BSO inhibitor

Vulnerability of glutathione-depleted Crassostrea gigas oysters to Vibrio species

A B S T R A C T
Glutathione (GSH) is a major cellular antioXidant molecule participating in several biological processes, including immune function. In this study, we investigated the importance of GSH to oysters Crassostrea gigas immune response. Oysters were treated with the GSH-synthesis inhibitor buthionine sulfoXimine (BSO), and the function of immune cells and mortality were evaluated after a bacterial challenge with different Vibrio species. BSO caused a moderate decrease (20–40%) in GSH levels in the gills, digestive gland, and hemocytes. As ex- pected, lower GSH decreased survival to peroXide exposure. Hemocyte function was preserved after BSO treatment, however, oysters became more susceptible to challenges with Vibrio anguillarum, V. alginolyticus, or V. harveyi, but not with V. parahaemolyticus and V. vulnificus, indicating a species-specific vulnerability. Our study indicates that in natural habitats or in mariculture farms, disturbances in GSH metabolism may pre-dispose oysters to bacterial infection, decreasing survival.

1.Introduction
Bivalve mollusks have a wide distribution around the world and are highly abundant in marine coastal areas. They are particularly impor- tant for the marine coastal ecosystem, as well as a valuable food source for humans, being cultivated worldwide. Coastal environments are known to be severely impacted by human activities. Some of the major human-related effects in these environments involve excess input of nutrients and toXic chemicals (Lu et al., 2018). As a consequence, the high input of nutrients can lead to the proliferation of pathogenic or opportunistic bacteria, while toXic chemicals may affect important biological defense mechanisms. The combination of these stress factors can be highly detrimental to the health of wild and cultivated bivalves. An example of this scenario has become evident in the case of the Pacific oyster Crassostrea gigas. This oyster species is the most cultivated worldwide and has experienced massive mortality outbreaks in many countries in the past few years, leading to considerable economic losses for the aquaculture industry (EFSA AHAW Panel, 2015). Ongoing research has been pointing to a multi-factorial cause for these mortality events which involves pathogens, pollution, and genetic factors of oyster populations (EFSA AHAW Panel, 2015; de Lorgeril et al., 2018). Therefore, studies that approach the interplay between different stress factors that are commonly concurrent in the environment are of fundamental importance.Parameters related to the antioXidant system are frequently assessed in bivalves exposed to environmental stress factors. Among these pa- rameters, the levels of glutathione (GSH) –a major endogenous antioX- idant produced by the cells of most living organisms (Meister and Anderson, 1983)– are frequently reported to be modulated in aquatic organisms exposed to pollutants and environmental factors. For example, cellular GSH can be depleted in bivalves during exposures to metals (Conners and Ringwood, 2000) and organic contaminants (Cossu et al., 1997). Although GSH is very well known for its roles in the elimination of reactive oXygen species (ROS) and metabolism of elec- trophilic compounds (Meister and Anderson, 1983; Chen et al., 2000), GSH can also participate in other biological processes, such as DNA and protein synthesis, and cellular stress protection (Meister and Anderson, 1983), including resistance to bacterial and viral infections (Halliwell and Gutteridge, 2015; Forman and Torres, 2002; Townsend et al., 2003). Increasing research using mammalian models revealed GSH as an important modulator of several aspects of the immune system (Town- send et al., 2003; Haddad and Harb, 2005; Kim et al., 2004; Abdalla et al., 2011; Dro€ge et al., 1994), however, this relationship is still not very clear, especially concerning invertebrates. The importance of GSH for different biological processes has been frequently assessed using the nontoXic chemical buthionine sulfoXimine (BSO), which specifically and 2.3. Oyster BSO treatments irreversibly inhibits glutamate-cysteine ligase, the rate-limiting enzyme responsible for GSH synthesis leading to GSH depletion (Lu, 2013). This approach has been already successfully achieved in several animal models (Meister and Anderson, 1983; Morris et al., 2013; Al-Adhami et al., 2006; Paes et al., 2001; Mitchelmore et al., 2003; Canesi et al., 2000), including oysters (Conners and Ringwood, 2000; Ringwood and Conners, 2000).The purpose of this study was to investigate the consequences of in vivo GSH depletion on the immune competence, as wells as in the oyster survival after bacterial infection. After treating Pacific oysters C. gigas with the specific GSH-depleting agent BSO, we found a preserved function of immune cells, but oysters with lower GSH levels were more susceptible to some Vibrio species. This higher bacterial susceptibility observed with certain Vibrio species, but not others, indicate that GSH can provide species-specific protection against bacterial infection.

2.Materials and methods
2.1.Oysters
Adult C. gigas oysters (male and females; shell length: 4.7 0.6 cm; height: 8.5 0.4 cm; and wet tissue weight: 13.5 7.5 g) were pur- chased from an oyster farm located at Florianopolis, Brazil. Animals were acclimated for a minimum period of one week and kept in UV- treated and filtered seawater (1 L/animal) at 19–21 C with constant aeration. Seawater renewal and feeding with a commercial plankton diet (Sera® Marin Coraliquid) were done every two days.

2.2. Bacteria
Five Vibrio species commonly found associated with oyster natural beds and aquaculture farms (Sawabe et al., 2013; RouX et al., 2015; Pruzzo et al., 2005; Travers et al., 2015; Bramble and Anderson, 1997; Wu et al., 2013) and that display pathogenicity to immunosupressed hosts (Wu et al., 2013) were used in this study: V. alginolyticus (ATCC 17749), V. anguillarum (ATCC, 19264), V. harveyi (ATCC 14126),V. parahaemolyticus (ATCC 17802), and V. vulnificus (LAM 64 Canada). The V. alginolyticus strain was kindly provided by Dr. Rafael Diego da Rosa (Federal University of Santa Catarina, Florianopolis, Brazil), while the other four Vibrio strains were kindly provided by Dr. Maria Risoleta Freire Marques (Federal University of Santa Catarina, Florianopolis, Brazil). All Vibrio species were grown overnight in LB Broth (Sig- ma-Aldrich, Sa~o Paulo, Brazil) containing 3% NaCl, at 20–22 C. Bacteria were harvested by centrifugation (1000g, 10 min, 20 C), resuspended in filtered (0.22 μm) sterile seawater (FSSW) and used to assess hemocyte phagocytosis and for the survival challenges.

To promote GSH depletion, oysters were treated with DL-BSO (Sigma-Aldrich, Sa~o Paulo, Brazil) by injecting 250 μL of 60 mM BSO (15 μmol/injection/animal, diluted in FSSW) directly into adductor muscle through a notch in the shell, using a 23G needle attached to a 1 mL syringe. For the GSH content analysis, oyster tissues were collected 1, 2, 3 or 6 days after the first BSO injection (Fig. 1). Oysters corre- sponding to the 2- or 3-day treatments received two or three BSO in- jections, respectively, every 24 h (total of 30 or 45 μmol BSO/animal, respectively), while oysters corresponding to the 6-day treatment received a total of three BSO injections, one every 48 h (total of 45 μmol BSO/animal). Each BSO treatment had its own control group, which received the same volume (250 μL) of vehicle (FSSW) and the same number of injections on the same days. Only the 3-day BSO treatment was used for all other assays, with its corresponding control group. Oysters (n ¼ 12) were maintained in UV-treated and filtered seawater (1 L/animal) at 19–21 C with constant aeration and seawater renewal every 24 h. One tank was used per treatment containing 6 oysters each. EXperiments were repeated twice.

2.4.Sampling
Oyster hemolymph was collected from the adductor muscle using a 23G needle attached to a cold 1 mL syringe and maintained on ice until use. During collection, a droplet of each hemolymph was monitored through a light microscope. Samples that presented signs of contami- nation with microorganisms, gametes and/or hemocyte aggregation were discarded. Hemolymph samples were maintained on ice for a maximum of 1 h before use in further analyses, in order to prevent he- mocyte activation and aggregation. After hemolymph collection, oysters were shucked using an oyster knife, and the gill and digestive gland were dissected. Tissue samples were weighed and immediately processed for total GSH analysis.

2.5.Total GSH content
ApproXimately 50 mg (wet weight) of gills and digestive gland and 1 106 hemocytes (hemocyte count determined as described in section 2.6) were used to measure total GSH content (the sum of reduced an oXidized forms, GSH-t). Hemocytes were obtained from freshly extracted hemolymph (section 2.4) through centrifugation (800g, 10 min, 4 C) and suspended in 50 μL of 0.5 M perchloric acid (PCA) containing 0.3
mM EDTA. After intense pipetting to promote cellular disruption and protein denaturation, homogenates were pH-neutralized (pH 7.0) with a
0.3 M MOPS/2 M KOH solution, centrifuged (15000 g, 2 min, 4 C) and the supernatants were used for GSH-t determination. Freshly dissected tissues (section 2.4) were homogenized in 500 μL of 0.5 M PCA con- taining 0.3 mM EDTA using an Omni TH (Omni International, Kennesaw

Fig. 1. BSO treatments timeline. Syringe icons indicate the day oysters were injected with 15 μmol/animal of BSO or filtered sterile seawater for control groups. Stars indicate the day hemolymph and tissues were collected for GSH measurements. For further experiments, for which the 3-day treatment was chosen as a pre- treatment, the star indicates the day hemolymph was collected for hemocyte function analyses or the day oXidative stress or bacterial challenges started GA, USA) tissue homogenizer. Homogenates were further centrifuged (15,000 g, 2 min, 4 C), supernatants were recovered, neutralized by the addition of 0.1 M phosphate buffer pH 7.0 containing 1 mM EDTA,
and used to determine GSH-t levels. All samples were spectrophoto- metrically assayed for GSH-t determination on the same day of collec- tion at 412 nm using the enzymatic-coupled method described by Akerboom and Sies (1981). The assays contained 100 mM potassium phosphate buffer pH 7.0, 1 mM EDTA, 0.2 mM NADPH, 0.1 mM DTNB and 0.2 U/mL purified glutathione reductase.

2.6.Hemocyte parameters
2.6.1. Total and differential hemocyte counts
Hemocyte concentration in hemolymph (total hemocyte count or THC) was determined using the TC20™ Automated Cell Counter (Bio- Rad, Rio de Janeiro, Brazil) without previous dilution. Differential he- mocyte count (DHC) was assessed by cell sorting according to their relative size (forward scatter; FSC) and complexity (side scatter; SSC) using a FACSCanto II (BD Biosciences, Sa~o Paulo) flow cytometer and data analysis were made with FCS EXpress 5 Plus (De Novo Software). Hemocyte populations were classified into hyaline hemocytes, granular hemocytes, and aggregates, as shown in Fig. 2 cytometer tubes. Finally, 4 107 Vibrio cells of each species (multi- plicity of infection of 200) were added to the tubes. Suspensions were maintained on ice for 30 min in order to synchronize cell-bacteria binding, and phagocytosis was initiated by incubating samples at 20 C for 1 h in the dark. An additional set of tubes were incubated on ice
for 1 h in the dark, as a negative control. The percentage of hemocytes able to engulf propidium iodide-labeled bacteria was further assessed by flow cytometry (FACSCanto II, BD Biosciences, Sa~o Paulo, Brazil) through the PerCP channel. Data were analyzed with Flowing Software and the values obtained from negative controls were subtracted from each sample to account for bacteria bound to the cellular surface but not internalized.

Oysters were waterborne exposed (1 L/animal) for 96 h to 300, 1000 and 3000 μM cumene hydroperoXide (CHP; Sigma-Aldrich, Sa~o Paulo) or 10, 30 and 100 mM hydrogen peroXide (H2O2; Vetec Quimica, Duque de Caxias, Brazil). Due to the low solubility of CHP in seawater, the 3000 μM CHP solution was kept under constant stirring for at least 1 h before dilution in the exposure tank to achieve specified final concen- trations. Mortality was checked daily for 96 h and water was renewed and peroXide levels refreshed every 24 h. An additional set of experi- ments was carried out by pre-treating oysters with BSO or vehicle (FSSW) (3-day treatment, as described in section 2.3) followed by waterborne exposure to 100 and 300 μM CHP or 10 and 30 mM H2O2. Mortality was checked daily for 96 h, and water was renewed and peroXide levels refreshed every 24 h. During experiments, animals from all exposure groups were not fed. One tank was used per treatment containing 6 oysters each. EXperiments were repeated twice (n ¼ 12).The different Vibrio species, grown and harvested as described in section 2.2, were suspended in FSSW to OD600 ¼ 3.6, 1.2 or 0.4. Oysters were injected with 500 μL of each Vibrio suspension into the adductor muscle (through a notch in the shell using a 23G needle attached to a 1 mL syringe), and mortality was monitored daily for 96 h. Later, an additional set of experiments was carried out, where oysters were pre- treated with BSO or vehicle (FSSW) (3-day treatment, as described in section 2.3) and further injected with 500 μL of each Vibrio suspensions (OD600 3.6 or 1.2), and mortality was checked daily for 96 h. During all bacterial challenges, oysters were maintained in filtered seawater (1 L/animal) at 19–21 C with constant aeration. Half of the seawater was renewed each 24 h and animals were not fed during experiments. One tank was used per treatment containing 6 oysters each. EXperiments were repeated twice (n ¼ 12).Data were checked for normal distribution by the Kolmogorov- Smirnov test and for homoscedasticity by the Levene test. Data were analyzed by Student t-test for parametric data or Mann-Whitney U test for nonparametric data (THC and adhesion). Survival curves were analyzed by the Gehan-Breslow-WilcoXon test. Multiple curve compar- isons were made using the Bonferroni method. The significance level was set to 0.05 for all analyses.

3.Results
3.1.BSO treatments promote GSH depletion in oyster
GSH-t was detected in gills, digestive gland and hemocytes (Fig. 3), but not in cell-free hemolymph (data not shown). GSH depletion ranging from 27 to 46% was achieved with the 3 and 6-days BSO treatments in all tissues studied (Fig. 3). Gills and hemocytes responded very similarly (Fig. 3A and C), whereas, in the digestive gland, GSH-t levels increased (27%) with the 1-day treatment, but thereafter, followed the pattern similar to other tissues (Fig. 3B). For further experiments, we chose the 3-day treatment, which caused approXimately 30% GSH-t depletion in the studied tissues.To determine if a ~30% GSH content decrease in Pacific oyster tis- sues was physiologically relevant during oXidative challenges, animals were exposed to two pro-oXidants, CHP and H2O2, and mortality was monitored daily. Results from oysters exposed to a range of concentra- tions of each peroXide were used to determine a non-lethal concentra- tion (Fig. 4A and B). When these non-lethal CHP and H2O2 concentrations were used (0.3 and 10 mM, respectively), BSO pre- treatment significantly induced oyster mortality (Fig. 4C and D). Mor- tality was also increased by BSO pre-treatment when higher peroXide concentrations were used, however, the differences were not enough to reach statistical significance (Fig. 4E and F). As expected, GSH depletion increased the mortality induced by oXidative stress, though such effects were more evident in animals exposed to lower oXidative insults. Furthermore, these results indicate that a small loss of GSH, such as ~30%, is sufficient to increase oyster susceptibility to pro-oXidants.

Fig. 3. Total glutathione content (GSH-t) in the tissues of oysters C. gigas treated with the glutathione-depleting agent BSO. Animals subjected to the 1- day treatment received a single BSO (15 μmol/animal) injection, while those subjected to the 2 and 3-day treatments received injections every 24 h, and those subjected to the 6-day treatment received injections every 48 h. Three different tissues were analyzed: (A) Gills, (B) digestive gland, and (C) hemo- cytes. Data are presented as mean standard deviation (n 10–12), and as a percentage of the respective control group (Ctrl, dashed line). All Ctrl groups received the same volume of vehicle (filtered sterile seawater) and the number of injections. * (p < 0.05) and *** (p < 0.001) indicate significant differences relative to the respective Ctrl group. Actual GSH levels corresponding to 100% for each tissue are (A) 0.89 � 0.25 nmo/mg of tissue, (B) 1.19 � 0.30 nmo/mg of tissue, and (C) 11.06 � 13.23 nmo/mg of protein. proportion of subpopulations types in the hemolymph (THC and DHC, respectively), hemocyte viability through two different assays (NRR and MTT), as well as immune-related functions such as cellular adhesion and hemocyte capacity to perform phagocytosis of latex beads and heat- inactivated Vibrio. BSO pre-treatment did not significantly affect any of the parameters analyzed (Table 1). Phagocytosis of all species of heat- inactivated Vibrio was very low (2.7–8.5% of cells presenting PI-labeled Vibrio) regardless of the treatment (data not shown). These results indicate that the function of these cells is not endangered by a small GSH depletion, at least under laboratory conditions and in the absence of additional stressors. Fig. 4. BSO pre-treatment increases oyster susceptibility to peroXides. Oysters (n ¼ 12) were waterborne exposed to (A) 0.1–3 mM cumene hydroperoXide (CHP) or (B) 3–100 mM hydrogen peroXide (H2O2) for 96 h. (C–F) Oysters (n ¼ 12) pre-treated with the glutathione-depleting agent BSO (15μmol/injection/animal; 3-day treatment) or filtered sterile seawater (Control) were subsequently exposed for 96 h to peroXides: CHP (C) 0.3 mM, and (E) 1 mM; or H2O2 (D) 10 mM and (F) 30 mM. Curve comparisons were assessed by the Gehan-Breslow-WilcoXon test (n ¼ 10–12). In A and B significant differences (p < 0.05) against the control group are indicated by *, while in C–F significant differences (p < 0.05) are indicated by groups not sharing the same letter.Hemocyte parameters remain unaltered in BSO-treated oysters. Hemocytes obtained from oysters (n ¼ 12) pre-treated with the glutathione-depleting agent BSO (15 μmol/injection/animal; 3-day treatment) or filtered sterile seawater (Control) were used to estimate: total (THC) and differential (DHC) counts; cellular viability evaluated by the neutral red (NRR) and MTT assays; phagocytosis (percentage of hemocytes capable of engulfing three or more fluorescent microspheres); and cellular adhesion (percentage of hemocytes able to adhere to a polypropylene substrate). Data are presented as mean standard deviation. No significant differences were observed between Control and BSO groups (p > 0.05). conditions. At first, survival rates were determined in oysters C. gigas challenged with five Vibrio species. Animals were injected with a range of bacterial concentrations: OD600 0.4, 1.2 or was observed with all Vibrio species with the highest bacterial load, although statistical significance indicating increased mortality (p < 0.05) was achieved only after V. anguillarum (Fig. 5A) and V. parahaemolyticus (Fig. 5B) challenges. To confirm that lethality was not caused strictly by a septic shock from the bacterial injections, as opposed to bacterial proliferation and pathogenesis, oysters were injected with heat-inactivated of each Vibrio species (OD600 3.6) and no mortality was observed (data not shown). BSO pretreatment significantly increased the mortality of oysters challenged with V. anguillarum (Fig. 6A) and a trend for higher mortality was observed with the V. alginolyticus (Fig. 6C) and V. harveyi (Fig. 6D) challenges. This trend was noticed because the combination of BSO pre- treatment and Vibrio challenge caused significant mortality when comparing to the control group, which was not achieved when oysters were individually exposed to these two Vibrio species. BSO did not in- crease oyster susceptibility during V. parahaemolyticus (Fig. 6B) andV. vulnificus (Fig. 6E) infections. Overall, these data indicate that GSH depletion may increase the susceptibility of oysters to some pathogens, but not others. 4.Discussion In the present study, we used the GSH-synthesis inhibitor BSO (Meister and Anderson, 1983) to evaluate the importance of this key antioXidant molecule to the immune system and survival of oysters C. gigas under bacterial challenges. This is the first study to investigate the importance of GSH for bivalve immunity using a specific GSH-depleting agent. BSO has already been successfully administered in bivalves to decrease GSH levels in gonads or digestive gland (including in vitro studies with isolated digestive gland cells) (Conners and Ringwood, 2000; Canesi et al., 2000; Ringwood and Conners, 2000). In these pre- vious studies, GSH depletion (45–65%) did not interfere with oysters C. virginica gamete’s fertilization rate and embryo development (Ring- wood and Conners, 2000) and did not affect basal cellular or biochem- ical parameters (e.g. metallothionein content, lipid peroXidation, and lysosomal membrane stability) in the digestive gland of this species (Conners and Ringwood, 2000). In isolated mussel digestive gland cells, a 30–40% GSH depletion did not alter the ability of growth factors to stimulate key glycolytic enzymes (Canesi et al., 2000). A previous study from our group indicated that in vitro GSH depletion (close to 100%) using the electrophile compound 1-chloro-2,4-dinitrobenzene (CDNB) did not affect C. gigas hemocyte phagocytosis, laminarin induced-ROS production, cellular adhesion and viability (Mello et al., 2015). Simi- larly to these reports, in this study, hemocyte parameters were preserved in oysters presenting ~30% GSH depletion throughout different tissues. Altogether, these studies suggest that decreased GSH levels did not lead to the dysfunction of many biological parameters in healthy adult bi- valves. In fact, organisms of many phyla are able to survive under sys- temic GSH depletion (Eriksson et al., 2015), indicating the existence of compensatory mechanisms. The known antioXidant function of GSH prompted us to test if a modest GSH depletion would affect oyster survival under pro-oXidant challenges. Indeed, we found that BSO-treated oysters were more sus- ceptible to both peroXides, especially at non-lethal concentrations, indicating the importance of GSH for oyster survival under moderate oXidative stress. The survival of animals exposed to higher peroXide concentrations was not increased by GSH depletion, suggesting that the GSH system is already overwhelmed by the high levels of oXidants. Oysters and mussels presenting GSH depletion and decreased activities of thiol reductases, such as glutathione reductase (GR) and thioredoXin reductase (TrxR), have decreased in vivo peroXide detoXification ca- pacity and, as consequence, increased mortality induced by oXidative stress (Trevisan et al., 2012, 2014). A possible reason for oXidative stress-induced mortality after GSH depletion would be related to an impaired peroXide degradation based on in vitro (Liddell et al., 2006) and in vivo studies (Trevisan et al., 2014). Besides the increased susceptibility to peroXides, we found that BSO-treated oysters were more susceptible to bacterial infections. Similarly, mice constitutively lacking GR, the enzyme responsible for the reduction of oXidized glutathione, are viable and do not present any major ab- normalities in blood chemistry, hematological parameters or organ histology (Yan et al., 2012). However, when these mice are submitted to a bacterial challenge, 100% mortality was observed, while wildtype animals survived (Yan et al., 2012). Moreover, the study of Eriksson Fig. 5. Survival rates of oysters C. gigas challenged with five Vibrio species. Oysters (n 12) were injected with different Vibrio species (OD600 0.4, 1.2 or 3.6) or filtered sterile seawater (Control), and survival was monitored every 24 h. Curve comparisons were assessed using the Gehan-Breslow-WilcoXon test. Significant differences are indicated by * p < 0.05, #p 0.1, as compared to the control group. Fig. 6. BSO pre-treatment increases oysters C. gigas susceptibility to some Vibrio species. Oysters (n ¼ 12) pre-treated with the glutathione-depleting agent BSO (15 μmol/injection/animal; 3-day treatment) or filtered sterile seawater were subsequently injected with different Vibrio species (OD600 ¼ 0.4, 1.2 or 3.6) or filtered sterile seawater (Control), and survival was monitored every 24 h (A–E). Curve comparisons were assessed using the Gehan-Breslow-WilcoXon test. Groups not sharing letters are significantly different (p < 0.05). et al. (2015) discusses that redoX homeostasis can be maintained in mice lacking the GSH and thioredoXin (Trx) reducing systems, however, these animals are at the ‘brink of failure’. This scenario could also be true for oysters with partial GSH depletion, as oyster mortality increased upon subsequent stress, either oXidative stress or bacterial infection.When the innate immune system is activated by bacterial pathogens, phagocytes undergo a phenomenon called oXidative burst, which con- tributes to microorganism killing but drastically increases ROS pro- duction (Schmitt et al., 2011). These oXidants, and particularly H2O2 due to its high permeability, have the potential to damage the phago- cytes themselves as well as surrounding cells (Halliwell and Gutteridge, 2015; Segal, 2008). Thus, the host must be able to constantly repair such cellular damages during exposure and recovery from bacterial in- fections, which makes GSH-associated defenses and other antioXidant-related enzymes crucial to protect cells against oXidative stress that occurs during these immune-related events (Forman and Torres, 2002; Farooqui and Farooqui, 2011; Trachootham et al., 2008; Kiss, 2010). Indeed, it has already been reported in bivalves that Vibrio challenges enhanced gene transcription, protein expression, and protein activity of many antioXidant enzymes (Canesi et al., 2010; Genard et al., 2013; Huan et al., 2011). These findings are in line with the idea that the antioXidant system has a crucial role in acting conjointly with the im- mune system to combat bacterial infections, favoring animal recovery. Therefore, we suggest that systemic depletion of GSH (i.e., gills, diges- tive gland, and hemocytes) may exacerbate cellular damage and prevent a proper recovery during and after infections, rather than directly affecting hemocyte function during pathogen clearance. The observed importance of GSH in host defense against bacterial pathogens may be particularly relevant to bivalves. Oysters and other coastal organisms face daily harsh conditions of heat and dehydration as well as hypoXia and reoXygenation events due to tidal cycles, which is conceivably possible to produce ROS and consume GSH, and according to our results, become less resistant to harmful bacterial infections. Mortality events in hatcheries of oysters, as well as other bivalves, were already reported to be associated with Vibrio species used in this study (Dubert et al., 2017), particularly V. anguillarum and V. alginolyticus, which we demonstrated to present higher virulence to GSH-depleted oysters. Mortality outbreaks affecting C. gigas have increased in terms of intensity and geographic distribution in the past years (Barbosa-Solomieu et al., 2015). Studies suggest that these mortalities outbreaks cannot be explained by a single triggering factor, rather, by complex interactions between the physiological and/or genetic state of the host and environmental stress factors, allied to the presence of viral and bacterial pathogens (Barbosa-Solomieu et al., 2015; Romalde and Barja, 2010; EFSA, 2010; Morley, 2010; Moreau et al., 2015; Gagnaire et al., 2007). Several environmental stressors may decrease GSH levels in bivalves, such as exposure to metals (Kovaˇrova and Svobodova, 2009), diesel oil (Marques et al., 2014), polycyclic aromatic hydrocar- bons (Grintzalis et al., 2012), complex pollutant miXtures (Jena et al., 2009; Machado et al., 2014; Wan et al., 2015), electrophilic substances (Trevisan et al., 2012, 2016a), and elevated pCO2 (Wang et al., 2015). Thus, these reports, together with our findings, suggest that the presence of GSH-depleting stressors in the environment can be a risk factor involved in oyster mortality outbreaks. The oysters and Vibrio species used in this study were obtained from the coast of Santa Catarina. This is also where other bacteria and virus associated with mass mortality events (Mello et al., 2018) and several pollutants (Trevisan et al., 2016b; Souza et al., 2012) were previously detected, suggesting their simulta- neous occurence in the environment. This indicates that, although massive mortality events are not frequently observed on the Santa Catarina coast, wild and cultivated oysters from the area could be on the brink of failure. This may be true for other animal species present in the Santa Catarina coast and any other location around the globe where pathogens and GSH-depleting agents co-occur. Integrative field studies investigating the presence of pathogens, pollutants and organism GSH levels should contribute to validate these assumptions.Altogether, our findings, alongside with others (Conners and Ring- wood, 2000; Ringwood and Conners, 2000; Eriksson et al., 2015; Yan et al., 2012), reinforce that antioXidant depletion may not cause major detrimental biological effects to living organisms unless there is an additional stressor. This highlights the importance of studying the ef- fects of multiple BSO inhibitor stressors, an area that is still underrepresented among the ecotoXicology and immunotoXicity fields.