Notice: Wiley Online Library will be unavailable on Saturday 27th February from 09:00-14:00 GMT / 04:00-09:00 EST / 17:00-22:00 SGT for essential maintenance. Apologies for the inconvenience.
Vibrio vulnificus has been shown to require a global transcription factor, NtrC for mature biofilm development via controlling the biosyntheses of lipopolysaccharide and exopolysaccharide (EPS). Biofilm formation and EPS production were dramatically increased in a medium including a tricarboxylic acid cycle-intermediate as a carbon source. These phenotypes required functional NtrC and were abolished by the addition of ammonium chloride. During the initial stage of biofilm formation, both expression of the ntrC gene and the cellular content of NtrC protein increased. Thus, the regulatory roles of NtrC in EPS biosynthesis were studied with three gene clusters for EPS biosyntheses. Transcriptions of the three clusters were positively controlled by NtrC and showed maximal expression at the early stage of biofilm development. Mutants deficient in one of the genes (VV1_2661, VV2_1579 and VV1_2305) in each cluster showed decreased production of EPS, attenuated ability to form biofilm and lowered cytoadherence to human epithelial cells. However, mutations in VV2_1579 and VV1_2305 resulted in lower cytotoxicity to human cells and mortality to mice than the mutation in VV1_2661. These results demonstrate that NtrC-regulated EPS are crucial in biofilm formation of V. vulnificus, and some EPS components play important roles in interacting with hosts.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Bacteria show surface-associated life styles in multicellular structures, which provide advantages in their survival against diverse environmental stresses. Specifically, the abilities of pathogenic bacteria to form multicellular structures on surfaces facilitate their host invasion via mediating adherence to host tissues and evasion from host defence mechanisms (Costerton et al., 1995). A mature biofilm is composed of aggregated microbial cells surrounded by a self-produced extracellular polymeric matrix (Sutherland, 2001). Extracellular polymeric matrix consists of various components including proteins, nucleic acids and polysaccharides. Extracellular polysaccharides include the exopolysaccharides (EPS) as well as the capsular polysaccharides (CPS), which tend to be more tightly associated with the bacterial envelope than EPS (Whitfield, 2006). In addition, firmly attached lipopolysaccharide (LPS) is an important component for biofilms made by Gram-negative bacteria (Nesper et al., 2001; Kim et al., 2007). EPS is suggested to stabilize bacteria–surface interactions and cell–cell interactions for three-dimensional architectures, and to restrict the diffusion of certain compounds into biofilm structures (Gilbert et al., 1997).
Many types of EPS are thought to be involved in the bacterial formation of biofilm. The gene products of alg, psl or pel clusters in Pseudomonas aeruginosa are necessary for EPS biosyntheses, and some are required for the formation of characteristic structured biofilms made by specific P. aeruginosa strains (Friedman and Kolter, 2004; Matsukawa and Greenberg, 2004). Recently, expression of the pel genes, which are responsible for the production of glucose-rich EPS, was found to be responsive to quorum-sensing signals via LasI and RhlI (Sakuragi and Kolter, 2007). In addition, FleQ, a cyclic-di-GMP (c-di-GMP) binding protein and a master regulator for flagella biosynthesis gene expression, represses transcription of the pel operon (Hickman and Harwood, 2008). In Vibrio spp., EPS is also essential for the development of mature biofilm structures and its production is mediated by various polysaccharide loci, as reviewed by Yildiz and Visick (2009). For example, Vibrio cholerae EPS are synthesized by the gene products of vps, which are clustered in two regions on its chromosomes, the vps-I and vps-II clusters. Expression of the vps genes is positively regulated by VpsR and VpsT and negatively regulated by the quorum-sensing regulator HapR (Yildiz et al., 2001; Hammer and Bassler, 2003; Casper-Lindley and Yildiz, 2004). In addition, c-di-GMP positively regulates vps expression and subsequently controls biofilm formation (Lee et al., 2007).
In contrast to CPS, a limited investigation has been performed on V. vulnificus EPS. The genes required for EPS synthesis were recently identified in strain 1003, and its production was shown to relate with the rugose colony variant of V. vulnificus forming a robust biofilm (Grau et al., 2005; 2008). V. vulnificus biofilm formation is positively regulated by a signal molecule, c-di-GMP, via regulating production of EPS (Nakhamchik et al., 2008). In addition, a global regulator CsrA inhibits biofilm formation of V. vulnificus possibly by repressing EPS synthesis (Jones et al., 2008).
In a previous report, we demonstrated that the global transcription regulator NtrC and its cognate alternative sigma factor, RpoN, are essential in biofilm formation via modulating gene expression of gmhD, of which the gene product constitutes biosynthetic enzymes for LPS biosynthesis (Kim et al., 2007). Interestingly, the gmhD mutation also resulted in a partial defect in EPS production. Therefore, we further studied the roles of EPS in biofilm formation by examining the phenotypes of various EPS-deficient mutants and the expression patterns of three EPS-related gene clusters. Relationships of EPS production with the pathogenicity of V. vulnificus were also examined.
Biofilm formation of V. vulnificus in growth media containing various carbon sources
Wild-type V. vulnificus, strain MO6-24/O, was statically grown in AB medium containing one of the various carbohydrates as a sole carbon source (1.0%). After 48 h incubation, biofilm formation was monitored by staining bacterial assemblages adhered to microtiter plates with crystal violet (Fig. 1A). As previously reported (Kim et al., 2007), growth in the AB-glycerol medium for 48 h results in biofilm-forming activity of about 2, which is the ratio of adhered bacterial mass (OD550) to the planktonic bacterial mass (OD595). While bacterial growth with glucose, maltose, mannitol, fructose, mannose, pyruvate or lactate resulted in similar OD550/OD595 ratios to bacterial growth with glycerol, the addition of organic acids comprising the tricarboxylic acid (TCA) cycle resulted in significantly higher OD550/OD595 ratios, which were 11–15 times more than that in the presence of glycerol (the P-values for the biofilm formation in the presence of TCA intermediates were all less than 0.002 when compared with that in the presence of glycerol, Student's t-test).
Planktonic growth rate of V. vulnificus in AB medium containing fumarate was almost the same as that in AB medium containing glycerol, and the maximal growth yield in AB-fumarate was slightly less than that in AB-glycerol (data not shown). At 48 h incubation of V. vulnificus in AB-glycerol and AB-fumarate media, the final pH were estimated to be 6.0 and 7.3 respectively. The survival and biofilm-forming ability of V. vulnificus, however, were similar under the pH ranges from 6 to 8 (data not shown). These results indicate the observed difference in biofilm formation in the presence of TCA intermediates and glycerol was not derived from the differential planktonic growth or from the different acidity in each growth condition used in this study. Cell numbers of bacteria consisting of biofilm structures in the presence of glycerol, fumarate or succinate were determined by resuspending adhered cells and subsequently plating them on agar plates (Fig. 1B). The total numbers of V. vulnificus cells within the biofilms formed in the presence of fumarate or succinate were about 3.1–3.5 times higher than that in medium containing glycerol (P-values ≤ 0.008, Student's t-test). In addition, cellular protein levels of biofilm assemblages in AB-fumarate or -succinate were estimated to be 2.8–3.0 times higher than those in AB-glycerol (P-values < 0.0001, Student's t-test). These results suggest that the increased biofilm in the presence of TCA intermediates was partially explained by increased biomass within the biofilm, and that other bacterial factor(s) might be responsible for increased OD550 values after staining with crystal violet.
Phenotypic characteristics of V. vulnificus grown in the presence of TCA intermediates
Vibrio vulnificus MO6-24/O was spot-inoculated on AB agar plates containing glycerol, fumarate or succinate, and bacterial colonies were observed. The areas around V. vulnificus colonies grown in the presence of fumarate or succinate showed a large area of white substance, suggesting more secretion of extracellular matrix than in the presence of glycerol (Fig. 2A). Growth in the other TCA intermediates also showed similar colony morphology (data not shown). To define the nature of the secreted materials, extracellular components of cells grown on AB agar plates were collected and treated to isolate EPS. The extracts were then separated on a 5% stacking polyacrylamide gel and visualized by Stains-All. For comparison of the amounts of EPS produced by V. vulnificus grown with glycerol, fumarate and succinate, each EPS was extracted from the same bacterial biomass, which was normalized by bacterial protein contents. V. vulnificus cells grown with fumarate or succinate showed more EPS than those grown with glycerol (Fig. 2B). Densitometric quantification of the EPS gel showed six times more EPS in extracts prepared from the fumarate- or succinate-grown cells than the glycerol-grown cells. Similar difference in EPS contents were observed when a polyacrylamide gel for EPS separation was stained by another dye, Alcian Blue (Cowman et al., 1984) (see Fig. S1).
In addition to analysis of EPS via gel electrophoresis and a subsequent staining, the carbohydrate contents in EPS extracts were measured by the phenol-sulphuric acid method (Dubois et al., 1956). Estimated concentrations of carbohydrates in EPS samples extracted from V. vulnificus cells grown in AB-glycerol, AB-fumarate and AB-succinate agar plates were 2.2 ± 0.5, 14.1 ± 0.5 and 13.1 ± 0.7 μg of carbohydrates per mg of bacterial protein respectively (Fig. 2B). This result also indicates that V. vulnificus cells grown with a TCA intermediate produced 6.0–6.5 times more EPS than those grown with glycerol (P-values < 0.0001, Student's t-test), as shown by gel analysis. It further suggests that the EPS fractions prepared in this study did not include detectable levels of CPS, since this method does not detect the V. vulnificus group 1 CPS, i.e. uronic acid.
Effect of the addition of an inorganic nitrogen on biofilm formation
The nitrogen-regulatory protein, NtrC has been found to be a key regulator for V. vulnificus biofilm formation, via controlling the biosyntheses of LPS and EPS (Kim et al., 2007). Thus, the effect of inorganic nitrogen on biofilm formation was investigated. When various concentrations of NH4Cl (ranged from 0 to 10 mM) were added to AB-fumarate medium, the degree of EPS production (data not shown) and biofilm formation (Fig. 3) by wild-type V. vulnificus decreased in a concentration-dependent manner. The addition of 10 mM NH4Cl to AB-fumarate decreased the biofilm-forming activity of wild type to the level of biofilm formation of the same strain grown in AB-glycerol. Biofilm formation by the ntrC mutant V. vulnificus, however, was not significantly affected by the addition of NH4Cl up to a concentration of 10 mM. In addition, the mutant defective in rpoN, which encodes the NtrC-cognate sigma factor N, was also insensitive to NH4Cl addition (at least up to 7 mM) in its biofilm formation.
These observations suggest that growth conditions in AB medium with a TCA intermediate may be sensed as a nitrogen-poor condition or a condition of high carbon-to-nitrogen ratio via NtrC, and bacterial growth under these conditions produce surplus EPS and therefore increase biofilm formation. In a subsequent experiment, cellular levels of NtrC itself were examined in a V. vulnificus population actively forming biofilm community. The wild-type strain containing the ntrC::luxAB transcriptional fusion was inoculated in AB medium with each carbon source (1.0%) in borosilicate tubes. Then luciferase activities of V. vulnificus in the biofilm samples, of which the planktonic population had been excluded, were monitored (Fig. 4A). The ntrC gene was highly expressed in the presence of TCA intermediates. The increased levels were about six times higher than ntrC expression in the AB-glycerol medium. The significant difference was seen during the early stage of biofilm formation, at 6–10.5 h after inoculation (P-values < 0.01, Student's t-test). The difference in luciferase activities of the ntrC::luxAB reporter in AB-glycerol and AB-fumarate was not derived from an artefact that might have influenced the bioluminescence reaction itself, since V. vulnificus carrying the other reporter, smcR::luxAB (Roh et al., 2006), showed the basically same luciferase activity patterns during the whole period of biofilm formation assay in either AB-glycerol or AB-fumarate media (see Fig. S2). In addition to gene expression, the cellular contents of the NtrC protein (molecular size is 52.5 kDa: Fig. 4B) were monitored using polyclonal anti-NtrC antibodies. NtrC was highly increased in 6–12 h biofilm samples grown in AB-fumarate and maintained the high level until the 24 h sample, while the NtrC band was barely seen in biofilm samples grown in AB-glycerol (Fig. 4C).
Expression of EPS biosynthesis gene clusters
Since EPS production was induced under nitrogen-poor conditions, the role of NtrC in EPS biosynthesis was examined. Genomes of V. vulnificus strain CMCP6 (GenBank Accession No. NC_004459 and NC_004460) and strain YJ016 (GenBank Accession No. NC_005139 and NC_005140) were searched in silico for the presence of tentative gene clusters for EPS biosyntheses (Fig. S3A–C). A total of three clusters were discernable in both strains: VV1_2658-2672, VV2_1574-1582 and VV1_2302-2312 (gene numbers were followed by the strain CMCP6 annotations). The cluster EPS-I (VV1_2658-2672) includes genes highly homologous to the syp genes involved in biosynthesis of the symbiosis-related polysaccharide in Vibrio fischeri (Darnell et al. 2008). A similar cluster is also present in Vibrio parahaemolyticus genome (VP1462-1476: GenBank Accession No. NC_004603) as reported by Yildiz and Visick (2009), but three ORFs are missing in this species when compared with V. vulnificus EPS cluster I. The cluster EPS-II (VV2_1574-1582) was recently reported to be responsible for the production of rugose EPS in V. vulnificus (the wcr locus: Grau et al., 2008). A similar cluster is present in V. cholerae (VC0934-0939: GenBank Accession No. NC_002505). VC0934, named vpsL, has been found to be involved in a phase variation of V. cholerae O1 El Tor colony morphology via rugose polysaccharide biosynthesis (Yildiz et al., 2001). V. cholerae EPS (VPS), of which production is mediated by this cluster, is involved in formation of matrix, pellicle and mature biofilm, in addition to rugose colony formation (Yildiz and Visick, 2009). The highly homologous cluster is also present in V. parahaemolyticus (VP1403-1413), but it was reported to be involved in production of CPS, rather than an EPS (Enos-Berlage and McCarter, 2000). A third EPS cluster, EPS-III (VV1_2302-2312), shows homology to the tentative cluster in V. fischeri (VF0343-0352), but this cluster has not yet been studied in any Vibrio species. Interestingly, each cluster includes its own EPS exporter and the tentative ATPase-typed EPS biosynthesis protein at the third and fourth loci respectively.
Another cluster (that includes 18 ORFs from wza to wbfV; Fig. S3D) was previously found in strain MO6-24/O (Chatzidaki-Livanis et al., 2006), and is reported to be involved in the biosynthesis of the group I CPS. V. vulnificus CPS is known as a determinant of bacterial colony morphology and a potent virulence factor to humans (Wright et al., 1990). Mutation of the wbfY-homologous gene, VP0235, was found to cause defect in normal biofilm development by V. parahaemolyticus (Enos-Berlage et al. 2005).
To investigate the regulatory role of NtrC in the expression of EPS clusters, the upstream regions, including the tentative promoters for the selected EPS clusters and the CPS cluster, were used for constructing luxAB-transcriptional fusions. Wild-type, ntrC and rpoN strains carrying EPS-I::luxAB, EPS-II::luxAB or EPS-III::luxAB fusions were grown in AB-glycerol broth and their expressions were monitored. Expressions of all the EPS cluster fusions were severely decreased in the ntrC and rpoN mutants (data not shown), indicating that expression of the EPS clusters were positively regulated by both NtrC and RpoN. However, CPS::luxAB expression in AB-glycerol broth was not influenced by mutation in ntrC nor rpoN (data not shown).
Transcriptional patterns of EPS clusters (pCB011, pCB012 and pCB013) and a CPS cluster (pCB014) were also investigated in V. vulnificus cells within the biofilm (Fig. 5). Expression of all EPS clusters were maximally induced at the early stages (i.e. at 6–8 h post inoculation) of biofilm formation and the maximal expression was 2.5 times higher in the presence of fumarate (or succinate; data not shown) than glycerol, when cellular NtrC was highly produced (Fig. 4C). In addition, their expression was barely detected in the ntrC mutant. Expression patterns showed a temporal difference between the EPS clusters and the CPS cluster, since pCB014 was maximally expressed after expressions of all three EPS clusters were declined. However, expression of the CPS cluster was not influenced by carbon sources added to the growth medium nor mutation at the ntrC locus (Fig. 5). These results demonstrated a tight relationship among the carbon/nitrogen sources, NtrC, EPS and biofilm formation.
Effect of mutation in each EPS cluster on EPS biosynthesis and biofilm formation
To investigate the biological roles of EPS, V. vulnificus mutants deficient in one of the genes in each EPS biosynthesis clusters were constructed by deleting VV1_2661, VV2_1579 or VV1_2305 to produce CBmΔ1, CBmΔ2 and CBmΔ3 respectively (see Fig. S3). EPS extracts were prepared from the same biomasses of wild-type and mutant cells, which were normalized by their cellular protein levels, and visualized by staining upon electrophoresis. All three mutants showed decreased production of EPS (Fig. 6). To confirm the observed difference in intensities of stained EPS represented the difference in carbohydrate levels, the carbohydrate levels in EPS extracts were measured by the phenol-sulphuric acid method (Dubois et al., 1956). Estimated carbohydrate concentrations in EPS extracts of wild type, CBmΔ1, CBmΔ2 and CBmΔ3 were 14.1 ± 0.5, 4.0 ± 0.5, 2.5 ± 0.5 and 2.2 ± 0.1 μg of carbohydrates per mg of cellular protein respectively. EPS contents in each mutant were significantly reduced when compared the EPS produced by wild type (P-values for all mutants were less than 0.003).
Similarly, CBmΔ1, CBmΔ2 and CBmΔ3 strains were tested for their ability to form biofilms in comparison with the wild type (Fig. 7). The time-course assessment of biofilm formation showed that all mutants exhibited decreased abilities to form biofilms during the entire experiment period. The degrees of reduction in biofilm formation by CBmΔ1, CBmΔ2 and CBmΔ3 were significant (P-values < 0.004, Student's t-test), approximately 1.8-, 2.0- and 4.2-fold lower than the wild type at 72 h post inoculation respectively. A mutant deficient in three EPS clusters (CBtΔ123) showed the similar level of biofilm-forming activity exhibited by CBmΔ3 (data not shown).
To confirm that the decreased abilities for EPS production and biofilm formation by each mutant were caused by the disruption of the VV1_2661, VV2_1579 or VV1_2305 ORFs, each mutant was complemented with the corresponding intact gene in a broad-host-range vector, pRK415. The complemented mutant strains recovered their EPS production (the lanes numbered by 3 in Fig. 6) as well as restored ability for biofilm formation (Fig. 7) to the levels of wild type containing pRK415. Among the three EPS biosynthesis genes, VV2_1579 (in EPS cluster II) and VV1_2305 (in EPS cluster III) influenced EPS biosynthesis at greater extents than VV1_2661 (in EPS cluster I), and VV1_2305 influenced biofilm formation at a greatest extent compared with the other two genes.
Effect of mutation in each EPS cluster on motility
It has been previously reported that a mutation in LPS biosynthesis also results in alteration of both EPS production and motility in V. vulnificus (Kim et al., 2007). Thus, it was further investigated if a mutation in EPS biosynthesis might affect bacterial motility, which is one of the determinants for biofilm formation by V. vulnificus (Lee et al., 2004). Wild type, CBmΔ1, CBmΔ2, CBmΔ3 and a triple mutant CBtΔ123 V. vulnificus were examined for their motility on soft agar plate. All the strains showed the same degree of swimming motility (Fig. 8). Thus, the reduction in biofilm formation by each EPS mutant was not derived from the mutational effect on motility.
Effect of mutation in each EPS cluster on V. vulnificus cytoadherence to the HEp-2 cell line
Based on the observation that all types of EPS synthesized by EPS-I, -II or -III play a role in forming biofilm on abiotic surfaces, we extended our experiments to reveal the function of each EPS in interactions of V. vulnificus with human cells. Human epithelial cells (HEp-2) were challenged with V. vulnificus strains for 30 min at a multiplicity of infection (moi) of 10, and then the number of bacterial cells adhered to HEp-2 was determined (Fig. 9). Bacterial adherence to abiotic surface of the well was also measured by incubating wild type in the adherence assay wells without HEp-2 cells as a negative control, and it was less than 0.2% of the initial inoculum. Approximately 4% of wild-type cells incubated with HEp-2 cells adhered to human cells. Adherence abilities shown by CBmΔ2 and CBmΔ3 were reduced more than sixfold compared with that of the wild type (P-values < 0.007, Student's t-test). However, adherence ability shown by CBmΔ1 only decreased threefold, compared with that of the wild type (P-value of 0.01, Student's t-test). Each mutant complemented with the corresponding intact gene recovered ability of adherence to about 40% of the wild-type level. The partial complementation observed in this experiment might be derived from the defects in the stable maintenance of the plasmids or in the expression of complemented gene products in this specific medium without the appropriate antibiotic, tetracycline.
Effect of the exogenous addition of EPS on V. vulnificus cytoadherence to HEp-2 cells
Although three mutants of EPS biosynthesis showed decreased abilities in both EPS biosynthesis and bacterial cytoadherence, it was not clear if the reduction in cytoadherence by each EPS mutant was directly caused from the defect in biosynthesis of some of EPS components or if it was derived from the pleiotropic effects of the mutations. To differentiate these possibilities, EPS was extracted from wild type (EPSwt) and included in the adherence assay of a triple mutant CBtΔ123 to HEp-2 cells at an moi of 50. Various amounts of EPSwt extract ranged from 0 to 10 μg of carbohydrates were added to the reaction mixture (1 ml) to see the effect on adherence, and the results showed that bacterial adherence increased in a dose-dependent manner (the P-values for EPSwt-added incubations were all less than 0.0003, Student's t-test) (Fig. 10A). The addition of 10 μg of EPSwt resulted in more than 20 times increased adherence compared with the control without added EPS.
Since EPS derived from wild-type V. vulnificus (EPSwt) would include all the EPS components produced by EPS-I, -II and -III clusters, the extent of each EPS on cytoadherence may be distinguished by preparing EPS from double mutants, CBdΔ12 (CBmΔ1/CBmΔ2), CBdΔ13 (CBmΔ1/CBmΔ3) or CBdΔ23 (CBmΔ2/CBmΔ3) to isolate EPSIII, EPSII and EPSI respectively. The control extract was also prepared from a triple mutant, CBtΔ123. To validate the comparison of their effect, the same amount of EPS (2 μg of carbohydrates in 1 ml of adherence mixture) was included in the adherence assay of CBtΔ123 to HEp-2 cells at an moi of 50. An addition of the control extract resulted in 1.0 ± 0.11% adherence of CBtΔ123, which is almost the same as the adherence without any EPS added (0.7 ± 0.13%). EPSII and EPSIII showed a significant positive effect on the cytoadherence of CBtΔ123 to HEp-2, which was estimated to be a 4.4- and 5.9-fold increase in adherence compared with the control respectively (P-values ≤ 0.0004, Student's t-test) (Fig. 10B). In contrast, EPSI showed a relatively minor contribution, and its effect on adherence was a 2.9-fold increase in adherence compared with the control. These results suggest that each EPS might have a differential role in the pathogenic interaction of V. vulnificus, and EPS synthesized by VV1_2661 (in EPS cluster I) did not seem to be a major component for bacterial adherence to abiotic surfaces as well as to host cells.
Effect of mutation in each EPS cluster on V. vulnificus cytotoxicity to human cells and virulence to mice
In subsequent experiments, we examined whether the defects of EPS mutants in biofilm formation and cytoadherence may further affect bacterial virulence towards various host cells. After HEp-2 cells were infected with wild-type V. vulnificus at an moi of 50 for 15–30 min, the viability of host cells was monitored by staining them with propidium iodide (PI) and followed by FACS analysis (Table 1). HEp-2 cells showed 50% and 79% death after 15 and 30 min incubation respectively. In addition to the HEp-2 cell line, V. vulnificus cytotoxicity was tested on other human cell lines, human T-lymphocytes (Jurkat T) and human monocytes (THP-1). THP-1 cells were the most susceptible to V. vulnificus and showed 87% death at an moi of 5 after 15 min, while Jurkat T cells showed 11% death under the same conditions. When these host cell lines were infected with EPS mutants, overall effects of each mutant were similar to each other, although each host cell line showed a differential susceptibility to V. vulnificus. The effect of VV1_2662 (in EPS cluster I) on the death of three cell lines were insignificant, since the mutant CBmΔ1 showed a similar degree of host death caused by the wild type (Table 1). In contrast, CBmΔ2 (in EPS cluster II) and CBmΔ3 (in EPS cluster III) mutants showed a dramatic decrease in cytotoxicity towards host cells; death of HEp-2 (at 15 min), Jurkat T (at 30 min), and THP-1 (at 15 min) by these two mutants were estimated to be 12–17%, 10–11% and 22–23% respectively. These values were three- to fourfold decreased death caused by wild type.
Table 1. Cytotoxicity of wild-type and EPS mutant V. vulnificus strains to various human cell lines.
V. vulnificus-induced cell death of human cell lines were determined by PI-staining and a subsequent FACS analysis, as described in the Experimental procedures. Cytotoxicity assays were performed in triplicates and the representative data set were presented in this table.
The role of V. vulnificus EPS in mouse lethality was then examined. Wild type and the four EPS mutants were injected intraperitoneally into 7-week-old specific pathogen free female mice without pretreatment with iron-dextran. The number of live mice was determined over a period of 48 h after the injection of 5.0 × 105 cells of each strain (Fig. 11). Compared with mice infected with the wild type, mice infected with CBmΔ2 (EPS cluster II) and CBmΔ3 (EPS cluster III) mutants showed significant attenuation in mouse lethality (P-values between 0.02 and 0.003: Log rank test). A triple mutant CBtΔ123 exhibited almost the same level of attenuation in mouse mortality shown by CBmΔ3. A mutation in the EPS cluster I (CBmΔ1), however, resulted in only a slight attenuation in mouse mortality, which is consistent with the observations of its minor effects on EPS production (Fig. 6), biofilm formation (Fig. 7), cytoadherence (Fig. 9) and cytotoxicity (Table 1).
Microorganisms show a wide range of niche specialization for free-living planktonic life style or attached life style to various surfaces, which include mutualistic or pathogenic interactions, particularly in the case of host-interacting microorganisms such as Vibrio spp. (Ruby and Lee, 1998; Yildiz and Visick, 2009). Bacterial abilities to form biofilms are thought to play a central role in their interaction with specific environments and their surface polysaccharides are considered to be a prerequisite for the adaptive advantage of biofilm formation (Costerton et al., 1995). The surface polysaccharides of Gram-negative bacteria include EPS, CPS and LPS, among which EPS is the most prevalent component of extracellular polymeric substances in their mature biofilms (Sutherland, 2001).
Difference in EPS and CPS is ultimately defined by carbohydrate composition, genetics and biosynthetic pathway. However, EPS tends to be loosely associated on the extracellular surface of bacterial cells in contrast to CPS, which is generally a tightly attached polysaccharide to bacterial surfaces (Whitfield, 2006). In the present study, the extraction procedures for EPS and CPS were examined, prior to the investigation of the genetic components and the functional roles of multiple extracellular polysaccharides in V. vulnificus. EPS was extracted from the supernatant of bacterial suspensions after vigorously shaking (Whitfield and Paiment, 2003), and CPS was extracted from the whole cells after their loosely associated polysaccharides have been washed away. EPS and CPS were then observed by electrophoresis on a stacking polyacrylamide gel and subsequent staining with dyes (Enos-Berlage and McCarter, 2000).
The surface polysaccharides of V. vulnificus strains, including wild type, a triple mutant of three EPS clusters (CBtΔ123), and a CPS mutant (CB504: wbpP::miniTn5), have been separately extracted and compared side by side (Fig. S4). The amount of EPS extracted from CBtΔ123 was much less than wild type and CB504 (Fig. S4A), but the triple mutant produced a minimally detectable level of EPS, suggesting the presence of another EPS gene cluster in its genome. However, this fraction extracted from the mutant CBtΔ123, if it were an EPS, did not show any detectable role in adherence because an exogenous addition of this fraction (extractcontol in Fig. 10B) did not increase V. vulnificus adherence to HEp-2 cells and resulted in a similar level of adherence shown by the distilled water control (DW in Fig. 10B). The EPS fraction extracted from a CPS mutant (CB504) showed a slightly reduced level of EPS compared with that of wild type, implying that some EPS components may be shared with those of CPS, as shown in other bacterial CPS/EPS (Whitfield and Paiment, 2003). However, the possibility of cross-contamination between EPS and CPS cannot be excluded during the elution process to separate the loosely associated EPS and the tightly attached CPS fractions.
Vibrio vulnificus CPS extracts contain higher-molecular-weight polysaccharides than EPS extracts, as observed in other bacterial CPS (Rahn et al., 2003). CPS extracted from CBtΔ123, a triple mutant of three EPS clusters, was similar to the level of wild-type CPS (Fig. S4B). In addition, a CPS mutant showed residual levels, indicating the presence of another CPS gene cluster in the V. vulnificus MO6-24/O genome in addition to the cluster shown in Fig. S3D. For example, V. vulnificus strains 1003 and 27562 have different gene clusters for the biosynthesis of the group 4 CPS (Nakhamchik et al., 2007), although this cluster is not observed in strains of CMCP6 or YJ016. CPS has been reported to negatively influence biofilm formation by V. vulnificus MO6-24/O (Joseph and Wright, 2004) and similarly, CPS mutant CB504 showed an increased ability to form biofilms about five times higher than that of the wild type (as described in the legend of Fig. S4). While V. vulnificus CPS was reported to be involved in the phase variation of colony morphology (opaque versus translucent morphotypes; Yoshida et al., 1985), the mutations in all three EPS clusters did not change the opacity of the colonies (data not shown). These results suggest that V. vulnificus produces a heterogeneous mixture of surface polysaccharides exhibiting different functions.
Until now, only the wcr operon of V. vulnificus strain 1003 has been identified to be involved in EPS synthesis (Grau et al., 2008), which is the EPS cluster II in this study (Fig. S3B). It is highly conserved in V. parahaemolyticus (the cps locus) and V. cholerae (the vps locus) (Yildiz and Visick, 2009). EPS cluster I has homology to clusters in the genomes of V. fischeri (the syp locus) and V. parahaemolyticus. The ORFs composed of EPS cluster III are found in the genome of V. fischeri only. This in silico analysis indicates that there is a tremendous difference in genetic organization for surface polysaccharide biosyntheses, as in other reports (Chen et al., 2003; Yildiz and Visick, 2009). At least three kinds of EPS (EPSI, EPSII and EPSIII) produced by V. vulnificus MO6-24/O, although their chemical structures have not yet been identified, are required for biofilm formation during bacterial attachment to abiotic surfaces at the initial stage. They are also important in bacterial aggregation to increase biofilm mass during the biofilm-maturing stage. Furthermore, two kinds of EPS, EPSII and EPSIII, play important roles in determining the pathogenicity of this bacterium during host interactions, as shown in Figs 9–11 and Table 1.
The production of EPS was greatly affected by the kind of carbon source and the concentration of inorganic nitrogen in the growth media for V. vulnificus (Figs 2 and 3). In the presence of any TCA intermediate as a carbon (C) source, expression of the ntrC gene, which encodes a global transcription factor, was induced at the early stage of biofilm formation (Fig. 4A), which resulted in increased NtrC protein levels (Fig. 4C). NtrC is known as a transcriptional activator acting under nitrogen (N)-deprived conditions (Reitzer, 2003). High concentrations of TCA intermediates appear to make bacterial cells sense its condition under the high C/N ratio, due to the changes in pools of organic acids and amino acids, and the reorientation of C-metabolic pathways, as shown in Escherichia coli, P. aeruginosa and Staphylococcus epidermidis (Nishijyo et al., 2001; Leigh and Dodsworth, 2007; Sadykov et al., 2008). In fact, the addition of inorganic nitrogen, such as ammonium chloride, relieves the effect of TCA intermediates on EPS production and biofilm formation by V. vulnificus (Fig. 3). Thus, the following cascade for EPS production is proposed: NtrC protein activation (i.e. phosphorylation) in the presence of high concentration of TCA intermediates, induction of NtrC-regulons including the gln-ntrB-ntrC operon (Fig. 4A) and three EPS clusters (Fig. 5), increased production of the NtrC protein (Fig. 4C) and biosynthesis of EPS molecules (Fig. 2), and finally formation of mature biofilm (Fig. 1). However, the induction of the ntrC gene was maximal 8–10 h after the initiation of biofilm formation, and the increased NtrC level was maintained for up to 24 h. This result suggests a presence of other regulatory factors to repress the expression of ntrC and NtrC-regulons, such as EPS clusters, during the late stage of the biofilm formation process.
The regulatory control connecting TCA intermediates to NtrC-mediated regulation was further confirmed by observing another parameter involved in biofilm formation. V. vulnificus swimming motility is regulated by NtrC (H.-S. Kim and K.-H. Lee, unpubl. data) and thus, the effects of the addition of TCA intermediates or an inorganic nitrogen were examined by inoculating wild-type cells on soft agar plates (Fig. S5). Motility was enhanced in the presence of succinate or fumarate, and their effects were abolished by exogenously adding ammonium chloride, in a concentration-dependent manner.
In other Vibrio spp., quorum-sensing regulators, such as V. harveyi LuxR-homologous proteins, have been shown to differently control the production of extracellular polysaccharides (Yildiz and Visick, 2009); for example, OpaR positively regulates cps expression in V. parahaemolyticus, but HapR negatively regulates vps expression in V. cholerae. A V. vulnificus strain deficient in luxR-homologous gene (smcR) showed a translucent colony morphotype as an opaR mutant of V. parahaemolyticus and a reduced biofilm formation (Lee et al., 2007). However, expression of smcR gene in AB-glycerol medium showed the similar levels of expression in AB-fumarate medium (Fig. S2), where NtrC was highly induced and biofilm formation was significantly enhanced. In addition, a triple EPS mutant CBtΔ123 exhibited opaque colony type as shown by wild-type V. vulnificus (data not shown). Thus, it seems that V. vulnificus EPS production may not be regulated by SmcR. However, it requires further investigation of the effects of diverse regulatory components on gene expression of V. vulnificus EPS and/or CPS clusters during biofilm growth period, to integrate NtrC regulatory pathways with other regulatory mechanisms, such as quorum-sensing regulation.
Salmonella enterica serovar Typhimurium produces EPS that contributes to the formation of biofilm on HEp-2 cells (Ledeboer and Jones, 2005). In P. aeruginosa, mucoid EPS is involved in adherence to the host via acting as an adhesin (Ramphal and Pier, 1985). In addition to its role in adherence, EPS is known to exert immuno-stimulating activity on host cells. In Bacteroides fragilis, its zwitterionic polysaccharides are able to attenuate certain immune defences; for example, the reduction of CD4+-T cell proportion in the lymphocyte and dysregulation of systemic cytokine production (Mazmanian and Kasper, 2006). Xanthomonas sp., a plant pathogen, forms polyanionic polysaccharides that suppress host signalling systems via chelating calcium ions (Aslam et al., 2008). Diverse functions of EPS prompt us to be interested in the roles of V. vulnificus EPS in bacteria–host interactions, in addition to the role in biofilm formation. V. vulnificus EPS produced by clusters II and III play critical roles in cytotoxicity and mouse mortality, whereas EPS made by cluster I showed minimal effects on bacterial virulence to the host (Table 1 and Fig. 11). Therefore, it is not speculated that three EPS clusters selected in this study are involved in the production of a single type of extracellular polysaccharide layer of V. vulnificus MO6-24/O. In a future study, biochemical characterization and chemical identification of each EPS would be necessary to interpret different or differential functions of each EPS in interactions with hosts.
Bacterial strains and media
The strains and plasmids used in this study are listed in Table S1. E. coli strains used for plasmid DNA preparation and for conjugational transfer were grown in Luria–Bertani (LB) medium supplemented with appropriate antibiotics at 37°C. V. vulnificus strains were grown at 30°C in LBS (LB medium containing NaCl at a final concentration of 2.5%) or AB medium (300 mM NaCl, 50 mM MgSO4, 0.2% vitamin-free casamino acids, 10 mM potassium phosphate, 1 mM L-arginine, pH 7.5; Greenberg et al., 1979) supplemented with various carbon sources (1.0%). All medium components were purchased from Difco, and the chemicals and antibiotics were purchased from Sigma. Antibiotics were used at the following concentrations for cultivation of E. coli: ampicillin at 100 μg ml−1, chloramphenicol at 20 μg ml−1, kanamycin at 100 μg ml−1 and tetracycline at 15 μg ml−1. The following concentrations of antibiotics were used for cultivation of V. vulnificus: ampicillin at 500 μg ml−1, streptomycin at 500 μg ml−1, chloramphenicol at 4 μg ml−1, kanamycin at 300 μg ml−1 and tetracycline at 3 μg ml−1.
Extraction and analyses of EPS
Bacterial cells grown on AB-glycerol, -fumarate or -succinate agar plates were suspended in the phosphate buffered saline solution (PBS: 100 mM NaCl, 20 mM sodium phosphate, pH 7.3) to include the same bacterial biomass in each resuspension by adjusting the volumes of PBS via normalizing the bacterial cellular protein determined by the Bradford method (Bio-Rad). Samples were vigorously shaken (200 r.p.m.) for 1 h at 30°C to elute loosely associated extracellular matrix. After cells and debris were removed by centrifugation at 10 000 g for 15 min, supernatants were treated at 37°C for 8 h with RNase A (50 μg ml−1) and DNase I (50 μg ml−1) in the presence of 10 mM MgCl2. Proteinase K (200 μg ml−1) was subsequently added to the reaction mixtures, and incubated at 37°C for 17 h. The remained polysaccharide fractions were extracted twice with phenol-chloroform, precipitated with 2.5× volumes of ethanol, washed with 70% ethanol, and resuspended in water. EPS resuspension was run on a 5% stacking polyacrylamide gel. The gels were stained with a dye, 1-ethyl-2-[3-(1-ethylnaphtho(1,2d)-thiazolin-2-ylidene)-2-methylpropenyl]-naphtho(1,2d)thiazolium bromide (Stains-All; Sigma) or Alcian Blue (see Fig. S1) as described (Cowman et al., 1984; Enos-Berlage and McCarter, 2000). The relative amounts of EPS were quantified by densitometric reading of the stained EPS gel using GelDoc 1000/2000 and Quantity One software (Bio-Rad). The carbohydrate contents in each EPS extract were measured by the phenol-sulphuric acid method developed by Dubois et al. (1956) using glucose as a carbohydrate standard. Estimated carbohydrate concentration was expressed as μg per mg of bacterial protein as described (Enos-Berlage and McCarter, 2000).
Biofilm formation assay
Overnight cultures of the bacterial strains were seeded into the fresh AB medium containing various carbon sources (1.0%) in a 96-well microtiter plate and incubated at 30°C without agitation. At various time points, planktonic cell densities were monitored by measuring OD595. Once the medium and planktonic cells were removed, each well was washed with artificial seawater solution [0.6 M NaCl, 0.1 M MgSO4, 0.02 M CaCl2, 0.02 M KCl, 50 mM Tris-HCl (pH 8.3); Ruby and Nealson, 1977] and stained with 1.0% crystal violet for 30 min. The stained cells were then washed with distilled water, air-dried and resuspended in 100% ethanol (O'Toole and Kolter, 1998). Amount of crystal violet in the stained bacterial mass was quantified by spectrometric reading at 550 nm and bacterial ability for biofilm formation was expressed as a normalized value as described (Kim et al., 2007): OD550 divided by the planktonic cell density (OD595) of each well. To estimate the bacterial cell numbers and biomasses in the biofilm structure, V. vulnificus was incubated in the borosilicate tubes containing 3 ml of AB-glycerol, -fumarate or -succinate. At 48 h incubation, planktonic phases were discarded and biofilm structures on the wall were washed three times with PBS. Then assemblage within a biofilm was resuspended in PBS and subjected to viable cell count and protein quantification using the Bradford method (Bio-Rad).
Western blot analysis of NtrC
A pair of oligonucleotides, ntrCover-BF (5′-CGGGATCCATGAGTAAAGGCTATGTTTGGG-3′; underlined sequence denotes a BamHI restriction site) and ntrCover-PR (5′-AAACTGCAGTGAGTGTGGTTAGTACAGCTC-3′; underlined sequence denotes a PstI restriction site), were used to amplify a 1419-base-paired DNA fragment containing the full sequence of the ntrC gene from the genomic DNA of V. vulnificus MO6-24/O. Recombinant NtrC protein (rNtrC) was overexpressed in E. coli JM109 carrying pQE30-ntrC as a His6-tagged form and purified using a Ni-NTA affinity column as directed by the manufacturer (Qiagen). Purified rNtrC was used to raise polyclonal antibodies by three immunizations of Sprague-Dawley rats (200 μg rNtrC per each immunization) at 3 week intervals. Biofilm assemblages of wild-type V. vulnificus on the borosilicate tubes were obtained by discarding the planktonic fraction of the culture and subsequently washing with artificial seawater solution twice. Lysates of bacterial cells in the biofilm state were prepared by sonication in TNT buffer [10 mM Tris-HCl, 150 mM NaCl and 0.05% (v/v) Tween 20, pH 8.0; Sambrook and Russell, 2001]. One hundred micrograms of each bacterial lysate was fractionated by SDS-PAGE, and transferred to a Hybond P membrane (Amersham). The membrane was incubated with polyclonal antibodies against rNtrC (1:5000 dilution), and then with alkaline phosphatase-conjugated rabbit anti-rat IgG immunoglobulin G (1:5000; Sigma). Immunoreactive protein bands were visualized using an NBT (nitroblue tetrazolium)-BCIP (5-bromo-4-chloro-3-indolylphosphate) system (Promega).
Construction of the EPS mutants
Construction of the EPS-I mutant (CBmΔ1). A region of 1990 bp upstream from the ORF of VV1_2661 was amplified from the genomic DNA of V. vulnificus MO6-24/O using the following two primers: epsA-upF (5′-GGGCCCAGAGATTGGCCCGATTTATGTCGC-3′; underlined sequence denotes an ApaI restriction site) and epsA-upR (5′-CGGGATCCCAAGAATGAATCGTTCAGCCAGCG-3′; underlined sequence denotes a BamHI restriction site). The PCR product was then cloned into pBluescript SKII (+) to produce pCB031. A DNA fragment containing 1727 bp downstream of the VV1_2661 ORF was generated using primers, epsA-downF (5′-CGGGATCCTGCTCGATGCCCGATCCATTCATC-3′; underlined sequence denotes a BamHI restriction site) and epsA-downR (5′-ATTCATTGGCGAGCTCAGCTTGGG-3′; underlined sequence denotes a SacI restriction site). The fragment was cloned into the corresponding sites of pCB031 to result in pCB032. A 1.2 kb kanamycin-resistance gene was isolated from pUC4K (Pharmacia) and inserted into the BamHI site of pCB032 to produce pCB033. A 4.9 kb DNA fragment from pCB033, digested with ApaI and SacI, was ligated to a suicide vector pDM4 (Table S1) to generate pCB034. An E. coli SM10 λpir strain carrying pCB034 was conjugated to V. vulnificus MO6-24/O, and the exconjugants were then selected on thiosulphate citrate bile sucrose medium supplemented with chloramphenicol. Colonies with characteristics indicating a double homologous recombination event (resistance to 5% sucrose, sensitivity to chloramphenicol and resistance to kanamycin) were further confirmed by PCR using epsA-comF (5′-CCCAAGCTTGGAGACACCATCTACATTCCAGAG-3; underlined sequence denotes a HindIII restriction site) and epsA-comR (5′-CGGGATCCTCTTTGGCTTTGGCGATGGCTTGG-3′; underlined sequence denotes a BamHI restriction site), and named Δ12661.
Construction of EPS-II mutant (CBmΔ2). A region of 1270 bp upstream from the VV2_1579 ORF was amplified from V. vulnificus genomic DNA using the following primers: epsB-upF (5′-GGGCCCACCTCAAGAGCATACGACCTGGAC-3′; underlined sequence denotes an ApaI restriction site) and epsB-upR (5′-CGGGATCCCGTAAAGTGCGGATGATTGGTAAG-3′; underlined sequence denotes a BamHI restriction site). The PCR product was then cloned into pBluescript SKII (+) to produce pCB035. Then, 1470 bp DNA fragment downstream of VV2_1579 was generated using primers, epsB-downF (5′-CGGGATCCTTTAGCGAAGCGAAGCGTGTCGCTCCATC-3′; underlined sequence denotes a BamHI restriction site) and epsB-downR (5′-TTCGAGCTCAATACAGGCAACCCT-3′; underlined sequence denotes a SacI restriction site). The fragment was cloned into the corresponding sites of pCB035 to result in pCB036. A 1.2 kb kanamycin-resistance gene, nptI, was inserted into the BamHI site of pCB036 to produce pCB037. A 4.0 kb DNA fragment from pCB037, digested with ApaI and SacI, was ligated to a suicide vector pDM4 to generate pCB038. The mutant Δ21579 was isolated as described above, and further confirmed by PCR using epsB-comF (5′-AACTGCAGCTCCTGGCGATATTG-3; underlined sequence denotes a PstI restriction site) and epsB-comR (5′-CGGGATCCTGCGCGCCACCATTTTTAGATAGG-3′; underlined sequence denotes a BamHI restriction site).
Construction of EPS-III mutant (CBmΔ3). A flanking region of the VV1_2305 ORF (2982 bp) was amplified using the following primer set: epsC-F (5′-CATCGTATGGCCAACAATGACCGC-3′) and epsC-R (5′-AAATAATGGATGAGCCTGCGACCG-3′). The PCR product was then cloned into pGEMT to produce pCB039. A deletion mutation in VV1_2305 was created by using two NruI restriction sites within the VV1_2305 coding region in pCB039, thus yielding pCB040. About a 4.0 kb DNA insert from pCB040, digested with ApaI and SacI, was ligated to a suicide vector, pDM4 to generate pCB041. The same procedure was used to construct the Δ12305 mutant as described above. The mutant was confirmed by PCR using epsC-comF (5′-AACTGCAGTTGAACTCACGCACCCTGTGAACC-3; underlined sequence denotes a PstI restriction site) and epsC-comR (5′-CGGGATCCAAATAATGGATGAGCCTGCGACCG-3′; underlined sequence denotes a BamHI restriction site).
Complementation of EPS genes
An intact VV1_2661 ORF was amplified using the primers, epsA-comF and epsA-comR. The amplified VV1_2661 DNA fragment was digested with HindIII and BamHI, and then cloned into a broad-host-range plasmid, pRK415 (Table S1), predigested with HindIII and BamHI, resulting in positioning of VV1_2661 ORF downstream of the lac promoter in pRK415 to produce pRK415-12661. Similarly, VV2_1579 and VV1_2305 DNA were amplified by using the primer sets, epsB-comF/epsB-comR and epsC-comF/epsC-comR respectively. Amplified DNA fragments were cloned to pRK415 utilizing the appropriate restriction sites in each primer, to produce pRK415-21579 and pRK415-12305. Resultant plasmids were transformed to E. coli SM10 λpir (Table S1), and then transferred into each EPS mutant strain by conjugation.
Construction of the luxAB-based transcriptional fusions ntrC transcriptional fusion. ntrC gene is a part of an operon of glnA-hypothetical ORF-ntrB-ntrC genes. A DNA fragment ranging from −3939 to −1080 relative to the translation initiation codon (IC) for the ntrC gene, was amplified using ntrC-BF (5′-CGGGATCCTGTAAAGTGCCCGATTGCTGCAGC-3′; underlined sequence denotes a BamHI restriction site) and ntrC-BR (5′-CGGGATCCTCATCATCCAAGATGAGCGTCGC-3′; underlined sequence denotes an BamHI restriction site). It includes all the possible promoters for this operon; the NtrC-binding site and RpoN-consensus site locate at −211 and −124 relative to IC for glnA respectively. This DNA fragment was digested with BamHI, and ligated to BamHI-digested pHK0011, which contained the promoterless luxAB genes (Table S1), to produce pCB010.
EPS cluster transcriptional fusions. V. vulnificus genome database [GenBank Accession no. NC_004459.2 (V. vulnificus strain CMCP6) and GenBank Accession no. NC_005139.1 (V. vulnificus strain YJ016)] show that the intergenic spaces between EPS cluster I (VV1_2658-2672) and its upstream ORF (VV1_2657), EPS cluster II (VV2_1582-1574) and its upstream ORF (VV2_1583), and EPS cluster III (VV1_2302-2312) and its upstream ORF (VV1_2301) are composed of 671, 414 and 627 nucleotides respectively. All appear to have tentative, but discernable, promoters. A DNA fragment containing the EPS cluster I promoter region ranged from −3537 to +42 (relative to IC for VV1_2661) was amplified using the epsA-FK (5′-GGGGTACCCCCGCCATGTTCACGTCCAGTATC-3′; underlined sequence denotes a KpnI restriction site) and epsA-RX (5′-GCTCTAGAGAGCTGTTCAATTTCGGCGTGAGT-3′; underlined sequence denotes an XbaI restriction site). A DNA fragment containing the EPS cluster II promoter region from −471 to +84 (relative to IC for VV2_1582) was amplified using epsB-FK (5′-GGGGTACCCACAGCGATGGTCGTGCTTGTAAG-3′; underlined sequence denotes a KpnI restriction site) and epsB-RX (5′-GCTCTAGAGCCGATGATGATCGTTAGATCTGC-3′; underlined sequence denotes an XbaI restriction site). A DNA fragment containing the EPS cluster III promoter region from −670 to +47 (relative to IC for VV1_2302) was amplified using epsC-FK (5′-GGGGTACCCTGAAGCGTTGAACGCCACAGTAC-3′; underlined sequence denotes a KpnI restriction site) and epsC-RX (5′-GCTCTAGAGTAGCGTATCCATCATCGGCAAAC-3′; underlined sequence denotes an XbaI restriction site). These DNA fragments were digested with KpnI and XbaI, and ligated to KpnI/XbaI-digested pHK0011, to produce the reporter fusions, pCB11–13.
CPS cluster transcriptional fusion. CPS cluster of V. vulnificus MO6-24/O has been reported previously (GenBank Accession no. DQ360502.1; Chatzidaki-Livanis et al., 2006). A DNA fragment ranging from −492 to +273 relative to IC for wza (the second ORF of CPS cluster) was amplified using wza-FK (5′-GGGGTACCGTCGGAACTTTAATAGCTCCGGTG-3′; underlined sequence denotes a KpnI restriction site) and wza-RX (5′-GCTCTAGACGATTGAGCGTTTAGCGGATACAG-3′; underlined sequence denotes an XbaI restriction site), and then ligated to KpnI/XbaI-digested pHK0011, to produce pCB014.
Each resultant plasmid was mobilized into wild type, ntrC mutant or rpoN mutant by conjugation and the exconjugants were grown in AB medium supplemented with tetracycline. The light produced by these cells was measured in the presence of 0.006% n-decyl aldehyde using a luminometer (TD-20/20 Luminometer, Turners Designs). Specific bioluminescence was calculated by normalizing the relative light units with respect to cell mass (OD595) as described (Park et al., 2004).
Cell line adherence assay
Adherence assays of V. vulnificus were performed with the HEp-2 cell line derived from human laryngeal epithelial cells (Korean Cell Line no. 10023) in 24-well culture plates by following the procedure as described previously (Goo et al., 2006). Each well of the culture plates was seeded with 105 HEp-2 cells and grown overnight at 37°C in the presence of 5% CO2. After removing the medium and washing the cells twice with PBS, 1 ml of serum-free Dulbecco's modified Eagle medium (Gibco-BRL) was added to the HEp-2 cells. Various V. vulnificus strains (wild-type and EPS mutants carrying pRK415, and EPS mutants carrying complementation plasmids) were grown at 30°C in AB-fumarate medium supplemented with tetracycline (3 μg ml−1). Cell monolayers were then inoculated in triplicate with 50 μl of the diluted bacterial resuspension to give an moi of 10 and incubated at 37°C in 5% CO2 for 30 min. The monolayer was then washed five times with prewarmed PBS to remove nonadherent bacteria. Following the last wash, the HEp-2 cells were lysed with 0.1% Triton X-100 treatment for 15 min. Bacterial cells were recovered from these cells with PBS, serially 10-fold diluted and then plated onto LBS agar. As a negative control, bacterial adherence to abiotic surface of the well was also estimated by incubating the same number of the wild-type V. vulnificus to the adherence assay wells not including HEp-2 cells. For a cytoadherence assay in the presence of various EPS extracted from wild type and several EPS mutants grown on AB-fumarate agar plates, the carbohydrate content in each extract was quantified by the method of Dubois et al. (1956) and each extract was appropriately diluted to make all EPS extracts include the same concentration of EPS. Then, the same volume of each extract was added to adherence assay tubes before bacterial inoculation.
Cytotoxicity of V. vulnificus to three human cell lines were performed using PI, as described previously (Goo et al., 2006). HEp-2 cells were seeded at 4 × 105 cells per well into 48-well culture plates, challenged with V. vulnificus at an moi of 50, and then incubated at 37°C for 15–30 min in the presence of 5% CO2. Besides attached cell such as HEp-2, the resuspended cells, Jurkat T-cell (ATCC No. TIB-152) and THP-1 (ATCC No. TIB-202), were also seeded to 48-well plate with approximately 4 × 105 cells per well and infected with V. vulnificus at an moi of 5 under the same condition as used for HEp-2 cell culture, as describe by Kim et al. (2008). Dead cells were detected by staining them with PI at a final concentration of 1 μg ml−1, and the degree of cell death was assessed using fluorescence-activated cell sorting (FACS) analysis (FACScan; BD Biosciences). FACS analysis was performed at least 10 000 cells per sample.
Mouse mortality experiment
For mouse mortality experiment, specific pathogen-free, 7-week-old, female ICR mice (CrljOri:CD1[ICR]) were used without an iron-dextran pretreatment. Bacteria grown overnight in LBS medium were freshly inoculated to the same medium, and harvested when bacterial growth reached an OD600 of 1.0. One hundred microlitres of 5 × 105 cfu of the bacterial suspension in PBS was then injected intraperitoneally into 10 mice, and then numbers of dead mice were counted for 48 h.
Results were expressed as the means ± standard deviations from at least three independent experiments. Statistical analysis was performed using Student's t-test (systat program, SigmaPlot version 9; Systat Software) to estimate P-values. Log rank test ( http://bioinf.wehi.edu.au/software/russell/logrank/) was performed to estimate statistical significance of the mouse mortality results (Fig. 11).
This study was supported by a grant from the Extreme Molecular Genomics Research Program of the Ministry of Marine Affairs and Fisheries (to K.-H.L.), and by Basic Science Research Program through the National Research Foundation funded by the Ministry of Education, Science and Technology (2009-0070681 to K.-H.L.), Republic of Korea. Authors thank W.-H. Kim (Yonsei Univ. Coll. Medical) for assisting the cytotoxicity assays using various human cell cultures.