Immunomodulation of Crassostrea gigas and Crassostrea virginica cellular defense mechanisms by Perkinsus marinus.

ABSTRACT The eastern oyster is an economically and ecologically important species whose vitality is threatened by the protozoal parasite Perkinsus marinus. To better understand which cellular defense mechanisms impart resistance to P. marinus, resistant (Crassostrea gigas) and susceptible (Crassostrea virginica) oyster species were challenged by an experimental infection with P. marinus and their cellular responses were quantified and compared. Both in vivo and in vitro infection trials measured hemocyte phagocytosis, respiratory burst, apoptosis at 1, 3 and 7 days postinfection (in vivo) or 1-h postco-incubation (in vitro). Total parasite body burden concentrations were also measured at the end of in vivo infections. Infections were significantly more severe in C. virginica than C. gigas at 3 and 7 days postinfection confirming the resistance of C. gigas and validating the experimental model. There was more phagocytosis in infected C. virginica than infected C. gigas three days postinfection. In vitro, C. virginica granulocytes phagocytized significantly more parasites and fluorescent latex beads than C. gigas granulocytes, and infection increased bead phagocytosis in both species, equally in cells with or without intracellular parasites. Neither in vivo nor in vitro infections significantly increased respiratory burst activity. While in vitro infections suppressed hemocyte apoptosis in both species, in vivo infections increased hemocyte apoptosis frequency in C. gigas at 3 days postinfection. In vivo infection increased hemocyte apoptosis in C. virginica at 7 days postinfection but not at three days postinfection. From those experiments, we concluded that the increased phagocytosis without concomitant increase in respiratory burst activity seen in infected C. virginica might exacerbate infections. Also, while in vitro P. marinus infection suppresses hemocyte apoptosis in both species, C. gigas appeared to overcome that suppression faster than C. virginica upon in vivo infection, suggesting that hemocyte apoptosis may be an effective oyster defense response against P. marinus infection. The combination in vitro and in vivo infections in P. marinus disease resistant and susceptible oyster species with multiple time points and assays allowed the identification of apoptosis as the cellular defense mechanism most likely to play an important role in defense against P. marinus. This information may provide more accurate predictive criteria far disease resistance, allowing for the testing and selection of more disease resistant oysters.

KEY WORDS: immunomodulation, infection, Perkinsus marinus, Crassostrea virginica, Crassostrea gigas, apoptosis, phagocytosis

INTRODUCTION

The eastern oyster, Crassostrea virginica, is an economically and ecologically important species of the eastern seaboard of the United States but has undergone massive population declines in the last half of the twentieth century. During that period, production declined considerably with Chesapeake Bay landings alone dropping from 3 million to 6,000 bushels per year (Andrews 1996). These oyster population losses have resulted in increased water turbidity with likely adverse effects on phytoplankton and estuarine water quality dynamics (Newell et al. 2002). Though these population declines may also be attributed to management (Hargis & Haven 1988) and environmental variables (Ford et al. 2000), the apicomplexan protozoal parasite Perkinsus marinus (Perkins 1976, Levine 1978) is currently the most widespread and lethal pathogen of the eastern oyster (Burreson & Ragone Calvo 1996).

Poor resistance to infection with P. marinus in Crassostrea virginica has made it difficult to identify cellular and humoral defense components that may collectively or individually contribute to disease resistance. Molluscan innate immunity includes cellular functions such as phagocytosis, respiratory burst (Anderson et al. 1992), apoptosis (Lacoste et al. 2002) and diapedesis (Cheng 1981), as well as humoral factors including lysozymes, aminopeptidases, phospholipase C, antibacterial proteins, lectins and protease inhibitors (Mohandas & Cheng 1985, Kamiya et al. 1989, Cheng et al. 1995, Faisal et al. 1999).

Cellular mechanisms normally fill a crucial role in the destruction of infectious agents in hemolymph. Pathogens removed from circulation by phagocytosis (Cheng 1981a) are destroyed by the generation of reactive oxygen intermediate (ROI) free radicals (Adema et al. 1991, Anderson et al. 1994, Wojcik & Paynter 1996, Goedken & De Guise 2004) or eliminated outside the host via diapedesis (Stauber 1950). Apoptosis may be a defense mechanism if P. marinus infected oyster cells terminate infections with intracellular pathogens resistant to ROI destruction by initiating their own death (cell suicide) for later removal by other phagocytic cells, as is the case with other protozoan parasites (Luder et al. 2001). Phagocytosis and respiratory burst are not efficient at destroying P. marinus. Phagocytized P. marinus proliferate in host cells (Bushek et al. 1997) and disseminate via circulating hemocytes to cause systemic infection (Perkins 1976b). The recent demonstration of superoxide dismutase in P. marinus may explain its capacity to scavenge host ROIs and resist this killing mechanism (Ahmed et al. 2003). The only study investigating apoptosis in P. marinus-infected C. virginica reported reduced apoptosis of hemocytes in naturally infected oysters (Sunila & LaBanca 2003).

Although humoral mechanisms play an important role in host defense (Chu 1988a), their significance in resistance to P. marinus infection is unclear. Perkinsus marinus has extracellular products (ECP) and proteases that reduce oyster lysozyme activity and hemolymph agglutination (Garreis et al. 1996). Complementary expression of antiproteolytic factors specific for P. marinus proteases by oysters suggests an important role for the humoral response (Faisal et al. 1998, Oliver et al. 1998). Despite this, there are numerous reports suggesting that humoral factors do not mediate disease resistance. Hemolymph lysozyme and protein concentrations were not linked to P. marinus infection in C. virginica (Chu & La Peyre 1989). Several hemolymph serum agglutinins (lectins) had no role in defense against P. marinus, with concurrent high serum protein levels related to pathology and not disease resistance (Chintala et al. 1994). P. marinus trophozoites cultured in media supplemented with C. gigas plasma proliferated whereas those supplemented with C. virginica plasma did not, suggesting that serum factors may not be associated with C. gigas's resistance (Gauthier & Vasta 2002).

Numerous experiments have documented higher resistance to P. marinus in C. gigas than in C. virginica (Meyers et al. 1991, Barber & Mann 1994, La Peyre et al. 1995a, Chu et al. 1996, Calvo et al. 1999). Comparison of cellular immune functions have shown that neither species mounts a significant hemocyte ROI response (La Peyre et al. 1995b) but that C. virginica hemocytes phagocytize more P. marinus trophozoites than C. gigas (Gauthier & Vasta 2002). The higher resistance of C. gigas may be attributed to an intracellular environment unfavorable for parasite propagation (La Peyre et al. 1995a) or higher inhibitory activity against P. marinus proteases (Faisal et al. 1999). Although it is clear that C. gigas is more resistant than C. virginica to P. marinus infection, the mechanisms responsible for imparting such resistance have not been fully elucidated.

The goal of this study is to better understand the role of oyster cellular defense mechanisms in resistance to P. marinus by measuring phagocytosis, respiratory burst, apoptosis and infection levels in experimentally infected and control C. gigas and C. virginica.

MATERIALS AND METHODS

Materials

NaCl, Tris buffer, magnesium chloride, phorbol-12-myristate-13-acetate (PMA), Dulbecco modified eagle medium (DME) and Hams F-12 media, Ray's fluid thioglycollate medium (RFTM), sodium bicarbonate, glucose, galactose, trehalose, yeast extract, streptomycin sulfate, phenol red, potassium iodide, goat serum, bovine serum albumin and sterile seawater were obtained from Sigma-Aldrich Chemicals (St. Louis, MO). Propidium iodide (PI), oxacarbocyanine D275 (DiO), 2'7'-dichiorofluorescin diacetate (DCFH-DA), 1-[micro]m yellow-green fluorescent latex beads and 1-[micro]m red carboxylate-modified fluorescent latex beads were purchased from Molecular Probes (Eugene, Oregon). Phosphate buffered saline (PBS), RPMI-1640 medium, HEPES, L-glutamine and penicillin G were obtained from Gibco (Grand Island, New York). Ethanol was obtained from Pharmco (Brookfield, Connecticut), and paraformaldehyde was purchased from Ted Pella, Inc. (Redding, California). FITC-conjugated recombinant human Annexin-V, Annexin-V binding buffer and streptavidin-tri-color conjugate were from Caltag Laboratories (Burlingame, California). Apoptag Fluorescein Apoptosis Detection Kit was obtained from Intergen (Purchase, New York). Synthetic sea salt and water quality test kits were purchased from Aquarium Systems (Mentor, Ohio). Algae were purchased from Reed Mariculture, Inc. (San Jose, California). Perkinsus marinus isolate ATCC50862 (Dungan & Hamilton 1995) and biotinylated antiP, marinus IgG (Rb04 antiP. marinus CRTW-3HE meront) were provided by Cooperative Oxford Laboratory collaborators (Oxford, Maryland).

Experimental Infections

Oysters

Six- to eight-centimeter shell-height P. marinus-free C. virginica and C. gigas were obtained from Taylor Shellfish Farms (Shelton, Washington) between November 2003 and June 2004. All animals were fed algae (Reed Mariculture Inc., San Jose, California) and acclimated and maintained in artificial seawater (910 mOsm) in a refrigerated (16[degrees]C) 400-gallon tank with mechanical and biologic filtration for 3 or more weeks prior to infection trials. Oysters for in vivo trials were further acclimated an additional week prior to infection in 250-gallon tanks with artificial seawater (910 mOsm) kept at 21[degrees]C.

Ten- to 14-centimeter shell-height C. ariakensis were provided by the Virginia Institute of Marine Science (Gloucester Point, Vaginia).

P. marinus

Perkinsus marinus were propagated at 20[degrees]C in 10 mL of DME: Ham's F-12 media (Dungan & Hamilton 1995) supplemented with 0.25 [micro]L/mL filtered C. ariakensis hemolymph (Earnhart et al. 2004). Parasites were harvested in log phase growth.

In vivo Infections

Challenge with experimental infections on 36 oysters per species were initiated in parallel with sham-infections (seawater without P. marinus from tanks with identical oyster source, seawater, filtration and aeration systems. Temperature, salinity, ammonia, nitrate, nitrite, pH and salinity were measured every 3 days with 50% of the water replaced if water quality values were abnormal or differed between tanks. Water changes were done once during the acclimation period and at day 4 during the infection trials. Shells were notched with limited traumatic vibration using a 0.2 cm-thick abrasive wheel in a rotary (Dremel) tool. Oysters were experimentally infected with either [10.sup.6] P. marinus in 910 mOsm synthetic seawater (isotonic seawater) or 100-[micro]L isotonic seawater via mantle cavity injections with a 1-ml syringe and a 1", 25-gauge needle. After infection, the valves were held together with rubber bands 8 h following injection (Chintala et al. 2002). Twelve oysters per group were tested at 1, 3 and 7 days postinfection and sacrificed for tissue collection.

In vitro Infections

Experimental hemolymph infections for each oyster were performed in parallel with sham-infection on subsamples from the same oyster. After concentrations of P. marinus in cultures were adjusted, [10.sup.6] parasites were coincubated for 1 h at 20[degrees]C with equal numbers of hemocytes collected from C. gigas and C. virginica as previously described (Goedken & De Guise 2004). Light microscopy was used to confirm that P. marinus was phagocytized by hemocytes. To assure that immune functions were assessed on hemocytes and not free parasites, free P. marinus were labeled with antiP. marinus biotinylated IgG antibody (Ragone Calvo et al. 2003) for exclusion prior to quantifying immune functions by flow cytometry. P. marinus that were phagocytized were not accessible to the antibodies and therefore not labeled. For labeling of free P. marinus, samples were blocked with 1% goat serum in PBS for 1 h at 20[degrees]C and washed with isotonic seawater containing 0.05% Tween-20. Cells were incubated in 16 [micro]g/mL antiP. marinus antibody (determined as optimal in previous experiments) for 1 h at 20[degrees]C and washed with isotonic seawater with 0.05% Tween-20. Cells were then incubated for 1 h at 20[degrees]C with 1 [mu]/mL streptavidin-tricolor conjugate and washed with isotonic seawater. To establish accurate placement of electronic gates, populations of unlabeled and labeled parasites were used as negative and positive controls, respectively.

HISTOLOGY

Three-millimeter transverse body sections of oysters used for in vivo experiments were fixed in 10% neutral buffered formalin, embedded in paraffin and stained with hematoxylin and eosin. Sections were examined and gonadal development was classified using a previously described technique (Seed 1969).

PERKINSUS MARINUS TOTAL BODY BURDEN

Total body burden assays were performed on tissues of oysters used for in vivo infection studies. After hemolymph (1-2.5 mL), gill (0.5 g), digestive gland (0.5 g) and histology (3-mm transverse body section) samples were removed, the remaining tissues were minced, incubated in RFTM (Ray 1952) and processed for hypnospore quantification as previously described (Bushek 1993, Nickens et al. 2002). Samples were coded and read in a blind manner to eliminate bias. Final parasite densities were expressed as hypnospores per gram wet tissue weight.

PHAGOCYTOSIS

In vivo

Four hundred microliters hemolymph samples were collected from the adductor muscle sinuses of infected and control oysters (Chu 1988b) for quantification of hemocyte phagocytosis as previously described (Goedken & De Guise 2004). Briefly, hemocytes were incubated for 1 h at 20[degrees]C with a suspension of 2.5 x [10.sup.7] 1-[micro]m diameter, yellow-green fluorescent latex beads per mL hemolymph, and cell-associated fluorescence was evaluated using a Becton Dickinson FACScan flow cytometer. Phagocytosis was measured at 530 nm (FL-1) as the fluorescence of hemocytes relative to that of beads (cells will acquire a fluorescence equal to that of the beads they engulf, allowing the quantification of the number of beads phagocytized by each cell), using the Cell Quest software (Becton Dickinson Immunocytometry Systems, San Jose, California). A forward scatter threshold excluded nonhemocyte-associated fluorescence (free beads). Phagocytic potential for each group was measured as the percentage of granulocytes, which are the most active phagocytic cell (Foley & Cheng 1972, Goedken & De Guise 2004), and the percentage of hyalinocytes engulfing one or more beads.

In vitro

Prior to hemocyte/parasite coincubations, P. marinus were centrifuged (8 minutes, 210g), the pellet was resuspended in 1 mL of 30 [micro]M DiO solution, incubated for 20 min at 20[degrees]C and washed twice with isotonic seawater. Following experimental infection, infected and uninfected samples were further incubated for 1 h at 20[degrees]C with 6-[micro]m in diameter red fluorescent latex beads at a ratio of approximately 100 beads per hemocyte. Cells were fixed in 1% buffed formalin at the end of the 1-h incubation. Three-color flow cytometry was used to distinguish hemocytes with beads and parasites from extracellular parasites using a Becton Dickinson FAC-Scan flow cytometer and the Cell Quest software. After free parasites were excluded based on their fluorescence at 670 nm (FL-3) on labeling with an anti P. marinus antibody, the number of labeled P. marinus and/or beads phagocytized by each hemocyte was quantified at 530 nm (FL-1) and 585 nm (FL-2), respectively.

RESPIRATORY BURST

In vivo

Respiratory burst was quantified as previously described (Goedken & De Guise 2004). Briefly, after adjusting hemocyte concentration to 2 x [10.sup.6] cells/mL in ice-cold isotonic seawater, samples were split into phorbol myristate acetate (PMA) stimulated and unstimulated aliquots. Aliquots were incubated at 0[degrees]C for 30 min with 5 mM nonpolar, nonfluorescent 2'7'-dichlorofluorescin diacetate (DCFH-DA), washed, further incubated with or without 16.6 nM PMA for 1 h at 0[degrees]C and fixed in 1% formalin. Granulocytes were morphologically differentiated from hyalinocytes based on relative cell size and complexity (Goedken & De Guise 2004) and respiratory burst-associated increase in 2'7'-dichlorofluorescein (DCF) associated fluorescence, from the oxidation of DCFH-DA upon production of reactive oxygen species (ROIs), was calculated at 530 nm (FL-1) as the difference between the mean fluorescence of PMA-stimulated and unstimulated granulocytes.

In vitro

Following in vitro infections with P. marinus, cells were incubated with DCFH-DA with or without PMA, as described earlier, fixed in 1% formalin and free P. marinus were labeled. Free parasites were excluded based on their fluorescence at 670 nm (FL-3) upon labeling with an anti P. marinus antibody, and hemocyte respiratory burst activity was measured as the difference in mean DCF-associated cell fluorescence between stimulated and control aliquots at 530 nm (FL-1).

APOPTOSIS

In vivo

Hemocytes (1 x [10.sup.6]) were resuspended in 1 mL of Caltag binding buffer adjusted to 910 mOsm with NaCl. Five [micro]l of Caltag Annexin-V/FITC and 0.5 [micro]g PI were added to 1 x [10.sup.5] hemocytes (in 100 [micro]l binding buffer). After a 15-min incubation at 20[degrees]C, the cell suspensions were diluted 1:4 with binding buffer. Apoptosis-associated fluorescence was identified at 530 nm (FL-1) and necrosis-associated fluorescence, as PI incorporation, at 585 nm (FL-2) using a FACScan flow cytometer and CellQuest software. Negative control hemocyte suspensions were incubated with isotonic seawater instead of PI and/or Annexin-V to allow appropriate placement of quadrants. As previously validated for use in oysters (Lacoste et al. 2002), apoptotic cells were defined as Annexin-V-positive and PI-negative staining cells.

In vitro

Following in vitro infection of C. gigas and C. virginica hemocytes with P. marinus, infected and noninfected aliquots were sequentially labeled for free P. marinus, Annexin-V/FITC and PI after free parasites were excluded. Samples were acquired using three-color flow cytometry to detect apoptosis frequencies in conjunction with free parasite and necrotic hemocyte exclusion.

STATISTICAL ANALYSES

In vivo

Significance for differences in gonadal development, apoptosis frequencies and phagocytosis of one or more beads was established with 2-way analysis of variance to assess the influence of species and infection. Significance for differences in total body burden parasite concentrations and mean fluorescence associated with respiratory burst were validated with Mann-Whitney rank sum tests. Statistical significance of treatment and species differences were established with P < 0.05.

In vitro

Significant statistical differences between infected and control groups were established for each species with paired student's t-test and P < 0.05. The increase in bead phagocytosis on infection was compared between species as the ratio of beads in infected and uninfected subsamples, the statistical significance of which was established with a student's t-test and P < 0.05.

RESULTS

In vivo Infections

Perkinsus Marinus Total Body Burdens

Infection intensities, expressed as hypnospores per gram body weight, were calculated with a whole body burden assay. While there were no significant differences in parasite burdens one day postinfection, body burdens in C. virginica were significantly higher (2-3-fold) than in C. gigas at 3 and 7 days postinfection (Fig. 1).

[FIGURE 1 OMITTED]

Phagocytosis

Phagocytosis was measured as the proportion of hemocytes ingesting fluorescent latex beads. Hyalinocyte phagocytosis was 57% that of granulocytes, and was not significantly different between species or upon infection at any of the time points (data not shown). Whereas there were no significant differences in granulocyte phagocytosis between species or treatments at 1 and 7 days postinfection (Figs. 2a, 2c), granulocyte phagocytosis was significantly higher (approximately a 20% increase) in infected C. virginica than in infected C. gigas 3 days postinfection (Fig. 2b).

[FIGURE 2 OMITTED]

Respiratory Burst

Respiratory burst was measured as the difference in granulocyte oxidation of nonfluorescent DCFH-DA to the highly fluorescent DCF between PMA stimulated and control samples. There were no significant differences associated with species or treatment for any time point (Fig. 3).

[FIGURE 3 OMITTED]

Apoptosis

Hyalinocytes and granulocytes were defined by their relative size (forward scatter) and complexity (side scatter), and two-color flow cytometry was used to distinguish viable hemocytes from Annexin-V-FITC-stained apoptotic and PI-stained necrotic hemocytes at the subpopulation level. There were no significant differences in hyalinocyte apoptosis between species or upon infection 1 day postinfection (Fig. 4a), but infection with P. marinus significantly increased (by approximately 50%) the frequency of apoptosis in C. gigas hyalinocytes at 3 days postinfection (Fig. 4b). P. marinus also significantly increased (by approximately 50%) the hyalinocyte apoptosis in C. virginica at 7 days postinfection (Fig. 4c) but not at 3 days postinfection (Fig. 4b). In addition, the frequency of hyalinocyte apoptosis was significantly higher in infected C. gigas than in infected C. virginica at 3 days postinfection. Granulocyte apoptosis did not differ significantly between species or upon infection at any time in this study.

[FIGURE 4 OMITTED]

Histology

There were no significant differences in reproductive tract development between species or infection status for any time point during this study (data not shown). All oysters were classified as ripe, the full maturation stage preceding spawning.

In vitro Infections

Confirmation of In vitro Infections

Light microscopy was used to confirm that P. marinus (generally one, occasionally a few) was frequently observed within hemocytes (Fig. 5).

[FIGURE 5 OMITTED]

Labeling of Free P. marinus

Labeling of parasites with antibodies after coincubation with hemocytes allowed the identification and exclusion of free P. marinus from further analysis of cell function in infected hemocytes. Free parasite labeling was performed prior to measurement of respiratory burst (Fig. 6), phagocytosis and apoptosis (data not shown). Using positive and negative controls, regions were established to delineate infected and noninfected hemocytes (RI) from free parasites (R2).

[FIGURE 6 OMITTED]

Phagocytosis

After excluding free parasites, two-color flow cytometry allowed the differentiation of hemocytes with (right quadrants) or without (left quadrants) DiO-labeled P. marinus while simultaneously identifying hemocytes with (upper quadrants) or without (lower quadrants) red fluorescent beads (Fig. 7). The proportion of hemocytes that engulfed parasites was significantly higher (almost 2-fold) in C. virginica than C. gigas (Fig. 8a). Phagocytosis of beads was also quantified in hemocyte subsamples incubated with or without parasites. Samples infected with parasites had significantly greater bead phagocytosis than uninfected samples for both species (Fig. 8b). Phagocytosis of beads was also greater in C. virginica than in C. gigas, whether they were infected or not (Fig. 8b). The ratio of beads phagocytized by infected and uninfected subsamples in each species were compared (using a t-test) to quantify immunomodulation caused by infection with P. marinus. No significant differences in the magnitude of up-regulation of bead phagocytosis were observed (data not shown). When looking at individual cells within the infected subsamples, there were no significant differences in bead phagocytosis between cells with or without intracellular parasites in both species (Fig. 8c).

[FIGURES 7-8 OMITTED]

Respiratory Burst

In both species, flow cytometry was used to quantify the influence of in vitro P. marinus infection on the ability of hemocytes to mount a respiratory burst after free parasite exclusion. Respiratory burst activity was measured as the difference in DCF-associated cell flourescence between PMA-stimulated and nonstimulated samples. Experimental in vitro infection with P. marinus significantly reduced respiratory burst activity (approximately 3-fold) in C. gigas but not C. virginica (Fig. 9).

[FIGURE 9 OMITTED]

Apoptosis

Three-color flow cytometry was used to quantify the influence of in vitro P. marinus infection on the apoptotic frequency of hemocytes. After free parasites were excluded, Annexin-V-positive and PI-negative apoptotic cells were differentiated from PI-positive necrotic cells and Annexin-V-negative, PI-negative viable cells. Infection with P. marinus significantly reduced apoptotic frequencies (approximately 2-3-fold) in both species (Fig. 10).

[FIGURE 10 OMITTED]

DISCUSSION

Successful establishment of infections in susceptible hosts requires means of avoiding host responses soon after initiating infection (Yoshino & Vasta 1996). For that reason we chose to examine the crucial early phase of host response (relative to the duration of disease progression) and most likely time frame for immunomodulation. This time frame allowed the observation of differences in the severity of infection (parasite total body burden) between the two species. Whereas examining moribund animals provides information about disease progression (Chintala et al. 1994), examining early stages of infection provides insights into the mechanisms involved and their relative importance.

Transverse body sections were fixed and embedded for histologic examination to assess reproductive status. Gonadal development was not significantly different between species or upon infection (data not shown) and therefore likely not a factor influencing defense mechanisms. No treatment-associated lesions were observed.

Perkinsus marinus is readily phagocytized by oyster hemocytes wherein they grow and spread throughout the host as an intracellular parasite. Oysters have lectin surface receptors that function to recognize nonself material (Vasta et al. 1982, Yoshino & Vasta 1996). Perkinsus marinus may use C. virginica's ability to recognize and phagocytize it as an invasion mechanism. Such a mechanism has been demonstrated in other pathogens, such as Actinobacillus actinomycetemcomitans, which elicits its own uptake into host cells as a means to gain access to other tissues (Yoshino & Vasta 1996, Fives-Taylor et al. 1999). In vivo assays showed that there were no significant differences in baseline phagocytosis (uninfected) between species (Fig. 2) whereas in vitro assays indicated that C. virginica had more phagocytic ability than C. gigas for both P. marinus (see Fig. 7a) and beads (see Fig. 7b). Higher phagocytosis in C. virginica when compared with C. gigas has been demonstrated before (La Peyre et al. 1995b). There was no significant difference between the ratio of beads phagocytized in infected and uninfected subsamples for each species (see Fig. 7b), suggesting that P. marinus significantly upregulates phagocytosis similarly for both species. The significantly greater bead phagocytosis exhibited by infected C. virginica granulocytes may be attributable to their inherently higher phagocytic capability when compared with C. gigas. In vitro studies at the cellular level allowed the comparison of phagocytosis between cells with and without intracellular parasites (see Fig. 7c). The lack of differences in bead phagocytosis between cells with or without intracellular pathogens suggests that modulation of phagocytosis may result from either a nonspecific parasite factor or the stimulation of non-infected hemocytes by infected ones via cytokine-like or other soluble molecules. This increased phagocytosis likely brings more parasites into infected cells and exacerbates disease severity. Because the parasite evades intracellular destruction, phagocytosis of the parasite may not only be an ineffective defense mechanism but could exacerbate infection. By default, C. gigas's resistance may be a consequence of it's "inviting" fewer parasites into the nutrient-rich and immunoprivileged intracellular environment required for propagation and spread.

Phagocytosis followed by the production of free radicals to kill protozoal parasites is a common cellular defense mechanism of metazoans (Babior 1978). P. marinus circumvents this host response by neutralizing superoxide anion imparting the parasite with the ability to thrive in hemocytes for eventual systemic disease and host death (Mackin 1962, Anderson et al. 1995, Volety & Chu 1995, Anderson 1999a, Anderson 1999b, Ahmed et al. 2003). Although our assay sensitivity was increased by electronically gating only on granulocytes, which generate the most ROIs (Cheng 1984, Goedken & De Guise 2004), results showed no significant increase in in vivo and in vitro ROI activity between species or upon infection. In fact, in vitro-infected C. gigas had significantly less respiratory burst activity than their noninfected controls further supporting the theory that ROI activity is not an effective defense mechanism against P. marinus (see Fig. 9). These results are consistent with in vitro studies reporting a lack of respiratory burst, as measured by chemiluminescence (CL), in P. marinus stimulated C. virginica and C. gigas hemocytes (La Peyre et al. 1995b, Anderson 1999a, Anderson 1999b), but contrast an in vitro P. marinus infection study reporting that infected C. gigas had significantly greater ROI-associated activity than infected C. virginica (La Peyre et al. 1992, Anderson 1999b). This discrepancy with the later studies may be a consequence of their use of zymosan, which relies on a more complex receptor-ligand binding activation cascade (Vuorte et al. 1996, Cotran et al. 1999) that may interfere with the true immunomodulatory influence of P. marinus. The parasite's ability to scavenge host ROI, and suppress it in the more resistant species, suggests that respiratory burst activity is not an effective defense mechanism against P. marinus.

Apoptosis was investigated as a source of differential species resistance because it is a common host defense against protozoal parasites (Williams 1994), and neither cellular nor humoral mechanisms have adequately explained C. gigas's resistance to P. marinus infection. Thorough studies of apoptosis and its crucial role in innate immunity have been undertaken in organisms ranging from nematodes to mammals. Host cells infected with protozoa, bacteria or viruses may counteract intracellular infections by initiating their own death (cell suicide) for removal by phagocytic cells (Luder et al. 2001). Nevertheless, this host defense mechanism is a double-edged sword, because several intracellular pathogens have evolved mechanisms to manipulate host cell apoptosis to their advantage. Pathogens may either inhibit host apoptosis to create an intracellular sanctuary for enhanced replication, as is the case of Toxoplasma gondii and herpes viruses, or increase host apoptosis with the goal of destroying immune cells and/or inducing systemic infection, as is seen with Bordetella pertussis and vesicular stomatitis virus (Heussler et al. 2001). Although several other publications have reported on apoptosis in C. gigas and C. virginica hemocytes (Lacoste et al. 2002, Sunila & LaBanca 2003, Terahara et al. 2003), none have reported on mechanisms involved in differences in immunomodulation in different oyster species. In vitro infections suppressed hemocyte apoptotic frequencies in both species. This is not surprising because host apoptosis suppression is a common intracellular parasite strategy to confer and prolong the advantage of a beneficial living environment to the pathogen. The fact that hemocytes appear undisturbed up until the multiplying parasites rupture them (Perkins 1976b) is consistent with P. marinus's ability to circumvent hemocytes' last attempt to eradicate them (Teodoro & Branton 1998). Our results show that after 3 days, infected C. gigas had significantly higher apoptotic frequencies than their control group or infected C. virginica (see Fig. 4b). A similar significant increase in apoptotic frequency occurred 7 days postinfection for C. virginica (see Fig. 4c). Although both species respond similarly to P. marinus infection by overcoming the parasite's suppression of apoptosis and up-regulating hyalinocytes apoptosis during early phases of host response, C. virginica took longer than C. gigas to do so. Also, hemocyte apoptosis is suppressed in C. virginica chronically infected with P. marinus (Sunila & LaBanca 2003). The suppression of hyalinocyte apoptosis may be attributed to either intracellular P. marinus (as hyalinocytes do phagocytize P. marinus, although less than granulocytes) or to cytokine-like soluble factors secreted by neighboring cells (either extracellular P. marinus or P. marinus-infected cells), in relation with or independently from the modulation of phagocytosis. Differences between species in modulation of apoptosis in association with differences in severity of infection suggest that apoptosis may be a previously undocumented defense mechanism contribution important for C. gigas's superior resistance to P. marinus.

This study provides evidence of the role of cellular defense mechanisms in resistance to P. marinus. Hemocyte responses to in vitro infection were isolated to the cellular level, whereas in vivo infections allowed measurement of defense mechanisms over time within the physiologic conditions present at the whole oyster level. The parasites ability to circumvent intracellular killing via scavenging ROIs is key to the importance of both phagocytosis and apoptosis. Phagocytosis may not be a resistance mechanism but a susceptibility factor, because the infected susceptible species nonspecifically takes in increased numbers of parasites, which then thrive and spread throughout the host. Apoptosis, the final effort to expel intracellular pathogens after ROI killing fails, is a function parasites attempt to suppress to maintain intracellular sanctuary and hosts up-regulate to remove parasites from their cytoplasm. Whereas both species overcome that suppression and increase the frequency of apoptosis, the resistant species reacts sooner. This condition may be pivotal to C. gigas's resistance. This report is the first to give evidence that apoptosis is involved in oyster resistance to P. marinus. Further studies to confirm these findings and elaborate on the mechanism by which the parasite modulates apoptosis will be important for further clarification of the relationship between host and parasites, a necessary step to help in the subsequent selection for more disease resistant C. virginica.

LITERATURE CITED

Adema, C. M., W. P. Vanderknapp & T. Sminia. 1991. Molluscan hemocyte-mediated cytotoxicity: the role of reactive oxygen intermediates. Rev. Aquat. Sci. 4:201-223.

Ahmed, H., E. J. Schott, J. D. Gauthier & G. R. Vasta. 2003. Superoxide dismutases from the oyster parasite Perkinsus marinus: purification, biochemical characterization, and development of a plate microassay for activity. Anal. Biochem. 318:132-141.

Anderson, R. S., L. M. Oliver & L. L. Brubacher. 1992. Superoxide anion generation by Crassostrea virginica hemocytes as measured by nitrotetrazolium reduction. J. Invert. Pathol. 59:303-307.

Anderson, R. S., L. L. Brubacher & K. T. Paynter. 1994. Comparison of oxyradical generation by hemocytes from common estuarine bivalve mollusks. J. Shellfish Res. 13:297-298.

Anderson, R. S., E. M. Burreson & K. T. Paynter. 1995. Defense responses of hemocytes withdrawn from Crassostrea virginica infected with Perkinsus marinus. J. Invert. PathoL 66:82-89.

Anderson, R. S. 1999a. Lack of hemocyte chemiluminescence stimulation by Perkinsus marinus in eastern oysters Crassostrea virginica with dermo disease. J. Aquat. Anim. Health 11:179-182.

Anderson, R. S. 1999b. Perkinsus marinus secretory products modulate superoxide anion production by oyster (Crassostrea virginica) haemocytes. Fish & Shellfish Immunol. 9:51-60.

Andrews, J. D. 1996. History of Perkinsus marinus, a pathogen of oysters in Chesapeake Bay 1950-1984. J. Shellfish Res. 15:13-16.

Babior, B. M. 1978. Oxygen-dependent microbial killing by phagocytes (first of two parts). N. Engl. J. Med. 298:659-668.

Barber, B. J. & R. Mann. 1994. Growth and mortality of eastern oysters, Crassostrea virginica (Gmelin, 1791), and Pacific oysters, Crassostrea gigas (Thunberg, 1793) under challenge from the parasite, Perkinsus marinus. J. Shellfish Res. 13:109-114.

Burreson, E. M. & L. M. Ragone Calvo. 1996. Epizootiology of Perkinsus marinus disease of oysters in Chesapeake Bay, with emphasis on data since 1985. J. Shellfish Res. 15:17-34.

Bushek, D. 1993. Evaluation of Perkinsus marinus quantification techniques using fluid thioglycollate media. J. Shellfish Res. 12:126.

Bushek, D., S. K. Allen, Jr., K. A. Alcox, R. G. Gustafson & S. E. Ford. 1997. Response of Crassostrea virginica to in vitro cultured Perkinsus marinus: Preliminary comparisons of three inoculation methods. J. Shellfish Res. 16:479-485.

Calvo, G. W., M. W. Luckenbach, S. K. Allen, Jr. & E. M. Burreson. 1999. Comparative field study of Crassostrea gigas (Thunberg, 1793) and Crassostrea virginica (Gmelin, 1791) in relation to salinity in Virginia. J. Shellfish Res. 2:465-473.

Cheng, T. C. 1981. Bivalves. London: Academic Press. pp. 233-300.

Cheng, T. C. 1984. A classification of molluscan hemocytes based on functional evidences. Comp. Pathobiol. 5:33-50.

Cheng, T. C., J. J. Manzi & V. G. Burrell, Jr. 1995. Differences in lectin-binding by hemocytes of oysters (Crassostrea virginica) from three regions and further evidence for the correlation between the presence of lathyrose and the absence of Haplosporidium nelsoni. J. Shellfish Res. 14:477-481.

Chintala, M. M., S. E. Ford, W. S. Fisher & K. A. Ashton-Alcox. 1994. Oyster serum agglutinins and resistance to protozoan parasites. J. Shellfish Res. 13:115-121.

Chintala, M. M., D. Bushek & S. E. Ford. 2002. Comparison of in vitro-cultured and wild-type Perkinsus marinus. II. Dosing methods and host response. Dis. Aquat. Org. 51:203-216.

Chu, F. L. 1988a. Humoral defense factors in marine bivalves. Bethesda: Amer. Soc. Spec. Pub. pp. 178-188.

Chu, F. L. 1988b. Development and evaluation of techniques to study acquired immunity to Perkinsus marinus in the oyster Crassostrea virginica. J. Shellfish Res. 7:51-55.

Chu, F. L. & J. F. La Peyre. 1989. Effect of environmental factors and parasitism on hemolymph lysozyme and protein of American oysters (Crassostrea virginica). J. Invert. Pathol. 54:224-232.

Chu, F. L., A. K. Volety & G. Constantin. 1996. A comparison of Crassostrea gigas and Crassostrea virginica: Effects of temperature and salinity on susceptibility to the protozoan parasite, Perkinsus marinus. J. Shellfish Res. 15:375-380.

Cotran, R., V. Kumar & T. Collins. 1999. Robbins pathologic basis of disease. Philadelphia: W.B. Saunders Company. pp. 1-29.

Dungan, C. F. & R. M. Hamilton. 1995. Use of a tetrazolium-based cell proliferation assay to measure effects of in vitro conditions on Perkinsus marinus (Apicomplexa) proliferation. J. Eukar. Microbiol. 42: 379-388.

Earnhart, C. G., M. A. Vogelbein, G. D. Brown, K. S. Reece & S. L. Kaattari. 2004. Supplementation of Perkinsus marinus cultures with host plasma or tissue homogenate enhances their infectivity. Appl. Environ. Micro. 70:421-431.

Faisal, M., E. A. MacIntyre, K. G. Adham, B. D. Tall, M. H. Kothary & J. F. La Peyre. 1998. Evidence for the presence of protease inhibitors in eastern (Crassostrea virginica) and Pacific (Crassostrea gigas) oysters. Comp. Biochem. Physiol. pp. 161-168.

Faisal, M., J. L. Oliver & S. L. Kaatari. 1999. Potential role of protease-antiprotease interactions in Perkinsus marinus infection in Crassostrea spp. Bull. Eur. Assoc. Fish Pathol. 19:269-276.

Fives-Taylor, P. M., D. H. Meyer, K. P. Mintz & C. Brissette. 1999. Virulence factors of Actinobacillus actinomycetemcomitans. Peridontal. 2000 20:136-167.

Foley, D. A. & T. C. Cheng. 1972. Interaction on molluscs and foreign substances: the morphology and behavior of hemolymph cells of the American oyster Crassorea virginica; in vitro. J. Invert. Pathol. 19: 282-394.

Ford, S. E., R. Smolowitz & M. M. Chintala. 2000. Temperature and range extension by Perkinsus marinus. J. Shellfish Res. 19:598.

Garreis, K. A., J. F. La Peyre & M. Faisal. 1996. The effects of Perkinsus marinus extracellular products and purified proteases on oyster defense parameters in vitro. Fish & Shellfish Immunol. 6:581-597.

Gauthier, J. D. & G. R. Vasta. 2002. Effects of plasma from bivalve mollusk species on the in vitro proliferation of the protistan parasite Perkinsus marinus. J. Exp. Zool. 292:221-230.

Goedken, M. & S. De Guise. 2004. Flow cytometry as a tool to quantify oyster defense mechanisms. Fish & Shellfish Immunol. 16:539-552.

Hargis, W. J., Jr. & D. S. Haven. 1988. Rehabilitation of the troubled oyster industry of the lower Chesapeake Bay. J. Shellfish Res. 7:271-279.

Heussler, V. T., P. Kuenzi & S. Rottenberg. 2001. Inhibition of apoptosis by intracellular protozoan parasites. Int. J. Parasitol. 31:1166-1176.

Kamiya, H., K. Muramoto, R. Goto, M. Sakai, Y. Endo & M. Yamazaki. 1989. Purification and characterization of an antibacterial and antineoplastic protein secretion of a sea hare, Aplysia juliana. Toxicon 27: 1269-1277.

La Peyre, J. F., F.-L. E. Chu & W. K. Vogelbein. 1992. In vitro interaction of Perkinsus marinus with hemocytes from eastern and Pacific oysters, Crassostrea virginica and Crassostrea gigas. J. Shellfish Res. 11:200.

La Peyre, J. F., F. L. Chu & J. A. Movern. 1995a. Hemocytic and humoral activities of eastern and pacific oysters following challenge by the protozoan Perkinsus marinus. Fish Shellfish Immunol. 5:179-190.

La Peyre, J. F., F. I. Chu & W. K. Vogelbein. 1995b. In vitro interaction of Perkinsus marinus merozoites with eastern and Pacific oyster hemocytes. Dev. Comp. Immunol. 19:291-304.

Lacoste, A., A. Cueff & S. A. Poulet. 2002. P35-sensitive caspases, MAP kinases and Rho modulate beta-adrenergic induction of apoptosis in mollusc immune cells. J. Cell. Sci. 115:761-768.

Levine, N. D. 1978. Perkinsus gen.n. and other new taxa in the protozoan phylum Apicomplexa. J. Parasitol. 64:549.

Luder, C. G., U. Gross & M. F. Lopes. 2001. Intracellular protozoan parasites and apoptosis: diverse strategies to modulate parasite-host interactions. Trends Parasitol. 17:480-486.

Mackin, J. G. 1962. Oyster diseases caused by Dermocystidium marinum and other microorganisms in Louisiana. College Station: Texas A&M University. pp. 132-229.

Meyers, J. A., E. M. Burreson, B. J. Barber & R. Mann. 1991. Susceptibility of diploid and triploid oysters, Crassostrea gigas, to Perkinsus marinus. J. Shellfish Res. 10:304.

Mohandas, A. & T. C. Cheng. 1985. Release pattern of aminopeptidase from Biomphalaria glabrata hemocytes subjected to high-level bacterial challenge. J Invert Pathol 45:298-303.

Newell, R. I. E., J. C. Cornwell & M. S. Owens. 2002. Influence of simulated bivalve biodeposition and microphytobenthos on sediment nitrogen dynamics: A laboratory study. Limnol. Oceanog. 47:1367-1379.

Nickens, A. D., J. F. La Peyre, E. S. Wagner & T. R. Tiersch. 2002. An improved procedure to count Perkinsus marinus in Eastern oyster hemolymph. J. Shellfish Res. 21:725-732.

Oliver, J. L., M. Faisal & S. L. Kaattari. 1998. The effect of oyster hemolytic factors on Perkinsus marinus proteases. J. Shellfish Res. 17:334-335.

Perkins, F. O. 1976. Dermocystidium marinum infection in oysters. Mar. Fish. Rev. NMFS 38:19-21.

Ragone Calvo, L. M., C. F. Dungan, B. S. Roberson & E. M. Burreson. 2003. Systematic evaluation of factors controlling Perkinsus marinus transmission dynamics in lower Chesapeake Bay. Dis. Aquat. Org. 56:75-86.

Ray, S. M. 1952. A culture technique for the diagnosis of infections with Dermocystidium marinum (Mackin, Owen & Collier) in oysters. Science 116: 360-361.

Seed, R. 1969. The ecology of Mytilus edulis L. on exposed rocky shoes. I. Breeding and settlement. Oecologia 3: 277-335.

Stauber, L. A. 1950. The fate of india ink injected intracardially into the oyster, Ostrea virginica Gmelin. Biol. Bull. 98:227-241.

Sunila, I. & J. LaBanca. 2003. Apoptosis in the pathogenesis of infectious diseases of the eastern oyster Crassostrea virginica. Dis. Aquat. Org. 56:163-170.

Teodoro, J. G. & P. E. Branton. 1998. Regulation of apoptosis by viral gene products. J. Virol. 71:1739-1746.

Terahara, K., K. G. Takahashi & K. Mori. 2003, Apoptosis by RGD-containing peptides observed in hemocytes of the Pacific oyster, Crassostrea gigas. Dev. Comp. Immunol. 27:521-528.

Vasta, G. R., J. T. Sullivan, T. C. Cheng, J, J. Marchalonis & G. W. Warr. 1982. A cell membrane-associated lectin of the oyster hemocyte. J. Invert. Pathol. 40:367-377.

Volety, A. K. & F. L. Chu. 1995. Suppression of chemiluminescence of eastern oyster (Crassostrea virginica) hemocytes by the protozoan parasite Perkinsus marinus. Dev. Comp. Immunol. 19:135-142.

Vuorte, J, S. E. Jansson & H. Repo. 1996. Standardization of a flow cytometric assay for phagocyte respiratory burst activity. Scand. J. Immunol. 43:329-334.

Williams, G. T. 1994. Programmed cell death: a fundamental protective response to pathogens. Trends Microbiol. 2:463-464.

Wojcik, J. & K. T. Paynter. 1996. Myeloperoxidase activity from blood cells of the eastern oyster, Crassostrea virginica. J. Shellfish Res. 15: 532.

Yoshino, T. P. & G. R. Vasta. 1996. Parasite-invertebrate host immune interactions. In: E. L. Cooper, editor. Invertebrate immune responses: Cell activities and the environment. Heidelberg: Springer-Verlag. pp. 125-167.

MICHAEL GOEDKEN, (1) BRENDA MORSEY, (1) INKE SUNILA (2) AND SYLVAIN DE GUISE (1) *

(1) Department of Pathobiology and Veterinary Science, University of Connecticut, 61 N Eagleville Road, Storrs, Connecticut 06269; (2) State of Connecticut, Department of Agriculture, Bureau of Aquaculture, PO Box 97, Milford, Connecticut, 06460

* Corresponding author. E-mail: [e-mail]