Strains of Vibrio vulnificus biotype 2, isolated from internal organs of diseased European eels as pure cultures of opaque cells, together with some reference strains from Japanese eels, were used in this study. Spontaneous translucent-phase variants were obtained from the corresponding parent strains and compared for a variety of phenotypic traits related to virulence for eels. The rate of colony dissociation from opaque to translucent cells was higher (around 10-2) than that observed for translucent to opaque cells (10-3 to 10-4). Electron microscopy with ruthenium red revealed the presence of a capsule of variable thickness on opaque cells, whereas translucent-type colonies had no observable capsular materials. No differences in plasmid profiles were detected between the two cell types so that plasmids do not seem to be implicated in the mechanism of phase shift of biotype 2 strains. No apparent difference in outer membrane protein and lipopolysaccharide patterns could be observed between the cell types. Both isogenic morphotypes were able to grow in eel serum and minimal medium supplemented with ethylenediamine di(O-hydroxyphenyl-acetic acid) or transferrin. Therefore, the presence of capsule was not required for the acquisition of iron from iron chelators or for resistance to serum bactericidal action. Both morphotypes were highly virulent for elvers, although the 50% lethal dose for translucent cells was higher than that for the corresponding opaque cells. The latter observation, together with the overall data, suggests that the production of capsular materials by biotype 2 of V. vulnificus is not essential for the development of vibriosis in eels, at least when cells are injected intraperitoneally.

No abstract