Immunotoxicity of heavy metals (silver, cadmium, mercury and lead) on marine bivalve Mytilus edulis: in vitro exposure of hemocytes

P. Rault,1,2 M. Fortier,2 J. Pedelucq,2 E. Lacaze,2 P. Brousseau,2 M. Auffret,3 M. Fournier2 1Institut Universitaire Europeen de la Mer, Universite de Bretagne Occidentale, Plouzane, France; 2INRS-Institut ArmandFrappier, Laval, Quebec, Canada; 3Laboratoire des sciences de l’environnement marin, UMR CNRS 6539, Institut Universitaire Europeen de la Mer, Universite de Bretagne Occidentale, Plouzane, France


Introduction
Anthropogenic pressures lead to profound deleterious effects on coastal aquatic ecosystems. Many chemicals released in the environment generate toxicity in aquatic organisms. In response to these environmental disturbances, there is a growing concern about the consequences of environmental contaminants on the immune system of aquatic species. 1,2 Various environmental surveys have been set up to monitor these changes. They allow medium and long-term monitoring of coastal ecosystems exposed to chemical contamination through the study of selected sentinel species. As sessile filter-feeder, bivalve mollusks are directly exposed to environmental contaminants. They bioaccumulate most environmental pollutants, including heavy metals. 3,4,5 Therefore, they reflect the surrounding contamination by integrating the environmental exposure. Immune defense in bivalves is coordinated primarily by circulating hemocytes. Those multifunctional cells are able to recognize foreign particles and respond through phagocytic ingestion. 6 In mussels, hemocytes circulating in the hemolymph represent the main component of their immune system. 7,8 Hence, the blue mussel, Mytilus edulis, is considered as interesting sentinel specie in immunotoxicological studies.
The use of hemocytes and the monitoring of their phagocytic activity may represent an ideal biological endpoint. 9 The present study aims to validate the use of phagocytic responses in Mytilus edulis as biomarker of heavy metals contamination. The purpose of this present study is to establish dose-response curves of viability and phagocytic activity on marine bivalve Mytilus edulis exposed to various heavy metals [cadmium (Cd), lead (Pb), mercury (Hg), silver (Ag)].

Animals
Mytilus edulis were collected in the Baie de Plaisance from Îles de la Madeleine (47°29'N, 61°87'W). They are located in the center of Gulf of St Lawrence, a place exempt from industrial activities. Upon arrival to the laboratory, all bivalves were transferred and maintained in aquaria filled with salted artificial marine water (15°C, salinity 31-1 psu) and fed ad libitum.

Hemolymph collection and viability of cells
Hemolymph was collected individually from the posterior adductor muscle using a 3 mL syringe with a 23G gauge needle. Immediately after the hemolymph collection, hemocytes viability was evaluated by flow cytometry using the Guava PCA flow cytometer (Guava Technologies, Hayward, CA, USA) and the Viacount kit (Guava technologies) according to the supplier's instructions. Briefly, an aliquot of 50 mL of hemolymph was mixed with 200 mL of Viacount and at least 1000 events were recorded. Only individuals whose cell viability was higher than 75% were kept.

Heavy metals and exposure protocol
Heavy metals used in this study were reagent grade: cadmium chloride (CdCl 2 ), lead chloride (PbCl 2 ), mercuric chloride (HgCl 2 ) and silver nitrate (AgNO 3 ) (Sigma, ON, Canada). The concentrations tested were the same for all four metals, ranging from 10 -3 to 10 -9 mol. Working solutions of each metal were then prepared by serial dilutions in distilled water.

Phagocytosis and cell viability
For each metal, dose-response curves were performed on nine different mussels. Phagocytosis was assessed by determining the ingestion of latex fluorescent beads by flow cytometry according to Brousseau et al. 10 Briefly, 500 mL of hemolymph containing 100,000 cells were mixed with 25 mL of each metal working solution and exposed during 3 h at 15°C under constant agitation. After the incubation, latex fluorescent beads (Yellowgreen Fluoresbrite, Polysciences ® ) were added to each tube at a ratio of 100:1 beads/cell and suspensions were re-incubated for another 18 hours. At the end of the exposure period, the cells were layered over a 3% bovine serum albumin gradient in RPMI-1640 medium containing 31 g.l -1 of marine salt and centrifuged at 138 x g for 8 min at 4°C to remove free beads. The cell pellets were then re-suspended in 0.5 mL of 0.5% formaldehyde and 0.2% sodium azide in 31 g.l -1 salt water. The number of engulfed beads was determined using a FACSCalibur (Becton-Dickinson, CA, USA) flow cytometer. Fluorescence emission was read in FL1 (λ=530 nm), with at least 10,000 events in the region of interest.

Statistical analysis
Data are presented as mean standard deviation. The results were tested for homogeneity using an analysis of variance (ANOVA). Significant differences between treatments and control were determined using the SigmaStat™ Software version 3.5. (Jandel, Scientific Software, San Rafael, CA, USA). Then, we used the Holm-Sidak method to perform multiple comparisons versus control group to identify significant differences (P<0.05).

Effects of heavy metals on cell viability
Regarding hemocyte viability, as illustrated in Figure 1, silver nitrate appeared to be more cytotoxic than mercury for this population of cells. We noticed a significant decrease in viability at 10 -5 M for Ag ( Figure 1D) and Hg ( Figure 1C). Cadmium ( Figure 1A) and lead ( Figure 1B) showed similar results on cell viability with a significant decrease at 10 -3 M.

Effects of heavy metals on phagocytosis
In vitro analysis of metal toxicity on hemocytes ranked the metals as follows: mercury > silver > lead > cadmium. These heavy metals also lead to a significant increase in the immune response at low concentrations. These concentrations tested are much higher than those found in the ocean. Silver induced a significant decrease of phagocytosis at concentrations ranging from 10 -5 M to 10 -3 M ( Figure 1D). Dissolved Ag concentrations in seawater ranging from 1-20.10 -12 M (0.1-2.0 ng.L -1 ) in pristine offshore waters up to 300.10 -12 M (30 ng.L -1 ) in estuaries close to point source discharges. 11 A significant increase of phagocytosis at 10 -7 M is noticed. The IC 50 value obtained (calculated with a linear regression) for the hemocytes exposed to silver nitrate is 9.10 -6 M ( Table 1). Mercury induced a significant decrease of phagocytosis at concentrations ranging from 10 -6 M to 10 -3 M ( Figure 1C). We can also notice a significant increase of phagocytosis at concentrations ranging from 10 -9 M to 10 -7 M. The IC 50 value calculated for the hemocytes exposed to mercury chloride is 3.10 -6 M ( Table 1). Total Hg concentrations are low and range from 0.2 to 4.10 -12 M (40 to 802 pg.L -1 ), with the higher concentrations found near the coast likely due to riverine inputs. 12 Cadmium and lead induced a significant decrease of phagocytosis at 10 -3 M ( Figure 1A and B). For both metals the IC 50 values calculated are greater than 10 -3 M (Table 1). CdCl 2 and PbCl 2 compounds significantly increased the immune response at low concentrations ranging from 10 -9 M to 10 -6 M. In seawater, dissolved lead is between 7,2 and 72.10 -11 M (15-150 ng.L -1 ) in the North Atlantic and total dissolved Cd exists at low concentrations usually less than 1.8.10 -6 M (0.2 mg.L -1 ). 13

Discussion
Environmental concentrations do not induce short-term responses, so we tested higher concentrations ranges. These concentrations test-ed are found in oceans for cadmium, while silver, lead and mercury are present in lesser amounts. Hemocyte cells of Mytilus edulis showed a classical pattern of the in vitro toxicity of heavy metals. At low concentrations, there is a lack of cell cytotoxicity and stimulation of the phagocytosis then followed with normal phagocytosis. At moderate concentrations, phagocytosis is dramatically reduced which is accompanied by low changes in cell viability. At high concentrations, cytotoxicity is relatively high, accompanied by a drastic inhibition of the phagocytosis. Similar results support the observation that exposure to low levels may stimulate phagocytic activity whereas increasing concentrations induce suppression. 9,14 The average sensitivity for phagocytosis can be ranked in decreasing order of toxicity: HgCl 2 , AgNO 3 , and CdCl 2 . The impact of lead on phagocytosis was never analysed in the blue mussel and was ranked between Hg and Cd. In Mya arenaria, it was suggested by others 9 that hemocyte mortality only partially explained the decrease of the phagocytic index. Similarly, for the blue mussel, cytotoxicity is not the sole mechanism for phagocytosis inhibition. For example, cells incubated with HgCl 2 at 10 -5 M show an almost normal level of viability (80% of control), while phagocytosis is strongly inhibited with only 20% of the control value. The phagocytosis dose-response curves for AgNO 3 , HgCl 2 suggest that the toxicity tends to be threshold rather than linear. For example, AgNO 3 concentrations lower than 10 -6 M, HgCl 2 lower than 10 -7 M had no apparent negative effect on phagocytosis. Yet, next higher dose shows inhibition levels sometimes exceeding 60%. Compared to lethal or acute bioassays, the measurement of phagocytosis by flow cytometry in bivalves represents a sensitive endpoint to study the adverse effects of heavy metals at sublethal concentrations. 15 However, exposure was done under acute and short-term conditions which differ from environmental exposures i.e., under chronic and low concentrations.