Effect of in vitro and in vivo organotin exposures on the immune functions of murray cod (Maccullochella peelii peelii)

Authors

  • Andrew J. Harford,

    1. Key Centre for Toxicology, School of Medical Sciences, RMIT University, P.O. Box 71, Plenty Road, Bundoora, Victoria 3083, Australia
    Search for more papers by this author
  • Kathryn O'Halloran,

    1. Landcare Research, P.O. Box 40, Lincoln 7640, New Zealand
    Search for more papers by this author
  • Paul F. A. Wright

    Corresponding author
    1. Key Centre for Toxicology, School of Medical Sciences, RMIT University, P.O. Box 71, Plenty Road, Bundoora, Victoria 3083, Australia
    • Key Centre for Toxicology, School of Medical Sciences, RMIT University, P.O. Box 71, Plenty Road, Bundoora, Victoria 3083, Australia
    Search for more papers by this author

Abstract

Murray cod (Maccullochella peelii peelii) is an iconic native Australian freshwater fish and an ideal species for ecotoxicological testing of environmental pollutants. The species is indigenous to the Murray-Darling basin, which is the largest river system in Australia but also the ultimate sink for many environmental pollutants. The organotins tributyltin (TBT) and dibutyltin (DBT) are common pollutants of both freshwater and marine environments and are also known for their immunotoxicity in both mammals and aquatic organisms. In this study, TBT and DBT were used as exemplar immunotoxins to assess the efficiency of immune function assays (i.e., mitogen-stimulated lymphoproliferation, phagocytosis in head kidney tissue, and serum lysozyme activity) and to compare the sensitivity of Murray cod to other fish species. The organotins were lethal to Murray cod at concentrations previously reported as sublethal in rainbow trout (i.e., intraperitoneal [i.p.] lethal dose to 75% of the Murray cod [LD75] = 2.5 mg/kg DBT and i.p. lethal dose to 100% of the Murray cod [LD100] = 12.5 mg/kg TBT and DBT). In vivo TBT exposure at 0.1 and 0.5 mg/kg stimulated the phagocytic function of Murray cod (F = 6.89, df = 18, p = 0.004), while the highest concentration of 2.5 mg/kg TBT decreased lymphocyte numbers (F = 7.92, df = 18, p = 0.02) and mitogenesis (F = 3.66, df = 18, p = 0.035). Dibutyltin was the more potent immunosuppressant in Murray cod, causing significant reductions in phagocytic activity (F = 5.34, df = 16, p = 0.013) and lymphocyte numbers (F = 10.63, df = 16, p = 0.001).

INTRODUCTION

Murray cod (Maccullochella peelii peelii) is the largest and best-known Australian freshwater species. Its distribution and abundance has declined in the past 50 years because of the construction of dams, changes to river flows and temperatures, and pollution of their habitat from various sources ([1]; http://www.deh.gov.au/water/basins/murray-cod/). In contrast, Murray cod aquaculture is an emerging industry with a large potential for rapid growth. The industry now produces both fingerlings for the stocking of waterways and table-sized fish (500–800 g) for human consumption ([2]; http://www.abareconomics.com/publications_html/fisheries/fisheries_03/er03_aquaculture.pdf). Murray cod was identified by our group as a native Australian freshwater fish that is well suited to immunoecotoxicology testing of xenobiotics under laboratory conditions. They are readily available from fish farmers, adapt extremely well to indoor tanks, are robust to handling and stress, and feed on a standard commercial pellet diet, and a large amount of immune tissue can be obtained from one fish. Furthermore, they also have a high ecological, economic, recreational, and cultural value. Our research group has successfully adapted functional immune assays for Murray cod (i.e., lysozyme, mitogenesis, and phagocytosis), which have been previously used in mammals and other fish species to assess the immunotoxicity of environmental pollutants [3].

Lysozyme is one of three hydrolyase enzymes that have a defensive role in the circulatory system. In fish, it is found in the blood, mucus, and lymphomyeloid tissue, highlighting its role in fish innate immune systems, which are increasingly important, as their specific immune system is slower and less developed in comparison to mammals [4]. Lysozyme was chosen as an indicator of immunomodulation because of the easy and rapid assay method and its importance in the innate defense of fish. The lymphoproliferative response to mitogens is similar to the adaptive lymphocytic response to antigens that are presented by macrophages; however, it does not require the action of antigen-presenting cells and occurs rapidly in response to natural agents conserved in many foreign organisms, such as bacteria (e.g., lipopolysaccharide) and plants (e.g., phytohemagglutinin [PHA]). The mitogen-stimulated lymphoproliferation assay assesses the function of lymphocytes and has often been used in immunotoxicity testing protocols [5]. Phagocytosis is a primitive defense mechanism, conserved in both vertebrates and invertebrates. It has been used in a tiered system for the immunotoxicological assessment of environmental pollutants and immunostimulants used in aquaculture [6]. Our group has recently reported the use of flow cytometry to measure the phagocytic activity of head kidney cells from three native Australian freshwater fish [7].

Numerous studies from abroad have demonstrated that many aquatic pollutants are immunotoxic in exotic species of fish. However, only limited studies have applied standardized immune functional assays to assess the immunotoxicity of environmental pollutants in native Australian freshwater fish [8–10]. Tributyltin (TBT) and dibutyltin (DBT) contamination of harbors and marinas has been a significant environmental concern because of their disruption of endocrine and reproductive functions and immunotoxic properties at very low concentrations [11]. The majority of environmental organotin pollution is from biocidal antifouling paints that leach TBT to protect ship hulls from algal and mollusk growth. Although an international treaty effectively banned the use of TBT-based marine paints in 2003, organotins are expected to persist in the environment for many years [12]. Furthermore, TBT and DBT also enter both freshwater and marine environments through treated woods, runoff from landfill, sewage, and industrial discharges [13]. Once in the aquatic environment, they are bioaccumulated by invertebrates, fish, and aquatic mammals and can reach extremely high levels in the tissues of these organisms [14,15].

Environmental organotin residues in freshwater ecosystems are reported less frequently than levels in marine ecosystems. Nevertheless, international studies have shown that TBT and DBT contamination of freshwater ecosystems is significant. Residues of the organotins have been reported up to 54 and 220 ng/L in water columns [16], 143 and 520 ng/g in sediments [17,18], and 0.7 and 2.5 mg/kg in fish tissues [17,19] for DBT and TBT, respectively. Very few studies have investigated the contamination of TBT and DBT in Australian waters, especially in recent years. A study of TBT contamination along the Western Australia coast reported high levels of TBT in marine sediment (0.001–1.35 μg/g TBT) and mussel (Mytilus edulis) tissue (0.003–0.32 μg/g TBT), which correlated with areas of high boating activity [20]. In New South Wales, Sydney Harbour had levels of 0.220 μg/L TBT and 0.051 μg/L DBT, and Georges River had lower levels of 0.1 μg/L TBT and 0.04 μg/L DBT [16], while oysters in the Georges River had bioaccumulated levels of 0.019 mg/kg DBT and 0.234 mg/kg TBT [21]. No studies have been reported in the literature concerning organotin contamination of the Murray-Darling basin despite the expectation of some contamination in this freshwater environment.

In mammals, TBT and DBT have been known for their specific reduction of T-lymphocytes through apoptotic mechanisms [22], and their immunotoxic effects in aquatic organisms appear to be similar to those observed in mammals (i.e., thymus atrophy and a reduction in lymphocyte numbers [13]). Numerous studies have reported TBT and DBT as immunotoxins in fish, and our group has previously reported organotin-induced immunotoxicity in rainbow trout and in the Australian freshwater native silver perch (Bidyanus bidyanus) [23]. In vitro studies using isolated spleen and head kidney cells from juvenile rainbow trout showed that TBT and DBT suppressed mitogen-stimulated lymphoproliferation at concentrations ⩾ 50 μg/L. The results also indicated that DBT was the more potent immunosuppressant [23]. Furthermore, extensive in vitro studies have reported that TBT can either suppress or increase the phagocytic activity (i.e., chemiluminescence) of oyster toadfish (Opsanus tau) macrophages, depending on the dose given [24,25].

The immunotoxicity of in vivo TBT exposure in channel catfish (Ictalurus punctatus) and rainbow trout have been previously reported. Catfish exposed to 1.0 mg/kg TBT displayed decreased natural killer cell activity (at 3 and 7 d) and suppressed phagocyte oxidative burst (3 d after exposure) [26]. Hematological examination of the fish showed an increase in circulating monocytes and neutrophils but a decrease in hematocrit values and lymphocyte numbers. Additionally, humoral responses were suppressed in all groups exposed to 0.01, 0.1, or 1.0 mg/kg TBT [26]. In vivo studies using rainbow trout bathed in 4.0 μg/L TBT for 28 d have reported a doserelated lymphocytic depletion, a marked proliferation of reticuloendothelial cells, an increased erythrophagia in the spleen, severe lesions within epithelia of the gills, and pseudobranch epithelial cells [14]. Extensive field experiments involving a number of different chemicals linked TBT as the causative agent of an increased prevalence of lymphocystis virus infections in the flounder (Platichthys flesus) along the Dutch coastline [27].

Preliminary immunotoxicity studies by our group demonstrated that in vitro exposures to organotins resulted in a reduction of lymphocyte subpopulations in three large native Australian freshwater fish: silver perch, golden perch (Macquaria ambigua), and Murray cod. However, in the crimson-spotted rainbowfish (Melanotaenia fluviatilis), granulocytes were the most sensitive subpopulation. The phagocytic activity of head kidney granulocytes was suppressed in golden perch, Murray cod, and rainbowfish, and of the native fish species tested, silver perch head kidney cells were the least sensitive to in vitro exposures of TBT and DBT [9]. In the present study, TBT and DBT were used as exemplar immunotoxins to assess the validity of immune function assays (i.e., mitogen-stimulated lymphoproliferation and phagocytosis in head kidney tissue and serum lysozyme activity) and to compare the sensitivity of Murray cod to other fish species.

MATERIALS AND METHODS

Wet-laboratory water

All fish were held in the wet-laboratory facilities at RMIT University (Bundoora, Melbourne, Victoria, Australia). The laboratory was a flow-through design that was supplied with carbon-filtered water, preheated to 20 ± 1°C, which is ideal for Murray cod aquaculture. The water had an oxygen concentration of 7.7 to 8.0 mg/L, a pH of 6.8 to 7.2, and a conductivity of 100 to 120 μs.

Fish procurement, maintenance, and surgery

Murray cod (150–250 g) were purchased from Australian Aquaculture Products (Euroa, Victoria, Australia) and transported to the laboratory (2 h) in an aerated 1,000-L transportable tank on the back of a utility vehicle. Individual fish were transferred to 40-L tanks receiving aeration and a constant supply (5 L/h) of fresh carbon-filtered aquaria water at 20 ± 1°C. They were fed a commercially available pellet every second day (Skretting Australia, Cambridge, Tasmania, Australia), and tanks were cleaned on the days between feeding. Individual fish were acclimatized in the glass aquaria for four weeks before conducting experiments.

Prior to surgery, Murray cod were anesthetized by bathing for ⩽3 min in MS 222 (i.e., 3-aminobenzoic acid ethyl ester methanesulfonate salt, 150–250 mg/L; Sigma Chemical, St. Louis, MO, USA) in carbonate-buffered aquaria water. The fish were then killed by a blow to the skull and quickly bled out, and the head kidney and spleen were removed and weighed.

Cell and serum isolation

Fish were bled via the caudal vein using a 26-gauge, 6.35-mm needle with a 1- or 3-ml heparinized syringe (CSL, Melbourne, Victoria, Australia). To obtain serum, the blood was placed in microfuge tubes and centrifuged at 800 g for 15 min at 4°C. The head kidney cells were isolated using the method previously described by O'Halloran et al. [23]. Briefly, cells were disrupted from the head kidney tissue and passed through a 250-μm nylon mesh. Red blood cells were separated from the cell suspension by density gradient centrifugation, that is, 4 ml of cell suspension layered over 3 ml of histopaque (Sigma) and centrifuged at 400 g for 30 min at 22°C. The immune cells collected from above the histopaque layer were washed twice, counted microscopically (using a haemocytometer in the presence of trypan blue), and diluted to the desired cell concentration with tissue culture media consisting of RPMI 1640 with 20 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), 300 mg/L glutamine, and 100 μg/ml gentamycin sulfate (Sigma) supplemented with 10% fetal calf serum (CSL).

Exposures

In vitro exposures. The TBT and DBT were purchased from Sigma, and stock solutions were dissolved in dimethyl sulfoxide (DMSO; BDH Chemicals, Poole, UK). Aliquots of the pesticides were added to the cultures at the commencement of each assay at a final concentration of 500, 50, and 2.5 μg/L for TBT and DBT. The final DMSO concentration in each assay was 0.25% v/v.

In vivo exposures. Fresh stock solutions of the organotins were diluted in polyethylene glycol (Unilab, Sydney, New South Wales, Australia) and were kept at 4°C for no longer than one week. Fish were exposed with 2 ml/kg, and the final concentrations were 0.1, 0.5, 2.5, and 12.5 mg/kg for both TBT and DBT. Four fish were exposed to each organotin concentration (except where otherwise stated), and seven fish were exposed to the same volume of polyethylene glycol (solvent controls). The fish were administered a single i.p. injection, and their immune tissues were sampled 14 d later.

Lysozyme

Lysozyme activity was measured using the method previously described by Ellis [28]. Quadruplicate samples of Murray cod serum (25 μl) were added to the wells of a flat-bottom 96-well plate with a solution of Micrococcus lysodeikticus (0.66 mg/ml final concentration; Sigma) in a final volume of 200 μl. The plates were acclimatized to room temperature for 2 min, and then the absorbance in each well was measured at 450 nm for 10 min on a DIAS plate reader (Dynatech Laboratories, Chantilly, VA, USA). The data were collected using the Biolinx assay management software (Ver 2.1; Dynatech Laboratories) and saved on file for future analysis. Lysozyme activity was calculated from a hen egg white lysozyme standard curve (0–20 μg/ml; Sigma), and results were expressed as a percentage of the untreated control.

Mitogenesis

Standardized protocols for the Murray cod mitogenesis assay are described in Harford et al. [3]. Briefly, Murray cod head kidney cells (5.0 × 105 cells/ml) were incubated (in triplicate with a 96-well plate) with 0, 2.5, 5, and 10 μg/ml PHA (Sigma) in a final volume of 200 μl of HEPES-free tissue culture media. Cells were cultured for 5 d at 20°C in an atmosphere of 5% CO2 and 95% air and were pulsed with tritiated thymidine (18.5 KBq/well; Amersham International, Amersham, UK) 48 h before harvesting. The cells in each well were harvested onto fiberglass discs using a semiautomatic cell harvester (Skatron, Lier, Norway), and the radioactivity (i.e., counts per minute [cpm]) of each disc was analyzed in a scintillation counter (LKB Instruments, Wallac, Turku, Finland). Proliferation index was calculated as peak cpm (2.5 μg/ml of the PHA) per basal cpm (without mitogen) and then expressed as a percentage of the proliferation index untreated control.

Phagocytosis

Standardized protocols for the Murray cod phagocytosis assay are described in Harford et al. [3]. Briefly, all samples were tested in triplicate, and the final concentration of head kidney cells in each assay was 1 × 106 cells/ml. Fluorescent (fluorescein isothiocyanate [FITC]) latex beads (1 μm; Polysciences, Warrington, PA, USA) were added to the cell suspension at a cell-to-bead ratio of 1:25. Identical negative control tubes were prepared for all tests in which the beads were not added until immediately before flow cytometric analysis.

Cells were cultured for 48 h at 20°C in an atmosphere of 5% CO2 and 95% air (Linde Gas, Sydney, New South Wales, Australia) and tubes placed on ice at the end of the incubation period, prior to flow cytometry. Just before analysis (2–5 min), propidium iodide (at 1-μg/ml final concentration; Sigma) was added to each tube to measure the number of unviable cells. The flow cytometer (EPICS Elite II, Beckman-Coulter, Fullerton, CA, USA) was used to collect 10,000 events for each sample, and counting gates were set up in regions representing granulocytes, lymphocytes, and debris. Unengulfed beads were excluded from the counts by the “live-gating” option available in the Expo32 software (Beckman-Coulter), and data were saved as listmode files for future analysis. The percentage of granulocytes emitting fluorescence (i.e., FITC-positive granulocytes) and mean voltage output of FITC-positive events within the granulocytes (i.e., the intensity of the fluorescent signal) represented the number of granulocytes engaged in phagocytosis and the number of beads engulfed per granulocyte, respectively. The number of events in each gated region were expressed as a percentage of untreated control values.

Statistics

Statistics were performed using the computer package SPSS 11.0 (SPSS, Chicago, IL, USA). Analyses were performed on raw data, which were first analyzed for linear relationships and checked for normal distribution using Levene's test. Data were arcsine transformed if the distribution was not normal prior to analysis of variance (ANOVA). A one-way ANOVA was performed, and Tukey's compromise post hoc test was conducted to determine homogeneous subsets. A p value of less than 0.05 was considered to indicate a statistically significant difference.

RESULTS

In vitro exposures

Preincubating TBT and DBT in Murray cod serum before the addition of M. lysodeikticus solution had no effect on the activity of lysozyme within the serum. This demonstrates that TBT and DBT do not directly inhibit the lysozyme enzyme (Table 1).

At all the TBT concentrations tested, there was a reduction in both the basal and the PHA-stimulated mitogenesis of Murray cod head kidney lymphocytes (Table 1). Exposure to the organotins at 500 μg/L resulted in a significant reduction in both basal and PHA-stimulated incorporation of tritiated thymidine (F = 4.74, df = 15, p = 0.021 and F = 5.40, df = 15, p = 0.014, respectively). However, there was no significant difference in the proliferation index because of the large variation between fish. Exposure to 500 μg/L DBT caused a significant reduction in basal and PHA-stimulated incorporation of tritiated thymidine (F = 3.53, df = 15, p = 0.049 and F = 4.31, df = 15, p = 0.028, respectively), which was similar to the reduction caused by TBT. Both TBT and DBT reduced the proliferation index of Murray cod head kidney lymphocytes by approximately 30 to 40% at the highest concentration.

Table Table 1.. The effect of in vitro organotin exposure on the activity of Murray cod head kidney lymphocyte mitogenesis and serum lysozyme
 Organotin concn. (μg/L)
 02.550500
  1. a Tributyltin.

  2. b Data are expressed as a percentage of control samples, and values represent mean ± standard error of four fish. Control value for lysozyme was 10.09 ± 1.49 μg/ml hen egg white lysozyme equivalent.

  3. c Data are expressed as percentage of control samples, and values represent mean ± standard error of four fish. Control values were 1.29 ± 0.11 for proliferation index. cpm = counts per minute. * denotes significantly different from the solvent control (p < 0.05).

  4. d Dibutyltin.

TBTa    
  Lysozymeb104 ± 5%102 ± 5.0%102 ± 5%100 ± 4%
  Proliferation indexc103.6 ± 5.980.8 ± 16.075.9 ± 8.957.3 ± 14.3
  Peak cpm1,149 ± 279999 ± 225748 ± 18984 ± 23*
DBTd    
  Lysozymeb104 ± 5%104 ± 5%108 ± 8%105 ± 5%
  Proliferation indexc103.6 ± 5.9100.1 ± 3.6115.3 ± 3.369.0 ± 11.9*
  Peak cpm1,149 ± 2791,100 ± 2691,642 ± 465136 ± 40*

In vivo exposures

Body and organ weights. All Murray cod survived exposure to ⩽0.5 mg/kg TBT and DBT. At 2.5 mg/kg DBT, there was a 75% mortality incidence, whereas the equivalent dose of TBT did not affect survival. All fish exposed to 12.5 mg/kg TBT or DBT died (Table 2). Additional fish were exposed to 12.5 mg/kg TBT and 2.5 mg/kg DBT to offset the lack of tissue available for subsequent immunoassays. No significant difference was observed in the body weight change over the study period in Murray cod that were exposed to TBT and DBT compared to controls. In vivo organotin exposure did not significantly change the somatic indices of Murray cod spleens or head kidneys (Table 2). Murray cod exposed to 2.5 mg/kg TBT had significantly lower splenic cell yields (F = 7.22, df = 16, p = 0.004), and fish exposed to 0.1 and 0.5 mg/kg TBT had significantly lower head kidney cell yields compared to the solvent control (F = 5.41, df = 16, p = 0.012). Exposure to DBT had no significant effect on the cell yield of either the head kidney or the spleen.

Table Table 2.. The effect of in vivo organotin exposure on Murray cod body and immune organ weights and lysozyme activitya
 Organotin concn. (mg/kg)
 00.10.52.512.5
  1. a* denotes significantly different from the solvent control (p < 0.05). Values represent the mean ± standard error of four to seven fish (except DBT 2.5 mg/kg, n = 2). BW = body weight; SI = somatic index; HK = head kidney; ND = parameters not determined due to 100% mortality.

  2. b Tributyltin.

  3. c Dibutyltin.

  4. d Cells/g of tissue.

  5. e Data are expressed as percentage of untreated control lysozyme activity (n = 7), calculated from a hen egg white lysozyme standard curve (i.e., 7.12 ± 1.02 μg/ml).

Mortality     
TBTb0/7 fish, 0%0/4 fish, 0%0/4 fish, 0%0/4 fish, 0%6/6 fish, 100%
DBTc 0/4 fish, 0%0/4 fish, 0%6/8 fish, 75%1/1 fish, 100%
BW change (%)     
TBT-1.34 ± 1.471.38 ± 2.03.67 ± 3.11.56 ± 2.2ND
DBT -0.67 ± 3.3-4.67 ± 3.14.83 ± 0.2ND
Spleen SI     
TBT0.055 ± 0.0080.069 ± 0.0120.052 ± 0.0080.048 ± 0.005ND
DBT 0.057 ± 0.0100.064 ± 0.0100.037 ± 0.008ND
HK SI     
TBT0.226 ± 0.0240.284 ± 0.0250.207 ± 0.0190.227 ± 0.013ND
DBT 0.233 ± 0.0210.230 ± 0.0310.195 ± 0.023ND
Spleen cell yieldd     
TBT800 ± 109812 ± 1181,173 ± 145278 ± 148*ND
DBT 1,787 ± 5441,549 ± 407413 ± 207ND
HK cell yieldd     
TBT688 ± 145292 ± 24*310 ± 72*561 ± 108ND
DBT 535 ± 52517 ± 140754 ± 333ND
Lysozymee     
TBT94.0 ± 9.4129.8 ± 11.6100.0 ± 9.068.0 ± 20.7ND
DBT 136.5 ± 18.287.7 ± 16.562.5 ± 9.4ND
Figure Fig. 1..

The effect of in vivo tributyltin (TBT) (a and b) and dibutyltin (DBT) (c and d) exposure on the lymphoproliferation profile (a and c) and proliferative index (b and d) of Murray cod head kidney cells. Proliferation is expressed as a percentage of untreated control (i.e., 1.46 ± 0.12, n = 8), and values represent the mean ± standard error of four to seven fish. * denotes significantly different from the solvent control (p < 0.05).

Lysozyme

Although an apparent increase in serum lysozyme activity was found in Murray Cod exposed to 0.1 mg/kg organotins, there was no statistically significant difference from the control group (Table 2).

Mitogenesis

Murray cod head kidney cells from fish exposed in vivo to 2.5 mg/kg TBT had a significantly lower proliferation index compared to the solvent control (F = 3.66, df = 18, p = 0.035; Fig. 1). However, this occurred as a result of an increase in basal cpm rather than a depression of peak cpm. Fish exposed to 0.5 mg/kg TBT also had a slightly higher basal cpm, resulting in a lower proliferation index, while those exposed to 0.1 mg/kg TBT had a lower peak cpm. Exposure to 0.1 and 0.5 mg/kg DBT did not result in any significant changes in the proliferation index or the basal and PHA-stimulated incorporation of tritiated thymidine into Murray cod head kidney lymphocytes. Variation was observed in the 0.5-mg/kg DBT group that was caused by one fish with a proliferation index 171% of the control, while the other three fish had proliferation indices of 77 to 88% of the control. The two fish that survived the 2.5-mg/kg DBT exposure had very low proliferation indices, as their basal cpm was higher than their PHA-stimulated cpm (Fig. 1).

Phagocytosis

In vivo exposure to TBT resulted in some changes in the phagocytic activity and leucocyte populations of Murray cod head kidney cells (Fig. 2). Fish exposed to 0.5 mg/kg TBT had a statistically significant increase in the percentage of gran ulocytes that had engulfed beads (i.e., FITC-positive granulocytes) compared to the control group (F = 6.89, df = 18, p = 0.004; Fig. 2a). A significant rise was observed in the granularity (i.e., side-scatter increase) of Murray cod head kidney granulocytes in fish exposed to 0.1 and 0.5 mg/kg TBT (F = 7.80, df = 18, p = 0.001; Fig. 2b). The increase in granularity is not likely due to engulfed beads, as the increase in side scatter was not accompanied by a concomitant increase in mean voltage of the granulocytes. Both side scatter and the number of FITC-positive granulocytes with beads returned to control levels in fish exposed to 2.5 mg/kg TBT. The Murray cod that were exposed to 2.5 mg/kg TBT displayed a significant reduction in the number of head kidney lymphocytes compared to controls (F = 7.92, df = 18, p = 0.002; Fig. 2c). As the data comprise a proportional count of 10,000 events from the head kidney, there was a concomitant increase in the proportion of granulocytes in the sample analyzed. These results suggest that head kidney lymphocytes are more sensitive to TBT than are the granulocytes. No significant change was observed in the number of events in the debris-gated region, which reflects a uniform cell isolation technique between the sample groups (Fig. 2d). In addition, there was no significant change in the number of unviable cells, although the number of nonviable cells was reduced in the 0.1 and 0.5 mg/kg TBT compared to that found in control groups.

Figure Fig. 2..

Murray cod head kidney phagocytic function and subpopulation counts following in vivo exposure to tributyltin (TBT). (a) Phagocytosis (i.e., fluorescein isothiocyanate [FITC]-positive cells), (b) number of beads per cell (i.e., mean voltage and side-scatter [SS] increase), (c) granulocytes and lymphocyte subpopulation counts, (d) debris counts and nonviable cells (propidium iodide [PI] positive). Data are expressed as a percentage of untreated control samples (n = 8), and values represent mean ± standard error of four to seven fish. Untreated control values were 10 ± 1% for FITC-positive total, 25 ± 4% for FITC-positive granulocytes, 130 ± 27 V for mean voltage, 123 ± 14 V for SS increase, 4,050 ± 200 events for granulocytes, 3,350 ± 200 events for lymphocytes, 1,600 ± 150 for debris, and 800 ± 150 for PI positive. * denotes significantly different from the solvent control (p < 0.05).

Figure Fig. 3..

Murray cod head kidney phagocytic function and subpopulation counts following in vivo exposure to dibutyltin (DBT). (a) Phagocytosis (fluorescein isothiocyanate [FITC]-positive cells), (b) number of beads per cell (i.e., mean voltage and side-scatter [SS] increase), (c) granulocytes and lymphocyte subpopulation counts, (d) debris counts and nonviable cells (propidium iodide [PI] positive). Data is expressed as percentage of untreated control samples (n = 8), and values represent mean ± standard error of four to seven fish (except DBT 2.5 mg/kg, n = 2). Untreated control values were 10 ± 1% for FITC-positive total, 25 ± 4% for FITC-positive granulocytes, 130 ± 27 V for mean voltage, 123 ± 14 V for SS increase, 4,050 ± 200 events for granulocytes, 3,350 ± 200 events for lymphocytes, 1,600 ± 150 events for debris, and 800 ± 150 for PI positive. * denotes significantly different from the solvent control (p < 0.05).

Dibutyltin showed clear signs of immunotoxic activity in the head kidney cells of exposed Murray cod (Fig. 3). In vivo DBT exposure resulted in a dose-dependent decrease in the number of granulocytes with beads, which reached statistical significance at 2.5 mg/kg DBT (F = 5.34, df = 16, p = 0.013; Fig. 3a). In addition, there was a significantly increased number of beads engulfed within each granulocyte from fish exposed to 2.5 mg/kg DBT (as shown by the increase in mean voltage in FITC-positive events, F = 3.48, df = 16, p = 0.047; Fig. 3b). A significant reduction was observed in the number of lymphocytes (F = 10.63, df = 16, p = 0.001) and a concomitant significant increase in granulocytes (F = 10.13, df = 16, p = 0.001), which reached statistical significance at 0.5 mg/kg DBT (Fig. 3c). No significant difference was observed in the number of propidium iodide-positive events (i.e., nonviable cells), but there were significantly lower numbers of debris-gated events in the highest-dosed group (F = 12.94, df = 16, p = 0.000). This result reflects the fact that there were more granulocytes in those samples, which are more robust and survive the incubation period better than lymphocytes (Fig. 3d).

DISCUSSION

Justification for in vivo exposures

The primary aim of the present study was to use the organotins as exemplar immunotoxicants to validate the immune function assays. Therefore, it was desirable to choose a route of exposure where animals were exposed to a controlled and known concentration. Intraperitoneal injection has been chosen as the route of administration by many aquatic toxicologists and immunotoxicologists [26,29,30] because it ensures that fish receive the desired dose and avoids variation in uptake and accumulation that is encountered when attempting to administer doses via the water or diet. By reducing the variation in the exposure received by each fish, the number of animals required in an experiment is also reduced. Additionally, lower amounts of chemical may be used overall, and the toxicants are metabolized to some degree before excretion, which reduces both the hazard of handling the chemicals and the contamination of wastewater.

Because of their lipophilic nature, organotins tend to bioaccumulate in the tissues of aquatic organisms. Bioconcentration factors have been reported in the range of 400 to 11,000, although no figures are available for Murray cod. Based on bioaccumulation studies in rainbow trout [14,31], the i.p. in vivo exposure of 0.1, 0.5, and 2.5 mg/kg used in this study would represent environmentally relevant waterborne TBT concentrations of 0.04, 0.2, and 1 μg/L, respectively (employing a bioconcentration factor of 2,500). Moreover, the dose range used in this study reflects TBT and DBT residues detected in the tissues of fish and reported in the literature [17]. The same doses and route of exposure were used in our previous studies investigating the immunotoxicity of TBT in rainbow trout, thus enabling a direct comparison to be made between exotic and native species.

The present study used the organotins to test the ability of Murray cod immune function assays to identify immunotoxicity. The study found that Murray cod head kidney cells are sensitive to in vitro and in vivo exposures of both TBT and DBT. The lymphocyte population was the main target of organotin exposure; however, granulocytes and their phagocytic function were also affected.

Murray cod serum lysozyme was unaffected by in vitro exposures to the organotins, but the lymphoproliferation of head kidney lymphocytes was significantly reduced by both TBT and DBT with in vitro exposures of 500 μg/L. Previous reports from our group have shown that in vitro doses of TBT and DBT reduced the PHA-stimulated proliferation of head kidney lymphocytes isolated from both rainbow trout and silver perch [10,23].

Murray cod that were dosed with 12.5 mg/kg TBT all died within 96 h of injection, and the one fish dosed with 12.5 mg/kg DBT died within 120 h. In addition, six of eight fish dosed 2.5 mg/kg DBT did not survive the 14-d study period, surviving an average of 9.4 d. This result is contrary to the majority of findings in other species, including the rainbowfish and rainbow trout studies conducted in our laboratory, that have shown TBT to be more acutely toxic than DBT [32,33]. Dibutyltin's greater toxicity may be attributed to a difference in the molecular weight of the compounds and the subsequent higher molar concentrations; that is, 12.5 mg/kg is equal to 52.5 and 63.1 mmol/kg for TBT and DBT, respectively. The highest sublethal dose previously used by our group to study organotin immunotoxicity in rainbow trout was 12.5 mg/kg. The results from this study demonstrate that Murray cod is more sensitive to acute organotin exposure compared to rainbow trout, which is a commonly used test species in ecotox-icological risk assessments. The increased sensitivity of native species implies that risk assessments based on nonnative species may not provide an adequate level of protection for Australian freshwater ecosystems.

Thymus atrophy and reduced numbers of lymphocytes have been a commonly reported effect of TBT and DBT in both fish and mammals [13,26,34]. The present study found that, although there was only an insignificant reduction in somatic indices of immune organs, there was a depleted immunocyte population within these organs. It also appears that the function of lymphocytes may be altered by in vivo exposure to TBT, whereas DBT had no effect on lymphoproliferation with in vivo exposures up to 0.5 mg/kg. The highest concentration of TBT significantly reduced the proliferation index of Murray cod lymphocytes; however, this was due not to the reduction of peak cpm but to a rise in basal cpm. The reason for this pattern is unclear; however, these results indicate that in vivo TBT exposure is disrupting the profile of mitogen-stimulated proliferative responses of Murray cod head kidney lymphocytes. Although the biological significance of the results of the in vivo experiment is somewhat equivocal, the assay showed excellent promise in the in vitro exposure experiments.

The flow cytometric analysis of head kidney cell subpopulations and the function of their granulocytes showed much greater sensitivity. Most striking of the results was the effect of in vitro and in vivo DBT exposure on subpopulation counts and granulocyte phagocytosis of FITC beads. A clear significant reduction was observed in the number of lymphocytes and a concomitant significant increase in the number of granulocytes. This effect was an expected result from in vivo DBT exposure, as the literature agrees that DBT is a potent thymolytic compound that reduces the number of T-lymphocytes in test animals [34]. Previous experiments have also found that DBT is more effective at killing lymphocytes than TBT [23,34]. In this in vivo study, TBT also reduced the number of lymphocytes but only at the highest exposure of 2.5 mg/kg, which indicates that DBT is the more potent immunosup-pressant.

In addition to DBT's effect on the subpopulation counts in Murray cod head kidneys, it appears to inhibit the ability of granulocytes to phagocytose FITC-latex beads. Conversely, in vivo TBT exposure resulted in a significant rise in the phagocytic activity of Murray cod granulocytes. Fish exposed to 0.5 mg/kg TBT displayed a significant increase in the percentage of active granulocytes and their side-scatter characteristics. This indicates not only that there were more granulocytes involved in engulfing beads but also that the granulocyte population as a whole had more particulate matter in their cytoplasm. More beads internalized would cause this effect; however, this may indicate that in vivo exposure to TBT changed the morphology of the granulocytes, perhaps by activating them, which could make the cells more granular. The fact that the mean voltage did not rise significantly suggests that the latter scenario is more likely. Other in vitro and in vivo studies in both fish and rats have reported similar biphasic response to TBT exposure. Specifically, high concentrations of TBT suppressed phagocytic function, while lower concentrations enhanced the activity. Rice and Weeks [24] reported that in vitro exposures of TBT increased the phagocytic functions (i.e., chemiluminescence) of macrophages from the toadfish before causing immunosuppression at higher doses. In a series of in vitro and in vivo studies, these researchers found that TBT stimulated chemiluminescence in fish macrophages by inducing calcium influx into the cell, resulting in a cascade of events. These events cause a temporary rise in reactive oxygen species that is observed as an increase in chemiluminescence when fish are exposed to low concentrations of TBT [35]. Long-term in vivo studies in rats also reported that TBT at high concentrations suppressed macrophage function, while lower concentrations increased phagocytosis [36]. Further support for a biphasic response to TBT is provided by the data from both the in vitro and the in vivo exposure studies, which concurred for the flow cytometry assay. Specifically, in vitro exposure of 50 μg/L TBT caused a slight rise in the number of granulocytes with beads followed by suppression at 500 μg/L TBT, while in vivo TBT exposure caused a significant rise in the number of granulocytes with beads at 0.5 mg/kg TBT followed by a return to control levels at 2.5 mg/kg TBT. Moreover, the in vitro and in vivo DBT exposure data also concurred, and both studies displayed a dose-dependent decrease in the number of granulocytes with beads.

In conclusion, these adapted assays, along with the general somatic index and cell yield data, provided a clear indication of the immunotoxic properties of TBT and DBT in Murray cod. More information is required to determine if environmental levels of organotins would reach sufficient concentrations in the Murray-Darling basin to cause immunotoxicity in Murray cod. This study found a no-observed-effect concentration of 0.1 mg/kg for DBT but could not determine a noobserved-effect concentration for TBT because of a significant increase in head kidney cell granularity (i.e., side-scatter increase) at the lowest concentration of 0.1 mg/kg. This result suggests that waterborne concentrations of TBT as low as 0.04 μg/L (based on a bioconcentration factor of 2,500) may have an impact on the immune functions of Murray cod, and while environmental levels within the basin are unknown, studies in the Georges River have reported TBT levels up to 0.1 μg/L. Furthermore, the organotins have a propensity to bioaccumulate, and further studies are warranted to determine the environmental levels of organotins within the Murray-Darling basin as well as their bioaccumulation and chronic immunotoxicity.

To date, there has been no ecotoxicological data reported on Murray cod, which is surprising considering the importance of these species in the economy and ecology of the Murray-Darling basin. This study has shown that Murray cod are more sensitive to some anthropogenic contaminants than rainbow trout, which is commonly used in the setting of Australia's water quality guidelines. Such a finding emphasizes the recommendations of the Australian water quality guidelines ([37]; http://www.deh.gov.au/water/quality/nwqms/index.html), specifically the need for the testing of chemicals using fish native to the ecosystems that are being polluted. This study has shown that Murray cod are an excellent species for ecotoxicological testing and that assays adapted from exotic fish species may be applied with Murray cod.

Acknowledgements

This research was funded by an Australian Research Council Large Grant (19803567) awarded to P.F.A. Wright.

Ancillary