Vibrio species isolated from diseased farmed sole, Solea

senegalensis (Kaup), and evaluation of the potential

virulence role of their extracellular products
I Zorrilla, S Arijo, M Chabrillon, P Diaz, E Martinez-Manzanares, M C Balebona
and M A Moriigo
Department of Microbiology, Faculty of Sciences, University of Malaga, Spain
Abstract
Bacteria isolated from an outbreak with moderate
mortalities in farmed sole, Solea senegalensis (Kaup),
in the south of Spain were identied as Vibrio
harveyi and V. parahaemolyticus. Only bacterial
strains showing swarming were virulent in sole and
caused mortalities in experimentally inoculated sh.
However, the signs of the disease were only repro-
duced with V. harveyi. The intramuscular inocula-
tion of the extracellular products (ECPs) of both
species produced mortalities in inoculated sh and
the appearance of surface ulcers in the case of
V. harveyi. However, the inoculation of sublethal
doses of ECPs to sh showed a protective effect
against V. harveyi.
Keywords: extracellular products, Solea senegalensis,
Vibrio spp., virulence.
Introduction
Marine aquaculture in southern Europe is an
expanding industry in which production has been
concentrated on species such as gilthead sea bream,
Sparus aurata L., and sea bass, Dicentrarchus labrax
(L.). However, diversication in culture is required.
A new species currently being cultured in the
southwestern and Mediterranean areas of Spain
and Portugal is the sole, Solea senegalensis (Kaup),
a at sh well adapted to temperate waters and
with a high economic value. A number of studies
on its developmental biology have been carried
out (Ribeiro, Sarasquete & Dinis 1999; Fernan-
dez-Diaz, Yufera, Canabate, Moyano, Alarcon &
Diaz 2001). However, unlike other sh, such as
gilthead sea bream, in which the incidence of
disease and the types of bacterial pathogens have
been well documented (Real, Deniz, Acosta &
Oros 1993; Balebona, Zorrilla, Morinigo &
Borrego 1998), information concerning the isola-
tion and characterization of pathogenic bacteria
from sole is scarce (Zorrilla, Balebona, Morinigo,
Sarasquete & Borrego 1999). Identication of
pathogenic micro-organisms is important for
future research on vaccine development, as well
as for other means of controlling and treating
disease.
In February 2001, an outbreak associated with
moderate mortalities (20%) in populations of sole
cultured in a farm in the south of Spain was
observed, the main external signs of the disease
being skin ulcers and haemorrhagic areas near the
ns and mouth. The aim of the present study was to
isolate and characterize the potential causative
agent(s) of the disease outbreak, and to demonstrate
its implication in the disease by carrying out
experimental challenges with the isolated bacteria.
Similarly, because extracellular products (ECPs) are
important virulence factors of some sh pathogenic
microorganisms (Fouz, Barja, Amaro, Rivas &
Toranzo 1993; Biosca & Amaro 1996; Balebona,
Andreu, Bordas, Zorrilla, Morinigo & Borrego
1998) we have evaluated the role of ECPs from the
isolated micro-organisms and assessed their poten-
tial use as a vaccine.
Journal of Fish Diseases 2003, 26, 103108
Correspondence Dr M A Morinigo, Department of Microbio-
logy, Faculty of Sciences, University of Malaga, 29071 Malaga,
Spain
(e-mail: morinigo@uma.es)
103
2003
Blackwell Publishing Ltd
Materials and methods
Sampling and processing of diseased sh
Six diseased 70 g sole from one sh farm located in
Granada (southern Spain) were sampled for bac-
teriology. These sh were sampled immediately
after their death. The sh were kept in tanks
receiving openow sea water at 33& salinity and
18 C. Diseased sh exhibited haemorrhagic lesions
on both body surfaces and pale livers. Samples from
ulcers, head, kidney, spleen and liver were cultured
on tryptone soya agar (TSA) and in tryptone soya
broth (TSB) (Oxoid Ltd, Hampshire, UK), both
supplemented with 1.5% NaCl (TSAS and TSBS,
respectively) and Flexibacter maritimus medium
(FMM) (Pazos, Santos, Mac as, Nunez & Toranzo
1996). All inoculated media were incubated at
22 C for between 24 and 48 h, and a represen-
tative number of the different colony types grown
on the plates were picked and biochemically
identied.
Identication of bacterial strains
Pure cultures of the isolates were identied bio-
chemically and physiologically following the criteria
proposed by Alsina & Blanch (1994) and Ortigosa,
Garay & Pujalte (1994). The physiological and
biochemical tests carried out are summarized in
Table 1. After the initial identication of the
isolates as Vibrio harveyi and V. parahaemolyticus,
their proles were compared with two reference
strains from the Spanish Type Culture Collection
(V. harveyi CECT 525 and V. parahaemolyticus
CECT 511).
Virulence testing
The virulence of all the bacterial isolates for sole
(Table 2) was determined in vivo following the
technique described by Amaro, Biosca, Esteve, Fouz
& Toranzo (1992). The strains assayed were grown
in TSBS at 22 C for 24 h. Then the cells were
recovered by centrifugation (10 000g for 30 min)
and the pellets resuspended in phosphate-buffered
saline (PBS) (pH 7.2). Groups of 20 sh (1015 g
body weight), maintained in tanks of 50 L sea water
at a salinity of 35& and temperature of 20 C,
were inoculated intraperitoneally (i.p.) and intra-
muscularly (i.m.) with bacterial doses ranging from
10
4
to 10
8
cfu mL
)1
. The same number of sh
were inoculated with 0.1 mL PBS and used as
controls. Inoculated sh were observed daily for
14 days after inoculation and mortalities recorded.
Mortalities were considered to be due to the
inoculation only when the bacterial strain was
reisolated in pure culture. These experiments were
carried out in triplicate. Lethal dose 50% (LD50)
represents the number of bacteria needed to kill
50% of the inoculated sh.
ECP toxicity testing
Bacterial ECPs from the virulent strains of
V. harveyi and V. parahaemolyticus were obtained
by the method of Amaro et al. (1992). Briey, tubes
containing 5 mL of TSBS were inoculated with
bacteria from one colony from a 24-h culture on
TSAS and incubated at 22 C for 2448 h. A
0.2 mL volume of the culture was spread onto a
sterile cellophane sheet overlying a TSAS plate and
incubated at 22 C for 2448 h. Bacterial cells were
harvested in PBS and the cell suspensions were
centrifuged at 13 500g for 20 min at 4 C. The
supernatants were ltered through 0.45 and 0.2 lm
membrane lters and used as crude ECP prepara-
tions. The total protein contents of the ECP
samples were determined using a commercially
available protein determination reagent (Biorad
Laboratories, Hercules, CA, USA) and bovine
serum albumin (Sigma Chemical Co., St Louis,
MO, USA) as a standard.
The role of the ECP in the pathogenicity of these
micro-organisms was assayed with the virulent
strains V. harveyi LG-16.00 and V. parahaemolyticus
LG-4.00, following the methodology described by
Amaro et al. (1992). Groups of 20 sh were
inoculated i.m. with 0.1 mL of a solution contain-
ing 120 or 60 lg of protein mL
)1
for V. harveyi
and 180 or 90 lg of protein mL
)1
for V. parahae-
molyticus. The same number of sh was inoculated
with 0.1 mL of PBS and used as a control group.
Three replicates of this experiment were carried out.
The sh were observed for 14 days for the presence
of mortalities and harmful effects of the ECP.
Vaccination study
The potential use of the ECP as a vaccine against
strains LG-16.00 and LG-4.00 was examined in
two experiments. Groups of 20 sh of 510 g body
weight were inoculated i.m. with sublethal doses
of raw ECP from strains LG-16.00 and LG-4.00
(60 and 90 lg protein mL
)1
, respectively). After
104
2003
Blackwell Publishing Ltd
Journal of Fish Diseases 2003, 26, 103108 I Zorrilla et al. Pathogenic Vibrio species from farmed sole
1 month, these sh were challenged i.p. with
10
5
cfu g
)1
sh of V. harveyi LG-16.00 or
V. parahaemolyticus LG-4.00. The same number
of sh were inoculated i.p. with the LD50 of these
strains and used as controls. Mortalities and clinical
signs (ulcers or haemorrhagic areas) were monitored
over 14 days. The results were evaluated by deter-
mining the percentage mortality in the control and
challenged groups, and the relative percentage of
survival (RPS) was calculated: [1)(% specic loss
controls)% specic loss challenged/% specic loss
controls)] 100.
Statistical analysis
Results were analysed by a Students t-test. Differ-
ences were considered statistically signicant at
P < 0.05.
Table 1 Biochemical and morphological characterization of strains of Vibrio harveyi and V. parahaemolyticus isolated from Solea
senegalensis and V. harveyi CECT 525 and V. parahaemolyticus CECT 511
V. harveyi
(n 5)
V. harveyi
CECT 525
V. parahaemolyticus
(n 3)
V. parahameolyticus
CECT 511
Growth on TCBS Y
a
Y G
b
G
Gram stain ) ) ) )
Swarming V
c
+ V )
Cytochrome oxidase + + + +
O/129 sensitivity + + + +
Growth at % NaCl
0% ) ) ) )
6% + + + +
8% + + + +
Growth at
4 C ) ) ) )
22 C + + + +
37 C + + + +
ONPG test ) ) ) )
Arginine dihydrolase ) ) ) )
Lysine decarboxylase + + + +
Ornithine decarboxylase + + + +
Citrate V ) V +
Production of H
2
S ) ) ) )
Urease V + ) )
Tryptophan deaminase ) ) ) )
Indol + + + +
Voges-Proskauer test ) ) ) )
Gelatin liquefaction + + + +
Acid from
Glucose + + + +
Mannose + + + +
Mannitol V + + +
Inositol ) ) ) )
Sorbitol V ) V )
Rhamnose ) ) ) )
Sucrose + + ) )
Melobiose ) + ) )
Amydaline + + + +
Arabinose ) ) ) +
Lactose ) ) ) )
Hydrolysis of
Casein + + + +
Starch + + + +
Tween 80 + + + +
Lipase (egg yolk) + + + +
Hydrolysis of aesculine ) ) ) )
TCBS: thiosulphate-citrate-bile salt-sucrose agar.
a
Yellow colonies.
b
Green colonies.
c
Variable result.
105
2003
Blackwell Publishing Ltd
Journal of Fish Diseases 2003, 26, 103108 I Zorrilla et al. Pathogenic Vibrio species from farmed sole
Results
Two different types of colonies were isolated on
TSAS and FMM from the diseased sole. These
bacteria were Gram-negative, straight rods, motile,
oxidase and catalase-positive, sensitive to the vibri-
ostatic agent O129 (150 lg) and fermentative on
glucose, and one colonial type was green on
thiosulphate-citrate-bile salt-sucrose medium
(TCBS), whilst the other was yellow. The bio-
chemical and physiological characteristics of these
isolates corresponded to V. harveyi and V. parahae-
molyticus. Both colony types were always recovered
from surface ulcers of the diseased sh, but only V.
harveyi was occasionally isolated from ulcers, liver,
spleen and kidney. Flexibacter spp. was not detected
on FMM.
The biochemical and physiological characteristics
of the strains isolated are summarized in Table 1.
The strains of V. harveyi and V. parahaemolyticus
were biochemically homogeneous, although some
differences were observed. Thus, strains of V. harveyi
showed identical biochemical proles except for
swarming, the use of citrate, production of acid from
mannitol and sorbitol and urease activity, whilst
V. parahaemolyticus strains differed in swarming,
citrate and acid from sorbitol.
Infectivity assays were carried out with the isolated
strains, but only those strains of V. harveyi and
V. parahaemolyticus which showed swarming were
virulent for sole, independent of the route of
administration. The LD50s of the pathogenic strains
ranged between 7.4 10
4
and 2.5 10
6
cfu g
)1
sh (Table 2). In all cases, virulent strains of
V. harveyi showed signicant lower LD50 values
(P < 0.05) than those of V. parahaemolyticus and, in
general, the LD50s obtained by i.m. inoculation
were lower than those observed by i.p. inoculation
(P < 0.05). Healthy sh inoculated with cells of
virulent strains of V. harveyi developed skin ulcers,
similar to those observed on the surface of naturally
diseased sole, within 10 days, whereas sh inoculated
with V. parahaemolyticus strain LG-4.00 did not
develop ulcers. Inoculation of the ECP from the
Figure 1 Percentage cumulative mortality after the inoculation of sole with different concentrations of ECPs of Vibrio harveyi
LG-16.00 (j, 120 lg mL
)1
; m, 60 lg mL
)1
) and V. parahaemolyticus LG-4.00 (d, 180 lg mL
)1
; r, 90 lg mL
)1
).
Table 2 Lethal dose (LD50), expressed as cfu g
)1
sh, of Vibrio
harveyi and V. parahaemolyticus isolated from sole
Route of administration
Intraperitoneal
a
Intramuscular
a
Swarming
V. harveyi
LG-6.00 >10
8
>10
8
)
LG-7.01 >10
8
>10
8
)
LG-14.00 2.1 10
5
9.8 10
4
+
LG-15.00 >10
8
>10
8
)
LG-16.00 7.4 10
4
9 10
4
+
V. parahaemolyticus
LG-4.00 6.3 10
5
10
5
+
LG-12.00 2.5 10
6
5 10
5
+
LG-13.00 >10
8
>10
8
)
a
Standard deviations of the replicates were lower than 15%.
106
2003
Blackwell Publishing Ltd
Journal of Fish Diseases 2003, 26, 103108 I Zorrilla et al. Pathogenic Vibrio species from farmed sole
same strains of V. harveyi and V. parahaemolyticus
produced mortalities of 20% and 50% and 0% and
10%, respectively, depending on the doses (Fig. 1).
Ulcers were only seen on the surface of the sh after
inoculation with ECP from V. harveyi.
The cumulative mortalities during vaccine testing
and the RPS after i.p. challenges of the sh
inoculated with sublethal doses of the ECPs of
V. harveyi LG-16.00 are summarized in Table 3. In
all cases the cumulative mortalities were signicantly
lower (P < 0.01) than those in the control groups.
The RPS obtained for V. harveyi were higher
(76.4% and 83.3%) than those obtained for
V. parahaemolyticus LG-4.00 (31.6% and 36.8%).
Discussion
The strains isolated from diseased sole in this study
were characterized and identied as Vibrio harveyi
and V. parahaemolyticus. V. harveyi has caused large-
scale losses of larval and juvenile penaeids (Liu, Lee
& Chen 1996; Alvarez, Austin, Alvarez & Reyes
1998). It has also been associated with ocular
lesions in sh held in captivity (Hispano, Nebra &
Blanch 1997) and mortalities in farmed gilthead sea
bream (Real et al. 1993). On the other hand, there
are few reports of disease caused by V. parahaemo-
lyticus (Alcaide, Amaro, Todol & Oltra 1999; Li,
Yie, Foo, Ling, Xu & Woo 1999), although this
bacterium has also been isolated from shrimp with
red disease (Alapidetendencia & Durez 1997) and
cultured small abalone with withering syndrome
(Liu, Chen, Huang & Lee 2000).
Only those bacterial strains that developed
swarming produced mortalities in experimentally
inoculated sh. Possibly the ability to develop
swarming may be a marker to differentiate between
virulent and avirulent strains of both Vibrio species.
Some authors have demonstrated that the expres-
sion of virulence factors such as immunoglobulinA-
degrading metalloprotease of the urinary tract path-
ogen Proteus mirabilis are expressed together with
other proteins and virulence factors during swarmer
cell differentiation (Walker, Mognaddamejafari,
Lockatell, Johnson & Belas 1999). However, our
results are based on a low number of strains and
therefore more isolates from other disease outbreaks
need to be examined before reaching any conclusion.
The LD50 of the ECP of V. harveyi LG-16.00
and V. parahaemolyticus LG-4.00 corresponded to
120 and >180 lg protein mL
)1
, respectively. The
initial signs observed after the inoculation of the
ECP of both strains were the inactivity of the sh
and anorexia. In the case of the ECP of V. harveyi
LG-16.00, haemorrhagic areas in the ns and
mouth, similar to those signs observed during the
natural outbreaks, were observed 24 h after inocu-
lation, whilst the appearance of ulcers was detected
after 7 days. Thus, despite a lower protein content
the ECP of V. harveyi produced higher sh mortal-
ities and lesions than V. parahaemolyticus ECP.
Therefore, in this study, the ECP from V. harveyi
was more toxic than V. parahaemolyticus ECP.
Additional ECP preparations from isolates from
other disease outbreaks should be examined to
conrm this assumption. Liu et al. (1996) also
concluded that the pathogenicity of different
isolates of V. harveyi in tiger and kuruma prawns
is associated with their ECP. Similarly, Zhang &
Austin (2000) demonstrated that the most patho-
genic isolate of V. harveyi to salmonids produced
ECP with a maximum effect on salmonids.
The role of the ECP as a protective factor for sole
against V. harveyi and V. parahaemolyticus was
demonstrated in this study. Thus, those sh which
were inoculated with the ECP of V. harveyi showed
a high RPS when they were challenged with virulent
V. harveyi strains. In the case of the ECP of
V. parahaemolyticus the RPS values observed were
lower. This could indicate that the V. parahaemo-
lyticus ECP is not as signicant for the virulence of
this bacterium as in the virulent strains of V. harveyi.
This is the rst description of V. harveyi and
V. parahaemolyticus as pathogenic bacteria for
cultured Senegalese sole. Although virulent strains
of V. harveyi and V. parahaemolyticus were isolated
from the original disease outbreak and each species
independently produced sh mortality under labor-
atory conditions, only virulent V. harveyi strains
and their ECP reproduced similar signs to those
observed during the natural outbreak.
Table 3 Percentage of mortality and relative percentage of
survival (RPS) of sole vaccinated and unvaccinated with sublethal
doses of ECP of Vibrio harveyi and V. parahaemolyticus
Percentage of mortality
No. sh/
group
Vaccinated
sh
Unvaccinated
sh RPS
V. harveyi
Experiment no. 1 20 20 85 76.4
Experiment no. 2 20 15 90 83.3
V. parahaemolyticus
Experiment no. 1 20 65 95 31.6
Experiment no. 2 20 60 95 36.8
107
2003
Blackwell Publishing Ltd
Journal of Fish Diseases 2003, 26, 103108 I Zorrilla et al. Pathogenic Vibrio species from farmed sole
The importance of the inclusion of ECP in
vaccine designs has been reported by several authors
(Magarinos, Romalde, Santos, Casal, Barja &
Toranzo 1994; Collado, Fouz, Sanjuan & Amaro
2000). The results obtained in this study showed
that ECP could have an important role in the
pathogenesis of V. harveyi, but that this role is not
so important in the case of V. parahaemolyticus.
However, ECP could be considered as protective
antigens to design potential vaccines against these
pathogenic micro-organisms.
Acknowledgement
We thank Tracey Coffey for her assistance in the
English revision of the manuscript.
References
Alapidetendencia E.V. & Durez L.A. (1997) Isolation of Vibrio
spp from Penaeus monodon (Fabricius) with red disease syn-
drome. Aquaculture 154, 107114.
Alcaide E., Amaro C., Todol R. & Oltra R (1999) Isolation and
characterization of Vibrio parahaemolyticus causing infection in
iberian toothcarp Aphanius iberus. Diseases of Aquatic Organ-
isms 35, 7780.
Alsina M. & Blanch A.R. (1994) Improvement and update of a
set of keys for biochemical identication of Vibrio species.
Journal of Applied Bacteriology 77, 719721.
Alvarez J.D., Austin B., Alvarez A.M. & Reyes H. (1998) Vibrio
harveyi: a pathogen of penaeid shrimps and sh in Venezuela.
Journal of Fish Diseases 21, 313316.
Amaro C., Biosca E.G., Esteve C., Fouz B. & Toranzo A.E.
(1992) Comparative study of phenotypic and virulence
properties in Vibrio vulnicus biotypes 1 and 2 obtained from
a European eel farm experiencing mortalities. Diseases of
Aquatic Organisms 13, 2935.
Balebona M.C., Andreu M.J., Bordas M.A., Zorrilla I., Morinigo
M.A. & Borrego J.J. (1998) Pathogenicity of Vibrio alginoly-
ticus for cultured gilt-head sea bream (Sparus aurata L.).
Applied and Environmental Microbiology 64, 42694275.
Balebona M.C., Zorrilla I., Morinigo M.A. & Borrego J.J.
(1998) Survey of bacterial pathologies affecting farmed gilt-
head sea bream (Sparus aurata L.) in southwestern Spain from
1990 to 1996. Aquaculture 166, 1935.
Biosca E.G. & Amaro C. (1996) Toxic and enzymatic activities
of Vibrio vulnicus biotype 2 with respect to host specicity.
Applied and Environmental Microbiology 62, 23312337.
Collado R., Fouz B., Sanjuan E. & Amaro C. (2000) Effectiveness
of different vaccine formulations against vibriosis caused by
Vibrio vulnicus serovar E (biotype 2) in European eels
Anguilla anguilla. Diseases of Aquatic Organisms 43, 91101.
Fernandez-Diaz C., Yufera M., Canabate J.P., Moyano F.J.,
Alarcon F.J. & Diaz M. (2001) Growth and physiological
changes during metamorphosis of Senegal sole reared in the
laboratory. Journal of Fish Biology 58, 10861097.
Fouz B., Barja J.L., Amaro C., Rivas C. & Toranzo A.E. (1993)
Toxicity of the extracellular products of Vibrio damsela iso-
lated from diseased sh. Current Microbiology 27, 341347.
Hispano C., Nebra Y. & Blanch A. (1997) Isolation of Vibrio
harveyi from an ocular lesion in the short sunsh (Mola mola).
Bulletin of the European Association of Fish Pathologists 17,
104107.
Li J., Yie J., Foo R.W.T., Ling J.M.L., Xu H.S. & Woo N.Y.S.
(1999) Antibiotic resistance and plasmid proles of Vibrio
isolates from cultured silver sea bream, Sparus sarba. Marine
Pollution Bulletin 39, 245249.
Liu P.C., Chen Y.C., Huang C.Y. & Lee K.K. (2000) Virulence
of Vibrio parahaemolyticus isolated from cultured small aba-
lone, Haliotis diversicolor supertexta, with withering syndrome.
Applied Microbiology 31, 433437.
Liu P.C., Lee K.K. & Chen S.N. (1996) Pathogenicity of dif-
ferent isolates of Vibrio harveyi in tiger prawn, Penaeus
monodon. Letters in Applied Microbiology 22, 413416.
Magarinos B., Romalde J.L., Santos Y., Casal J.F., Barja J.L. &
Toranzo A.E. (1994) Vaccination trials on gilthead seabream
(Sparus aurata) against Pasteurella piscicida. Aquaculture 120,
201208.
Ortigosa M., Garay E. & Pujalte M.J. (1994) Numerical tax-
onomy of Vibrionaceae isolated from oysters and seawater
along an annual cycle. Systematic Applied Microbiology 17,
216225.
Pazos F., Santos Y., Mac as A.R., Nunez S. & Toranzo A.E.
(1996) Evaluation of media for the successful culture of
Flexibacter maritimus. Journal of Fish Diseases 19, 193197.
Real F., Deniz S., Acosta B. & Oros J. (1993) Pathological and
pathogenicity study of Vibrio harveyi on Sparus aurata. In:
Proceedings of VI International Conference of the European
Association of Fish Pathologists, pp. 21. Brest, France (Abstracts).
Ribeiro L., Sarasquete C. & Dinis M.T. (1999) Histological and
histochemical development of the digestive system of Solea
senegalensis (Kaup, 1858) larvae. Aquaculture 171, 293308.
Walker K.E., Mognaddamejafari S., Lockatell C.V., Johnson D.
& Belas R. (1999). Zapa, the IgA-degrading metalloprotease
of Proteus mirabilis, is a virulence factor expressed specically
in swarmer cells. Molecular Microbiology 32, 825836.
Zhang X.H. & Austin B. (2000) Pathogenicity of Vibrio harveyi
to salmonids. Journal of Fish Diseases 23, 93102.
Zorrilla I., Balebona M.C., Morinigo M.A., Sarasquete C. &
Borrego J.J. (1999) Isolation and characterization of the cau-
sative agent of pasteurellosis, Photobacterium damsela ssp. pis-
cicida, from sole, Solea senegalensis (Kaup). Journal of Fish
Diseases 22, 167172.
Received: 23 April 2002
Accepted: 17 October 2002
108
2003
Blackwell Publishing Ltd
Journal of Fish Diseases 2003, 26, 103108 I Zorrilla et al. Pathogenic Vibrio species from farmed sole