536
that other serogroups, especially serogroup O2, which produces a 50 kDa porin
(Davey et al. 1998 ) may be more aggressive to fi sh than serogroup O1.
The presence of common antigens between V. anguillarum and other taxa would
explain the cross protection observed with fi sh vaccines (e.g. Nakai et al. 1989 ).
What is the nature of these cross protecting antigens? From capsular antigens
(Rasmussen and Larsen 1987 ), attention focused on the OMP (Chart and Trust
1984 ) – a porin of ~40 kDa molecular weight, which is now regarded as a common
antigen (Simón et al. 1996 ). This porin, Om1, was examined by immunoblotting,
ELISA and dot blot; antiserum to Om1 of serogroup 01 cross reacted with Vibrio
spp., but not with other genera – except Plesiomonas shigelloides (Simón et al.
1998 ). Detailed investigation described a 38 kDa OMP, which was considered to be
involved in environmental adaptation and resistance to bile (Wang et al. 2003 ).
Debate has centred over the possible interaction of exotoxins and/or endotoxins
(Bullock and Conroy 1971 ; Abe 1972 ; Umbreit and Ordal 1972 ; Grischkowsky
1973 ; Inamura et al. 1984 ; De La Cruz and Muroga 1989 ). Umbreit and Ordal
( 1972 ) reported mortalities in goldfi sh following injection of fi lter-sterilised supernatant derived from 24 h broth cultures. Thus, the debilitating effect of bacterial
ECP was suggested (but not proven!) at this stage; but later held responsible for
virulence [a neurotoxic acetylcholinestase has been isolated from the ECP; Peréz
et al. 1998 ] (Lamas et al. 1994a , b ). This experiment was repeated, and resulted in
>70 % mortalities among a group of goldfi sh (Umbreit and Tripp 1975 ).
Unfortunately, the 3-year interval between publications did not achieve any signifi -
cant improvement in knowledge. A similar basic theme was used by Abe ( 1972 ),
who injected ‘endotoxins’ into chinook salmon and recorded the presence of haemorrhaging lesions at the point of injection. A signifi cant development stems from the
work of Wolke ( 1975 ) and Roberts ( 1976 ), who suggested that ‘haemolytic toxins’
might be responsible for the anaemic response in infected fi sh. Subsequent efforts
by Munn ( 1978 , 1980 ) demonstrated conclusively that haemolysins were involved.
These were described as thermolabile enzymes (activity was lost by heating to
50 °C for 10 min) with optimum pH of 7.2–7.4, which were inactivated by gangliosides. The molecular weight of one of these enzymes was estimated as 191 kDa
(Munn 1980 ). In the ensuing experiments, haemolytic activity was fi rst detectable
in cultures (fi ltrates) after 19 h incubation at 20 °C. Production peaked at 39 h, and
then declined. One explanation for this result is that haemolysin production only
occurs during the stationary phase of growth. Another possibility is that production
of the enzyme occurs intracellularly, with release into the environment taking place
at a later time, perhaps during autolysis of the bacterial cells. Munn postulated that
inactive haemolysins may also be secreted by the bacteria. These enzymes could
then be re-activated by as-yet unknown phenomena, at a later period. A haemolysin
gene, vah1, has been identifi ed, but does not account for the haemolytic activity.
Subsequently, a rtx operon was studied, and concluded to represent a second haemolysin gene cluster. Both vah1 and the RtxA activator have cytotoxic activity to
Atlantic salmon kidney cells, with the former causing cell vacuolation, and the later
causing rounding of the cells (Li et al. 2008 ). Proteases of 36 kDa molecular weight
have also been implicated with virulence (Inamura et al. 1984 , 1985; Kodama et al.
10 Vibrios
that other serogroups, especially serogroup O2, which produces a 50 kDa porin
(Davey et al. 1998 ) may be more aggressive to fi sh than serogroup O1.
The presence of common antigens between V. anguillarum and other taxa would
explain the cross protection observed with fi sh vaccines (e.g. Nakai et al. 1989 ).
What is the nature of these cross protecting antigens? From capsular antigens
(Rasmussen and Larsen 1987 ), attention focused on the OMP (Chart and Trust
1984 ) – a porin of ~40 kDa molecular weight, which is now regarded as a common
antigen (Simón et al. 1996 ). This porin, Om1, was examined by immunoblotting,
ELISA and dot blot; antiserum to Om1 of serogroup 01 cross reacted with Vibrio
spp., but not with other genera – except Plesiomonas shigelloides (Simón et al.
1998 ). Detailed investigation described a 38 kDa OMP, which was considered to be
involved in environmental adaptation and resistance to bile (Wang et al. 2003 ).
Debate has centred over the possible interaction of exotoxins and/or endotoxins
(Bullock and Conroy 1971 ; Abe 1972 ; Umbreit and Ordal 1972 ; Grischkowsky
1973 ; Inamura et al. 1984 ; De La Cruz and Muroga 1989 ). Umbreit and Ordal
( 1972 ) reported mortalities in goldfi sh following injection of fi lter-sterilised supernatant derived from 24 h broth cultures. Thus, the debilitating effect of bacterial
ECP was suggested (but not proven!) at this stage; but later held responsible for
virulence [a neurotoxic acetylcholinestase has been isolated from the ECP; Peréz
et al. 1998 ] (Lamas et al. 1994a , b ). This experiment was repeated, and resulted in
>70 % mortalities among a group of goldfi sh (Umbreit and Tripp 1975 ).
Unfortunately, the 3-year interval between publications did not achieve any signifi -
cant improvement in knowledge. A similar basic theme was used by Abe ( 1972 ),
who injected ‘endotoxins’ into chinook salmon and recorded the presence of haemorrhaging lesions at the point of injection. A signifi cant development stems from the
work of Wolke ( 1975 ) and Roberts ( 1976 ), who suggested that ‘haemolytic toxins’
might be responsible for the anaemic response in infected fi sh. Subsequent efforts
by Munn ( 1978 , 1980 ) demonstrated conclusively that haemolysins were involved.
These were described as thermolabile enzymes (activity was lost by heating to
50 °C for 10 min) with optimum pH of 7.2–7.4, which were inactivated by gangliosides. The molecular weight of one of these enzymes was estimated as 191 kDa
(Munn 1980 ). In the ensuing experiments, haemolytic activity was fi rst detectable
in cultures (fi ltrates) after 19 h incubation at 20 °C. Production peaked at 39 h, and
then declined. One explanation for this result is that haemolysin production only
occurs during the stationary phase of growth. Another possibility is that production
of the enzyme occurs intracellularly, with release into the environment taking place
at a later time, perhaps during autolysis of the bacterial cells. Munn postulated that
inactive haemolysins may also be secreted by the bacteria. These enzymes could
then be re-activated by as-yet unknown phenomena, at a later period. A haemolysin
gene, vah1, has been identifi ed, but does not account for the haemolytic activity.
Subsequently, a rtx operon was studied, and concluded to represent a second haemolysin gene cluster. Both vah1 and the RtxA activator have cytotoxic activity to
Atlantic salmon kidney cells, with the former causing cell vacuolation, and the later
causing rounding of the cells (Li et al. 2008 ). Proteases of 36 kDa molecular weight
have also been implicated with virulence (Inamura et al. 1984 , 1985; Kodama et al.
10 Vibrios