR and Southern hybridization analysis

 
DNA isolation.
Chromosomal DNA was extracted from each isolate using the G-nome DNA isolation kit (BIO 101). A single colony was inoculated into 3 ml LB broth and incubated overnight, with shaking, at 37 °C. The cells were pelleted at 3000 r.p.m. for 5 min and resuspended in 1·85 ml suspension solution. The cells were lysed and treated with RNase and protease solutions. Following incubation at 37 °C for 2 h, any remaining proteins were precipitated out and the sample was centrifuged. Supernatants were transferred to a clean Eppendorf tube (1·5 ml) and treated with both 100 % ethanol and TE buffer. The DNA was spooled out and air-dried to remove excess alcohol. Following drying, the DNA was dissolved in TE buffer, pH 8, and stored at –20 °C.

Molecular analysis.
To assay for the presence of VPI-2 among V. mimicus isolates, 17 primer pairs that span the entire 57·3 kb VPI-2 region and two primer pairs that flank the island were used for PCR analysis (Jermyn & Boyd, 2002). PCR was performed in volumes of 20 µl containing 100 ng genomic DNA, 200 µM primer, 1·25 mM dNTPs, 50 mM MgCl2, and 1U Taq DNA polymerase (Bioline). The amplification conditions were pre-incubation at 96 °C for 60 s, followed by 30 cycles of 94 °C for 30 s, 48–55 °C (depending on the primer pair) for 30 s and 72 °C for a time chosen based on the size of the expected fragment (60 s kb–1). All PCRs were carried out in a PTC-200 Peltier Thermal Cycler (MJ Research). PCR products were resolved by electrophoresis on a 0·8 % agarose gel made with 1x TAE (consisting of 40 mM Tris/acetate and 1 mM EDTA) and containing ethidium bromide. PCR amplification products were visualized by UV transillumination.

Southern hybridization was carried out with DNA probes generated from PCR products from the reference V. cholerae strain N16961 as a template (Heidelberg et al., 2000). PCR products were purified using the ConceRT-PCR purification kit (Gibco, BRL). For Southern hybridization analysis, DNA from each strain was digested with the restriction enzyme EcoRI (Roche Molecular Biochemicals), and the fragments were separated by electrophoresis in 0·6 % TAE agarose. The gel was depurinated with 0·25 M HCl, denatured with a solution of 5 M NaCl and 10 N NaOH, and neutralized with 1 M Tris and 20x SSC. The fragments were transferred to nitrocellulose membranes by a posiblotter (Stratagene). DNA was fixed to the membrane by UV cross-linking. Approximately 100 ng probe DNA was conjugated to horseradish peroxidase using the ECL direct nucleic acid labelling system (Amersham Pharmacia Biotech), and after hybridization overnight, was detected by the ECL chemiluminescent substrate. Hybridization membranes were washed according to manufacturer's instructions.

Pulsed-field gel electrophoresis.
To determine whether the ORF VC1758 encoding an integrase was located in the proximity of VC1773–VC1787, we carried out pulsed-field gel electrophoresis (PFGE). Genomic DNA was prepared from the V. mimicus strains PT5 and 523-80 as follows. The bacteria were grown in LB broth overnight at 37 °C. The cells were centrifuged and resuspended in SE buffer (consisting of 75 mM NaCl and 25mM EDTA) to an OD620 of 1·0. An equal volume of the cell suspension and 2 % agarose was mixed and cast in a mould. The individual agarose inserts were treated with PEN buffer (0·5 M EDTA, 1 %, w/v, N-lauroylsarcosine, 1 mg Pronase ml–1) and washed with TE buffer (100 mM Tris/Cl, 10 mM EDTA). After processing, the inserts were incubated for at least 6 h with SfiI at 50 °C. Electrophoresis was performed with a CHEF-DR II electrophoresis system (Bio-Rad) in 1x TBE buffer (45 mM Tris, 45 mM boric acid, 1 mM EDTA) at 13 °C. Pulse ramps were 3 to 80 s for 21 h at 160 V in 1·0 % agarose. After electrophoresis, the gel was stained with ethidium bromide, washed and treated for Southern hybridization analysis.

DNA sequencing.
The primer pairs used for PCR amplification and subsequent sequencing are listed in Table 2. The nucleotide sequence of an internal fragment of six genes spanning the VPI-2 region was obtained: VC1764, encoding a methyl accepting subunit; nanE (VC1781), encoding N-acetylmannosamine-6-phosphate 2-epimerase; nagC (VC1783), encoding N-acetylglucosamine-6-phosphate deacetylase; nanH (VC1784), encoding neuraminidase; VC1799, encoding an integrase; and VC1806, encoding a replicase. For the nanH gene, fragments were amplified from the chromosomal DNA of 22 V. cholerae strains and three V. mimicus strains, whereas for the remaining five genes, five V. cholerae strains were examined, which included a classical strain O395, an O139 serogroup strain MO2, and three non-O1/non-O139 strains V46, V47 and V52, and a V. mimicus strain PT5. PCR was performed in a 20 µl reaction mixture using conditions as described earlier. PCR amplicons were purified using the ConceRT-PCR purification kit (Gibco-BRL) in accordance with manufacturer's instructions. Following purification, an aliquot of 10 µl was used as a sequencing template. All sequencing was carried out by MWG-Biotech.


View this table:
[in this window]
[in a new window]
 
Table 2. Oligonucleotide primer sequences used for PCR analysis and Southern hybridization probe generation

 
Bioinformatic and phylogenetic analysis.
The VC1764, nanE, nagC, nanH, VC1799 and VC1806 nucleotide sequences were aligned with the CLUSTAL W multiple sequence alignment program (http://www.ebi.ac.uk/clustalw/) (Higgins et al., 1996). The housekeeping locus mdh was also included in the analysis for the same set of strains (O'Shea et al., 2004a). Phylogenetic analysis was performed on the nanH and the mdh sequences with the Molecular Evolutionary Genetics Analysis (MEGA) version 2.1 (Kumar et al., 2001). Phylogenetic trees were constructed using the neighbour-joining method based on synonymous nucleotide sites, with corrections for multiple substitutions by the Jukes–Cantor method (Jukes & Cantor, 1969; Saitou & Nei, 1987; Kumar et al., 2001). Rates (per sites) of synonymous (kS) and non-synonymous (kN) substitutions were calculated by the method of Nei and Gojobori (1986). We also calculated the kN/kS ratio, which is a measure of the selective constraints on a gene. The VPI-2 region was analysed using the Basic Local Alignment Search Tool (BLAST) at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/COG/).


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Distribution of VPI-2 (ORFs VC1758–VC1809) among V. mimicus
Previously, we identified the 57·3 kb VPI-2 region among toxigenic V. cholerae isolates (Fig. 1). To determine whether the VPI-2 region was present in V. mimicus, PCR assays were carried out on 12 clinical and five environmental V. mimicus isolates using 17 primer pairs spanning the VPI-2 region (Tables 1 and 2). In all 17 V. mimicus isolates examined, positive PCR products were obtained with four of the 17 primer pairs: 148F/146B (encompassing VC1773–VC1776), 147F/146A (encompassing VC1777–VC1779), nagC1/nagA2 (encompassing VC1782–VC1783) and nanH3/nanH4 (encompassing VC1784). Thus, with primer pair 148F/146B, a 4·2 kb PCR band was obtained, and similarly with primer pair 147F/146A, a 2·3 kb PCR product was obtained from all strains examined (Fig. 2). The primer pair nagC1/nagA2 gave a 2·1 kb PCR product band and primer pair nanH3/nanH4 gave a 0·7 kb PCR product from all V. mimicus strains examined (Fig. 2). In addition, 10 clinical V. mimicus isolates and one environmental V. mimicus isolate gave positive PCR products with three additional primer pairs: nint3/nint4 (encompassing VC1758), VC3A/VC4A (encompassing VC1804–VC1807) and VC3B/VC4B (encompassing VC1806–VC1809). With the primer pair nint3/nint4, a 1·0 kb product was obtained; with the primer pairs VC3A/VC4A and VC3B/VC4B, 2·5 kb and 2·3 kb products were obtained, respectively (Fig. 2). To verify the PCR results, we carried out Southern hybridization analysis (Fig. 3). Using probes 148F, 147F, nagC and nanH, positive hybridization fragments were obtained from all 17 isolates, as expected, which span the region of interest. The probe RadC, spanning VC1786–VC1789, however, also gave a positive hybridization fragment in all isolates, even though PCR analysis failed to amplify this region. Using DNA probe nint spanning VC1758, and DNA probes VC3A and VC3B spanning VC1804–VC1809, Southern hybridization confirmed the presence of these regions in 11 V. mimicus isolates only. No hybridization fragments were obtained using probes specific for regions VC1759–VC1772 and VC1789–VC1803, demonstrating their absence in all V. mimicus isolates examined. In summary, examination of the results of the PCR and Southern hybridization analysis indicate that the VPI-2 in 11 V. mimicus isolates was 19·4 kb in size and consisted of 22 ORFs (VC1758, VC1773–VC1787 and VC1804–VC1809), whereas in six V. mimicus isolates the VPI-2 region consisted of a 14·1 kb region encompassing 15 ORFs VC1773–VC1787.



View larger version (25K):
[in this window]
[in a new window]
 
Fig. 1. Schematic representation and genetic organization of VPI-2 in V. cholerae O1 and O139 serogroups and V. mimicus isolates. ORFs are represented by open arrows indicating the direction of transcription. The spotted filled arrows indicate ORFs VC1773–VC1787. The dashed arrows indicate VC1758 (int gene) and VC1804–VC1809. Unfilled arrows represent the remaining VPI-2 ORFs. A black arrow represents the tRNA serine.

 


View larger version (49K):
[in this window]
[in a new window]
 
Fig. 2. PCR analysis of the VPI-2 region among 16 V. mimicus strains. Lanes: 1 and 20, 1 kb molecular mass marker; lane 2, V. cholerae strain N16961; lanes 3–18 V. mimicus isolates; lane 19, V. cholerae strain O395. (a) PCR amplification using the primer pair 148F and 146B; (b) PCR amplification using the primer pair nanH3 and nanH4; (c) PCR amplification using the primer pair VC3B and VC4B.

 


View larger version (45K):
[in this window]
[in a new window]
 
Fig. 3. Southern hybridization analysis of EcoRI-digested chromosomal DNA to verify the presence of the VPI-2 in clinical and environmental V. mimicus isolates. Lanes: 1, N16961; 2, PT5; 3, 9583; 4, 523-80; 5, 531-90; 6, CS16; 7, CS18; 8, CS61; 9, ES28; 10, ES26; 11, ES24; 12, ES13; 13, ES6. Lane 1 contained the V. cholerae strain N16961, whereas lanes 2–13 contained V. mimicus strains. (a) Hybridization analysis with DNA probe 148F encompassing ORFs VC1773–VC1776; (b) hybridization analysis with DNA probe VC3A encompassing ORFs VC1804–VC1807. No hybridization band was detected in five V. mimicus strains (lanes 6, 9–11 and 13), indicating that the ORFs VC1804–VC1807 were absent in these strains.

 
We used PCR analysis and DNA sequencing to establish whether the region VC1773–VC1787 was contiguous with VC1804–VC1809. Using the PCR primer pair NF2 (located within the ORF VC1784) and 1806R (located within the ORF VC1806), we obtained a PCR product of 2·9 kb from the 11 isolates containing the 19·4 kb VPI-2 region. Nucleotide sequencing of the 2·9 kb PCR product from V. mimicus strain PT5 showed the presence of the ORFs VC1785, VC1786, VC1787, VC1804 and VC1805. Analysis of the region VC1785–VC1805 in V. cholerae strain N16961 revealed a 7 bp sequence AACTATT in the 3' end of the ORF VC1787 and an identical 7 bp sequence in the 5' end of the ORF VC1804. However, examination of the sequence in V. mimicus strain PT5 revealed that the last 5 bp of VC1787 and the first 12 bp of VC1804 were deleted, so only one copy of the 7 bp sequence AACTATT remained intact.

Genomic mapping of the VPI-2 in V. mimicus
To determine whether the 19·4 kb VPI-2 in V. mimicus was inserted in the same insertion site as that in V. cholerae, we carried out PCR analysis using primers designed from the insertion site of the VPI-2 in V. cholerae. Using the primer pair tRNAF (located within the serine tRNA) and int4 (located within ORF VC1758), a PCR product of 1·3 kb was identified in 11 isolates containing the 19·4 kb VPI-2 region. Nucleotide sequencing of the 1·3 kb product from V. mimicus strain PT5 showed that VPI-2 was located adjacent to a serine tRNA. Sequencing revealed that there was an intergenic region of 313 bp between the serine tRNA and VC1758, the first ORF of VPI-2 in V. mimicus.

PFGE was used to determine if VC1758 was located in close proximity to the ORFs VC1773–VC1787, since PCR analysis failed to amplify any product using a primer pair designed between these regions. Using the probes nint3/nint4 (encompassing VC1758) and 148F/146B (encompassing VC1773–VC1776), an identical hybridization fragment of approximately 200 kb was obtained from V. mimicus strain PT5. Similarly for V. mimicus strain 523-80, an identical hybridization fragment of approximately 120 kb was obtained from each probe. This indicated that the ORFs VC1758 and VC1773–VC1776 were located in proximity to each other, on the same hybridization fragment, on the V. mimicus chromosome.

Genetic variation at the nanH locus
To determine the evolutionary history of the VPI-2 region among V. cholerae and V. mimicus, we sequenced a 611 bp (nucleotides 177–788) fragment of the nanH gene from 22 V. cholerae strains and three V. mimicus isolates. Sequencing revealed variation at 49 nucleotide sites among the 22 V. cholerae nanH sequences examined, 41 of which were synonymous substitutions (silent sites) and eight of which were non-synonymous substitutions (amino acid replacement sites) (Table 3, Fig. 4). No nucleotide differences were found between the epidemic V. cholerae O1 and O139 isolates examined. Of the 49 polymorphic sites identified, there were 14 singleton sites found among V. cholerae non-O1/non-O139 serogroup isolates. For example, V. cholerae O141 serogroup strain V46 contained seven singleton sites, O54 serogroup strain SG14 contained three singleton sites and O37 serogroup strain 151 contained two singleton sites. The maximum nanH sequence difference between V. cholerae strains was 5·23 % between strains 151 and V46.


View this table:
[in this window]
[in a new window]
 
Table 3. Sequence variation in the 611 bp nanH gene fragment of VPI-2 and in the 648 bp fragment of the chromosomally encoded mdh gene derived from V. cholerae and V. mimicus strains

Abbreviations: Total poly., total polymorphic sites; Syn., synonymous sites; Non-syn., non-synonymous sites; kS, rate per site of synonymous substitution; kN, rate per site of non-synonymous substitution.

 


View larger version (31K):
[in this window]
[in a new window]
 
Fig. 4. (a) Distribution of the polymorphic bases within the nanH gene of V. cholerae and V. mimicus. The nanH sequence of V. cholerae strain N16961 was used as the consensus sequence and only polymorphic sites in the nanH gene are shown. Nucleotides identical to those of N16961 are represented by dots. The position of each nucleotide in the gene is indicated by the value at the top of the alignment. The 1, 2 or 3 at the bottom of the alignment indicates the position of each base within a codon. The asterisks indicate non-synonymous substitutions resulting in amino acid changes. Alleles numbered 1 to 8 indicate grouping of identical sequences. (b) Distribution of the polymorphic amino acid residues within the NanH protein sequence of V. cholerae and V. mimicus.

 
All three V. mimicus strains exhibited identical nanH nucleotide sequence to one another and contained three unique polymorphic sites compared to the V. cholerae nanH sequence (nucleotides 314, 549 and 609) (Table 3, Fig. 4). The mean pairwise percentage difference among the V. cholerae and V. mimicus nanH sequences was 2·00 %.

Of the 203 NanH amino acid sites examined, nine polymorphic amino acid sites were found among the V. cholerae and V. mimicus strains (Fig. 4). Eight amino acid polymorphisms were found among V. cholerae isolates and three among V. mimicus isolates, with one amino acid polymorphic site unique to V. mimicus. Six of the nine amino acid replacement polymorphisms were conservative and involved substitution of amino acids with similar properties. The remaining three amino acid polymorphisms all involved substitution of a hydrophobic residue with a hydrophilic one.

Sequence analysis of the 22 V. cholerae mdh sequences revealed variation at 24 nucleotide sites, of which 23 were synonymous substitutions and one was a non-synonymous substitution (Table 3). All three V. mimicus strains exhibited identical mdh sequences to one another, and contained 58 unique polymorphic sites compared to V. cholerae mdh sequences, two of which resulted in amino acid changes (Table 3). The mean pairwise percentage difference among the V. cholerae and V. mimicus mdh sequences was 3·73 % and the maximum difference was 12·03 %.

The variability observed at the nanH locus was greater than that observed at the mdh locus for the same set of isolates. To determine whether the selective constraints at these loci are different, that is, whether nanH is under positive selection or not, we examined the synonymous (kS) and non-synonymous (kN) substitution rates and calculated the kN/kS ratio for each gene (Table 3). Amino acid replacements in a protein that incur an advantage to the cell are fixed at a higher rate than silent site substitutions resulting in a kN/kS ratio greater than 1. Amino acid replacements in a protein that are deleterious to the cell are fixed at a lower rate by purifying selection, with a kN/kS ratio less than 1. Substitutions in a protein that are neither selectively advantageous nor deleterious are neutral, with a kN/kS ratio equal to 1. The nanH gene kN/kS ratio was not high, indicating that this gene is not under positive selection (Table 3).

Evolutionary genetic relationships of VPI-2 among V. cholerae and V. mimicus
To reconstruct the evolutionary history of the VPI-2 region from V. cholerae and V. mimicus, a phylogenetic tree based on the nanH gene was constructed. The nanH sequences from 25 isolates were analysed using the Jukes–Cantor distance method and a neighbour-joining gene tree was constructed based on synonymous polymorphic sites only (Fig. 5). From Fig. 5 it can be seen that all V. cholerae O1 and O139 serogroup strains clustered together on the nanH gene tree. This cluster also included two V. cholerae non-O1/non-O139 serogroup strains: V52, a toxigenic O37 serogroup strain; and V45, a rough isolate, which gave an identical nanH sequence to the O1 and O139 serogroup isolates. The other V. cholerae non-O1/non-O139 serogroup strains were grouped into two divergent clusters on the nanH gene tree. The first cluster contained V. cholerae strains V46, V47, V51 and SG14. The second cluster contained the strains V54, 9581, 9582, 151 and E714 in addition to V. mimicus strains PT5, PT48 and 9583.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 5. Evolutionary relationships based on synonymous site variation in the nanH and mdh genes among V. cholerae and V. mimicus isolates. Phylogenetic trees were constructed by the neighbour-joining method based on the Jukes–Cantor distance method (Jukes & Cantor, 1969; Saitou & Nei, 1987). Bootstrap values based on 1000 computer-generated trees are indicated at the branch node.

 
A comparison of the nanH gene tree with the species tree, based on the chromosomal housekeeping gene mdh, revealed a similar clustering of strains in each tree, with the exception of the V. mimicus isolates (Fig. 5). The V. cholerae O1 and O139 serogroup isolates formed a cluster, which included strains V45, V52 and V54 on the mdh gene tree. Similarly, V. cholerae non-O1/non-O139 strains V46, V47, V52 and SG14 clustered together, as did strains 151, 9581, 9582 and E714, on separate branches of the mdh tree. However, a notable feature of the gene trees is the placement of V. mimicus isolates: unlike the nanH gene tree, where V. mimicus isolates clustered with V. cholerae isolates, on the mdh gene tree V. mimicus isolates formed a separate distinct divergent lineage. The clustering of the nanH sequence from V. mimicus with V. cholerae indicates that horizontal transfer of nanH between V. cholerae and V. mimicus has occurred. Interestingly, the branch lengths of the mdh tree among V. cholerae isolates are short, which suggests that V. cholerae isolates are closely related and diverged from one another only very recently. The branch lengths on the nanH tree are longer than the mdh gene tree, which suggests that this region is evolving faster (Fig. 5).

To further determine the evolutionary history of the entire VPI-2 region among V. cholerae and V. mimicus isolates, we sequenced gene fragments from five additional ORFs, VC1764, VC1781, VC1783, VC1799 and VC1806, which spanned VPI-2. Nucleotide analysis of the five ORFs, for six V. cholerae strains, N16961, O395, MO2, V46, V47 and V52, provided similar findings to the nanH (VC1784) gene sequence analysis (Table 4). Strains N16961, O395, MO2 and V52 exhibited identical nucleotide sequences for the five ORFs VC1764, VC1781, VC1783, VC1799 and VC1806. Four ORFs, VC1781, VC1783, VC1799 and VC1806, of strains V46 and V47 exhibited different nucleotide sequences when compared to the consensus sequence in strain N16961. In contrast, the nucleotide sequence of the ORF VC1764 was identical between V46, V47 and N16961. Comparison of the nucleotide sequence of three ORFs, VC1781, VC1783 and VC1806, between V. mimicus strain PT5 and V. cholerae strain N16961 revealed similar findings to the nanH (VC1784) nucleotide comparison.


View this table:
[in this window]
[in a new window]
 
Table 4. Number of variable sites in selected genes of VPI-2 between the consensus sequence of V. cholerae O1 strain N16961 and the sequences of V. cholerae and V. mimicus strains

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In this study, we identified regions homologous to VPI-2 from V. cholerae in V. mimicus isolates. Of the 17 V. mimicus clinical and environmental isolates examined, all isolates contained a 14·1 kb region of VPI-2, which encompassed ORFs VC1773–VC1787, encoding homologues of neuraminidase and genes involved in sialic acid metabolism (Fig. 1). This region in V. mimicus was inserted adjacent to a serine tRNA similar to the VPI-2 region in V. cholerae. In addition, 11 of the 17 V. mimicus isolates examined also contained VPI-2 ORFs VC1758 (encoding an integrase) and VC1804–VC1809. Further genomic mapping of these regions in V. mimicus indicated that ORFs VC1773–VC1787 were located upstream of VC1804–VC1809; however, PCR analysis failed to amplify any region between VC1758 and VC1773–1787 or VC1804–1809. Nonetheless, by PFGE analysis VC1758 was found located on the same genomic fragment as VC1773–VC1787 and VC1804–VC1809, indicating that additional DNA may be present between VC1758 and VC1773 which was outside the range of PCR amplification. Previously, we showed that, among V. cholerae O139 serogroup isolates, VPI-2 was truncated, with only a 20 kb region (VC1758–1760 and VC1789–VC1809) present, which included a number of Mu phage-like genes (Jermyn & Boyd, 2002). These data and the data presented in this study suggest that the VPI-2 is highly unstable. The size differences and instability of VPI-2 may be associated with the presence of a Mu-like phage, since Mu phage insertion and replication occurs by transposition and thus can induce chromosomal rearrangements such as deletion and insertion events.

The deletion of the type 1 restriction modification gene cluster and the nannag gene cluster from the V. cholerae O139 serogroup isolates may be the reason why this serogroup is no longer a dominant cause of epidemic cholera on the Indian subcontinent. The retention of VPI-2 in toxigenic V. cholerae O1 serogroup isolates suggests that this region may provide a competitive advantage in vivo, whereas the presence of the nannag region in all V. mimicus isolates examined suggests that it may be an essential region for survival in the environment. Interestingly, the nannag gene cluster is also present in the genomes of Vibrio vulnificus strains CMCP6 and YJ016 as well as Vibrio fischeri strain ES114, but is absent from the genome of Vibrio parahaemolyticus strain RIMD2210633 (W. S. Jermyn, unpublished data). Among Vibrio species sequenced to date, only V. cholerae and V. mimicus have been shown to encode neuraminidase (nanH), which cleaves sialic acid from sialoglycoconjugates (Corfield, 1990). V. vulnificus, V. parahaemolyticus and V. fischeri, however, encode N-acetylneuraminic acid synthase, which can synthesize sialic acid de novo, whereas V. cholerae and V. mimicus must acquire sialic acid from an external source.

There were 49 polymorphic substitutions among the nanH sequences examined in V. cholerae, which was approximately twice as many polymorphic sites compared to mdh in the same set of strains (Table 3). This could indicate that nanH is under some selective pressure for change or that there is relax pressure at this locus. Analysis of the nanH sequence demonstrated that most of the polymorphic sites were silent site substitutions that do not result in amino acid changes, and the kN/kS ratio at this locus was much lower than 1, indicating that mutations in this gene are probably deleterious to the host (Table 3). In fact, there was a limited number of replacement polymorphisms at this locus, indicating functional constraints on this protein. Analysis of the sequenced nanH fragment revealed that a number of key features and motifs, which are essential for enzyme function, were conserved. For example, in the active site of NanH, two arginines that stabilize the carboxylic acid group of sialic acid were conserved. Also the FRIP region, which is involved in substrate affinity, was conserved. This indicated that the enzyme function was maintained between Vibrio isolates.

In order to elucidate the evolutionary history of VPI-2 among V. cholerae and V. mimicus, comparative sequence analysis of the nanH gene tree with the species tree, deduced from the housekeeping gene mdh, was carried out. In a number of studies, the mdh locus has been shown to be an excellent indicator of the overall genetic diversity of a species, since it is not under selective pressure and is present in all isolates of a species (Boyd et al., 1994, 1996; Byun et al., 1999; O'Shea et al., 2004a; Reen & Boyd, 2004). Clustering of V. cholerae isolates on the nanH gene tree was similar to the clustering of isolates on the mdh gene tree; however, the congruence between the trees did not extend to nanH from V. mimicus. On the nanH gene tree, V. mimicus did not form a separate divergent branch from V. cholerae as it does on the mdh tree, but instead clustered with V. cholerae non-O1/non-O139 isolates (Fig. 5). We examined five additional loci on the VPI-2 among six V. cholerae isolates and one V. mimicus isolate (Table 4). These five loci, which span the nannag gene cluster and regions in the 5' and 3' of VPI-2, overall gave a similar pattern of divergence and clustering as nanH. This indicates that nanH, and by extension the nannag genes, were horizontally transferred between V. mimicus and V. cholerae. We favour an evolutionary scenario in which VPI-2 was present in the most recent common ancestor of V. mimicus and horizontally transferred to V. cholerae soon after these species diverged. Horizontal transfer between V. cholerae and V. mimicus has previously been documented (Boyd et al., 2000b; Faruque et al., 1999). The presence of both CTX{pi} and VPI in V. mimicus natural isolates has been demonstrated (Boyd et al., 2000b). Nucleotide sequence identity of genes from both CTX{pi} and VPI derived from V. mimicus and V. cholerae has been reported, which strongly suggests that contemporary horizontal transfer of DNA between these species has occurred (Boyd et al., 2000b). Both V. cholerae and V. mimicus occupy similar environmental niches and it is possible that V. mimicus acts as an environmental reservoir of novel DNA for V. cholerae. Horizontal gene transfer between Vibrio species appears to be an emerging theme in the evolution of this genus, since transfer of a 8·5 kb region encoded on the Vibrio seventh pandemic island-II (VSP-II) between V. cholerae and V. vulnificus has also recently been described (O'Shea et al., 2004b).


   ACKNOWLEDGEMENTS
 
Research in E. F. B.'s laboratory is funded by a PRTLI-3 grant from the Higher Education Authority Ireland and an Enterprise Ireland basic research grant.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Acuna, M. T., Diaz, G., Bolanos, H., Barquero, C., Sanchez, O., Sanchez, L. M., Mora, G., Chaves, A. & Campos, E. (1999). Sources of Vibrio mimicus contamination of turtle eggs. Appl Environ Microbiol 65, 336–338.[Abstract/Free Full Text]

Albert, M. J., Siddique, A. K., Islam, M. S., Faruque, A. S., Ansaruzzaman, M., Faruque, S. M. & Sack, R. B. (1993). Large outbreak of clinical cholera due to Vibrio cholerae non-O1 in Bangladesh. Lancet 341, 704.[Medline]

Berche, P., Poyart, C., Abachin, E., Lelievre, H., Vandepitte, J., Dodin, A. & Fournier, J. M. (1994). The novel epidemic strain O139 is closely related to the pandemic strain O1 of Vibrio cholerae. J Infect Dis 170, 701–704.[Medline]

Bickle, T. A. & Kruger, D. H. (1993). Biology of DNA restriction. Microbiol Rev 57, 434–450.[Medline]

Bik, E. M., Bunschoten, A. E., Gouw, R. D. & Mooi, F. R. (1995). Genesis of the novel epidemic Vibrio cholerae O139 strain: evidence for horizontal transfer of genes involved in polysaccharide synthesis. EMBO J 14, 209–216.[Abstract]

Boyd, E. F., Nelson, K., Wang, F.-S., Whittam, T. S. & Selander, R. K. (1994). Molecular genetic basis of allelic polymorphism in malate dehydrogenase (mdh) in natural populations of Escherichia coli and Salmonella enterica. Proc Natl Acad Sci U S A 91, 1280–1284.[Abstract]

Boyd, E. F., Wang, F.-S., Whittam, T. S. & Selander, R. K. (1996). Molecular genetic relationships of the salmonellae. Appl Environ Microbiol 62, 804–808.[Abstract]

Boyd, E. F., Heilpern, A. J. & Waldor, M. K. (2000a). Molecular analysis of a putative CTX{pi} precursor and evidence for independent acquisition of distinct CTX{pi}s by toxigenic Vibrio cholerae. J Bacteriol 182, 5530–5538.[Abstract/Free Full Text]

Boyd, E. F., Moyer, K. L., Shi, L. & Waldor, M. K. (2000b). Infectious CTX{pi} and the Vibrio pathogenicity island prophage in Vibrio mimicus: evidence for recent horizontal transfer between V. mimicus and V. cholerae. Infect Immun 68, 1507–1513.[Abstract/Free Full Text]

Byun, R., Elbourne, L. D., Lan, R. & Reeves, P. R. (1999). Evolutionary relationships of pathogenic clones of Vibrio cholerae by sequence analysis of four housekeeping genes. Infect Immun 67, 1116–1124.[Abstract/Free Full Text]

Campos, E., Bolanos, H., Acuna, M. T., Diaz, G., Matamoros, M. C., Raventos, H., Sanchez, L. M., Sanchez, O. & Barquero, C. (1996). Vibrio mimicus diarrhea following ingestion of raw turtle eggs. Appl Environ Microbiol 62, 1141–1144.[Abstract]

Cholera Working Group (1993). Large epidemic of cholera-like disease in Bangladesh caused by Vibrio cholerae non-O139 Bengal. Lancet 342, 387–390.[Medline]

Chowdhury, M. A., Hill, R. T. & Colwell, R. R. (1994). A gene for the enterotoxin zonula occludens toxin is present in Vibrio mimicus and Vibrio cholerae O139. FEMS Microbiol Lett 119, 377–380.[CrossRef][Medline]

Corfield, T. (1990). Bacterial sialidases - roles in pathogenicity and nutrition. Glycobiology 2, 509–521.

Davis, B. M. & Waldor, M. K. (2000). CTX{pi} contains a hybrid genome derived from tandemly integrated elements. Proc Natl Acad Sci U S A 97, 8572–8577.[Abstract/Free Full Text]

Davis, B. M., Kimsey, H. H., Kane, A. V. & Waldor, M. K. (2002). A satellite phage-encoded antirepressor induces repressor aggregation and cholera toxin gene transfer. EMBO J 21, 4240–4249.[Abstract/Free Full Text]

Davis, B. R., Fanning, G. R., Madden, J. M., Steigerwalt, A. G., Bradford, H. B., Jr, Smith, H. L., Jr & Brenner, D. J. (1981). Characterization of biochemically atypical Vibrio cholerae strains and designation of a new pathogenic species, Vibrio mimicus. J Clin Microbiol 14, 631–639.[Medline]

Dziejman, M., Balon, E., Boyd, D., Fraser, C. M., Heidelberg, J. F. & Mekalanos, J. J. (2002). Comparative genomic analysis of Vibrio cholerae: genes that correlate with cholera endemic and pandemic disease. Proc Natl Acad Sci U S A 99, 1556–1561.[Abstract/Free Full Text]

Faruque, S. M., Rahman, M. M., Asadulghani, Nasirul Islam, K. M. & Mekalanos, J. J. (1999). Lysogenic conversion of environmental Vibrio mimicus strains by CTXPhi. Infect Immun 67, 5723–5729.[Abstract/Free Full Text]

Faruque, S. M., Asadulghani Kamruzzaman, M., Nandi, R. K., Ghosh, A. N., Nair, G. B., Mekalanos, J. J. & Sack, D. A. (2002). RS1 element of Vibrio cholerae can propagate horizontally as a filamentous phage exploiting the morphogenesis genes of CTXphi. Infect Immun 70, 163–170.[Abstract/Free Full Text]

Galen, J. E., Ketley, J. M., Fasano, A., Richardson, S. H., Wasserman, S. S. & Kaper, J. B. (1992). Role of Vibrio cholerae neuraminidase in the function of cholera toxin. Infect Immun 60, 406–415.[Abstract]

Heidelberg, J. F., Eisen, J. A., Nelson, W. C. & 23 other authors (2000). DNA sequence of both chromosomes of the cholera pathogen Vibrio cholerae. Nature 406, 477–483.[CrossRef][Medline]

Higgins, D. G., Thompson, J. D. & Gibson, T. J. (1996). Using CLUSTAL for multiple sequence alignments. Methods Enzymol 266, 383–402.[Medline]

Jermyn, W. S. & Boyd, E. F. (2002). Characterization of a novel Vibrio pathogenicity island (VPI-2) encoding neuraminidase (nanH) among toxigenic Vibrio cholerae isolates. Microbiology 148, 3681–3693.[Medline]

Jukes, T. H. & Cantor, C. R. (1969). Evolution of Protein Molecules. New York: Academic Press.

Karaolis, D. K., Johnson, J. A., Bailey, C. C., Boedeker, E. C., Kaper, J. B. & Reeves, P. R. (1998). A Vibrio cholerae pathogenicity island associated with epidemic and pandemic strains. Proc Natl Acad Sci U S A 95, 3134–3139.[Abstract/Free Full Text]

Kovach, M. E., Shaffer, M. D. & Peterson, K. M. (1996). A putative integrase gene defines the distal end of a large cluster of ToxR-regulated colonization genes in Vibrio cholerae. Microbiology 142, 2165–2174.[Medline]

Kumar, S., Tamura, K., Jakobsen, I. B. & Nei, M. (2001). MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17, 1244–1245.[Abstract/Free Full Text]

Nei, M. & Gojobori, T. (1986). Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol Biol Evol 3, 418–426.[Abstract]

O'Shea, Y. A., Reen, F. J., Quirke, A. M. & Boyd, E. F. (2004a). Evolutionary genetic analysis of the emergence of epidemic Vibrio cholerae isolates based on comparative nucleotide sequence analysis and multilocus virulence gene profiles. J Clin Microbiol 42, 4657–4671.[Abstract/Free Full Text]

O'Shea, Y. A., Finnan, S., Reen, F. J., Morrissey, J., O'Gara, F. & Boyd, E. F. (2004b). The Vibrio seventh pandemic island-II is a 26·9 kb genomic island present in Vibrio cholerae El Tor and O139 serogroup isolates that shows homology to a 43·4 kb island in V. vulnificus. Microbiology 150, 4053–4063.[CrossRef][Medline]

Ramamurthy, T., Albert, M. J., Huq, A., Colwell, R. R., Takeda, Y., Takeda, T., Shimada, T., Mandal, B. K. & Nair, G. B. (1994). Vibrio mimicus with multiple toxin types isolated from human and environmental sources. J Med Microbiol 40, 194–196.[Abstract]

Reen, F. J. & Boyd, E. F. (2004). Molecular typing of epidemic and non-epidemic Vibrio cholerae isolates, and V. cholerae and V. mimicus isolates by PCR-Single Strand Conformation Polymorphism Analysis. J Appl Microbiol (in press).

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

Shi, L., Miyoshi, S., Hiura, M., Tomochika, K., Shimada, T. & Shinoda, S. (1998). Detection of genes encoding cholera toxin (CT), zonula occludens toxin (ZOT), accessory cholera enterotoxin (ACE) and heat-stable enterotoxin (ST) in Vibrio mimicus clinical strains. Microbiol Immunol 42, 823–828.[Medline]

Spira, W. M. & Fedorka-Cray, P. J. (1983). Production of cholera toxin-like toxin by Vibrio mimicus and non-O1 Vibrio cholerae: batch culture conditions for optimum yields and isolation of hypertoxigenic lincomycin-resistant mutants. Infect Immun 42, 501–509.[Medline]

Spira, W. M. & Fedorka-Cray, P. J. (1984). Purification of enterotoxins from Vibrio mimicus that appear to be identical to cholera toxin. Infect Immun 45, 679–684.[Medline]

Stroeher, U. H., Parasivam, G., Dredge, B. K. & Manning, P. A. (1997). Novel Vibrio cholerae O139 genes involved in lipopolysaccharide biosynthesis. J Bacteriol 179, 2740–2747.[Abstract]

Taylor, R. K., Miller, V. L., Furlong, D. B. & Mekalanos, J. J. (1987). Use of phoA gene fusions to identify a pilus colonization factor coordinately regulated with cholera toxin. Proc Natl Acad Sci U S A 84, 2833–2837.[Abstract]

Vimr, E. R., Kalivoda, K. A., Deszo, E. L. & Steenbergen, S. M. (2004). Diversity of microbial sialic acid metabolism. Microbiol Mol Biol Rev 68, 132–153.[Abstract/Free Full Text]

Waldor, M. K. & Mekalanos, J. J. (1994). Emergence of a new cholera pandemic: molecular analysis of virulence determinants in Vibrio cholerae O139 and development of a live vaccine prototype. J Infect Dis 170, 278–283.[Medline]

Waldor, M. K. & Mekalanos, J. J. (1996). Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 272, 1910–1914.[Abstract]

Received 13 September 2004; revised 30 September 2004; accepted 30 September 2004.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Jermyn, W. S.
Articles by Boyd, E. F.
Articles citing this Article
PubMed
PubMed Citation
Articles by Jermyn, W. S.
Articles by Boyd, E. F.
Agricola
Articles by Jermyn, W. S.
Articles by Boyd, E. F.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.