-
DISEASES OF AQUATIC ORGANISMSDis Aquat Org
Vol. 79: 95–105, 2008doi: 10.3354/dao01902
Published April 1
INTRODUCTION
The eastern oyster Crassostrea virginica is an impor-tant
ecological and economical resource of the Gulf ofMexico and
Atlantic coasts of North America (Cas-tagna et al. 1996). Several
parasitic pathogens, includ-ing Perkinsus marinus, causative agent
of dermo dis-ease, and Haplosporidium nelsoni, causative agent
ofmultinucleated sphere X (MSX) disease, have seri-ously hindered
culture and restoration efforts of C. vir-ginica (Ford & Tripp
1996). Shellfish production is alsooften affected by pathogenic
bacteria, leading to highmortality rates that occur most frequently
during the
larval and juvenile stages (Paillard et al. 2004). Mortal-ities
of larval C. virginica caused by members of thegenus Vibrio were
reported in the 1960s and 70s (Tubi-ash et al. 1965, Elston et al.
1980, Brown 1981, Brown &Tettelbach 1988). Another bacterial
species, Roseovar-ius crassostreae, has been recently described as
thecausative agent of juvenile oyster disease (JOD)(Boettcher et
al. 2005), and the disease is now referredto as Roseovarius oyster
disease (ROD) (Maloy et al.2007). ROD causes losses that may exceed
90% of totalproduction at enzootic sites in the northeastern
USA(Boettcher et al. 2006). Gross signs of the diseaseinclude
organic deposits in the inner valves of the oys-
© Inter-Research 2008 · www.int-res.com*Corresponding author.
Email: [email protected]
Survival of eastern oysters Crassostrea virginicafrom three
lines following experimental challenge
with bacterial pathogens
Javier Gómez-León1, Luisa Villamil1, Scott A. Salger1, Rachel H.
Sallum1,Antonio Remacha-Triviño1, Dale F. Leavitt2, Marta
Gómez-Chiarri1, 3,*
1Department of Fisheries, Animal, and Veterinary Science,
University of Rhode Island, Kingston, Rhode Island 02881, USA2Roger
Williams University, Bristol, Rhode Island 02809, USA
3University of Rhode Island, 23 Woodward Hall, Kingston, Rhode
Island 02881, USA
ABSTRACT: Shellfish production is often affected by bacterial
pathogens that cause high losses inhatcheries and nurseries. We
evaluated the relative survival of larvae and juveniles of 3
Crassostreavirginica oyster lines: (1) GHP, a Rhode Island line;
(2) NEHY, a line resistant to dermo and multinu-cleated sphere X
diseases; and (3) FLOWERS, a line resistant to Roseovarius oyster
disease, experi-mental challenge with Vibrio spp. isolates RE22 and
RE101, causative agents of bacillary necrosis inPacific oyster
larvae, and the type strain of Roseovarius crassostreae, causative
agent of Roseovariusoyster disease. All of the isolates were able
to induce significant mortalities in oyster larvae and juve-niles.
Susceptibility to bacterial challenge in larvae was significantly
higher at 25°C than at 20°C.Susceptibility decreased with oyster
age; mean survival time ranged from 24 h in oyster larvae tomore
than 6 wk in juveniles. Significant differences in susceptibility
to bacterial challenge wereobserved between oyster lines; NEHY was
the most resistant line overall. Extracellular products(ECPs) from
Vibrio sp. RE22 and R. crassostreae, as well as viable bacteria,
were toxic to hemocytesfrom the 3 oyster lines, suggesting that
ECPs are involved in pathogenesis and that external andmucosal
barriers to infection are major contributors to resistance to
bacterial challenge. These proto-cols will be useful in the
elucidation of mechanisms of bacterial pathogenesis and resistance
to infec-tion in oysters.
KEY WORDS: Disease resistance · Pathogenesis · Hemocyte
viability · Juvenile oyster disease ·Roseovarius crassostreae ·
Vibriosis
Resale or republication not permitted without written consent of
the publisher
-
Dis Aquat Org 79: 95–105, 2008
ter that form a light to dark brown ring typicallylocated inside
the valve margins (Bricelj et al. 1992,Ford & Borrero 2001).
The impact of ROD may varyfrom year to year; individuals
-
Gómez-León et al.: Oyster survival following bacterial challenge
97
ria were also included. Each experimental group wasdone in
triplicate. Plates were incubated for 48 h at 20or 25°C. Survival
was determined by direct observa-tion using an inverted microscope
(Leica Dmil, LeicaMicrosystems) using the following criteria: (1)
live lar-vae include swimming larvae and larvae with valvesclosed
but showing internal movement, and (2) deadlarvae include closed
larvae without internal move-ment. In the case of the juveniles,
groups of 50 oystersranging from 4 to 6 mm in shell height for the
FLOW-ERS line and from 5 to 9 mm for the GHP and NEHYlines were
placed in circular flat-bottom 100 ml con-tainers. Larger oyster
sizes ranging from 15 to 22 mmin shell height from the 3 lines were
placed in 1 l con-tainers. Triplicate groups of oysters were
experimen-tally challenged by bath with either Vibrio sp. RE22 orR.
crassostreae CV919-312T at a final concentration inthe bath water
of 5 × 105 CFU ml−1. Control containerswithout bacteria inoculation
were also included. Oys-ters were maintained in SSW at 28 to 30‰ at
25°C withpartial aeration (12 h d−1), and fed with Instant
Algae(Reed Mariculture). The water was partially changed(50%)
weekly. Survival rates were determined weeklycounting dead and live
oysters in each container. Size(shell heights in mm) of the oysters
were also deter-mined. Results were expressed as mean percent
sur-vival + SD. Recently dead and moribund oysters weresampled to
re-isolate and identify associated bacteria.In all cases,
mortalities were attributed to the inocu-lated bacterial strain if
it was the predominant bacter-ial species recovered from gaping or
dead challengedoysters.
Condition index. Thirty oysters from each line withsizes ranging
from 15 to 22 mm in shell height wereprocessed for determination of
the condition indexfollowing the method of Abbe & Albright
(2003).
Histopathological examination. Selected samples ofoysters from
the experimental challenges were fixed inDavidson’s fixative (Shaw
& Battle 1957) for 24 h. Tis-sue samples were processed in an
automatic tissueprocessor, embedded in paraffin wax blocks, and
cuton a microtome. Sections of 5 µm were
deparaffinized,rehydratated, and stained with hematoxylin and
eosin(H&E). Histopathological examinations of all sampleswere
performed with a light microscope (NikonEclipse E-600) and a SPOT
Insight 2 digital camerawith SPOT software, v4.6 (Diagnostic
Instruments).
Detection of Roseovarius crassostreae in oystertissues by
immunofluorescence. Histological sectionsmounted on Superfrost Plus
slides (Fisher Scientific)were deparaffinized, rehydrated through
ethanolgraded series, equilibrated in phosphate bufferedsaline
(PBS), and incubated for 1 h at room tempera-ture in blocking
buffer (BlockHen, Avies Labs). Slideswere washed in washing buffer
(PBS with 0.05% [v/v]
Tween-20 [PBST]) and incubated in a humid chamberat room
temperature for 1 h with a 1:250 dilutionin PBS of a chicken
anti-Roseovarius crassostreae(CV919-312T) polyclonal antibody
(Boardman 2005).Slides were washed in PBST, incubated for 1 h at
roomtemperature with a 1:200 dilution of goat anti-chickenantibody
labeled with Alexa Fluor 546 (MolecularProbes), washed, and cover
slipped using ProlongGold (Molecular Probes) antifade and
mountingmedium. Negative controls were performed by omit-ting
either the primary antibody, the secondary anti-body, or
substituting the primary antibody with pre-immune serum. Sections
were examined using a ZeissAxioPlan 2 epifluorescent microscope
with a ZeissAxioCam digital camera and Zeiss AxioVision v4.5imaging
software (Carl Zeiss).
Preparation of bacterial extracellular products(ECPs). Bacterial
ECPs from Vibrio sp. RE22 and Roseo-varius crassostreae CV919-312T
were obtained usingthe cellophane plate technique (Liu 1957), by
spreading0.1 ml of a 24 h broth culture over sterilized
cellophanesheets placed on TSAS. Plates were incubated for 24 hat
room temperature and the cells washed of the cello-phane with
phosphate buffered saline. Suspensionswere centrifuged at 10 000 ×
g for 30 min at 4°C, and su-pernatants were filtered through 0.45
µm membranesand stored at −80°C until required. The protein
concen-tration of the ECPs was evaluated with Coomasie Bril-liant
Blue assays (Bio Rad Laboratories).
Hemolymph extraction and effect of bacteria onhemocyte
viability. Adult oysters from the 3 oysterlines were maintained
separately in 50 l tanks of aer-ated artificial seawater at 15°C
and 28‰ of salinity andfed daily with Instant Algae. Oyster shells
werenotched and 1 ml of hemolymph was withdrawn fromeach oyster by
the adductor muscle sinus with a dis-posable syringe. The number of
viable hemocytes ineach oyster’s hemolymph was counted with a
hemacy-tometer after staining with trypan blue. Since the
con-centration of hemocytes in each oyster was similar(around 1 ×
106 hemocytes ml−1 of hemolymph), hemo-cytes were left in the
hemolymph to avoid further cellmanipulation. Hemolymphs extracted
from 10 oysterswere pooled; 3 pools of 10 oysters each were used
foreach treatment. Treatments included viable or heat-killed
(100°C, 2 h) bacteria at doses of 10, 103, and 106
CFU ml−1, and heat-treated (100°C, 2 h) or non-treatedbacterial
ECPs at 75, 150, and 300 µg ml−1. After 4 and24 h of incubation
with the different treatments,respectively, aliquots were taken and
cell viabilitiesdetermined by trypan blue exclusion assays with
3replicate counts per sample. Data were expressed aspercent cell
viability. The effect of treatments on cellmorphologies was
determined by observation with aninverted microscope.
-
Dis Aquat Org 79: 95–105, 2008
Statistics. Differences in hemocyte viabilities be-tween oyster
lines were tested by 2-way ANOVA usingSigmastat 3.1 software
(Systat). Data collected as per-centages were transformed (arcsine
of the square root)before analysis. Differences in survival between
oysterlines were tested with the Kaplan-Meier
log-rank(nonparametric test) survival analysis using SigmaStat3.1.
Multiple comparisons were done using the Holm-Sidak post-hoc
method. Results were deemed signifi-cant at p < 0.05.
RESULTS
Experimental challenges
Larvae
Experimental challenges by bath exposure showedthat Vibrio spp.
isolates RE22 and RE101, and Roseo-varius crassostreae CV919-312T
induced significantmortalities in larvae of Crassostrea virginica
after 24 hof exposure at either 20 or 25°C (Fig. 1). Low or
absentlarval mortalities were observed in unchallenged con-trol
larvae. Significantly higher mortalities of GHP lar-vae were
observed at 25°C than at 20°C when exposedto Vibrio sp. isolate
RE101 and R. crassostreae CV919-312T. Significant differences
between oyster lines inpercent survival to challenge with Vibrio
sp. RE101were observed at 20°C (mean survival time ± SD: 37 ±3 h
for GHP, 43 ± 1 h for NEHY, p < 0.001), as well as tochallenge
with R. crassostreae CV919-312T (27 ± 2 hfor GHP, 38 ± 2 h for
NEHY, p < 0.001), but not with thehighly pathogenic Vibrio sp.
isolate RE22 (24 ± 1 h forall oyster lines). Vibrio sp. isolate
RE22 was signifi-cantly more pathogenic to NEHY oysters than
isolateRE101 and R. crassostreae CV919-312T. No
statisticaldifferences in survival were observed between
oysterlines, or between bacterial isolates, at 25°C, probablydue to
the severity of the challenge (Fig. 1b).
Microscopic examination of larvae from the GHPand NEHY lines at
regular intervals during experimen-tal challenges showed that the
first sign of disease wasa reduction of motility, followed by an
abnormal circu-lar pattern of swimming, and, finally, the inability
toswim. The abnormal velum of these moribund larvaewas stalk-like
with clumped cilia (Fig. 2b). At the peakof mortalities, bacteria
swarming inside and arounddead and moribund larvae were observed
(not shown).No lesions or abnormal swimming were observed
innon-infected control individuals (Fig. 2a). Histopatho-logical
examination revealed the presence of rod-shaped bacteria and
phagocytic cells in the visceralcavity of larva infected with
Vibrio spp. RE22(Fig. 2c,d) and RE101. No lesions were observed
in
larvae infected with R. crassostreae CV919-312T ornon-infected
control larvae (not shown).
Juvenile oysters
No significant differences were observed betweenthe condition
indices of the 3 oyster lines (not shown).Experimental challenge of
oyster juveniles (4–9 mmshell height) with Vibrio sp. RE22 and
Roseovariuscrassostreae CV919-312T at 25°C resulted in mortali-ties
in the 3 oyster lines. Significant differences in sur-vival were
detected; FLOWERS was the least resistantline since no surviving
individuals remained 27 d post-challenge (Fig. 3). According to the
survival analysis,NEHY was significantly more resistant than GHP
andFLOWERS to challenge with R. crassostreae CV919-312T, with a
mean survival time of 28 ± 1 d compared to
98
0
10
20
30
40
50
60
70
80
90
100
% S
urvi
val
GHPNEHY
0Control
RE101 RE22 CV919-312
Challenged Control Challenged Control Challenged
10
20
30
40
50
60
70
80
90
100 (b) 25°C
00
(a) 20°C
ControlRE101 RE22 CV919-312
Challenged Control Challenged Control Challenged
Fig. 1. Crassostrea virginica. Percent survival of oyster
larvaechallenged by bath in 5 × 105 CFU ml−1 of Vibrio spp.
isolatesRE101 and RE22, and Roseovarius crassostreae CV919-312T
for 24 h at (a) 20°C and (b) 25°C. Data are expressed as mean+
SD of % of oyster survival of 3 replicate groups of 10 to 12larvae
per treatment. For treatments without mortalities,
mortality bars are located by zeros
-
Gómez-León et al.: Oyster survival following bacterial
challenge
9 ± 1 d for both FLOWERS and GHP (p < 0.001). In thecase of
challenges with Vibrio sp. RE22, survival for alllines was
significantly different (p < 0.05), with NEHYoysters showing the
highest mean survival time (24 ±1 d), then GHP (15 ± 1 d), and
FLOWERS (10 ± 1 d) oys-ters. For a particular line, no significant
differences insurvival to experimental challenge with isolates
RE22or CV919-312T were observed, with the exception ofthe GHP line,
which was more resistant to Vibrio sp.RE22 than to R. crassostreae
CV919-312T. Althoughlow levels of mortality occurred in
unchallenged oys-ters, neither Vibrio sp. RE22 nor R.
crassostreaeCV919-312T were isolated from these oysters. No
sig-nificant relationship between oyster size and time tomortality
was observed in infected oysters in this sizerange (r2 = 0.002, F =
0.422).
In the case of larger juvenile oysters (15−22 mm shellheight), a
longer time was required to induce mortali-
ties after experimental challenge with Vibrio sp. RE22and
Roseovarius crassostreae CV919-312T at 25°C thanin smaller
juveniles (Fig. 4). No significant differencesin survival after
bacterial challenge with Vibrio sp.RE22 or R. crassostreae
CV919-312T were observedbetween the 3 oyster lines. Following
challenges withVibrio sp. RE22, mean survival time ranged from 8.2
±0.5 wk for GHP oysters to 9.3 ± 0.3 wk for both NEHYand FLOWERS
oyster lines. Following challenges withR. crassostreae CV919-312T,
mean survival time was7.8 ± 0.6 wk for FLOWERS, 9.3 ± 0.4 for NEHY,
and9.7 ± 0.3 wk for GHP oysters.
Oyster juveniles experimentally challenged withVibrio sp. RE22
presented histological lesions charac-terized by disorganization of
muscle fibers, hemocyticinfiltration, and necrosis in the mantle
(not shown). Inthe case of oyster juveniles infected with
Roseovariuscrassostreae CV919-312T, lesions were characterized
99
Fig. 2. Crassostrea virginica. Larvae experimentally challenged
for 24 h with Vibrio spp. isolates RE22 and RE101, or
Roseovariuscrassostreae CV919-312T. (a) Unchallenged larva; (b)
challenged larva showed deformed vela with clumped cilia (arrow);
(c & d)H&E-stained histological section of larva challenged
with Vibrio sp. RE22 showed rod-shaped bacteria (arrowhead) and
phagocytic cells in the visceral cavity, (d) is a higher
magnification of the area boxed in (c). Scale bars = 10 µm
-
Dis Aquat Org 79: 95–105, 2008
by degeneration and erosion of the mantle associatedwith
hemocytic infiltration and the presence of organicdeposits
(conchiolin), as well as the presence of densespherical bodies in
the mantle (Fig. 5a,b). No histolog-ical lesions were observed in
unchallenged controloysters (not shown). Immunofluorescent labeling
ofhistological sections from juvenile oysters challengedwith R.
crassostreae CV919-312T with an anti-R. cras-sostreae antibody
showed the presence of rod-shaped
bacteria on the surface of the mantle and within theconchiolin
(Fig. 5c,d), but not within oyster tissues, norin unchallenged
oysters. Macroscopically, oysters (15−22 mm shell height)
challenged with R. crasso-streae CV919-312T showed conchiolin
deposits oninterior valve surfaces 3 to 4 wk post-challenge,
whileno gross signs of ROD were observed in oyster juve-niles less
than 15 mm shell height or in control oysters(not shown).
100
Fig 3
0102030405060708090
100
0 8 14 20 27
Days
% S
urvi
val
FLOWERS
0102030405060708090
100
0 8 14 20 27
Days
% S
urvi
val
FLOWER
0102030405060708090
100
0 8 14 20 27
Days
% S
urvi
val
FLOWER
0102030405060708090
100
0 8 14 20 27
Days
FLOWERS
10
20
30
40
50
60
70
80
90
100
% S
urvi
val
NEHY
00 8 14 20 27 35 43
0 8 14 20 27 35 43
102030405060708090
100
RE22JODControl
GHP
Fig. 3. Crassostrea virginica. Percent survival of oyster
juve-niles (4–9 mm shell height) from 3 lines experimentally
chal-lenged by bath in 5 × 105 CFU ml−1 of Vibrio sp. RE22
orRoseovarius crassostreae CV919-312T for 43 d at 25°C.
Dataexpressed as mean ± SD of % oyster survival of 3 replicate
groups of 50 oysters per line and treatment
FLOWERS
0102030405060708090
100
Weeks
NEHY
0102030405060708090
100
% S
urvi
val
GHP
0
10
20
30
40
50
60
70
80
90
100
0 1 2 3 4 5 6 7 8 9 10 11 12 13
0 1 2 3 4 5 6 7 8 9 10 11 12 13
0 1 2 3 4 5 6 7 8 9 10 11 12 13
RE22
Control
CV919-312
Fig. 4. Crassostrea virginica. Percent survival of oyster
juve-niles (15–22 mm shell height) from the 3 lines
experimentallychallenged by bath in 5 × 105 CFU ml−1 of Vibrio sp.
RE22 orRoseovarius crassostreae CV919-312T for 14 wk at 25°C.
Dataexpressed as mean ± SD of % oyster survival of 3 replicate
groups of 50 oysters per line and treatment
-
Gómez-León et al.: Oyster survival following bacterial
challenge
Effect of bacteria on oyster hemocyte viability
The viabilities of hemocytes from oysters of theGHP, NEHY, and
FLOWERS lines were significantly re-duced after 4 h (data not
shown) and 24 h of incubationwith viable bacterial cells (10, 103,
and 106 CFU ml−1) ofVibrio sp. RE22 or Roseovarius crassostreae
CV919-312T (Fig. 6). Significantly lower survival to
bacterialchallenge (103 and 106 CFU ml−1 of isolates RE22
orCV919-312T) occurred among hemocytes from GHP incomparison to
hemocytes from FLOWERS and NEHYoysters (p < 0.05). For hemocytes
from each oyster line,viabilities were inversely proportional in
nominal doseresponses to challenge concentrations of live
bacterialcells. No significant differences were observed be-tween
survivals of unchallenged control hemocytes andhemocytes challenged
with heat-killed bacteria.
Incubation of hemocytes with ECPs from Vibrio sp.RE22 and
Roseovarius crassostreae CV919-312T signif-
icantly reduced their survival relative to unchallengedhemocytes
(Fig. 7). No significant differences betweenoyster lines were
detected in the effect of bacterialECPs on hemocyte survival. For
hemocytes from eachoyster line, viabilities were inversely
proportional innominal dose responses to concentrations of
non-heated bacterial ECPs. The toxic effects of all the ECPson
hemocytes were eliminated when ECPs were heat-treated.
Differences in hemocyte morphologies were ob-served following
incubation with viable Vibrio sp.RE22 bacteria, as well as with
non-heated ECPs.These changes were characterized by a high
propor-tion of rounded refringent hemocytes (Fig. 8b) thatwere not
observed in untreated cells (Fig. 8a). Simi-lar results were
observed following incubation ofoyster hemocytes to live bacteria
and non-heatedECPs from Roseovarius crassostreae CV919-312T
(notshown).
101
Fig. 5. Crassostrea virginica. Representative photomicrographs
of oyster juveniles experimentally challenged with
Roseovariuscrassostreae CV919-312T. H&E-stained sections of a
challenged oyster showing: (a) degeneration and erosion of the
mantleassociated with hemocytic infiltration (arrows) and the
presence of conchiolin (arrowheads) (scale bar = 100 µm); (b)
bacteria(arrows) and dense spherical bodies (arrowhead) in the
mantle (scale bar = 5 µm); (c & d) immunofluorescent labeling
of R. cras-sostreae in histological sections of a challenged oyster
showing the presence of rod-shaped labeled bacteria (arrowhead)
within
conchiolin deposits, (d) is a magnification of the area boxed in
(c) (scale bars = 10 µm)
-
Dis Aquat Org 79: 95–105, 2008
DISCUSSION
Experimental challenges with bacterial pathogenshave been
successfully used to evaluate host−patho-gen interactions in
oysters (e.g. Labreuche et al. 2006).In this work, we have
successfully applied a modifica-tion of the experimental challenge
protocol developedby Estes et al. (2004) in Pacific oysters
Crassostreagigas to test for differences in susceptibility to
bacterialchallenge between different lines of eastern oystersC.
virginica. Factors affecting levels of resistance tobacterial
challenge in oysters included temperature,bacterial isolate,
age/size of the oyster, and oyster line.These experimental
challenges will provide a useful
102
log10 (CFU) ml–1
Vibrio sp. RE22
01 3 6 6 Control
1 3 6 6 Control
102030405060708090
100
% S
urvi
val
NEHYFLOWERSGHP
R. crassostreae CV919-312
0102030405060708090
100
Live Killed bacteria
Fig. 6. Crassostrea virginica. Viabilities of oyster
hemocytesincubated for 24 h at 20°C with viable and heat-killed
Vibriosp. RE22 and Roseovarius crassostreae CV919-312T cellsat
concentrations of 10, 103, and 106 CFU ml−1. Data ex-pressed as
mean ± SD of % viable hemocytes in 3 hemo-lymph pools per
experimental group. The experiment was
performed twice
ECPs, µg ml–1
Heat-inactivated
Active
Vibrio sp. RE22
0102030405060708090
100
% S
urvi
val
NEHYFLOWERSGHP
R. crassostreae CV919-312
075 150 300 300 Control
75 150 300 300 Control
102030405060708090
100
Fig. 7. Crassostrea virginica. Viability of oyster hemocytes
in-cubated for 24 h at 20°C with ECPs of Vibrio sp. RE22
andRoseovarius crassostreae CV919-312T at concentrations of 75,150,
and 300 µg ml–1. Data expressed as mean ± SD of %viable hemocytes
in 3 hemolymph pools per experimental
group. The experiment was performed twice
Fig. 8. Crassostrea virginica. Representative
phase-contrastphotomicrographs showing the effect of bacteria on
hemo-cytes of oysters from the FLOWERS line after 24 h of
in-cubation with Vibrio sp. RE22. (a) Control hemocytes thatare
predominantly spread on the culture well surface; (b) he-mocytes
treated with 150 µg ml–1 bacterial ECPs show-ing phase-contract
refringence due to rounding up (arrow).
Scale bars = 10 µm
-
Gómez-León et al.: Oyster survival following bacterial
challenge
model for studying host-pathogen interactions andmechanisms of
resistance to bacterial infection inoysters.
We first provide evidence that 2 Vibrio spp. strainsisolated
from diseased Crassostrea gigas larvae (Esteset al. 2004) were able
to induce mortalities in C. vir-ginica larvae and juveniles.
Consistent with obser-vations in C. gigas, Vibrio sp. RE22 was more
patho-genic to C. virginica larvae than Vibrio sp. RE101. Thefact
that Vibrio species isolated from C. gigas are ableto induce
mortalities in C. virginica is not surprisingsince this bacterial
genus has been implicated in larvalmortalities of different bivalve
species in hatcheries(Paillard et al. 2004). The histological
lesions observedafter experimental challenge of C. virginica larvae
andjuveniles with Vibrio spp. RE22 and RE101 resembledthose
previously described for larval vibriosis in oys-ters (Tubiash et
al. 1965, Elston et al. 1980, Estes et al.2004), clams (Gomez-Leon
et al. 2005), and cockles(Fujiwara et al. 1993), suggesting common
mecha-nisms of Vibrio spp. pathogenesis in bivalve species.
Furthermore, we provide further evidence thatRoseovarius
crassostreae is the causative agent ofJOD, now called Roseovarius
oyster disease (Maloy etal. 2007). This disease was first observed
in Cras-sostrea virginica in the northeastern USA in the late1980s.
Several causative agents have been evaluatedsince then, including
bacteria of various genera,including Vibrio spp. (Lee et al. 1996,
Paillard etal. 1996), Aeromonas and Pseudomonas spp. (Paillardet
al. 1996), as well as protozoan parasites (Boettcher etal. 2006).
Recently, a novel species of alphaproteobac-terium, R.
crassostreae, was identified as the etiologi-cal agent of ROD based
on the observation that R. cras-sostreae is consistently the
dominant bacterial speciesassociated with JOD-affected animals
(Boettcher et al.1999, 2000, 2005) and the successful reproduction
ofdisease signs after challenge of oyster juveniles byinjection of
R. crassostreae into the pallial cavity(Maloy et al. 2007). We have
been able to cause mor-talities in oyster larvae and juveniles by
bath exposureto R. crassostreae, and have reproduced the
character-istic signs of ROD in oyster juveniles between 15 and22
mm in shell height. Those clinical signs includedmantle lesions
characterized by degeneration, erosionand the presence of dense
spherical bodies termed‘coccoid bodies’, and the presence of
conchiolindeposits in the interior valve margins (Bricelj et
al.1992, Ford & Borrero 2001). Consistent with
recentlypublished research in ROD-affected oysters (Board-man et
al. 2008), the presence of R. crassostreae inexperimentally
challenged juveniles was restricted tothe outer edge of the mantle
and the conchiolin. Simi-larly, in brown ring disease, a bacterial
pathology firstdescribed in the clam Ruditapes philippinarum,
the
etiological agent Vibrio tapetis can not be detected
his-tologically within clam tissues (Paillard et al. 1994,Paillard
& Maes 1995).
Our results also confirm the important role oftemperature on the
pathogenesis of bacterial infection inoyster larvae and juveniles;
temperatures at or above25°C were necessary for Roseovarius
crassostreae tocause significant mortalities among juvenile
oysters.Furthermore, higher temperatures (25 versus 20°C)
alsoresulted in significantly higher mortalities when larvaewere
challenged with Vibrio spp. isolates RE22 andRE101. These results
are in agreement with observationsin oyster hatcheries that
indicate higher incidence ofbacillary necrosis at warmer
temperatures (Ford &Borrero 2001), and observations in the
field that showthat ROD mortalities occur when water
temperaturesincrease (Bricelj et al. 1992). Warmer temperatures
couldresult in higher mortalities by favoring bacterial
prolifer-ation and secretion of extracellular virulence factors.As
observed in previous research (Ford & Borrero 2001),warm
temperatures were unlikely to be the direct causeof ROD, as holding
the control oyster at 25°C didnot caused unusual mortalities or
conchiolin depositson the interior of the valves.
Our results are also consistent with observations inthe field
that show that resistance to bacterial infectionsignificantly
increases with oyster age and size (Briceljet al. 1992, Ford &
Borrero 2001). The capacity forrepair as well as immune defenses
including externalbarriers such as the shell may be more efficient
at pro-tecting the oyster from bacteria invasion as the
oysterincreases in size (Mount et al. 2004).
We demonstrate here that experimental challengesare particularly
useful in evaluating differences in sur-vival to bacterial
challenge between oyster lines selec-tively breed for resistance to
diseases caused by bacte-rial (ROD) or protistan (dermo and MSX)
pathogens. Ingeneral, the MSX and dermo disease-resistant
oysterline NEHY showed the highest levels of resistanceto bacterial
challenge of the 3 lines that we tested.Interestingly, NEHY oysters
were more resistant toRoseovarius crassostreae challenge than
oysters fromFLOWERS at the larval and early juvenile stages(4–9 mm
shell height). This is consistent with observa-tions in the field
that indicate that hybrids betweenNEHY and FLOWERS lines were more
susceptible toROD than the NEHY line (Guo et al. 2003). Althoughthe
FLOWERS line was not evaluated in this field trial,the field
results are in agreement with our laboratoryobservations that the
NEHY line is more resistant toROD than the FLOWERS line. These
observations sug-gest that the FLOWERS line may have lost
resistanceto ROD, possibly due to decreased disease pressure,
orthat resistance to ROD is dependent on the strain ofR.
crassostreae to which oysters are exposed. The find-
103
-
Dis Aquat Org 79: 95–105, 2008
ing that the NEHY line is relatively ROD-resistant isinteresting
as NEHY oysters have probably never beenexposed to ROD, although it
is often exposed to Vibriospp. infections in the hatchery (X. Guo
pers. comm.).These findings suggest that oysters may use
commonmechanisms of resistance to defend themselvesagainst
infection by different bacterial pathogens.
As a first step in the elucidation of the potentialmechanisms of
bacterial pathogenesis in oysters, weevaluated the effect of
bacterial challenge and expo-sure to bacterial ECPs on the survival
of oyster hemo-cytes from the 3 oyster lines. Hemocytes are
majoreffectors of the immune system in oysters, and are
alsoinvolved in other functions like digestion and woundhealing
(Bachere et al. 2004). Bacterial interactionswith hemocytes are
inevitable during invasive infec-tions or when bacteria are
ingested during the normalfiltration and feeding processes. The
virulence of Vib-rio spp. and their capacity to induce mortalities
duringlarval and juvenile stages has been correlated withtheir
ability to produce extracellular toxins (Elston etal. 1980, Nottage
& Birkbeck 1987, Riquelme et al.1996, Lambert et al. 2001,
Gomez-Leon et al. 2005) thatin some cases have ciliostatic
activity, and are able toinvade the bivalve tissues directly,
causing necrosis(Nottage et al. 1989). The role of extracellular
toxins inthe pathogenicity of other bacterial genera (such
asRoseovarius crassostreae) has been poorly studied. Inthe present
work, we show for the first time that theECPs of R. crassostreae
(CV919-312T) can contribute tothe development of the ROD
pathogenesis since theyhave cytotoxic activity that can
significantly diminishoyster hemocytes survival. It is also
possible that thepresence of a possible toxin with ciliostatic
activitycould have a detrimental action in the infected
oysters,since feeding impairment has been observed in
exper-imentally infected animals (Boettcher et al. 2000),
con-sistent with the ‘starved’ appearance of naturallyinfected
animals. The results obtained in the presentwork indicate that the
ECPs of both Vibrio sp. RE22and R. crassostreae CV919-312T are
heat-labile, sug-gesting that toxicity is not solely due to the
lipopolysac-charide content of the ECPs (Gomez-Leon et al.
2005).The fact that no major differences in hemocyte survivalafter
treatment with bacteria or ECPs (or at least notconsistent with
differences in oyster survival to bacter-ial challenge) were
observed between the differentoyster lines suggests that the
differences in survivalbetween these oysters are due to factors
other than thetoxic effects of bacteria on oyster hemocytes.
Further-more, the fact that these pathogenic bacteria are toxicto
hemocytes from adult oysters suggests that externaland mucosal
barriers to infection are major contribu-tors to the higher
resistance to bacterial challengeobserved in oysters as they
age.
In summary, the use of in vivo experimentalchallenges by bath,
which do not bypass mechanicalbarriers to infection, combined with
in vivo challengesby injection and in vitro challenges of hemocytes
willbe useful in the elucidation of mechanisms of patho-genesis as
well as the study of the mechanisms ofresistance to bacterial
challenge. Furthermore, futurecomparison of the results from the
experimental chal-lenges with the overall performance of the
differentoyster lines in the field would indicate the potential
ofusing experimental challenges as a tool in the develop-ment of
selectively-bred lines of oysters resistant tobacterial
pathogens.
Acknowledgements. The authors thank K. Boettcher, R.Elston, and
G. DeBrosse for providing bacterial strains, anti-bodies, and
oysters, and X. Guo and K. Boettcher for helpfuldiscussions. We
thank K. Tammi at the RWU hatchery and H.Giddings at URI for
support in the maintenance of the oysters.This research was
supported by grants RIAI03-001 from theRhode Island Aquaculture
Initiative and 2002-34438-12688from the US Department of
Agriculture. R.H.S. was supportedby a Coastal Fellowship for
Undergraduate Research. Thisresearch was also made possible in part
by use of the RhodeIsland Genomics and Sequencing Center, supported
by theNSF under EPSCoR Grant No. 0554548, and the RI-INBREResearch
Core Facility supported by Grant No. P20 RR16457from NCRR, NIH.
LITERATURE CITED
Abbe GR, Albright BW (2003) An improvement to the deter-mination
of meat condition index for the eastern oysterCrassostrea virginica
(Gmelin 1791). J Shellfish Res 22:747–752
Bachere E, Gueguen Y, Gonzalez M, de Lorgeril J, Garnier
J,Romestand B (2004) Insights into the anti-microbial de-fense of
marine invertebrates: the penaeid shrimps andthe oyster Crassostrea
gigas. Immunol Rev 198:149–168
Barber BJ, Davis CV, Crosby MA (1998) Cultured
oysters,Crassostrea virginica, genetically selected for fast
growthin the Damariscotta River, Maine, are resistant to
mortalitycaused by juvenile oyster disease (JOD). J Shellfish
Res17:1171–1175
Boardman C (2005) Host-pathogen interactions between east-ern
oysters (Crassostrea virginica) and the bacterial agentof juvenile
oyster disease (Roseovarius crassostreae). MSthesis, University of
Maine
Boardman CL, Maloy AP, Boettcher KJ (2008) Localization ofthe
bacterial agent of juvenile oyster disease
(Roseovariuscrassostreae) within affected eastern oysters
(Crassostreavirginica). J Invertebr Pathol 97:
Boettcher KJ, Barber BJ, Singer JT (1999) Use of
antibacterialagents to elucidate the etiology of juvenile oyster
disease(JOD) in Crassostrea virginica and numerical dominanceof an
alpha-proteobacterium in JOD-affected animals.Appl Environ
Microbiol 65:2534–2539
Boettcher KJ, Barber BJ, Singer JT (2000) Additional evi-dence
that juvenile oyster disease is caused by a memberof the
Roseobacter group and colonization of nonaffectedanimals by Stappia
stellulata-like strains. Appl EnvironMicrobiol 66:3924–3930
104
-
Gómez-León et al.: Oyster survival following bacterial
challenge
Boettcher KJ, Geaghan KK, Maloy AP, Barber BJ (2005)Roseovarius
crassostreae sp. nov., a member of the Roseo-bacter clade and the
apparent cause of juvenile oyster dis-ease (JOD) in cultured
Eastern oysters. Int J Syst EvolMicrobiol 55:1531–1537
Boettcher KJ, Smolowitz R, Lewis EJ, Allam B and others(2006)
Juvenile oyster disease (JOD) in Crassostrea vir-ginica: synthesis
of knowledge and recommendations.J Shellfish Res 25:683–686
Bricelj VM, Ford SE, Borrero FJ, Perkins FO and others
(1992)Unexplained mortalities of hatchery-reared, juvenileoysters
Crassostrea virginica (Gmelin). J Shellfish Res11:331–347
Brown C (1981) A study of two shell-fish-pathogenic
Vibriostrains isolated from a Long Island hatchery during arecent
outbreak of disease. J Shellfish Res 1:83–87
Brown C, Tettelbach LP (1988) Characterization of a non-motile
Vibrio sp. pathogenic to larvae of Mercenariamercenaria and
Crassostrea virginica. Aquaculture 74:195–204
Castagna M, Gibbons MC, Kurkowski K (1996) Culture:applications.
In: Kennedy VS, Newell RI (eds) The easternoyster, Crassostrea
virginica. Maryland Sea Grant Col-lege, College Park, MD, p
675−690
Davis CV, Barber BJ (1994) Size-dependent mortality
inhatchery-reared populations of oysters, Crassostrea vir-ginica,
Gmelin 1791, affected by juvenile oyster disease.J Shellfish Res
13:137–142
Davis CV, Barber BJ (1999) Growth and survival of selectedlines
of eastern oysters, Crassostrea virginica (Gmelin1791) affected by
juvenile oyster disease. Aquaculture178:253–271
Elston R, Leibovitz L, Laurent PJ (1980) Pathogenesis
ofexperimental vibriosis in larval American oysters, Cras-sostrea
virginica. Can J Fish Aquat Sci 37:964–978
Estes RM, Friedman CS, Elston RA, Herwig RP (2004)
Patho-genicity testing of shellfish hatchery bacterial isolates
onPacific oyster Crassostrea gigas larvae. Dis Aquat
Org58:223–230
Ford SE, Borrero FJ (2001) Epizootiology and pathology
ofjuvenile oyster disease in the eastern oyster,
Crassostreavirginica. J Invertebr Pathol 78:141–154
Ford SE, Tripp MR (1996) Diseases and defense mechanisms.In:
Kennedy VS, Newell RI, Eble AE (eds) The eastern oys-ter,
Crassostrea virginica. Maryland Sea Grant, CollegePark, MD, p
581−660
Fujiwara M, Ueno Y, Iwao A (1993) A Vibrio sp. associatedwith
mortalities in cockle larvae Fulvia mutica (Mollusca:Cardidae).
Fish Pathol 28:83–89
Gomez-Leon J, Villamil L, Lemos ML, Novoa B, Figueras A(2005)
Isolation of Vibrio alginolyticus and Vibrio splen-didus from
aquacultured carpet shell clam (Ruditapesdecussatus) larvae
associated with mass mortalities. ApplEnviron Microbiol
71:98–104
Guo X, Ford SE, DeBrosse G, Smolowitz R (2003) Breedingand
evaluation of eastern oyster strains selected for MSX,dermo and JOD
resistance. J Shellfish Res 22:333–334
Labreuche Y, Lambert C, Soudant P, Boulo V, Huvet A, Nico-las JL
(2006) Cellular and molecular hemocyte responsesof the Pacific
oyster, Crassostrea gigas, following bacterialinfection with Vibrio
aestuarianus strain 01/32. MicrobesInfect 8:2715–2724
Lambert C, Nicolas J, Bultel V (2001) Toxicity to
bivalvehemocytes of pathogenic Vibrio cytoplasmic extract.J
Invertebr Pathol 77:165–172
Lee M, Taylor GT, Bricelj VM, Ford SE, Zahn S (1996) Eva-luation
of Vibrio spp. and microplankton blooms ascausative agents of
juvenile oyster disease in Crassostreavirginica (Gmelin). J
Shellfish Res 15:319–329
Lewis EJ (2001) Juvenile oyster disease (JOD) and manage-ment
strategies: a review. Bull Natl Res Inst Aquac (Suppl5):101–109
Liu PV (1957) Survey of hemolysin production among speciesof
pseudomonads. J Bacteriol 74:718–727
Maloy AP, Ford SE, Karney RC, Boettcher KJ (2007) Roseo-varius
crassostreae, the etiological agent of juvenile oysterdisease (now
to be known as Roseovarius oyster disease)in Crassostrea virginica.
Aquaculture 269:71–83
Mount AS, Wheeler AP, Paradkar RP, Snider D (2004)
Hemo-cyte-mediated shell mineralization in the eastern
oyster.Science 304:297–300
Nottage AS, Birkbeck TH (1987) Production of proteinase dur-ing
experimental infection of Ostrea edulis L. larvae withVibrio
alginolyticus NCMB 1339 and the antigenic relation-ship between
proteinases produced by marine vibrios path-ogenic for fish and
shellfish. J Fish Dis 10:265–273
Nottage AS, Sinclair PD, Birkbeck TH (1989) Role of
low-molecular-weight ciliostatic toxins in vibriosis of
bivalvemollusks. J Aquat Anim Health 1:180–186
Paillard C, Maes P (1995) The brown ring disease in theManila
clam, Ruditapes philippinarum: I. Ultrastructuralalterations of the
periostracal lamina. J Invertebr Pathol65:91–100
Paillard C, Maes P, Oubella R (1994) Brown ring disease inclams.
Annu Rev Fish Dis 4:219–240
Paillard C, Ashton-Alcox K, Ford SE (1996) Changes in bacte-rial
densities and hemocyte parameters in eastern oysters,Crassostrea
virginica, affected by juvenile oyster disease.Aquat Living Resour
9:145–158
Paillard C, Le Roux F, Borreg JJ (2004) Bacterial disease
inmarine bivalves, a review of recent studies: trends andevolution.
Aquat Living Resour 17:477–498
Ragone Calvo LM, Calvo GW, Burreson EM (2003) Dual dis-ease
resistance in a selectively bred eastern oyster, Cras-sostrea
virginica, strain tested in Chesapeake Bay. Aqua-culture
220:69–87
Renault T, Chollet B, Cochennec N, Gerard A (2002) Shell
dis-ease in eastern oysters, Crassostrea virginica, reared
inFrance. J Invertebr Pathol 79:1–6
Riquelme C, Toranzo AE, Barja JL, Vergara N, Araya R
(1996)Association of Aeromonas hydrophila and Vibrio algi-nolyticus
with larval mortalities of scallop (Argopectenpurpuratus). J
Invertebr Pathol 67:213–218
Roch P (1999) Defense mechanisms and disease prevention infarmed
marine invertebrates. Aquaculture 172:125–145
Shaw BL, Battle HI (1957) The gross and microscopic anatomyof
the digestive tract of Crassostrea virginica (Gmelin).Can J Zool
35:325−347
Tubiash HS, Chanley PE, Leifson E (1965) Bacillary necrosis,a
disease of larval and juvenile bivalve mollusks. I. Etiol-ogy and
epizootiology. J Bacteriol 90:1036–1044
Yu ZN, Guo XM (2006) Identification and mapping of
disease-resistance QTLs in the eastern oyster, Crassostrea
vir-ginica Gmelin. Aquaculture 254:160–170
105
Editorial responsibility: Eugene Burreson,Gloucester Point,
Virginia, USA
Submitted: April 26, 2007; Accepted: January 31, 2008Proofs
received from author(s): March 14, 2008
cite4: cite5: cite8: cite10: cite11: cite12: cite13: cite14:
cite15: cite19: cite20: cite21: cite22: cite23: cite26: cite27:
cite28: cite29: cite30: cite31: cite32: cite35: cite36: cite37:
cite38: cite39: cite40: cite41: cite42: cite43: cite44: cite45:
cite46: cite47: cite48: cite49: cite50: