The extracellular metalloprotease of Vibrio tubiashii is a major virulence factor for pacific oyster (Crassostrea gigas) larvae

Abstract

Vibrio tubiashii is a recently reemerging pathogen of larval bivalve mollusks, causing both toxigenic and invasive disease. Marine Vibrio spp. produce an array of extracellular products as potential pathogenicity factors. Culture supernatants of V. tubiashii have been shown to be toxic to oyster larvae and were reported to contain a metalloprotease and a cytolysin/hemolysin. However, the structural genes responsible for these proteins have yet to be identified, and it is uncertain which extracellular products play a role in pathogenicity. We investigated the effects of the metalloprotease and hemolysin secreted by V. tubiashii on its ability to kill Pacific oyster (Crassostrea gigas) larvae. While V. tubiashii supernatants treated with metalloprotease inhibitors severely reduced the toxicity to oyster larvae, inhibition of the hemolytic activity did not affect larval toxicity. We identified structural genes of V. tubiashii encoding a metalloprotease (vtpA) and a hemolysin (vthA). Sequence analyses revealed that VtpA shared high homology with metalloproteases from a variety of Vibrio species, while VthA showed high homology only to the cytolysin/hemolysin of Vibrio vulnificus. Compared to the wild-type strain, a VtpA mutant of V. tubiashii not only produced reduced amounts of protease but also showed decreased toxicity to C. gigas larvae. Vibrio cholerae strains carrying the vtpA or vthA gene successfully secreted the heterologous protein. Culture supernatants of V. cholerae carrying vtpA but not vthA were highly toxic to Pacific oyster larvae. Together, these results suggest that the V. tubiashii extracellular metalloprotease is important in its pathogenicity to C. gigas larvae.

FIG. 1.

Proteolytic and hemolytic analyses of bacterial culture supernatants. Quantitative data are shown for proteolytic (A) and hemolytic (B) activities of culture supernatants of V. tubiashii pathogenic strains RE22 and RE98 and the nonpathogenic bacterial isolate RE15. (C) Zymography analysis of these supernatants. Bacteria were grown in LB medium supplemented with 1% NaCl at 25°C and were harvested at an OD₆₀₀ of approximately 3.0. Proteolytic and hemolytic activities were determined using azocasein and sheep blood, respectively, as described in Materials and Methods. The error bars indicate standard deviations (n = 3). Data for proteolytic and hemolytic activities were evaluated by Student's t test (*, P < 0.01 compared with RE15). Bands of proteolytic activity in the zymogram gel are shown as clear protein bands in a dark background. The molecular masses (kDa) on the left indicate the positions of molecular size markers.

FIG. 2.

Relationship between growth and proteolytic (A) and hemolytic (B) activities of culture supernatants of V. tubiashii strain RE22. Bacteria were grown in LB medium supplemented with 1% NaCl at 25°C, and samples were harvested at different times during bacterial growth. The error bars indicate standard deviations (n = 3).

FIG. 3.

Effects of protease and hemolysin inhibitors on proteolytic and hemolytic activities and on toxicity to Pacific oyster larvae. Enzymatic activities and toxicity levels are shown as percentages of those for nontreated samples. For the toxicity assay, filter-sterilized supernatants were added to a final concentration of 1%. The error bars indicate standard deviations (n = 3). Data for proteolytic and hemolytic activities were evaluated by Student's t test (*, P < 0.01 compared with the nontreated control). Data for larval mortality were evaluated by chi-square test (*, P < 0.05 compared with nontreated control). NT, nontreated; TEP, tetraethylene pentamine; OPA, 1,10-phenanthroline; PMSF, phenylmethylsulfonyl fluoride; PPA, pepstatin A; CHOL, cholesterol.

FIG. 4.

Alignment of deduced V. tubiashii metalloprotease amino acid sequence with those of various bacterial species. Numbers on the right refer to the positions of the amino acid residues. The black bar indicates the previously identified region by Delston et al. (9). Arrows indicate critical residues for zinc binding. Black shaded areas indicate identical amino acids in all strains, and gray shaded areas indicate identical or similar amino acids in eight or more strains at any position. The following sequences were aligned using ClustalW: zinc metalloproteases of Vibrio sp. strain MED222 (GenBank accession no. NZ_AAND01000005), V. splendidus strain 12B01 (accession no. ZP_00990032), V. proteolyticus (accession no. AAA27548), Vibrionales bacterium strain SWAT-3 (ZP_01816166), Vibrio (Listonella) anguillarum strain M93Sm (accession no. AAR88093), Vibrio vulnificus strain YJ016 (accession no. NP_937521), V. cholerae strain 623-39 (accession no. ZP_01980763), V. aestuarianus strain 01/32 (accession no. AAU04777), V. angustum strain S14 (accession no. ZP_01236251), Photobacterium sp. strain SKA34 (accession no. ZP_01158654), and V. fluvialis strain AQ0005 (accession no. BAB86344).

FIG. 5.

Alignments of deduced amino acid sequences of VthA and VthB with V. vulnificus VvhA and VvhB. (A) Genetic organization of the vthA and vthB genes in V. tubiashii strain RE22. ClustalW alignments were done with VthB (B) and VthA (C) and their V. vulnificus homologs (GenBank accession no. AB124803). The black bar indicates the region previously identified by Kothary et al. (20).

FIG. 6.

Analyses of supernatants from a VtpA-deficient mutant of V. tubiashii strain RE22. Protease (A) and hemolysin (B) production, as well as toxicity to oyster larvae (C), was compared to that of the wild-type strain. Although the mutant strain, but not the wild-type strain, was grown in the presence of 50 μg/ml kanamycin, all cells were harvested at an OD₆₀₀ of approximately 3.0. For the toxicity assay, filter-sterilized supernatants were added to a final concentration of 0.5%. The error bars indicate standard deviations (n = 3). Data for proteolytic and hemolytic activities were evaluated by Student's t test (*, P < 0.01 compared with the wild type). Data for larval mortality were evaluated by chi-square test (*, P < 0.05 compared with the wild type).

FIG. 7.

Expression of vtpA and vthA in V. cholerae. Protease (A) and hemolysin (B) production, as well as toxicity to oyster larvae (C), was analyzed for V. cholerae strains O395N1 and 638 carrying the empty vector (pBAD-TOPO) or the V. tubiashii metalloprotease (pBAD-vtpA) or hemolysin (pBAD-vthA) gene. For the toxicity assay, filter-sterilized supernatants were added to a final concentration of 1%. The error bars indicate standard deviations (n = 3). Data for proteolytic and hemolytic activities were evaluated by Student's t test (*, P < 0.01 compared with the empty vector). Data for larval mortality were evaluated by the chi-square test (*, P < 0.05 compared with the empty vector).