Using Western blot to examine the nature of soluble antigens produced by Renibacterium salmoninarum, it was found that the major 57-kilodaIton (kDa) antigen was unstable. SDS-PAGE of extracellular product (ECP) fractions showed that degradation of the 57-kDa protein increased with time and increased temperature. Several lower molecular mass pcptides accumulated temporarily from this degradation. Phenylmethylsuiphonyl fluoride prevented breakdown of the 57-kDa protein suggesting a serine protease present in the ECPs was responsible. The results indicated that most, if not all, immunoreaetive bands in ECP fractions, other than the 57-kDa protein, arose as a result of degradation of this protein. Western blot analysis of two dimensional gels revealed that the presumptive proteolytic activity was associated with the 57-kDa antigen and several of the apparent degradation products. Many common peptide fragments appeared to be generated from heat-induced proteolysis of these protein moieties, confirming the familial relationship between much of the immunoreaetive material in ECP fractions. The results suggested that the 57-kDa antigen is autolytic. Western blot analysis of tissue samples from Atlantic salmon, Salmo salar L., infected with R. salmoninarum suggested that this lability of the 57'kDa antigen also occurred in situ.

Mutants with multiple low-level antibiotic resistance were isolated from virulent wild-type Aeromonas salmonicida strains exposed to a low concentration of any one of several low-molecular-mass (approximately 635 daltons or less) antibiotics. Multiple resistance was toward beta-lactam compounds (penicillin G, ampicillin, cloxacillin), quinolones (flumequine, oxolinic acid, nalidixic acid), tetracyclines, chloramphenicol, and novobiocin. Susceptibilities of the mutants toward several higher-molecular-mass (>700 daltons) hydrophobic or polycationic antibiotics such as rifampin, erythromycin, polymyxin B, and streptomycin sulfate were not affected. The mutants-were obtained at frequencies suggesting point mutations. Outer membrane protein profiles, examined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, revealed that all multiple low-level resistant mutants were deficient in a major protein of approximately 38.5 kilodaltons and contained a major protein of approximately 37 kilodaltons which was not present in significant amounts in the wild-type strains. In addition, these mutants lacked exoprotease activity. Furthermore, mutants isolated as deficient in exoprotease were found, with the exception of one avirulent strain, to exhibit multiple low-level antibiotic resistance and the outer membrane protein changes.

We have examined the possibile relationships between extracellular protease and haemolysin and virulence in Aeromonas salmonicida, and the possible plasmid-encoded origin of these extracellular enzyme activities. Variants, isolated from different strains, showed simultaneous loss of both protease and haemolysin. The frequency of spontaneous loss from one strain (0.1–0.4%) as well as the frequency of induced loss from four strains (0.3–18%) treated with ethidium bromide suggested a plasmid-encoded origin for these enzymes. However, plasmid analyses showed no detectable loss of plasmids correlated with the loss of the extracellular activities. There was no change in the LD100 when fish were exposed to one virulent strain or its protease, haemolysin-negative variants indicating that these extracellular activities are not necessary for virulence or pathogenicity in the acute form of furunculosis. Furthermore, both protease, haemolysin-negative and protease, haemolysin-positive clones isolated from a second virulent strain treated with ethidium bromide were avirulent (LD50 increased >4 logs). Thus, attenuation of A. salmonicida can occur without detected loss of the extracellular enzymes, plasmids, or the A-layer.

A simple, rapid PCR assay for the identification of Renibacterium salmoninarum in Atlantic salmon (Salmo salar L.) tissues detected DNA extracted from between 4 and 40 bacterial cells. PCR was at least as sensitive as culture when it was used to identify subclinically infected fish experimentally challenged with R. salmoninarum. However, PCR identified much higher numbers of kidney tissue and ovarian fluid samples from commercially reared broodstock fish to be positive for R. salmoninarum than did culture. This difference may be due to the antibiotic chemotherapy of broodstock fish used by the industry in 1994 to control the vertical transmission of R. salmoninarum. A much closer relationship between PCR and culture results was observed for ovarian fluid samples collected from broodstock fish in 1993. Also, PCR scored a much higher percentage of kidney tissue samples than ovarian fluid samples from 1994 broodstock fish positive for R. salmoninarum, which may reflect the uneven distribution of the pathogen in different fish tissues. Inclusion of a nested probe to identify the PCR-positive 1994 ovarian fluid samples increased the sensitivity of detection to between one and four cells and the number of samples that scored positive by almost threefold. These data indicate that many infected ovarian fluid samples contained very low numbers of R. salmoninarum cells and, because almost all these samples were culture negative, that PCR may have detected dead or otherwise unculturable bacterial cells.

No abstract