Variability of haemocyte and haemolymph parameters in European flat oyster Ostrea edulis families obtained from brood stocks of different geographical origins and relation with infection by the protozoan Bonamia ostreae

Fish & Shellfish Immunology (2008) 24, 551e563

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/fsi

Variability of haemocyte and haemolymph parameters in European flat oyster Ostrea edulis families obtained from brood stocks of different geographical origins and relation with infection by the protozoan Bonamia ostreae ˜a, Jose ´ Fuentes, Antonio Villalba* Patricia Mirella da Silva, Pilar Comesan Centro de Investigacio´ns Marin˜as, Consellerı´a de Pesca e Asuntos Marı´timos, Xunta de Galicia, Aptdo. 13, E-36620 Vilanova de Arousa, Spain Received 18 July 2007; revised 22 October 2007; accepted 9 November 2007 Available online 21 November 2007

KEYWORDS Ostrea edulis; Crassostrea gigas; Bonamia ostreae; Immune system; Haemocyte; Enzymes; Phenoloxidase

Abstract A research project to compare productive traits (growth and mortality), disease susceptibility and immune capability between Ostrea edulis stocks was performed. This article reports the results on the immune capability and its relation with infection by the intrahaemocytic protozoan Bonamia ostreae. Four to five oyster spat families were produced from each of four European flat oyster populations (one from Ireland, one from Greece and two from Galicia, Spain) in a hatchery. The spat were transferred to a raft in the Rı´a de Arousa (Galicia) for on growing for 2 years. Total haemocyte count (THC) and differential haemocyte count (DHC) were estimated monthly through the second year of growing-out. Three types of haemocytes were distinguished: granulocytes (GH), large hyalinocytes (LHH) and small hyalinocytes (SHH). Significant correlations between the mean relative abundance of GH and SHH of the families and the mean prevalence of B. ostreae, the overall incidence of pathological conditions and the cumulative mortality of the families were found; these correlations supported the hypothesis that high %GH and low %SHH would enhance oyster immune ability and, consequently, would contribute to lower susceptibility to disease and longer lifespan. Infection by B. ostreae involved a significant increase of circulating haemocytes, which affected more markedly the LHH type. The higher the infection intensity the higher the %LHH. This illustrates the ability of B. ostreae to modulate the immune responses of the O. edulis to favour its own multiplication. A significant reduction of the phenoloxidase activity in the haemolymph of oysters O. edulis infected by B. ostreae was observed. Nineteen enzymatic activities in the haemolymph of O. edulis and Crassostrea gigas (used as a B. ostreae resistant reference) were

* Corresponding author. Tel.: þ34 986500155; fax: þ34 986506788. E-mail addresses: [email protected] (P. Mirella da Silva), [email protected] (P. Comesan ˜a), [email protected] (J. Fuentes), [email protected] (A. Villalba). 1050-4648/$ - see front matter ª 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.fsi.2007.11.003

552

P. Mirella da Silva et al. measured using the kit api ZYM, Biomerieux. Qualitative and quantitative differences in enzyme activities in both haemocyte and plasma fractions between B. ostreae noninfected O. edulis from different origins were recorded. However, no clear positive association between enzyme activity and susceptibility to bonamiosis was found. The only enzyme detected in the resistant species C. gigas that was not found in the susceptible one O. edulis was b-glucosidase (in plasma). B. ostreae infected O. edulis showed significant increase of some enzyme activities and the occurrence of enzymes that were not detected in noninfected oysters. These changes could be due to infection-induced enzyme synthesis by the host or to enzyme synthesis by the parasite. ª 2007 Elsevier Ltd. All rights reserved.

Introduction The disease caused by the protozoan parasite Bonamia ostreae contributed to the exhaustion of most European flat oyster Ostrea edulis populations in Galicia and is the main constraint for oyster farming in that region. Unsuccessful strategies to fight against bonamiosis were implemented in France [1] and Holland [2,3], including zootechnical prophylaxis and eradication attempts. However, two selective breeding programmes for bonamiosis resistance, one in France [4e7] and another in Ireland [8], gave rise to encouraging results for the oyster industry, and highlighted the possibility to grow flat oysters in areas affected by bonamiosis, with higher survival and lower B. ostreae prevalence and infection intensity [9]. Accordingly, the Centro de Investigacio ´ns Marin ˜as decided to develop a selective breeding programme to produce oysters O. edulis with increased tolerance to bonamiosis [10]. As a previous stage to this programme, evaluation of the variability of productive traits, disease susceptibility, and immune capability through oyster populations was performed, because particular populations could be favourable for the programme. Four geographic origins were chosen for variability assessment, including extreme and intermediate locations in the European flat oyster geographic range: Greece (Eastern Mediterranean), Ireland (Northern Atlantic), and two areas in Galicia, Spain. Oysters from those locations were used as brood stocks and 19 full- or half-sib families were produced (4e5 families from each origin). The evaluation of productive traits and disease susceptibility showed significant differences in growth, mortality and susceptibility to bonamiosis and other diseases, both between origins and between families within origins; these results have been published elsewhere [10]. Evaluation of the immune capability was also considered in the experimental design because it could explain the differences in disease susceptibility that could be found. The immune system of bivalve molluscs relies upon haemocytes, which play an important role against parasites by phagocytosis or encapsulation, with subsequent destruction via hydrolytic enzymes [11], reactive oxygen intermediates (ROIs) [12] and antimicrobial peptides [13]; in addition, humoral factors consisting of non-self recognition molecules and immune effectors, such as lectins, opsonins, and components of the prophenoloxidase (proPO) system may play a key role in immune reactions [14]. Two main haemocyte types can be distinguished in the haemolymph of bivalves: granular haemocyte (GH) and hyaline haemocyte (HH) [15], which also applies to O. edulis [16].

This paper reports the inter (between origins) and intra (between families within origins) population variability in a number of immune parameters: total haemocyte count (THC), differential haemocyte count (DHC), 19 hydrolytic enzyme activities in the cell and plasma fractions of the haemolymph, and phenoloxidase activity. The association of these parameters with the B. ostreae infection was also analysed.

Materials and methods Production of oyster families Four oyster populations were selected as brood stock for the experiments: one in the North of Ireland (IR), one in Greece (GR) and two in Galicia (NW Spain), one from the Rı´a de Ortigueira (OR) and another from Coroso (CO) (Rı´a de Arousa). The oysters from each origin were brought to the hatchery facilities of the Centro de Investigacio ´ns Marin ˜as in December 2000 and handled as described by da Silva et al. [10] to produce various families of spat from each origin. Briefly, the oysters were distributed into five trays per origin with 15e20 individuals per tray and conditioned for spawning. Batches of recently spawned larvae originated from a single mother were collected from each tray. Thus, all the larvae within each batch were half or full sibs. Every larval family was separately reared. Once spat surpassed 1 cm in height they were transferred to a raft for on growing. A total of 19 families were produced, 5 from each origin, except for IR, for which only 4 families were obtained.

Oyster on growing In September 2001, approximately 4000 individuals from each family were transferred to a raft located near Cambados (Rı´a de Arousa, Galicia, NW Spain), in an oyster culture area heavily affected by bonamiosis since the 1980s [17]. Oysters were cultivated up to September 2003, being handled as described by da Silva et al. [10] through the on growing process. Samples of oysters were randomly taken (monthly up to June 2002 and quarterly since then) to estimate growth by measuring their height and whole weight. Estimation of mortality was accomplished monthly up to June 2002, quarterly since then, by counting dead and live individuals. A number of oysters (five oysters up to July 2002 and six since then) from each family were randomly taken monthly for disease diagnosis. Sampling for measuring

Immune capability and its relation with infection immune parameters started when oysters were large enough to allow haemolymph withdrawal from the adductor mussel, in September 2002. Since then, the six oysters per each family that were sampled monthly for disease diagnosis were also used for THC and DHC estimation. Three Irish families (IR1, IR2 and IR4) reached 100% cumulative mortality before the end of the experiment, thus no IR1 sample was available for measuring immune parameters and samples of IR2 and IR4 were not available in some months.

Haemolymph collection, total haemocyte count (THC), differential haemocyte count (DHC) and analysis of pathological conditions For haemolymph collection, the oyster shell was slightly opened with a wedge and the haemolymph was withdrawn from the adductor muscle using a 21-Gauge needle attached to a 2-mL syringe containing Alsever’s anti-aggregant solution [18]. The THC of each oyster was estimated as the number of haemocytes per mL of haemolymph by examining the haemolymph with light microscope, using a Malassez chamber. To estimate the DHC a haemolymph cell monolayer was prepared by haemolymph cytocentrifugation [19]; then the haemolymph cell monolayer was fixed and stained with a Hemacolor kit (Merck) and examined with light microscope to allocate 100 haemocytes chosen at random to haemocytic types. After haemolymph withdrawal, each oyster was shucked and a sagital, approximately 5 mm thick section of meat containing gills, visceral mass and mantle lobes was excised and fixed in Davidson’s solution and embedded in paraffin; 5 mm thick sections were stained with Harris’ hematoxylin and eosin [20]. The histological sections were examined under light microscope for disease diagnosis. Haemolymph cell monolayers were also used to diagnose the presence of B. ostreae. The monthly prevalence of B. ostreae, symbionts and other pathological conditions was calculated as the percentage of affected oysters in each monthly sample. The estimation of B. ostreae infection intensity involved ranking each oyster after the examination of histological sections using the scale proposed by da Silva and Villalba [19]: null (nondetected), light, moderate and heavy infection. The oysters in which infection was detected in haemolymph cell monolayers but was not in histological sections were allocated in the light infection class. Eleven symbionts/pathological conditions were detected through the study, which has been reported elsewhere [10]: B. ostreae, viral infection, Rickettsia-like organisms, Haplosporidium armoricanum, ciliates in digestive gland, ciliates in gills, copepods, disseminated neoplasia, extensive gill lesions, haemocytic infiltration, and granulocytomas. An index of the overall incidence of pathological conditions was obtained for each family as follows: the mean prevalence of each symbiont and pathological condition in each family for the whole study period was calculated. Then, families were ranked for each of the 11 symbionts/pathological conditions, from the lowest to the highest mean prevalence; that is each family had 11 ranks, 1 for each symbiont/pathological condition. The index of the overall incidence of pathological conditions of each family was calculated as the mean rank of the family. The index provides information

553 on how heavily a family was affected by symbionts/pathological conditions or, in other words, how susceptible was a family to disease: the higher the index the more susceptible. Families IR1, IR2 and IR4 were excluded from the ranking because they reached 100% cumulative mortality before the end of the study and thus samples of the last months could not be taken from these families.

Phenoloxidase activity (PO) A total of 33 oysters corresponding to 12 families, from all the origins except IR, were taken from the raft in June 2003, when the prevalence of B. ostreae was high. Those oysters were used to compare PO activity in the haemolymph between B. ostreae infected and noninfected oysters. PO activity was estimated spectrophotometrically by recording the formation of DOPA-chrome, after oxidation of the enzyme and the substrate L-DOPA [21]. Haemolymph from each oyster was withdrawn as described above. One hundred ml of haemolymph from each oyster were used to produce a haemolymph cell monolayer for B. ostreae diagnosis (see above). The remaining haemolymph was sonicated (45 s, 4  C) and then centrifuged (16,000  g, 20 min, 4  C). The supernatant, representing plasma plus haemocyte lysate supernatant (PHLS) was used in the assay. The total protein concentration present in the PHLS was measured according to the method of Bradford [22], using bovine serum albumin as a standard. Fifty ml of TBS (250 mM trisbaseeHCl, 400 mM NaCl, pH 7.6) and 50 ml of PHLS were added into micro-plate wells; then 50 ml of L-DOPA (2 mg mL1, Sigma) were added to each well. As control, samples (PHLS) were incubated for 15 min with 30 mM tropolone (specific inhibitor of PO) before adding L-DOPA. Blanks were prepared by replacing PHLS, tropolone and L-DOPA with TBS. Absorbance was read immediately after the L-DOPA addition at 492 nm each 5 min for 40 min, using a micro-plate reader (Expert 96, Asys hitech GesmbH). PO activity was expressed as the average variation of absorbance min1 mg protein1. One unit of PO activity is equivalent to an absorbance change of 0.001 min1 mg protein1 at 20  C.

Api ZYM test The colorimetric semi-quantitative method api ZYM (BioMe ´rieux), was used to estimate enzyme activity levels in oyster haemolymph. In February 2003, oysters O. edulis from all families (except IR1, IR2 and IR4, which were not available at that time) were taken from the raft to compare enzyme activity in the haemolymph between oysters from different geographic origins and between B. ostreae infected and noninfected oysters. In addition, Pacific oysters Crassostrea gigas (Thunberg) that were being cultured in the same raft were collected. C. gigas was involved in the comparison because this species is not considered as a B. ostreae susceptible species [23e25]. Oysters were kept in tanks with flowing seawater until used (max. 20 days). The assay was performed in triplicate, running each replicate in a different day. The haemolymph of 5e8 individuals of the same family was pooled to assure a number of haemocytes sufficient to detect low levels of enzyme activity.

554 One pool per O. edulis family plus one pool of C. gigas were used in each replicate. Each pool was used for one replicate only. One hundred ml of haemolymph from each pool was used to produce a haemolymph cell monolayer for B. ostreae diagnosis (see above). Then, a volume of haemolymph containing 2  106 haemocytes from each pool was centrifuged (755  g, 10 min, 4  C) to separate the haemocytes (pellet) from the plasma (supernatant). Both fractions were stored at 80  C until used. The pellet was resuspended in 1300 ml of distilled water and sonicated (45 s, 4  C) to obtain a haemocyte lysate suspension. The total protein concentration present in the plasma was measured as mentioned above. Samples (either plasma or haemocyte lysate suspension) were added to the reaction strips of the kit, 65 ml per well, and incubated at 37  C for 4 h; then the enzyme activity was revealed using the kit reagents. The intensity of reaction was evaluated according to the colour scale provided with the kit. The enzymatic activity was expressed as nanomoles of hydrolysed substrate per 106 haemocytes (for haemocyte lysate samples) or per mg of total protein (for plasma samples).

Statistical analysis Comparisons of THC (Sept 02e03) between origins and families were analysed using nested ANOVA (origin as main factor and family nested within origins) with replicates. The comparisons of THC between months were analysed by one-factor ANOVA. All comparisons for DHC (Sept 02e03) were analysed by the KruskaleWallis test. Correlation between the mean values of immune parameters of the families and their mean prevalence of B. ostreae, their overall incidence of pathologic conditions, and their cumulative mortality at the end of the experiment, respectively, was estimated with the Pearson correlation coefficient; the families IR1, IR2 and IR4 were excluded from the correlation analyses because of their reduced number of haemolymph samples. The comparisons of THC and DHC between infected and noninfected oysters were done by the KruskaleWallis test. One-factor ANOVA was applied to compare THC, DHC and PO activity between classes of B. ostreae infection intensity and the Tukey test was used to identify differences between groups. Correlation between PO activity and classes of B. ostreae infection intensity was estimated with the Pearson correlation coefficient. A KrulkaleWallis test was used to compare the enzymatic activities (EA) between origins and families; C. gigas was considered as one origin in the comparison. MINITAB 14 software was used for all the statistical analyses. Differences were considered statistically significant when P values were lower than 0.05.

Results Total haemocyte count (THC) and differential haemocyte count (DHC) The examination of haemolymph cell monolayers allowed distinguishing granular haemocytes (GH) and two types of hyaline haemocytes, small (SHH) and large (LHH) hyalinocytes, in the haemolymph of the European flat oyster. The

P. Mirella da Silva et al. haemograms of oysters from all origins showed temporal variation: THC was higher in the beginning of summer and lower in autumn and winter (Fig. 1). The %GH showed the highest values in December and August whereas the lowest mean value corresponded to May; the %SHH showed a rather opposite pattern (Fig. 2). The THC and DHC values obtained after grouping the records of B. ostreae noninfected oysters of the entire study period are summarised in Table 1 and Figs. 3 and 4. The differences between geographic origins in THC and DHC were not significant. The differences in THC between families under origins were not significant; however, in the case of DHC, significant differences were found between IR families in the %GH (P Z 0.011) and the SHH (P Z 0.024), and between GR families in the %SHH (P Z 0.014) (Table 1). A significant negative correlation was found between the mean %GH of the families and each of the following variables: the mean prevalence of B. ostreae, the index of overall incidence of pathological conditions, and the cumulative mortality of the families (Table 2); on the contrary, a significant positive correlation was detected between the mean %SHH of the families and each of the mentioned variables. The mean THC and the mean %LHH did not significantly correlate with any of the mentioned variables (Table 2). In the case of the origin CO, the family CO5, the one with the lowest B. ostreae mean prevalence, lowest overall incidence of pathological conditions and lowest cumulative mortality of this origin, showed the highest mean %GH and the lowest mean %SHH of the origin; on the contrary, the family CO3, with the highest B. ostreae mean prevalence and the highest overall incidence of pathological conditions of this origin and very high cumulative mortality, showed the lowest mean %GH and the highest mean %SHH of the origin (Table 2). The case of the family GR5 was also remarkable, because it was the family with the lowest B. ostreae mean prevalence, lowest overall incidence of pathological conditions and lowest cumulative mortality of the origin GR and showed the highest mean %GH and one of the lowest mean %SHH of this origin. The haemogram of oysters infected by B. ostreae showed a temporal variation similar to that of noninfected oysters (Figs. 1 and 2). Nevertheless, a significant increase in the THC (Figs. 1 and 3) was observed in infected oysters. Differences in DHC involved that infected oysters showed lower %GH and higher %LHH than noninfected ones (Figs. 2 and 4), although the differences were significant only between some origins (Fig. 4). Differences in %SHH were less important (Fig. 4). The comparison of DHC between classes of infection intensity showed that the THC and the %LHH increased significantly with the intensity of infection (Figs. 5 and 6).

Phenoloxidase activity (PO) PO activity was found in the haemolymph of O. edulis noninfected and infected by B. ostreae. The number of PO units was significantly reduced in infected oysters (Fig. 7), this reduction was significantly negatively correlated (r Z 0.49; P Z 0.003) with parasite infection intensity, the higher was the intensity the lower was the PO activity.

Immune capability and its relation with infection IR

555

GR

OR Non infected by B. ostreae

CO

5x107

Infected by B. ostreae

4x107

1x107

THC (no. haemocytes ml-1)

3x107 2x107

8x106

1x107

6x106

4x106

2x106

0

0 A

S

O

N

D

J

F

M

A

M

J

2002

J

A

S

2003

Figure 1 Temporal variation of total haemocyte counts (THCs) (mean  SE) in the haemolymph of the European flat oyster O. edulis. Bars correspond to B. ostreae noninfected oysters (left Y axis) from different geographic origins (different bar patterns). The line corresponds to B. ostreae infected oysters after grouping all the origins (right Y axis). IR: Ireland; GR: Greece; CO: Coroso and OR: Ortigueira.

Enzymatic activities of oyster haemolymph by api ZYM test

significant differences in the concentration of some enzymes in haemocytes and plasma were detected between oysters O. edulis from different origins and between O. edulis and C. gigas (Figs. 8 and 9). Thirteen enzymes were detected in O. edulis from all origins, except a-fucosidase that was not detected in the haemocytes of IR oysters (Fig. 8). In the haemocyte samples of O. edulis, a-chymotrypsin, a-glucosidase and b-glucosidase were not detected; trypsin

Sixteen enzymes were detected in haemocytes and 14 in plasma of the European flat oyster noninfected by B. ostreae (Figs. 8 and 9). Fourteen enzymes were detected in both haemocytes and plasma of the Pacific oyster C. gigas by using the commercial kit api ZYM. Qualitative and quantitative

Non infected by Bonamia ostreae GH

60

LHH

Infected by Bonamia ostreae GH

SHH

LHH

SHH

% Haemocytes

50

40

30

20

10

0 S

O

N

2002

D

J

F

M

A

M

J

J

A

S

2003

Figure 2 Temporal variation of the percentage of haemocyte types (mean  ES) from oysters O. edulis infected and noninfected by B. ostreae.

556

P. Mirella da Silva et al.

Table 1 Mean total haemocyte counts (THCs) (SE) and mean percentage of each haemocyte type (SE) corresponding to B. ostreae noninfected oysters O. edulis of each family after grouping the 12 monthly samples 6

THC (10 ) %GH %LHH %SHH THC (106) %GH %LHH %SHH 6

THC (10 ) %GH %LHH %SHH THC (106) %GH %LHH %SHH

IR2a

IR3

IR4b

Mean IR

4.59  2.931 32.8  4.22 42.4  3.56 24.8  2.62

1.87  0.196 48.1  1.28 35.0  0.85 16.9  1.09

4.02  0.900 48.6  1.29 34.4  1.14 17.0  1.00

2.79  0.391 47.3  1.01 35.3  0.70 17.4  0.77

GR1

GR2

GR3

GR4

GR5

Mean GR

3.19  0.732 47.8  1.21 35.0  1.17 17.3  0.76

3.42  0.722 47.7  1.35 33.9  1.15 18.4  0.91

2.43  0.337 46.1  1.41 34.6  1.22 19.3  1.39

2.28  0.364 50.0  1.71 34.8  1.47 15.2  0.93

2.53  0.341 50.8  1.11 33.9  1.08 15.3  0.80

2.75  0.227 48.6  0.62 34.3  0.54 17.1  0.46

CO1

CO2

CO3

CO4

CO5

Mean CO

3.12  0.549 48.2  0.74 36.1  1.11 15.7  0.92

3.05  0.369 49.9  0.67 34.6  1.06 15.5  1.02

2.66  0.315 46.7  1.64 36.3  1.64 17.1  1.31

3.99  0.695 49.1  0.73 35.4  1.04 15.5  0.98

2.79  0.437 51.1  0.95 33.4  0.95 15.5  0.79

3.16  0.234 49.2  0.51 35.1  0.44 15.7  0.36

OR1

OR2

OR3

OR4

OR5

Mean OR

2.64  0.475 49.6  1.07 34.4  0.96 16.0  0.66

3.06  0.414 48.9  1.13 34.4  0.89 16.7  0.99

4.73  1.442 46.9  1.21 36.9  0.98 16.2  0.76

2.87  0.266 49.4  1.01 33.4  0.88 16.2  0.70

2.80  0.383 51.4  1.41 32.9  1.1 15.7  0.75

3.20  0.314 49.2  0.53 34.6  0.43 16.2  0.36

The mean value after grouping all the families of each geographic origin is also included (IR, Ireland; GR, Greece; CO, Coroso; OR, Ortigueira). GH: granulocytes; LHH: large hyalinocytes and SHH: small hyalinocytes. a Only one monthly sample of this family was analysed. b Only seven monthly samples of this family were analysed.

and a-galactosidase were only detected in IR3 and a-mannosidase only in GR3. GR oysters showed the highest mean concentration of eight enzymes in the haemocytes, although differences were significant only for two enzymes (alkaline phosphatase and valine arylamidase) (Fig. 8). The enzymes

IR

1.8x107

THC (no. haemocytes ml-1)

1.6x107

*

GR

CO

OR

*

7

1.4x10

*

1.2x107 1.0x107 8.0x106 6.0x106 4.0x106 2.0x10

6

58

37

8

*

*

*

6

0.0

82 Infected

200 228 244 Non infected

Figure 3 Total haemocyte counts (THCs) (mean  SE) from oysters O. edulis infected and noninfected by B. ostreae from different geographical origins. The number of oysters analysed from each origin is indicated in the bars. IR: Ireland; GR: Greece; CO: Coroso and OR: Ortigueira. *, Significant differences between oysters infected and noninfected by B. ostreae.

trypsin, a-galactosidase and a-mannosidase were not detected in oysters from Galicia. Leucine arylamidase was the only enzyme that significantly varied between families; CO families presented the lowest concentration of this enzyme, except CO5 (data not shown). The 14 enzymes detected in C. gigas were also found in O. edulis, whereas trypsin and a-galactosidase were not detected in the former. C. gigas had higher concentrations of six enzymes, although differences were significant only for three enzymes (leucine arylamidase, b-glucuronidase, a-mannosidase). In the plasma samples (Fig. 9), trypsin and a-mannosidase were not detected in any of the oyster groups. IR oysters showed the highest mean concentration of six enzymes in the plasma, although differences were significant only for leucine arylamidase. GR oysters showed the highest mean concentration of four enzymes, although differences were significant only for acid phosphatase. Again, leucine arylamidase was the only enzyme that significantly varied between families: CO families presented the lowest concentration of this enzyme, except for CO5, whereas the highest values were observed in IR3 and GR5. The origin with more enzymes showing the highest concentration in the plasma was IR (six enzymes), followed by GR and C. gigas (four enzymes each). OR oysters showed the highest concentration of none enzyme. b-glucosidase was the only enzyme detected in the plasma of C. gigas that was not found in that of O. edulis. The analysis of haemolymph cell monolayers of each haemolymph pool showed that eight pools were infected by

Immune capability and its relation with infection

557 IR

60

*

50

% Haemocytes

60

Granulocytes

* *

*

* *

76 185 223 268

6

58 33

CO

OR 60

Large Hyalinocytes

* *

*

*

40

10

30

30

30

20

20

20

10

10

10

0

Non infected

Small Hyalinocytes

50

50

40

40

GR

Infected

0

Non infected

Infected

0

Non infected

Infected

Figure 4 Percentage of haemocyte types (mean  SE) from oysters O. edulis infected and noninfected by B. ostreae from different geographical origins. The number of oysters analysed from each origin is indicated in the bars. IR: Ireland; GR: Greece; CO: Coroso and OR: Ortigueira. *, Significant differences between oysters infected and noninfected by B. ostreae.

B. ostreae (Table 3). The comparison between B. ostreae infected and noninfected oysters O. edulis showed significant differences in the concentration of some enzymes (Figs. 10 and 11). In the haemocytes (Fig. 10), acid phosphatase and N-acetyl-b-glucosaminidase showed significantly higher concentrations in infected oysters than in noninfected ones; a-chymotrypsin was only detected in infected oysters (GR1). In the plasma (Fig. 11), acid phosphatase and b-glucuronidase presented significantly higher concentrations in infected oysters than in noninfected ones; a-glucosidase and a-mannosidase were only detected in infected oysters (CO3 and GR1, respectively).

Discussion Three types of haemocytes were distinguished in the haemolymph of O. edulis, as described by Cochennec-Laureau et al. [26]. The most frequent type was granulocyte, followed by large hyalinocyte and small hyalinocyte. Cochennec-Laureau et al. [26] found more large hyalinocytes than granulocytes in the haemolymph of O. edulis. The discrepancy could be due

to differences in the environmental conditions or could be genetically based, since the oysters analysed in each study came from different geographic areas. The haemograms of O. edulis from all origins showed a similar pattern of temporal variation. Likely, factors influencing this temporal pattern were more closely related with physiological issues (gonad development) than with seawater temperature, because in October and November 2002 the seawater temperature was as high as in summer 2003, but the THC and %LHH were lower. The increase of THC in summer coincided with the gonad reabsorption (data not shown), which could suggest a necessity of haemocytes to eliminate nonreleased gametes or even a post-spawning stress. Nevertheless, the relationship between circulating haemocytes and haemocytes in tissues is not well established. Fisher et al. [27] observed in the Eastern American oyster Crassostrea virginica (Gmelin) high values of THC after the spawning. The individual variability of THC found in oysters in this study was high and the differences between origins and families under origins were not statistically significant. However, in the case of DHC, significant correlations

Table 2 Correlation between the mean values of immune parameters of the oyster O. edulis families and their mean prevalence of B. ostreae, their overall incidence of pathologic conditions, and their cumulative mortality at the end of the experiment, respectively

THC %GH %LHH %SHH

Mean prevalence of B. ostreae

Overall incidence of pathologic conditions

Cumulative mortality

r Z 0.30; P Z 0.26 r Z 0.49; P Z 0.05 r Z 0.23; P Z 0.40 r Z 0.52; P Z 0.04

r Z 0.35; P Z 0.18 r Z 0.59; P Z 0.02 r Z 0.27; P Z 0.31 r Z 0.62; P Z 0.01

r Z 0.26; P Z 0.30 r Z 0.48; P Z 0.06 r Z 0.24; P Z 0.37 r Z 0.49; P Z 0.05

The mean values of the immune parameters were calculated excluding the oysters infected by B. ostreae. GH: granulocytes; LHH: large hyalinocytes; SHH: small hyalinocytes; r: Pearson’s correlation coefficient and P: Probability of r being different from zero.

558

P. Mirella da Silva et al.

2.0x107

250

c

Bonamia ostreae Non detected Light Moderate Heavy

Bonamia ostreae Non detected Light Moderate Heavy

a

c 200

1.5x107

1.0x107

PO Units

THC (no. haemocytes ml-1)

2.5x107

b

a

150

100 b

6

5.0x10

a

b

50

754

70

23

16 13

0.0

10

8

2

0

Figure 5 Total haemocyte counts (THCs) (mean  SE) from oysters O. edulis grouped in classes of B. ostreae infection intensity: nondetected, light, moderate and heavy. The number of oysters analysed from each class is indicated in the bars. Different letters indicate significant differences between classes of infection.

between the relative abundance of some haemocyte types (GH and SHH) of the families and the prevalence of B. ostreae, the index of overall incidence of pathological conditions and the cumulative mortality of the families were found. These correlations support the hypothesis that high %GH and low %SHH would enhance oyster immune ability and, consequently, would contribute to lower susceptibility to disease and longer lifespan. Significant differences between origins and between families within origins in the prevalence of B. ostreae, the index of overall incidence of pathological conditions and the cumulative mortality had been detected [10]. Thus, DHC could contribute to explain those differences rather than THC. In bivalve molluscs, granulocytes have a higher phagocytic ability than hyalinocytes [28e32]. Granulocytes have abundant lysosomes that contribute to intracytoplasmic pathogen degradation

60

% Haemocytes

50 40

a

b

b b

c

Bonamia ostreae Non detected Light Moderate Heavy

c

b 752 77 19 11

a

30 20

a

a

a

a

10 0

GH

LHH

SHH

Figure 6 Differential haemocyte counts (mean  SE) from oysters O. edulis grouped in classes of B. ostreae infection intensity: nondetected, light, moderate and heavy. The number of oysters from each class is indicated in the bars. GH: granulocytes; LHH: large hyalinocytes and SHH: small hyalinocytes. Different letters indicate significant differences between classes of infection.

Figure 7 Phenoloxidase activity (PO) (mean  SE) from oysters O. edulis grouped in classes of B. ostreae infection intensity: light; moderate and heavy. The number of oysters from each group is indicated in the bars. Different letters indicate significant differences between classes of infection.

[15,30,31] and also can release lysosomal enzymes that can contribute to extracellular pathogen destruction [15,33]. On the contrary, hyalinocytes and especially small hyalinocytes lack lysosomes [15,30,31]. Cochennec-Laureau et al. [26] reported that B. ostreae occurred more frequently and multiplied more successfully within hyalinocytes than in granulocytes. Infection by B. ostreae involved a significant increase of circulating haemocytes, which affected more markedly the LHH type. The higher the infection intensity, the higher the %LHH. THC increase was associated with the increase of haemocytic infiltration of the connective tissue of different organs (data not shown); heavy haemocytic infiltration of the connective tissue is characteristic of the infection by this parasite [34,35]. Cochennec-Laureaeu et al. [26] did not found significant differences in THC between B. ostreae infected and noninfected oysters, whatever the degree of infection intensity. However, in agreement with our results, those authors observed a high %LHH and low %GH in infected oysters in the three classes of infection intensity. The results suggest the hypothesis that B. ostreae induces proliferation of the haemocytic type within which divides more successfully. If so, the parasite would modulate the host immune response to favour its own multiplication. The lack of knowledge on oyster haemopoiesis avoids stating whether the parasite blocks differentiation from hyalinocytes to granulocytes after stimulating multiplication of a common stem cell or the parasite stimulates preferentially the multiplication of a hypothetical blast cell precursor of LHH, in case this type is a completely differentiated one. Ford et al. [36] found that an oyster C. virginica strain more susceptible to infection by Haplosporidium nelsoni had lower THC than a more tolerant strain; in addition, the granulocyte relative abundance was lower in oysters heavily infected, which belonged to the more susceptible strain. Anderson et al. [37] reported increased THC in oysters C. virginica moderately and heavily infected by the protozoan Perkinsus marinus. Chu and La Peyre [38] found that high THC, high %GH and high lysozyme concentration in

Immune capability and its relation with infection

559

350

300

Hydrolysed substrate (nanomoles)

HAEMOCYTES

IR GR CO OR C. gigas

*

250

200

150

100

50

*

*

*

*

*

*

0 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

Enzyme number

Figure 8 Enzymatic activities (mean of three replicates  SE) in haemocytes of European flat oysters O. edulis noninfected by B. ostreae from different geographical origins and of Pacific oysters C. gigas. The Y axis scale represents the nanomoles of hydrolysed substrate per 1.3  105 haemocytes (quantity added in each well). IR: Ireland; GR: Greece; CO: Coroso and OR: Ortigueira. The numbers 1e20 in the X axis correspond to control (1), alkaline phosphatase (2), esterase (3), esterase lipase (4), lipase (5), leucine arylamidase (6), valine arylamidase (7), cystine arylamidase (8), trypsin (9), a-chymotrypsin (10), acid phosphatase (11), naphtolAS-BI-phosphohydrolase(12), a-galactosidase (13), b-galactosidase (14), b-glucuronidase (15), a-glucosidase (16), b-glucosidase (17), N-acetyl-b-glucosaminidase (18), a-mannosidase (19), and a-fucosidase (20). *, Significant differences between oyster groups. 2.0 1.8

Hydrolysed substrate (nanomoles)

PLASMA

IR GR CO OR C. gigas

1.6

*

1.4 1.2 1.0 0.8 0.6

*

*

0.4 0.2

*

*

0.0 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

Enzyme number

Figure 9 Enzymatic activities (mean of three replicates SE) in plasma of European flat oysters O. edulis noninfected by B. ostreae from different geographical origins and of Pacific oysters C. gigas. The Y axis scale represents the nanomoles of hydrolysed substrate per mg of total protein. IR: Ireland; GR: Greece; CO: Coroso and OR: Ortigueira. The numbers 1e20 in the X axis correspond to control (1), alkaline phosphatase (2), esterase (3), esterase lipase (4), lipase (5), leucine arylamidase (6), valine arylamidase (7), cystine arylamidase (8), trypsin (9), a-chymotrypsin (10), acid phosphatase (11), naphtol-AS-BI-phosphohydrolase (12), a-galactosidase (13), b-galactosidase (14), b-glucuronidase (15), a-glucosidase (16), b-glucosidase (17), N-acetyl-b-glucosaminidase (18), a-mannosidase (19), and a-fucosidase (20). *, Significant differences between oyster groups.

560

P. Mirella da Silva et al.

Table 3 Percentage of O. edulis haemocytes infected by B. ostreae in the haemolymph cell monolayers prepared from the haemolymph pools of each oyster family that were used for the api ZYM test. The pools that were not infected are not shown Family pool GR1-replicate GR1-replicate GR1-replicate GR3-replicate CO1-replicate CO3-replicate CO3-replicate OR5-replicate

Infected haemocytes (%) 1 2 3 2 3 1 2 3

17 1 26 1 1 20 10 24

the haemolymph conferred on C. virginica a higher tolerance to the P. marinus infection. A significant reduction of the PO activity in the haemolymph of oysters O. edulis infected by B. ostreae was observed in this work. The involvement of PO in the immune system of arthropods is known [39]. Its role in the immune system of the bivalve molluscs is not well understood. Nevertheless, reduction of PO levels has been reported in various mollusc species when infected by different parasites: C.

350

virginica and the mussel Geukensia demissa when challenged P. marinus [40], R. philippinarum infected by Perkinsus olseni (Zatlanticus) [41], and Sydney rock oysters Saccostrea glomerata (Gould) infected by Marteilia sydneyi [42]. Oysters S. glomerata tolerant to M. sydneyi had more PO activity than oysters from wild population during the noninfective period, suggesting a protective function [43,44]. Qualitative and quantitative differences in enzyme activities in both haemocytes and plasma between B. ostreae noninfected oyster O. edulis origins were recorded. In all the cases in which significant differences in enzyme activity between O. edulis origins were found, the oysters from Ortigueira (the bonamiosis least susceptible origin and the one with the lowest mortality) showed the lowest levels of enzyme activity, whereas the oysters of foreign origins (Ireland and Greece, both with the highest mortality and with the highest susceptibility to bonamiosis) showed the highest levels of enzyme activity. The interpretation of these results in terms of immune significance is difficult because the implication of those enzymes in defence mechanisms is not clear enough. Comparison of enzyme activities between haemocytes and plasma was not suitable because the total protein concentration in haemocyte samples was not measured, which avoided a proper standardisation way. Comparison between B. ostreae noninfected O. edulis and the B. ostreae resistant species C. gigas revealed that the only enzyme detected in the resistant species that was not found in the susceptible one was b-glucosidase

HAEMOCYTES

300

Hydrolysed substrate (nanomoles)

Infected O. edulis Non infected O. edulis 250

C. gigas

200

150

*

100

* 50

* 0.5 0.0

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

Enzyme number

Figure 10 Enzymatic activities (mean of three replicates SE) in haemocytes of European flat oysters O. edulis infected and noninfected by B. ostreae and of Pacific oysters C. gigas. The Y axis scale represents the nanomoles of hydrolysed substrate per 1.3  105 haemocytes (quantity added in each well). The numbers 1e20 in the X axis correspond to control (1), alkaline phosphatase (2), esterase (3), esterase lipase (4), lipase (5), leucine arylamidase (6), valine arylamidase (7), cystine arylamidase (8), trypsin (9), a-chymotrypsin (10), acid phosphatase (11), naphtol-AS-BI-phosphohydrolase (12), a-galactosidase (13), b-galactosidase (14), b-glucuronidase (15), a-glucosidase (16), b-glucosidase (17), N-acetyl-b-glucosaminidase (18), a-mannosidase (19), and a-fucosidase (20). *, Significant differences between infected and noninfected oysters O. edulis (C. gigas is excluded from comparison).

Immune capability and its relation with infection

561

1.4

PLASMA Infected O. edulis

Hydrolysed substrate (nanomoles)

1.2

Non infected O. edulis C. gigas

1.0

0.8

0.6

* *

0.4

0.2

*

*

0.008 0.004 0.000 1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

Enzyme number

Figure 11 Enzymatic activities (mean of three replicates SE) in plasma of European flat oysters O. edulis infected and noninfected by B. ostreae and of Pacific oysters C. gigas. The Y axis scale represents the nanomoles of hydrolysed substrate per mg of total protein. The numbers 1e20 in the X axis correspond to control (1), alkaline phosphatase (2), esterase (3), esterase lipase (4), lipase (5), leucine arylamidase (6), valine arylamidase (7), cystine arylamidase (8), trypsin (9), a-chymotrypsin (10), acid phosphatase (11), naphtol-AS-BI-phosphohydrolase (12), a-galactosidase (13), b-galactosidase (14), b-glucuronidase (15), a-glucosidase (16), b-glucosidase (17), N-acetyl-b-glucosaminidase (18), a-mannosidase (19), and a-fucosidase (20). *, Significant differences between infected and noninfected oysters O. edulis (C. gigas is excluded from comparison).

(in plasma). The contribution of this enzyme to resistance to B. ostreae should be further tested. On the contrary, trypsin and a-galactosidase were detected in noninfected O. edulis (at very low concentration, in IR oysters) but were not in C. gigas. Therefore, trypsin, a serine protease, could be irrelevant in the defence against B. ostreae. Xue and Renault [45] compared the enzymatic activity in the haemolymph between O. edulis and C. gigas using the api ZYM method. Our results showed some differences with regard to those reported by the authors: they detected aglucosidase but did not detect either trypsin or a-galactosidase. In addition, they found a-galactosidase in C. gigas. Xue and Renault [45] reported significant higher levels for most enzymes in C. gigas and suggested that some of those enzymes could play a role in the resistance against B. ostreae. Discrepancies between both studies could be due to differences in technical procedure because Xue and Renault [45] used a higher number of haemocytes per sample. In addition, environmental factors and genetic differences between the oysters of the two studies could have contributed to result discrepancy, because the oysters used in the studies came from different areas. The comparison of the enzyme activities in the haemolymph between B. ostreae infected and noninfected oysters showed significant increase of some enzyme activities in infection. Infection-induced enzyme synthesis by the host or enzyme synthesis by the parasite [46] could explain the increase of concentration and the detection of new enzyme activities in the infected oysters.

Acknowledgements M.I. Seoane and V. Rodrı´guez led the work to produce oyster spat. The Company ‘‘Jose Maria Daporta Leiro e Hijos, S.L.’’, the shellfish farmers L. Nogueira, J.L. Nogueira and M. Nogueira and the ‘‘Confrarı´a de Pescadores Nosa Sen ˜ora do Carmen de Carin ˜o’’ helped with oyster on growing and provided brood stock oysters. E. Penas, I. Mele ´ndez, M. Andrade, I. Ferna ´ndez, A.I. Gonza ´lez, M.V. Gregorio, A.C. Iglesias and C. Rodrı´guez provided field and laboratory technical assistance. This work was partially supported by funds from the ‘‘Secretarı´a Xeral de Investigacio ´n e Desenvolvemento Tecnolo ´xico da Xunta de Galicia’’, through the project PGIDT-CIMA 01/1. P.M. da Silva was supported by successive scholarships from the ‘‘Agencia Espan ˜ola de Cooperacio ´n Internacional’’ (AECI) and the ‘‘Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo ´gico’’ (CNPq) of the Brazilian Government. The authors thank Dr. Philippe Soudant and Dr. Margherita Barracco for their comments on the manuscript.

References [1] Grizel H, Bache `re E, Mialhe E, Tige ´ G. Solving parasite-related problems in cultured molluscs. In: Howell MJ, editor. Parasitology, Quo vadit, Proceedings of the sixth international congress of parasitology. Canberra: Australian Academy of Science; 1986. p. 301e8.

562 [2] van Banning P. Management strategies to control diseases in the Dutch culture of edible oysters. Am Fish Soc Spec Publ 1988;18:243e5. [3] van Banning P. Observations on bonamiosis in the stock of the European flat oyster, Ostrea edulis, in the Netherlands, with special reference to the recent developments in Lake Grevelingen. Aquaculture 1991;93:205e11. [4] Martin AG, Ge ´rard A, Cochennec N, Langlade A. Selecting flat oysters, Ostrea edulis, for survival against the parasite Bonamia ostreae: assessment of the resistance of a first selected generation. In: Barnabe ´ G, Kestemont P, editors. Production, environment and quality, Bourdeaux aquaculture 92. Ghent, Belgium: European Aquaculture Society; 1993. p. 547e54. Spec Publ No. 18. [5] Baud JP, Ge ´rard A, Naciri-Graven Y. Comparative growth and mortality of Bonamia ostreae-resistant and wild flat oysters, Ostrea edulis, in an intensive system. Mar Biol 1997;130:71e9. [6] Naciri-Graven Y, Martin AG, Baud JP, Renault T, Ge ´rard A. Selecting the flat oyster Ostrea edulis (L.) for survival when infected with the parasite Bonamia ostreae. J Exp Mar Biol Ecol 1998;224:91e107. [7] Naciri-Graven Y, Haure J, Ge ´rard A, Baud JP. Comparative growth of Bonamia ostreae resistant and wild flat oyster Ostrea edulis in an intensive system II. Second year of the experiment. Aquaculture 1999;171:195e208. [8] Culloty SC, Cronin MA, Mulcahy MF. An investigation into the relative resistance of Irish flat oysters Ostrea edulis L. to the parasite Bonamia ostreae (Pichot et al., 1980). Aquaculture 2001;199:229e44. [9] Culloty SC, Cronin MA, Mulcahy MF. Potential resistance of a number of populations of the oyster Ostrea edulis to the parasite Bonamia ostreae. Aquaculture 2004;237:41e58. [10] da Silva PM, Fuentes J, Villalba A. Growth, mortality and disease susceptibility of oyster Ostrea edulis families obtained from brood stocks of different geographical origins, through on-growing in the Rı´a de Arousa (Galicia, NW Spain). Mar Biol 2005;147:965e77. [11] Pipe RK. Hydrolytic enzymes associated with the granular haemocytes of the marine mussel Mytilus edulis. Histochem J 1990;22:595e603. [12] Bache `re E, Hervio D, Mialhe E. Luminol-dependent chemiluminescence by hemocytes of two marine bivalves, Ostrea edulis and Crassostrea gigas. Dis Aquat Org 1991;11:173e80. [13] Mitta G, Vandenbulckes R, Hubert F, Salzet M, Roch P. Involvement of mytilins in mussel antimicrobial defense. J Biol Chem 2000;275:12954e62. [14] Vargas-Albores F, Barracco MA. Mecanismos de defensa de los moluscos bivalvos, con e ´nfasis en pectı´nidos. In: MaedaMartı´nez AN, editor. Los Moluscos pectı´nidos de Iberoame ´rica. Me ´xico: Ciencia y Acuicultura. Editorial Limusa; 2001. p. 127e 46. [15] Cheng TC. Bivalves. In: Ratcliffe NA, Rowley AF, editors. Invertebrate blood cells. London: Academic Press; 1981. p. 233e300. [16] Auffret M. Comparative study of the hemocytes of two oyster species: the European flat oyster, Ostrea edulis, Linnaeus, 1750 and the pacific oyster, Crassostrea gigas (Thunberg, 1793). J Shellfish Res 1989;8:367e73. [17] Montes J, Acosta CP, Guerra A. Oyster mortality in galicia (NW Spain). In: De Pauw N, Jaspers E, Ackefors H, Wilkins N, editors. Aquaculture a biotechnology in progress. Bredene, Belgium: European Aquaculture Society; 1989. p. 941e8. [18] Bache `re E, Chagot D, Grizel H. Separation of Crassostrea gigas hemocytes by density gradient centrifugation and counterflow centrifugal elutriation. Dev Comp Immunol 1988;12:549e59. [19] da Silva PM, Villalba A. Comparison of light microscopic techniques for the diagnosis of the infection of the European flat oyster Ostrea edulis by the protozoan Bonamia ostreae. J Invertebr Pathol 2004;85:97e104.

P. Mirella da Silva et al. [20] Howard DW, Smith CS. Histological techniques for bivalve mollusks NOAA technical memorandum NMFS-F/NEC-25. Massachussets: Woods Hole; 1983. [21] Asokan R, Arumugam M, Mullainadhan P. Activation of prophenoloxidase in the plasma and haemocytes of the marine mussel Perna viridis Linnaeus. Dev Comp Immunol 1997;21: 1e12. [22] Bradford MM. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein dye-binding. Anal Biochem 1976;72:248e54. [23] Chagot D, Boulo V, Hervio D, Mialhe E, Bache `re E, Mourton C, et al. Interactions between Bonamia ostreae (Protozoa: Ascetospora)and hemocytes of Ostrea edulis and Crassostrea gigas (Mollusca: Bivalvia): entry mechanisms. J Invertebr Pathol 1992;59:241e9. [24] Renault T, Cochennec N, Grizel H. Bonamia ostreae, parasite of the European flat oyster, Ostrea edulis, does not experimentally infect the Japanese oyster, Crassostrea gigas. Bull Eur Ass Fish Pathol 1995;15:78e80. [25] Culloty SC, Novoa B, Pernas M, Longsheaw M, Mulcahy MF, Feist SW, et al. Susceptibility of a number of bivalve species to the protozoan parasite Bonamia ostreae and their ability to act as vectors for this parasite. Dis Aquat Org 1999;37: 73e80. [26] Cochennec-Laureau N, Auffret M, Renault T, Langlade A. Changes in circulating and tissue-infiltrating hemocyte parameters of European flat oysters, Ostrea edulis, naturally infected with Bonamia ostreae. J Invertebr Pathol 2003;83:23e30. [27] Fisher WS, Oliver LM, Edwards P. Hematologic and serologic variability of eastern oysters from Apalachicola bay, Florida. J Shellfish Res 1996;19:555e64. [28] Fisher WS. Environmental influence on bivalve hemocyte function. Am Fish Soc Spec Publ 1988;18:225e37. [29] Cajaraville MP, Pal SG. Morphofunctional study of the haemocytes of the bivalve mollusc Mytilus galloprovincialis with emphasis on the endolysosomal compartment. Cell Struct Funct 1995;20:355e67. [30] Carballal MJ, Lo ´pez MC, Azevedo C, Villalba A. In vitro study of phagocytic ability of Mytilus galloprovincialis Lmk. haemocytes. Fish Shellfish Immunol 1997;7:403e16. [31] Lo ´pez C, Carballal MJ, Azevedo C, Villalba A. Differential phagocytic ability of the circulating haemocyte types of the carpet shell clam Ruditapes decussatus (Mollusca: bivalvia). Dis Aquat Org 1997;30:209e15. [32] Chu FE. Defense mechanisms of marine bivalves. In: Fingerman M, Nagabhushanam R, editors. Recent advances in marine biotechnology immunology and pathology. USA: Science Publishers Inc; 2000. p. 1e42. [33] Montes JF, Durfort M, Garcı´a-Valero J. Cellular defence mechanism of the clam Tapes semidecussatus against infection by the protozoan Perkinsus sp. Cell Tissue Res 1995;279:529e38. [34] Comps M, Tige ´ G, Grizel H. Etudes ultrastructurales sur un protiste parasite de l’huıˆtre plate Ostrea edulis, 209D. Paris: C R Acad Sci; 1980. p. 383e4. [35] Balouet G, Poder M, Cahour A. Haemocytic parasitosis: morphology and pathology of lesions in the French flat oyster, Ostrea edulis L. Aquaculture 1983;34:1e14. [36] Ford SE, Kanaley SA, Littlewood DTJ. Cellular responses of oysters infected with Haplosporidium nelsoni: changes in circulating and tissue-infiltrating hemocytes. J Invertebr Pathol 1993;61:49e57. [37] Anderson RS, Burreson EM, Paynter KT. Defense responses of hemocytes withdrawn from Crassostrea virginica infected with Perkinsus marinus. J Invertebr Pathol 1995;66:82e9. [38] Chu FE, La Peyre JF. Development of disease caused by the parasite, Perkinsus marinus and defense-related hemolymph factors in three populations of oysters from Chesapeake Bay, USA. J Shellfish Res 1993;12:21e7.

Immune capability and its relation with infection [39] So ¨derha ¨ll K, Cerenius L. Role of the prophenoloxidase activating system in invertebrate immunity. Curr Opin Immunol 1998; 10:23e8. [40] Jordan PJ, Deaton LE. Characterization of phenoloxidase from Crassostrea virginica hemocytes and the effect of Perkinsus marinus on phenoloxidase activity in the hemolymph of Crassostrea virginica and Geukensia demissa. J Shellfish Res 2005; 24:477e82. [41] Mun ˜oz P, Meseguer J, Esteban MA. Phenoloxidase activity in three commercial bivalve species. Changes due to natural infestation with Perkinsus atlanticus. Fish Shellfish Immunol 2006;20:12e9. [42] Peters R, Raftos DA. The role of phenoloxidase suppression in QX disease outbreaks among Sydney rock oysters (Saccostrea glomerata). Aquaculture 2003;223:29e39.

563 [43] Newton K, Peters R, Raftos D. Phenoloxidase and QX disease resistance in Sydney rock oysters (Saccostrea glomerata). Dev Comp Immunol 2004;28:565e9. [44] Bezemer B, Butt D, Nell J, Adlard R, Raftos D. Breeding for QX disease resistance negatively selects one form of the defensive enzyme, phenoloxidase, in Sydney rock oysters. Fish Shellfish Immunol 2006;20:627e36. [45] Xue Q, Renault T. Enzymatic activities in European flat oyster, Ostrea edulis and pacific oyster, Crassostrea gigas, hemolymph. J Invertebr Pathol 2000;76:155e63. [46] Hervio D, Chagot D, Godin P, Grizel H, Mialhe E. Localization and characterization of acid phosphatase activity in Bonamia ostreae (Ascetospora), an intrahemocytic protozoan parasite of the flat oyster Ostrea edulis (Bivalvia). Dis Aquat Org 1991;12:67e70.