- Research article
- Open Access
The genome sequence of the fish pathogen Aliivibrio salmonicida strain LFI1238 shows extensive evidence of gene decay
BMC Genomicsvolume 9, Article number: 616 (2008)
The fish pathogen Aliivibrio salmonicida is the causative agent of cold-water vibriosis in marine aquaculture. The Gram-negative bacterium causes tissue degradation, hemolysis and sepsis in vivo.
In total, 4 286 protein coding sequences were identified, and the 4.6 Mb genome of A. salmonicida has a six partite architecture with two chromosomes and four plasmids. Sequence analysis revealed a highly fragmented genome structure caused by the insertion of an extensive number of insertion sequence (IS) elements. The IS elements can be related to important evolutionary events such as gene acquisition, gene loss and chromosomal rearrangements. New A. salmonicida functional capabilities that may have been aquired through horizontal DNA transfer include genes involved in iron-acquisition, and protein secretion and play potential roles in pathogenicity. On the other hand, the degeneration of 370 genes and consequent loss of specific functions suggest that A. salmonicida has a reduced metabolic and physiological capacity in comparison to related Vibrionaceae species.
Most prominent is the loss of several genes involved in the utilisation of the polysaccharide chitin. In particular, the disruption of three extracellular chitinases responsible for enzymatic breakdown of chitin makes A. salmonicida unable to grow on the polymer form of chitin. These, and other losses could restrict the variety of carrier organisms A. salmonicida can attach to, and associate with. Gene acquisition and gene loss may be related to the emergence of A. salmonicida as a fish pathogen.
Aliivibrio salmonicida (formerly Vibrio salmonicida) is a facultative pathogen of fish responsible for causing cold-water vibriosis (CV) in farmed Atlantic salmon (Salmo salar), sea farmed rainbow trout (Oncorhynchus mykiss) and captive Atlantic cod (Gadus morhua) . At the peak of its prevalence in the 1980s infected fish farms suffered heavy losses reaching 50–90% . CV appeared to be effectively controlled in 1998  but before vaccination was introduced, A. salmonicida was estimated to have been responsible for over 80% of disease related losses to the Norwegian aquaculture industry . Although the impact of A. salmonicida on the aquaculture industry is primarily on salmonoids there is concern it poses a risk to new commercially important species for which farming is at an early stage or is planned. The decline in the wild Atlantic cod population has lead to a massive expansion of cod aquaculture. In Norway alone 7410 tons of farmed cod were sold in 2005, which is more than twice the amount from previous year . So far the cod farming industry has only suffered a few outbreaks of CV, and only in unvaccinated fish. However, despite this successful treatment the CV vaccine is administered by intraperitoneal injection and its use is associated with severe side-effects such as impaired growth, intra-abdominal lesions  and adhesions in the abdominal cavity of the fish that may affect physiological functions and reduce the quality of the final product . Hence, alternative approaches and vaccines are essential.
The halophilic and psychrophilic bacterium belongs to Vibrionaceae, which includes 85 species found in a wide range of aquatic environments in free-living forms and attached to both biotic and abiotic surfaces. Plankton organisms, mainly copepods, host large populations of bacteria. The attachment to zooplankton may enhance environmental survival of Vibrionaceae which are able to break down the chitinaceous exoskeleton and utilize the polysaccharides as an abundant source of carbon and nitrogen . Vibrionaceae are also found associated with, and are pathogens of, other aquatic organisms such as fish, mussels, corals, molluscs, seagrass, shrimps and squid . Currently the genome sequences of nine Vibrionaceae have been published. We report here the complete genome sequence of the first fish pathogenic Vibrionaceae.
During an infection A. salmonicida elicits tissue degradation, hemolysis and sepsis. Clinical symptoms such as severe anaemia and extensive haemorrhages on the surface of all internal organs of the fish are commonly observed. However, very little is known about the molecular mechanisms that produce the pathology of these infections and the genome should provide an insight into evolution and mechanisms involved in mediating the disease. The cod isolate A. salmonicida strain LFI1238 taken from the head kidney (lymphoid organ) of a diseased fish was chosen for sequencing in order to better understand pathogen-host interactions.
Results and discussion
I. General features of the genome
The general features of the A. salmonicida strain LFI1238 (LFI1238) genome are summarized in Table 1. The genomic G+C content of 39.6% is relatively low in comparison to other sequenced Vibrionaceae. Characteristically for members of Vibrionaceae  the A. salmonicida genome consists of two circular chromosomes of 3.3 and 1.2 Mb (chr I and chr II respectively) (Figure 1). The presence of essential genes on chr II indicates that this replicon is not a dispensable megaplasmid . However unlike the other Vibrionaceae sequenced LFI1238 also carries four circular plasmids designated pVSAL840 (83.5 kb), pVSAL320 (30.8 kb), pVSAL54 (5.4 kb) and pVSAL43 (4.3 kb) which represent 2.7% of the total genomic DNA and harbour 111 protein coding sequences (CDSs; Table 1 and [Additional file 1A]).
The functional distribution of CDSs between the chromosomes is similar to that reported for other Vibrionaceae : chr I carries the majority of CDSs needed for DNA replication, cell division, biosynthesis of amino acids and nucleotides. Conversely, the majority of CDSs involved in adapting to environmental changes, such as stress response functions, proteins associated with the cell envelope and proteins that could not be assigned any function are encoded on chr II (Figure 2). From similarity searches comparing all of the LFI1238 CDSs against the CDSs from the other published Vibrionaceae genomes, it is apparent that A. salmonicida shares more orthologous genes with Aliivibrio fischeri (70%) than the other Vibrionaceae compared (average 55–60% shared orthologs). These observations are consistent with 16S rRNA gene sequence analysis data  and support the reclassification of these two species, together with Aliivibrio wodanis and Aliivibrio logei as a separate genus .
The presence of multiple plasmids is characteristic of A. salmonicida  with many which are common to isolates from diverse geographical areas in the North Atlantic Ocean (Norway, Canada, the Shetland Islands, Faroe Islands). From plasmid profiles plasmids of the same size as pVSAL43, pVSAL54 and pVSAL320 are common to isolates from all of the above regions. However, LFI1238 pVSAL840 appears to be restricted to isolates from the northern parts of Norway where it is found in strains alongside either pVSAL43, pVSAL54 and pVSAL320, or together with pVSAL320 . pVSAL840 harbours a tra locus containing 21 CDSs with functions related to plasmid conjugation. This region is highly syntenic with the tra locus of the conjugation plasmid pYJ016 identified in Vibrio vulnificus  and plasmid pES100 in A. fischeri , and suggests a similar function involved in conjugation for pVSAL840.
Plasmids pVSAL43 and pVSAL54 are predicted to encode acyltransferases. Acyltransferases have the potential to change the acetylation state of the lipopolysaccharide (LPS) and so maybe important in providing antigenic variability of the cell surface to give better protection against the host antibody immune recognition . In a recent study, the expression of an iron ABC transporter harboured on pVSAL320 was shown to be dependent upon iron and probably regulated by the ferric uptake regulator Fur . pVSAL320 may therefore be important for the non-siderophore based uptake of ferrous iron. Valla and colleagues (1992) showed that a plasmid cured strain of A. salmonicida when injected through the intraperetoneal route was still able to cause CV in salmon . Therefore although these plasmids may contribute to colonisation and virulence they are not essential, at least by this route of infection.
The most striking feature of the A. salmonicida genome is the high number of insertion sequence (IS) elements relative to other Vibrionaceae. In total A. salmonicida carry 521 CDSs (12.2% of all CDSs) representing 288 whole and partial IS elements (Table 1), compared to only one IS element in A. fischeri. These IS elements can be subdivided into 20 different types (denoted VSa1 – VSa20), and fall into 12 different IS families based on sequence similarities with defined families in the IS Finder database  [see Additional file 2]. The relative proportion of transposases is slightly larger in chr II than in chr I (14.8 and 11.4%, respectively). IS element insertions have disrupted 183 CDSs (4.3% of the total CDSs) in the chromosomes and plasmids. Most of these "natural knock-outs" are probably not translated to give functional products. The distribution of the IS elements suggests that the IS elements present in high numbers on the chromosomes have spread to the plasmids by transposition. However, VSa3 and VSa4 are found exclusively on pVSAL840, VSa19 is restricted to pVSAL320 and pVSAL54 carries none of the A. salmonicida IS elements. This suggests that these plasmids do not tolerate insertions or that they are relatively recent acquisitions and that pVSAL54 is the most recently acquired. However, Codon Adaptation Index (CAI) analysis, which measures the relative adaptiveness of the codon usage of genes towards the codon usage of highly expressed genes , revealed that genes on pVSAL840 (0.48) deviate more from the average genome composition (0.58) than the other plasmids (Table 1). In addition the Codon Bias Index (CBI) versus CAI plot described in  for A. salmonicida clearly showed that genes on pVSAL840 deviate most from the genome background, suggesting that this is likely to be the most recent acquired plasmid [see Additional file 3].
II. Genome structure
Compositional asymmetries (GC deviation) in the leading and lagging strand of DNA, with bias towards G on the leading strand of the bidirectional replication fork, is a common characteristic of bacterial genomes . It is evident from Figure 1 that both of the A. salmonicida chromosomes show anomalies in their GC deviation. Significantly IS elements are found flanking all large regions showing an aberant GC deviation. Since their homologous DNA can serve as recombinational cross-over points they are likely to be largely responsible for the apparent anomalies . Consistent with this, whole genome comparison with A. fischeri also shows that these anomalous regions represent breaks in synteny [see Additional file 4].
By designing PCR primers to amplify across the borders of these anomalous regions we discovered that several genomic configurations may exist within a population of any given isolate (data not shown). It has been suggested that this type of interreplichore recombinations have an effect on the gene dosage, whereby the continual initiation of replication folks leads to genes closer to the origin being at a higher relative gene dosage than those at the terminus. It has also been shown that gene orientation is under selection, with essential genes being preferentially encoded on the leading strand; this is hypothesised to be due to avoidence of the deleterious effects of collisions between the transcription and translation machinery . How stable any given genomic configuration is and what affect this has on transcription in A. salmonicida is yet to be determined, but similar rapid rearrangements have been reported in other genomes with high IS element loads .
Interestingly in addition to mediating homologous recombination, IS elements also border three regions in the chromosome that are found duplicated in the plasmids [see Additional file 1B]. Two such regions, each encoding three CDSs from pVSAL840 and pVSAL320 respectively, are the flanking parts of the genomic island GI-VS1 (Table 2). The duplicated CDSs displayed nucleotide sequence identity up to 100%, and the functions of most are unknown. The third region carries four CDSs of which two encode a hemolysin co-regulated protein (Hcp) and a VgrG protein. Both Hcp and VgrG are virulence effector proteins secreted by the Type VI secretion system. Codon usage analysis clearly showed that the duplicated genes cluster more closely to the plasmid genes than to the genome background [see Additional file 3], which suggests that the genes originated from the plasmids. Thus this recombination between IS elements represents a mechanism by which to introduce new functions into the chromosome from a highly variable complement of plasmids.
1. Gene acquisition
In addition to the plasmids and the IS elements, the genome of A. salmonicida carries other mobile genetic elements, including nine prophages as well as 16 regions which have the characteristics of genomic islands (Table 2) . The tailed phage ϕ VS4 present on chr I has an overall GC content of 40.8%, slightly higher than the chromosome average (39.8%). The majority of the 43 CDSs show considerable homology and synteny to the K139 phage of Vibrio cholerae strain O139  [see Additional file 5], but this phage is not found in any of the other sequenced Vibrionaceae genomes. ϕ VS4 is likely to be the only complete prophage within the A. salmonicida genome (Table 2). The remaining 8 prophage-like regions are likely to be remnants.
In total, 25 regions larger than 5 kb were identified as having atypical DNA compositional and being present in A. salmonicida but absent the other Vibrionaceae genomes (Table 2). Although the majority of CDS encoded on these regions of difference are of no known function, some encode proteins involved in secretion, biosynthesis of capsular polysaccharides (CPS) and biosynthesis and uptake of siderophores [see Additional file 6].
In addition to the regions duplicated on the plasmids and chromosomes Chr I also carries an additional perfect duplication of approximately 29 kb. PCR analysis of 27 different A. salmonicida isolates confirmed the duplication at the same locations in all tested isolates [see Additional file 7]. Each duplicate contains 27 genes, the majority encoding products involved in the biosynthesis of constituents of the LPS. L-rhamnose is present in the O-antigens of Gram-negative bacteria . Four genes, rmlBADC, necessary for the conversion of D-glucose 1-phosphate to dTDP-L-rhamnose are present in the repeat. Seven genes are similar to those found in the wav gene cluster of V. cholerae. The wav genes are responsible for the synthesis of LPS core oligosaccharides . Nesper and colleges suggested that genetic exchange of wav genes could improve outer membrane stability by altering the structure of the core LPS. In such case, it would provide for better adaptation to different niches. However, the duplicates in LFI1238 are identical at the nucleotide level and homologous recombination would therefore not increase the variety of the surface molecules expressed in the bacteria. On the other hand, in Haemophilus influenzae genes involved in the capsule expression are located within an 18 kb cap locus. Up to five copies of the locus have been detected, and a relationship between the number of copies of the cap locus and the production of capsule has been demonstrated . We speculate that the amplification may increase the gene expression and lead to increased LPS production. Espelid et al. observed that ball-shaped aggregates containing the protein/lipopolysaccharide VS-P1, the dominant immunoreactive antigen of A. salmonicida, were released in large quantities from the bacterial membrane inside the host . It has been suggested that much of the specific immune response of the fish may be directed against this "smoke screen" . Increased LPS production by A. salmonicida could therefore be advantageous when entering a host.
3. Gene loss
In total we identified 185 pseudogenes (4.3% of the total CDSs) containing frameshift and nonsense mutations that might disrupt expression of functional products [see Additional file 3]. Loss of functions seems to occur across all functional classes of products, but a striking number of transposases, transport proteins and proteins associated with the cell envelope are included in this list (Figure 2 and [Additional file 3]).
The accumulation of pseudogenes is high for genes involved in the utilisation of the polysaccharide chitin. Chitin (GlcNAc)n is an insoluble homopolymer of N-acetyl-D-glucosamine (GlcNAc), and is highly abundant in marine environments as constituents of the exoskeleton of crustaceans and zooplankton. Chitin is important for the attachment of bacteria to a carrier organism such as copepods [36, 37] and known to be an important as a nutrient source . Furthermore in a recent study Hunt et al. showed that the majority of genes involved in chitin degradation are conserved among the Vibrionaceae .
In A. salmonicida seven of the pseudogenes represent key components in the chitinolytic cascade (Figure 3 and [Additional file 8]) including a methyl accepting chemotaxis gene (VSAL_I2601) that may be involved in motility toward chitin , three chitinases (VSAL_I1942, VSAL_I0902/I0763 and VSAL_I1414) involved in the extracellular breakdown of chitin to chitin oligosaccharides [41, 42], a chitoporin (VSAL_I2352) responsible for mediating transport of chitin oligosaccharides into the periplasma , and a chitodextrinase (VSAL_I1108) involved in the periplasmic breakdown of chitin oligosaccharides .
In addition, several genes involved in the chitinolytic cascade are regulated by chitin oligosaccharides and a two-component chitin catabolic sensor/kinase encoded by chiS [40, 45]. The gene regulation on the transcriptional level is not known, but the periplasmic chitin binding protein (CBP) is required for ChiS-regulation (Figure 3). The CBP orthologue in A. salmonicida (VSAL_I2576) contains a frameshift. The functional loss of genes thought to be regulated by ChiS/CBP is likely to have preceded, and perhaps facilitated, the degeneration of this gene.
To investigate whether the loss of these genes has impaired the ability of A. salmonicida to utilize chitin, six A. salmonicida isolates including LFI1238 were grown on a minimal media containing either α-chitin (GlcNAc)n or GlcNAc as the only source of carbon. As a control, A. wodanis and Vibrio splendidus were grown in parallel. None of the A. salmonicida isolates showed growth on (GlcNAc)n nor on GlcNAc [see Additional file 9]. In contrast, the majority of controls grew on both the homopolymeric and monomeric form of GlcNAc. This implies that the loss of seven genes involved in the chitinolytic cascade have probably affected processes such as sensing, degradation and transport of chitin and suggests that the ability to catabolise chitin is no longer required by A. salmonicida. Consistent with these findings preliminary studies looking for A. salmonicida in the environment have failed to find this species associated with copepods (personal communication B. Landfald). Accordingly, this could also confine the variety of carrier organisms A. salmonicida can attach to, and associate with.
It should be mentioned that programmed frameshifting and readthrough of premature stop codons are often used as methods of bacterial gene regulation . In addition, homopolymeric DNA tracts can give rise to slipped-strand mispairing during replication . It is therefore possible that some of the predicted pseudogenes could be translated into functional products, and are retained in the genome for selective reasons. Two flagellar biosynthesis genes, fliF (VSAL_I2308) and flaG (VSAL_I2316) are disrupted by premature stop codons. While the function of flaG is unknown, the product of fliF is the major component of the M-ring, a central motor component of the flagellum. Despite the disruption of fliF and flaG the sequenced strain is still motile. This could imply that these genes are not essential in A. salmonicida, or that the translational machinery is able to read through the premature stop codons and produce functional products.
IV. Quorum sensing
Bacterial cell-to-cell communication, or quorum sensing (QS) is a sophisticated mechanism that can allow for a synchronized gene expression of a whole community. Bacteria can respond to environmental changes by monitoring the presence of other bacteria in the surroundings by producing and responding to extracellular signal molecules (autoinducers). A. salmonicida has five QS systems (AinR/S, LuxI/R, VarS/A, LuxM/N and LuxS/PQ), which is more than reported in any other Vibrionaceae . However, there is extensive evidence of gene loss in these systems: luxN and luxP encoding the autoinducer receptors of the LuxM/N and LuxS/PQ systems, respectively are pseudogenes. In addition, A. salmonicida lacks luxM and luxL, required for the production of N-(3-hydroxylbutanyol)-L-homoserine lactone (HHL), the autoinducer of the LuxM/N system  further indicating that this system is non-functional. However, since the frameshift within luxN occurs within a homopolymeric tract (of 6 bp) it is possible that the function of this gene could be restored by programmed frameshifting. In the absence of LuxM and LuxL this would allow the system to function as a "mute" system monitoring the presence of HHLs produced by other bacteria.
V. Potential virulence factors
Little is known about the molecular mechanisms by which A. salmonicida causes disease. Through detailed analysis of the genome possible functions that may be associated with mediating CV have been predicted [see Additional file 10].
The roles of several important virulence factors have been described for other Vibrionaceae, such as the cholera toxin (CT) of V. cholerae , the thermostable hemolysin (TDH) of V. parahaemolyticus  and the metalloprotease (VVP) of V. vulnificus . Common for CT, TDH and VVP is that they act extracellularly, and are exported from the cell by various secretion mechanisms. Although none of these factors were found in A. salmonicida, the tissue damage observed in fish with CV suggests that A. salmonicida secrets proteins during an infection like these other pathogens. Several protein secretion systems were identified in the genome, including three Type I secretion systems (T1SS), one Type II secretion system (T2SS), two Type VI secretion systems (T6SS) and one Flp-type pilus system.
The CDSs of the Flp-type pilus are harboured on GI-VSA5 and show sequence similarities and high synteny to the Tad (tight adherence) macromolecular transport system of Actinobacillus actinomycetemcomitans. The tad system is widely distributed in bacteria and secrets a pilus that is involved in adherance to surfaces . This function is necessary for colonization and pathogenesis by A. actinomycetemcomitans. The tad genes are present and intact in A. salmonicida, A. fischeri,V. parahaemolyticus and an incomplete operon is found in both V. vulnificus strains sequenced.
Functional gene-loss is evident in one T1SS, and both T6SSs gene clusters [see Additional file 10]. The products of the pseudogenes of the T6SSs are not predicted to be structural components of the secretion apparatus . It is therefore possible that these systems are functional in A. salmonicida. T6SSI and T6SSII are located on GI-VS5 and GI-VS6, respectively. Both systems show sequence similarities as well as considerable synteny to the V. cholerae T6SS . Virulence effector proteins secreted by T6SS lack an N-terminal signal sequence and include a hemolysin co-regulated protein (Hcp) and a VgrG protein . By sequence similarity we identified three VgrG (VSAL_I1358, VSAL_p840_36 and VSAL_I1744) and three Hcp (VSAL_I1357 and VSAL_I1202) homologs in the genome. VSAL_I1744 is disrupted by the insertion of an IS-element and is probably not expressed.
Among the predicted CDSs with the potential to cause tissue degradation and hemolysis in the fish, we have identified two CDSs, VSAL_I0993 and VSAL_I0411 with 77% and 52% sequence identity to V. anguillarum hemolysins VAH2 and VAH5 respectively [see Additional file 10]. VAH2 and VAH5 showed hemolytic activity against fish erythrocytes and are suggested to contribute to the hemolytic activity of V. anguillarum . To what extent VSAL_I0993 and VSAL_I0411 can cause hemolysis of fish blood cells, as observed in fish with CV is unknown. Similar to VAH2 and VAH5, no export signal sequence was found for VSAL_I0993 and VSAL_I0411. It is possible that the two putative hemolysins are exported by one or several of the A. salmonicida T1SS. In E. coli, export of hemolysin HlyA is mediated by the hemolysin secretion system, which has been described as one of the prototypes of T1SS .
A. salmonicida uses the siderophore bisucaberin to acquire iron . A complete siderophore biosynthesis/acquisition system is contained on GI-VSA3 (VSAL_I0141-I0135), but whether it could be responsible for the production of bisucaberin remains to be clarified. We have also identified a heme uptake system with high sequence similarity and synteny to that of many other Vibrionaceae . Both transport of heme complexes, and ferric-siderophores across the outer membrane require a functional TonB system. Several members of Vibrionaceae possess two TonB systems [59, 60]. A. salmonicida harbours three TonB systems, named TonB1, TonB2 and TonB3. In V. cholerae both TonB systems corresponding to A. salmonicida TonB1 and TonB2 are capable of mediating the transport of heme and siderophores , while in V. anguillarum only the TonB system homologous to A. salmonicida TonB3 is essential for the ferric-siderophore transport and virulence . In the TonB1 system, tonB1 (VSAL_I1751) contains a translational frameshift and is probably not translated into a functional product. All three TonB systems in A. salmonicida are located adjacent to CDSs with functions associated with iron-uptake. This indicates that more than one system may be involved in iron acquisition.
The A. salmonicida genome displays a mosaic structure (Figure 1) caused by large intra-chromosomal rearrangements, gene acquisition, deletion and duplication of DNA within the chromosomes and between the chromosomes and the plasmids. From our sequence analysis it is clear that many of these events are mediated by homologous recombination between IS elements.
Multiple lines of evidence, such as compositional sequence differences, were used to identify recent gene acquisitions. The majority of the horizontally acquired DNA is flanked by IS elements. Although the direct influence the gene acquisitions have had on the evolution and adaptation of A. salmonicida is not clear, some of the GIs carry genes that may have provided new functions to the bacteria. For example, two T6SS and one Flp-type pilus system that are involved in the export of proteins are located on DNA segments that have the typical characteristics of GIs. T6SS have been recognized as a major virulence determinant in other pathogens were they have been shown to be involved in the extracellular translocation of proteins required for cytotoxicity [54, 61]. The Flp-pilus system is similar to the Tad macromolecular transport system of A. actinomycetemcomitans. The Tad system has been proposed to represent a new subtype of T2SS and is essential for biofilm formation, colonization and pathogenesis . Phylogenetic analysis of the Tad system shows a complex history of gene shuffling and multiple HGT among prokaryotes . Our findings support the hypothesis that the distribution of the tad genes is explained by their location on a mobile GI (widespread colonisation island, WCI) . Whether the protein secretion systems are important for the virulence towards fish remains to be elucidated.
Over 300 CDSs are disrupted by IS elements or contain point mutations causing frameshifts or premature stop codons [Additional file 11]. A large fraction of the degenerate CDSs have roles in the response to environmental changes and in modulating the host-cell interaction. The extensive loss of the same types of genes has been reported for the pathogen species Mycobacterium leprae, Salmonella Typhi, Bordetella pertussis, and others which have become host adapted [26, 63]. The DNA sequences of these CDSs are still intact, which indicates that the gene losses are relatively recent events in A. salmonicida. IS expansion has been related to genome reduction in the evolution and emergence of pathogenicity , and accumulation of pseudogenes has been described for several other host-restricted pathogens [26, 28, 65], supporting the hypothesis that A. salmonicida may have also become host-restricted through gene loss.
Taken together, the acquisition of novel genes and loss of old functions may be related to the emergence of A. salmonicida as a pathogenic species for salmonids. The outcome of the horizontal acquisition of genes could have allowed for an expansion to a previously unexplored niche, and the accumulation of pseudogenes and IS expansion resulting in massive loss of functional genes observed in A. salmonicida may be a result of selection against the expression of genes not required in the new niche, or a neutral process associated with the relaxation of selective pressure due to the evolutionary bottleneck associated with niche adaptation. The observations made for the A. salmonicida genome are similar to those of other recently-evolved host-restricted pathogens, suggesting that A. salmonicida has recently made the transition to the specific niche of fish pathogenicity.
We applied the whole-genome shotgun strategy to sequence an environmental isolate of A. salmonicida (strain LFI1238) from cod provided by Elin Sandaker at The Norwegian Institute of Fisheries and Aquaculture Ltd. A single colony of LFI1238 grown on blood agar containing 2.5% NaCl was transferred to marine broth and grown overnight with shaking at 12°C. Cells were collected and total DNA (10 mg) was isolated using proteinase K treatment followed by phenol extraction. The DNA was fragmented by sonication, and several libraries were generated in pUC19 and pMAQ1Sac using size fractions ranging from 2.2 to 4.0 kb and 4.0 to 12.0 kb, respectively. The whole genome was sequenced to a depth of 10 times coverage using dye terminator chemistry on ABI3700 automated sequencers. End sequences from larger insert plasmid (pBeloBACII, 50–70 kb insert size) libraries were used as a scaffold.
The sequence was annotated using Artemis software . Initial CDS predictions were performed using Orpheus  and Glimmer2  software. These predictions were amalgamated, and codon usage, positional base preference methods and comparisons to the non redundant protein databases using BLAST  and FASTA  software were used to refine the predictions. The entire DNA sequence was also compared in all six reading frames against the nonredundant protein databases, using BLASTX to identify any possible coding sequences previously missed. Protein motifs were identified using Pfam  and Prosite , transmembrane domains were identified with TMHMM , and signal sequences were identified with SignalP version 2.0 . Stable RNAs were identified using Rfam . GIs and bacteriophages were predicted using Alien Hunter . The sequence is available from EMBL/GenBank/DDBJ with the accession numbers [EMBL: FM178379, FM178380, FM178381, FM178382, FM178383 and FM178384].
Comparison of the genome sequences was facilitated by using the Artemis Comparison Tool (ACT) , which enabled the visualization of BLASTN and TBLASTX comparisons  between the genomes. Orthologous proteins were identified as reciprocal best matches using FASTA with subsequent manual curation. Pseudogenes had one or more mutations that would prevent correct translation and each of the inactivating mutations were subsequently checked against the original sequencing data.
In order to determine if duplicated genes originated from the plasmids or from the chromosomes and to predict the order in which the plasmids were acquired we performed a CAI and a CBI analysis: CAI and a CBI analysis: CAI used the Highly expressed genes (encoding all ribosomal proteins and tRNA synthetases in the genome) as the reference; CBI used the codon usage of all the genes in the genome and measured the adaptation of each gene to that. The CAI was done via EMBOSS cai and the CBI was done via EMBOSS codcmp .
Amplification of genes from other isolates was performed by PCR using Platinum Pfx DNA Polymerase (Invitrogen, Carlsbad, CA) according to the protocol supplied by the manufacturer. PCR amplification products were analyzed in 0.8% agarose gels stained with ethidium bromide.
Isolates of A. salmonicida, A. wodanis and V. splendidus were grown in LB medium containing 2.5% NaCl, diluted in A. salmonicida minimal medium (Vsmm [100 mM KH2PO4, 15 mM (NH4)2SO4, 3.9 μM FeSO4, 2.5% NaCl, 0.81 mM MgSO4, 2 mM Valin, 0.5 mM Isoleucin, 0.5 mM Cystein, 0.5 mM Methionin, 40 mM Glutamate]) and transferred to Vsmm agar supplemented with 10 mg/ml α-chitin (Sigma-Aldrich) or N-acetyl-α-D-glucosamine (Calbiochem). Plates were incubated from 2 to 7 days at 12°C (A. salmonicida and A. wodanis) and 22°C (V. splendidus) and growth evaluated by visual examination.
protein coding sequence
Codon Adaptation Index
Codon Bias Index
chitin binding protein
horizontal gene transfer
Schrøder MB, Espelid S, Jørgensen TØ: Two serotype of Vibrio salmonicida isolated from diseased cod (Gadus morhua L.); virulence, immunological studies and advanced experiments. Fish & Shellfish Immunology. 1992, 2: 211-221.
Hjeltnes B, Andersen K, Egidius E: Multiple antibiotic resistance in Vibrio salmonicida. Bulletin of the European Association of Fish Pathologists. 1987, 7 (4): 85-
Colquhoun DJ: Vibrio salmonicida, the causative agent of cold-water vibriosis: factors relating to pathogenesis and vaccine protection. 2002, Oslo: The Norwegian School of Veterinary Medicine
Poppe TT, Håstein T, Salte R: "Hitra Disease" (Haemorrhagic Syndrome) in Norwegian Salmon Farming: Present Status. Fish & Shellfish Pathology. 1985, 223-229.
Statistisk Sentralbyrå. Statistics Norway. [http://www.ssb.no/]
Midtlyng PJ, Lillehaug A: Growth of Atlantic salmon Salmo salar after intraperitoneal administration of vaccines containing adjuvants. Dis Aquat Organ. 1998, 32 (2): 91-97.
Midtlyng PJ: A field study on intraperitoneal vaccination of Atlantic salmon (Salmo salarL.) against furunculosis. Fish & Shellfish Immunology. 1996, 6 (8): 553-565.
Riemann L, Azam F: Widespread N-acetyl-D-glucosamine uptake among pelagic marine bacteria and its ecological implications. Appl Environ Microbiol. 2002, 68 (11): 5554-5562.
Thompson FL, Iida T, Swings J: Biodiversity of vibrios. Microbiol Mol Biol Rev. 2004, 68 (3): 403-431.
Okada K, Iida T, Kita-Tsukamoto K, Honda T: Vibrios commonly possess two chromosomes. J Bacteriol. 2005, 187 (2): 752-757.
Heidelberg JF, Eisen JA, Nelson WC, Clayton RA, Gwinn ML, Dodson RJ, Haft DH, Hickey EK, Peterson JD, Umayam L, Gill SR, Nelson KE, Read TD, Tettelin H, Richardson D, Ermolaeva MD, Vamathevan J, Bass S, Qin H, Dragoi I, Sellers P, McDonald L, Utterback T, Fleishmann RD, Nierman WC, White O, Salzberg SL, Smith HO, Colwell RR, Mekalanos JJ, Venter JC, Fraser CM: DNA sequence of both chromosomes of the cholera pathogen Vibrio cholerae. Nature. 2000, 406 (6795): 477-483.
Reen FJ, Almagro-Moreno S, Ussery D, Boyd EF: The genomic code: inferring Vibrionaceae niche specialization. Nat Rev Microbiol. 2006, 4 (9): 697-704.
Thompson FL, Gevers D, Thompson CC, Dawyndt P, Naser S, Hoste B, Munn CB, Swings J: Phylogeny and molecular identification of vibrios on the basis of multilocus sequence analysis. Appl Environ Microbiol. 2005, 71 (9): 5107-5115.
Urbanczyk H, Ast JC, Higgins MJ, Carson J, Dunlap PV: Reclassification of Vibrio fischeri, Vibrio logei, Vibrio salmonicida and Vibrio wodanis as Aliivibrio fischeri gen. nov., comb. nov., Aliivibrio logei comb. nov., Aliivibrio salmonicida comb. nov. and Aliivibrio wodanis comb. nov. Int J Syst Evol Microbiol. 2007, 57 (Pt 12): 2823-2829.
Sørum H, Myhr E, Zwicker BM, Lillehaug A: Comparison by plasmid profiling of Vibrio salmonicida strains isolated from diseased fish from different North European and Canadian areas of the Atlantic Ocean. Canadian Journal of Fish Aquatic Science. 1993, 50: 247-250.
Sørum H, Hvaal AB, Heum M, Daae FL, Wiik R: Plasmid profiling of Vibrio salmonicida for epidemiological studies of cold-water vibriosis in Atlantic salmon (Salmo salar) and cod (Gadus morhua). Appl Environ Microbiol. 1990, 56 (4): 1033-1037.
Chen CY, Wu KM, Chang YC, Chang CH, Tsai HC, Liao TL, Liu YM, Chen HJ, Shen AB, Li JC, Su TL, Shao CP, Lee CT, Hor LI, Tsai SF: Comparative genome analysis of Vibrio vulnificus, a marine pathogen. Genome Res. 2003, 13 (12): 2577-2587.
Ruby EG, Urbanowski M, Campbell J, Dunn A, Faini M, Gunsalus R, Lostroh P, Lupp C, McCann J, Millikan D, Schaefer A, Stabb E, Stevens A, Visick K, Whistler C, Greenberg EP: Complete genome sequence of Vibrio fischeri: a symbiotic bacterium with pathogenic congeners. Proc Natl Acad Sci USA. 2005, 102 (8): 3004-3009.
Slauch JM, Mahan MJ, Michetti P, Neutra MR, Mekalanos JJ: Acetylation (O-factor 5) affects the structural and immunological properties of Salmonella typhimurium lipopolysaccharide O antigen. Infect Immun. 1995, 63 (2): 437-441.
Ahmad R, Hjerde E, Hansen GA, Haugen P, Willassen NP: Prediction and Experimental Testing of Ferric Uptake Regulator Regulons in Vibrios. J Mol Microbiol Biotechnol. 2008
Valla S, Frydenlund K, Coucheron DH, Haugan K, Johansen B, Jorgensen T, Knudsen G, Strom A: Development of a gene transfer system for curing of plasmids in the marine fish pathogen Vibrio salmonicida. Appl Environ Microbiol. 1992, 58 (6): 1980-1985.
IS Finder database. [http://www-is.biotoul.fr/is.html]
Sharp PM, Li WH: The codon Adaptation Index – a measure of directional synonymous codon usage bias, and its potential applications. Nucleic Acids Res. 1987, 15 (3): 1281-1295.
Karlin S, Mrazek J, Campbell AM: Codon usages in different gene classes of the Escherichia coli genome. Mol Microbiol. 1998, 29 (6): 1341-1355.
Rocha EP, Danchin A, Viari A: Universal replication biases in bacteria. Mol Microbiol. 1999, 32 (1): 11-16.
Parkhill J, Sebaihia M, Preston A, Murphy LD, Thomson N, Harris DE, Holden MT, Churcher CM, Bentley SD, Mungall KL, Cerdeno-Tarraga AM, Temple L, James K, Harris B, Quail MA, Achtman M, Atkin R, Baker S, Basham D, Bason N, Cherevach I, Chillingworth T, Collins M, Cronin A, Davis P, Doggett J, Feltwell T, Goble A, Hamlin N, Hauser H, Holroyd S, Jagels K, Leather S, Moule S, Norberczak H, O'Neil S, Ormond D, Price C, Rabbinowitsch E, Rutter S, Sanders M, Saunders D, Seeger K, Sharp S, Simmonds M, Skelton J, Squares R, Squares S, Stevens K, Unwin L, Whitehead S, Barrell BG, Maskell DJ: Comparative analysis of the genome sequences of Bordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica. Nat Genet. 2003, 35 (1): 32-40.
Rocha EP, Danchin A: Gene essentiality determines chromosome organisation in bacteria. Nucleic Acids Res. 2003, 31 (22): 6570-6577.
Parkhill J, Wren BW, Thomson NR, Titball RW, Holden MT, Prentice MB, Sebaihia M, James KD, Churcher C, Mungall KL, Baker S, Basham D, Bentley SD, Brooks K, Cerdeno-Tarraga AM, Chillingworth T, Cronin A, Davies RM, Davis P, Dougan G, Feltwell T, Hamlin N, Holroyd S, Jagels K, Karlyshev AV, Leather S, Moule S, Oyston PC, Quail M, Rutherford K, Simmonds M, Skelton J, Stevens K, Whitehead S, Barrell BG: Genome sequence of Yersinia pestis, the causative agent of plague. Nature. 2001, 413 (6855): 523-527.
Dobrindt U, Hochhut B, Hentschel U, Hacker J: Genomic islands in pathogenic and environmental microorganisms. Nat Rev Microbiol. 2004, 2 (5): 414-424.
Kapfhammer D, Blass J, Evers S, Reidl J: Vibrio cholerae phage K139: complete genome sequence and comparative genomics of related phages. J Bacteriol. 2002, 184 (23): 6592-6601.
Li Q, Hobbs M, Reeves PR: The variation of dTDP-L-rhamnose pathway genes in Vibrio cholerae. Microbiology. 2003, 149 (Pt 9): 2463-2474.
Nesper J, Kraiss A, Schild S, Blass J, Klose KE, Bockemuhl J, Reidl J: Comparative and genetic analyses of the putative Vibrio cholerae lipopolysaccharide core oligosaccharide biosynthesis (wav) gene cluster. Infect Immun. 2002, 70 (5): 2419-2433.
Corn PG, Anders J, Takala AK, Kayhty H, Hoiseth SK: Genes involved in Haemophilus influenzae type b capsule expression are frequently amplified. J Infect Dis. 1993, 167 (2): 356-364.
Espelid S, Holm KO, Hjemeland K, Jørgensen T: Monoclonal antibodies against Vibrio salmonicida: the causative agent of coldwater vibriosis (Hitra disease) in Atlantic salmon, Salmo salar L. J Fish Dis. 1988, 11: 207-214.
Hjelmeland K, Stenvåg K, Jørgensen TØ, Espelid S: Isolation and characterization of a surface layer antigen from Vibrio salmonicida. Journal of Fish Diseases. 1988, 11: 197-205.
Montgomery MT, Kirchman DL: Role of Chitin-Binding Proteins in the Specific Attachment of the Marine Bacterium Vibrio harveyi to Chitin. Appl Environ Microbiol. 1993, 59 (2): 373-379.
Pruzzo C, Crippa A, Bertone S, Pane L, Carli A: Attachment of Vibrio alginolyticus to chitin mediated by chitin-binding proteins. Microbiology. 1996, 142 (Pt 8): 2181-2186.
Garay E, Arnau A, Amaro C: Incidence of Vibrio cholerae and related vibrios in a coastal lagoon and seawater influenced by lake discharges along an annual cycle. Appl Environ Microbiol. 1985, 50 (2): 426-430.
Hunt DE, Gevers D, Vahora NM, Polz MF: Conservation of the chitin utilization pathway in the Vibrionaceae. Appl Environ Microbiol. 2008, 74 (1): 44-51.
Meibom KL, Li XB, Nielsen AT, Wu CY, Roseman S, Schoolnik GK: The Vibrio cholerae chitin utilization program. Proc Natl Acad Sci USA. 2004, 101 (8): 2524-2529.
Orikoshi H, Nakayama S, Miyamoto K, Hanato C, Yasuda M, Inamori Y, Tsujibo H: Roles of four chitinases (chia, chib, chic, and chid) in the chitin degradation system of marine bacterium Alteromonas sp. strain O-7. Appl Environ Microbiol. 2005, 71 (4): 1811-1815.
Svitil AL, Kirchman DL: A chitin-binding domain in a marine bacterial chitinase and other microbial chitinases: implications for the ecology and evolution of 1,4-beta-glycanases. Microbiology. 1998, 144 (Pt 5): 1299-1308.
Keyhani NO, Li XB, Roseman S: Chitin catabolism in the marine bacterium Vibrio furnissii. Identification and molecular cloning of a chitoporin. J Biol Chem. 2000, 275 (42): 33068-33076.
Keyhani NO, Roseman S: The chitin catabolic cascade in the marine bacterium Vibrio furnissii. Molecular cloning, isolation, and characterization of a periplasmic chitodextrinase. J Biol Chem. 1996, 271 (52): 33414-33424.
Li X, Roseman S: The chitinolytic cascade in Vibrios is regulated by chitin oligosaccharides and a two-component chitin catabolic sensor/kinase. Proc Natl Acad Sci USA. 2004, 101 (2): 627-631.
Farabaugh PJ: Programmed translational frameshifting. Microbiol Rev. 1996, 60 (1): 103-134.
Torres-Cruz J, Woude van der MW: Slipped-strand mispairing can function as a phase variation mechanism in Escherichia coli. J Bacteriol. 2003, 185 (23): 6990-6994.
Milton DL: Quorum sensing in vibrios: complexity for diversification. Int J Med Microbiol. 2006, 296 (2–3): 61-71.
Bassler BL, Wright M, Showalter RE, Silverman MR: Intercellular signalling in Vibrio harveyi: sequence and function of genes regulating expression of luminescence. Mol Microbiol. 1993, 9 (4): 773-786.
Faruque SM, Albert MJ, Mekalanos JJ: Epidemiology, genetics, and ecology of toxigenic Vibrio cholerae. Microbiol Mol Biol Rev. 1998, 62 (4): 1301-1314.
Zhang XH, Austin B: Haemolysins in Vibrio species. J Appl Microbiol. 2005, 98 (5): 1011-1019.
Miyoshi S, Narukawa H, Tomochika K, Shinoda S: Actions of Vibrio vulnificus metalloprotease on human plasma proteinase-proteinase inhibitor systems: a comparative study of native protease with its derivative modified by polyethylene glycol. Microbiol Immunol. 1995, 39 (12): 959-966.
Tomich M, Planet PJ, Figurski DH: The tad locus: postcards from the widespread colonization island. Nat Rev Microbiol. 2007, 5 (5): 363-375.
Pukatzki S, Ma AT, Sturtevant D, Krastins B, Sarracino D, Nelson WC, Heidelberg JF, Mekalanos JJ: Identification of a conserved bacterial protein secretion system in Vibrio cholerae using the Dictyostelium host model system. Proc Natl Acad Sci USA. 2006, 103 (5): 1528-1533.
Rodkhum C, Hirono I, Crosa JH, Aoki T: Four novel hemolysin genes of Vibrio anguillarum and their virulence to rainbow trout. Microb Pathog. 2005, 39 (4): 109-119.
Andersen C: Channel-tunnels: outer membrane components of type I secretion systems and multidrug efflux pumps of Gram-negative bacteria. Rev Physiol Biochem Pharmacol. 2003, 147: 122-165.
Winkelmann G, Schmid DG, Nicholson G, Jung G, Colquhoun DJ: Bisucaberin – a dihydroxamate siderophore isolated from Vibrio salmonicida, an important pathogen of farmed Atlantic salmon (Salmo salar). Biometals. 2002, 15 (2): 153-160.
Mourino S, Osorio CR, Lemos ML: Characterization of heme uptake cluster genes in the fish pathogen Vibrio anguillarum. J Bacteriol. 2004, 186 (18): 6159-6167.
Seliger SS, Mey AR, Valle AM, Payne SM: The two TonB systems of Vibrio cholerae: redundant and specific functions. Mol Microbiol. 2001, 39 (3): 801-812.
Stork M, Di Lorenzo M, Mourino S, Osorio CR, Lemos ML, Crosa JH: Two tonB systems function in iron transport in Vibrio anguillarum, but only one is essential for virulence. Infect Immun. 2004, 72 (12): 7326-7329.
Schell MA, Ulrich RL, Ribot WJ, Brueggemann EE, Hines HB, Chen D, Lipscomb L, Kim HS, Mrazek J, Nierman WC, Deshazer D: Type VI secretion is a major virulence determinant in Burkholderia mallei. Mol Microbiol. 2007, 64 (6): 1466-1485.
Planet PJ, Kachlany SC, Fine DH, DeSalle R, Figurski DH: The Widespread Colonization Island of Actinobacillus actinomycetemcomitans. Nat Genet. 2003, 34 (2): 193-198.
Cole ST, Eiglmeier K, Parkhill J, James KD, Thomson NR, Wheeler PR, Honore N, Garnier T, Churcher C, Harris D, Mungall K, Basham D, Brown D, Chillingworth T, Connor R, Davies RM, Devlin K, Duthoy S, Feltwell T, Fraser A, Hamlin N, Holroyd S, Hornsby T, Jagels K, Lacroix C, Maclean J, Moule S, Murphy L, Oliver K, Quail MA, Rajandream MA, Rutherford KM, Rutter S, Seeger K, Simon S, Simmonds M, Skelton J, Squares R, Squares S, Stevens K, Taylor K, Whitehead S, Woodward JR, Barrell BG: Massive gene decay in the leprosy bacillus. Nature. 2001, 409 (6823): 1007-1011.
Siguier P, Filee J, Chandler M: Insertion sequences in prokaryotic genomes. Curr Opin Microbiol. 2006, 9 (5): 526-531.
Parkhill J, Dougan G, James KD, Thomson NR, Pickard D, Wain J, Churcher C, Mungall KL, Bentley SD, Holden MT, Sebaihia M, Baker S, Basham D, Brooks K, Chillingworth T, Connerton P, Cronin A, Davis P, Davies RM, Dowd L, White N, Farrar J, Feltwell T, Hamlin N, Haque A, Hien TT, Holroyd S, Jagels K, Krogh A, Larsen TS, Leather S, Moule S, O'Gaora P, Parry C, Quail M, Rutherford K, Simmonds M, Skelton J, Stevens K, Whitehead S, Barrell BG: Complete genome sequence of a multiple drug resistant Salmonella enterica serovar Typhi CT18. Nature. 2001, 413 (6858): 848-852.
Rutherford K, Parkhill J, Crook J, Horsnell T, Rice P, Rajandream MA, Barrell B: Artemis: sequence visualization and annotation. Bioinformatics. 2000, 16 (10): 944-945.
Frishman D, Mironov A, Mewes HW, Gelfand M: Combining diverse evidence for gene recognition in completely sequenced bacterial genomes. Nucleic Acids Res. 1998, 26 (12): 2941-2947.
Delcher AL, Harmon D, Kasif S, White O, Salzberg SL: Improved microbial gene identification with GLIMMER. Nucleic Acids Res. 1999, 27 (23): 4636-4641.
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. J Mol Biol. 1990, 215 (3): 403-410.
Pearson WR, Lipman DJ: Improved tools for biological sequence comparison. Proc Natl Acad Sci USA. 1988, 85 (8): 2444-2448.
Bateman A, Birney E, Cerruti L, Durbin R, Etwiller L, Eddy SR, Griffiths-Jones S, Howe KL, Marshall M, Sonnhammer EL: The Pfam protein families database. Nucleic Acids Res. 2002, 30 (1): 276-280.
Falquet L, Pagni M, Bucher P, Hulo N, Sigrist CJ, Hofmann K, Bairoch A: The PROSITE database, its status in 2002. Nucleic Acids Res. 2002, 30 (1): 235-238.
Krogh A, Larsson B, von Heijne G, Sonnhammer EL: Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol. 2001, 305 (3): 567-580.
Nielsen H, Engelbrecht J, Brunak S, von Heijne G: A neural network method for identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites. Int J Neural Syst. 1997, 8 (5–6): 581-599.
Griffiths-Jones S, Bateman A, Marshall M, Khanna A, Eddy SR: Rfam: an RNA family database. Nucleic Acids Res. 2003, 31 (1): 439-441.
Vernikos GS, Parkhill J: Interpolated variable order motifs for identification of horizontally acquired DNA: revisiting the Salmonella pathogenicity islands. Bioinformatics. 2006, 22 (18): 2196-2203.
Carver TJ, Rutherford KM, Berriman M, Rajandream MA, Barrell BG, Parkhill J: ACT: the Artemis Comparison Tool. Bioinformatics. 2005, 21 (16): 3422-3423.
MultiFun. Cellfunction assignment schema. [http://genprotec.mbl.edu/files/MultiFun.html]
We would like to acknowledge the support of the Wellcome Trust Sanger Institute core sequencing and informatics groups, particularly Zahra Abdellah, Rebecca Atkin, Tracey Chillingworth, Nancy Holroyd, Kay Jagels, Sharon Moule, Rob Squares and Sally Whitehead. We also would like to acknowledge Henning Sørum for providing access to his large collection of A. salmonicida isolates, and Christopher G. Fenton for his contribution on setting up the bioinformatical infrastructure. This work was partly supported by grants from The Research Council of Norway and the University of Tromsø.
EH: study conception, data analysis, research design, manuscript writing. MSL: research design, data collection, manuscript production. MTGH: research design, manuscript production. KS: data collection. SP: research design, manuscript production. NB: data collection. CC: data collection. DH: data collection. HN: data collection. MAQ: data collection. SS: data collection. ST: data collection. JP: study conception, manuscript production. NPW: study conception, manuscript production. NRT: research design, study conception, manuscript writing.