Haemolymph fluid osmolality influences the neutral-red retention assay in the eastern oyster<i>Crassostrea virginica</i>

The neutral-red retention assay (NRRA) is a biomarker of general stress that can be used to detect primary intracellular disturbances, i.e. lysosomal destabilization (Lowe, Soverchia & Moore, 1995; Izagirre & Marigomez, 2009). It is known that environmental variables (e.g. salinity and temperature) may impact lysosomal stability in bivalves (Camus et al., 2000; Ringwood et al., 2002; Izagirre, Ramos & Marigomez, 2008). Salinity and temperature fluctuations have been correlated with overall stress in oysters using the NRRA technique (Hauton et al., 1998; Song et al., 2007; Mayrand et al., 2013). However, the interpretation of NRRA results can be problematic, particularly in field-collected animals exposed to highly variable environmental conditions (Ringwood et al., 1999). In this study, we investigated another factor with potential to compromise the quantitative assessment of NRRA responses. NRRA measurements involve the use of a prescribed physiological saline solution (Schlieper, 1972) adapted to the salinity of the surrounding seawater. Specifically, haemocytes are sampled from the posterior adductor muscle sinus and mixed into an equal volume of the saline solution. We postulate that under low temperature (5 °C) and salinity (6‰) conditions, oysters are not isosmotic to their surrounding environment and we therefore question the robustness of the standard NRRA technique (Méthé et al., 2015). For example, in the southern Gulf of St Lawrence (Canada) water temperature reaches sub-zero values during winter and, in the upper reaches of the river systems, saline water is often absent during spring freshets and autumn floods. Under such conditions oysters close their valves and limit exchange with their surroundings (Loosanoff, 1958; Comeau et al., 2008), thereby presumably rendering them hyperosmotic to the ambient seawater. The general objective of this laboratory study was to compare the existing technique (NRRA) with a modified technique (NRRAMOD) wherein haemolymph fluid from Crassostrea virginica was used instead of the prescribed saline solution. It was hypothesized that using haemolymph fluid perfectly adjusted to the osmolality of the individual oyster should improve the accuracy of the assay. Oysters were subjected to different temperature and salinity regimes in order experimentally to manipulate haemolymph fluid osmolality. To gain additional insight into oyster behaviour, valve activity was monitored during the study. Oysters (77.5 ± 1.1 mm shell height) were collected from floating cages located in Shédiac Bay, New Brunswick on 5 December 2011, and transported to the Aquatic Animal Facility at the Atlantic Veterinary College, University of Prince Edward Island, Charlottetown, Prince Edward Island. Water temperature and salinity at the time of collection were 5 °C and 27‰, respectively. Within 2 h of collection, oysters were placed in a static aquarium (300 l) that was equipped with titanium chillers and contained aerated artificial seawater (Instant Ocean®, Kent Marine, Acworth). Environmental conditions were maintained as follows: temperature 5.0 ± 0.5 °C, salinity 27 ± 2‰, pH 7.5, dissolved oxygen [DO] > 70% and total ammonia < 1 ppm. Oysters were fed with algal paste once a day, at a concentration of 10,000 cells ml−1 (Innovative Aquaculture Products, Lasqueti Island, BC) and the water was changed on a weekly basis. Following a 30-d acclimation period, oysters were distributed among 20 22-l static tanks and acclimated to different experimental conditions by increasing or decreasing water temperature at a rate of 1–2 °C per day (consistent with observations from the collection site) using titanium chillers or heaters such that all groups reached the desired temperature on the same day (0, 5, 10, 15, 20 °C). Salinity was maintained at 27 ± 2‰ with the addition of freshwater as required to compensate for evaporation. Each tank (four tanks per temperature regime) contained 15 oysters (i.e. 1.5 l seawater per oyster). Oysters were fed algal paste daily (150,000 cells ml−1), the water was changed every 7 d and the photoperiod was maintained at ambient winter level (8 L:16D). Other water quality parameters were monitored daily throughout the study period. Salinity, pH and dissolved oxygen were measured using a handheld YSI Pro 2030 meter. Total ammonia was determined with a HACH kit (Hach Company, Loveland, CO). Dissolved oxygen and ammonia levels were 83.0–111.5% and 0.00–0.51 ppm, respectively. These values correspond to a stress-free holding environment (New Zealand Seafood Industry Council, 2006; Keppler, 2007). Mortality was also monitored daily. The experiment was carried out in two phases (I and II). During Phase I of the experiment (4 weeks), oysters were subjected to temperature treatments (0, 5, 10, 15, 20 °C) at a fixed salinity (27‰). During Phase II (2 weeks), ambient salinity was dropped to 6‰ over a 12-h period in two of the four tanks per temperature regime by regulating freshwater input. In Phase I, four randomly sampled oysters per temperature regime (0, 5, 10, 15, 20 °C; salinity 27‰) were wired to individual sensors and monitored every minute for valve opening (gaping), whereas in Phase II two randomly sampled oysters from each temperature–salinity combination were monitored (0, 5, 10, 15, 20 °C at 6‰; 0, 5, 10, 15, 20 °C at 27‰; Comeau et al., 2012). In Phase I of the experiment, one oyster was randomly sampled from each of the 20 tanks (5 treatments × 4 replicates) every week for osmolality measurements (2 d week−1 or 8 sampling days). A total of 16 oysters (4 oysters tank−1) from each of the five treatment groups (0, 5, 10, 15, 20 °C) were sampled (n = 80 oysters). In Phase II, two oysters were randomly sampled from each of the 20 tanks (10 treatments × 2 replicates) each week for osmolality and NRRA measurements. Logistical challenges associated with the NRRA technique limited the number of oysters that could be analysed each day to 10; measurements were conducted over 4 d week−1 or 8 sampling days. A total of eight oysters (four oysters tank−1) from each of the 10 treatment groups (0, 5, 10, 15, 20 °C, at salinities of 6 or 27‰) were sampled (n = 80 oysters). At sampling time, 3 ml of water was collected from each tank and a subsample of 20 µl was taken for osmolality measurements (± 1 mOsm kg−1 H2O). Simultaneously one oyster was sampled from the same tank; the oyster was immediately opened and 20 μl of mantle fluid was collected using a plastic syringe fitted with a filter (0.2-µm MFS cellulose acetate filter, Ultident Inc., St-Laurent, QC). The difference in osmolality (ΔOSM) was calculated as haemolytic osmolality minus ambient water osmolality. Neutral-red dye (3-amino-7-dimethylamino-2-methylphenazine hydrochloride, F1631, Sigma-Aldrich, St Louis, MO) stock solution was prepared by dissolving 2.28 mg of dye in 1 ml of dimethyl sulfoxide (DMSO, D128 Thermo Fisher Scientific, Waltham, MA). For the standard NRRA, a working solution was prepared daily by dissolving 17 µl of the dye stock solution in 1 ml of saline solution (Schlieper, 1972) adjusted to the salinity of the experimental tank. For NRRAMOD, oyster haemolymph was used in the working solution, instead of the prepared saline solution. Immediately after sampling the mantle fluid for osmolality measurements, 0.2 ml of haemolymph was withdrawn from the pericardial cavity/adductor muscle sinus and transferred to a low-retention Eppendorf tube pre-filled with either 0.2 ml of pre-adjusted saline solution (NRRA) or mantle fluid (prefiltered, as described above; NRRAMOD) and processed according to Méthé et al. (2015). The lysosomal membrane destabilization index (LDI), which represents a count of haemocytes (out of a total of 100) having destabilized lysosomes (i.e. leakage of NR into the cytosol or enlarged lysosomes), was expressed as a percentage. The difference between the two NRRA outcomes (ΔLDI) was calculated as follows: ΔLDI = LDI (NRRA) – LDI (NRRAMOD). A statistical analysis of ΔOSM was carried out for Phase I (temperature effect), whereas in the case of Phase II (temperature–salinity effect) the primary focus was on the haemolymph values rather than the ΔOSM values, which were intentionally out of range. A statistical analysis of ΔLDI was carried out for Phase II. The ΔLDI estimate for each temperature–salinity combination was compared with a ΔLDI value of 0 using the Student's t-test. A linear mixed model with random tank effects was used, due to the hierarchical (nested) structure of the experimental design. The models took into account temperature, salinity (Phase II), day, week and their interactions. In the presence of significant interaction, pairwise comparison tests (with Bonferroni correction) or contrasts were performed. The significance level was set at P ≤ 0.05. Scatterplot of gaping activity (percentage of time for which valves open) of individual oysters held at different temperatures during Phase I (salinity 27‰, 810 mOsmo kg−1; n = 17) and Phase II (▴ spring freshet, salinity 6‰, 180 mOsm kg−1; Δ control, salinity 27‰, 810 mOsmo kg−1; n = 16). Haemolymph fluid vs ambient water osmolality at different temperatures during Phase I (salinity 27‰, 810 mOsmo kg−1; n = 16 samples per bar) and Phase II (salinity 6‰,180 mOsm kg−1; n = 8 samples per bar). Values are expressed as mean ± SE. Scatterplot of gaping activity (percentage of time for which valves open) of individual oysters in relation to haemolymph osmolality during Phase II (2 weeks) (♦ spring freshet, 6‰ salinity, 180 mOsm kg−1; ♢ control, 27‰ salinity, 810 mOsm kg−1). Mean (SE) osmolality of tank water and haemolymph fluid, and osmolality differential (ΔOSM) of oysters (n = 7–8) held at each temperature during Phase II (6‰, 180 mOsmo kg−1; 27‰, 810 mOsmo kg−1) based on pooled data. Mean (SE) osmolality of tank water and haemolymph fluid, and osmolality differential (ΔOSM) of oysters (n = 7–8) held at each temperature during Phase II (6‰, 180 mOsmo kg−1; 27‰, 810 mOsmo kg−1) based on pooled data. Estimates from final statistical model for haemolymph fluid osmolality from oysters held at different temperatures during Phase II (6‰, 180 mOsm kg−1). Abbreviations: Md, median; CI, 95% confidence interval. Values not sharing a common superscript (A, B and/or C) differ significantly (P < 0.05) based on pairwise comparison tests with Bonferroni correction. *P-value for temperature. Estimates from final statistical model for haemolymph fluid osmolality from oysters held at different temperatures during Phase II (6‰, 180 mOsm kg−1). Abbreviations: Md, median; CI, 95% confidence interval. Values not sharing a common superscript (A, B and/or C) differ significantly (P < 0.05) based on pairwise comparison tests with Bonferroni correction. *P-value for temperature. A 95% confidence interval for the difference between the lysosomal membrane destabilization index (ΔLDI, %) in the two NRRA outcomes at each temperature during Phase II (2 weeks) (♦ spring freshet, 6‰ salinity, 180 mOsm kg−1; ♢ control, 27‰ salinity, 810 mOsm kg−1). Asterisk indicates significant difference (P < 0.05) from Y-axis reference line. Scatterplot panels showing difference between the lysosomal membrane destabilization index (ΔLDI, %) of individual oysters (n = 8) in relation to osmolality differential (ΔOSM, mOsm kg−1) at each temperature during Phase II (spring freshet, 6‰ salinity, 180 mOsm kg−1). We would like to thank the staff of the Aquatic Animal Facility of the Atlantic Veterinary College for their commitment to the project. Sincere thanks are especially due to Nellie Gagné for her expertize and André Nadeau for technical assistance. Funding was provided by Fisheries & Oceans Canada (ACRDP MG-11-01-001) and the Elsipogtog First Nations (NB, Canada).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.