- Email: [email protected]
Metabolic Responses of Fish Following Exposure to Two Different Oil Spill Remediation Techniques
Ecotoxicology and Environmental Safety 48, 306}310 (2001) Environmental Research, Section B doi:10.1006/eesa.2000.2020, available online at http://www.idealibrary.com on
Metabolic Responses of Fish Following Exposure to Two Different Oil Spill Remediation Techniques Adam Cohen,* Dayanthi Nugegoda,* and Marthe Monique Gagnon*Department of Applied Biology and Biotechnology, RMIT, City Campus, GPO Box 2476V, Melbourne, Victoria 3001, Australia; and -School of Environmental Biology, Curtin University of Technology, Perth, Western Australia 6845, Australia Received July 24, 2000
To assess the impacts of two oil spill remediation techniques on 5sh metabolism, change in aerobic and anaerobic enzyme activities in juvenile Australian Bass, Macquaria novemaculeata, was examined. Changes in cytochrome C oxidase (CCO) and lactate dehydrogenase (LDH) activities were investigated following exposure to the crude oil water accommodated fraction (WAF) and chemically dispersed crude oil WAF. There was a signi5cant stimulation in CCO activity in the gills and livers of 5sh exposed to the WAF of Bass Strait crude oil and chemically dispersed crude oil, compared to the control treatment. In addition, LDH activity was signi5cantly stimulated in the liver of 5sh exposed to dispersed crude oil WAF, compared to the crude oil WAF. Fish exposed to the dispersed crude oil WAF treatment had signi5cantly higher oxygen consumption, as measured by oxygen depletion in a sealed chamber, than 5sh exposed to the crude oil WAF and control treatments. 2001 Academic Press Key Words: metabolic enzymes; cytochrome C oxidase; lactate dehydrogenase; crude oil; dispersed crude oil; Corexit 9527; biomarkers; oxygen consumption.
INTRODUCTION
Oil spills are caused by tankers running aground, collisions between vessels, accidents on exploration/production platforms, or at a re"nery or port area (Australian Institute of Petroleum, 1996). Spill events contribute to 12% of the total inputs of petroleum into the marine environment. O!shore oil slicks are often left to naturally disperse, while a variety of remediation techniques are available when the oil slicks threaten shorelines or valuable natural resources. Oil spill remediation options include mechanical removal, dispersing, burning, and leaving the oil to naturally degrade. Each technique results in di!erent compositions and concentrations of hydrocarbons entering the water column. Chemically dispersing crude oil places up to "ve times more total petroleum hydrocarbons (TPH) into the water column, compared to leaving the oil to naturally disperse (Fucik, 1994).
Aquatic organisms exposed to compounds such as oil may exhibit biochemical, physiological, and/or behavioral responses. In "eld studies, biomarkers of e!ects are preferred to biomarkers of exposure, because individual biological response can be extrapolated to impacts at a population level (Depledge, 1993). There has been limited research utilizing physiological parameters such as oxygen consumption, in complement with biochemical monitoring. The simultaneous measurement of both biochemical and physiological parameters could improve understanding of the impact of xenobiotics on organisms. Signi"cant increases in aerobic and decreases in anaerobic respiration have been reported following exposure to hydrocarbons (Ravindran, 1988). A key enzyme in aerobic metabolism is cytochrome C oxidase (CCO), the terminal step in the electron transport system. CCO is a mitochondrial enzyme that correlates well with actual oxygen consumption rates for di!erent tissues (Simon and Robin, 1971). Changes in carbohydrate metabolism have been reported in "sh chronically exposed to petroleum hydrocarbons (Dey et al., 1983). Lactate dehydrogenase (LDH) is found in the cellular cytoplasm and is active during glycolysis, converting pyruvate from glucose to lactic acid (Knox et al., 1994). Lactic acid is produced in "sh in larger amounts during periods of stress, exercise, and/or hypoxia (Hoar and Randall, 1970) and the LDH enzyme has been reported to increase with changes in growth rates (Pelletier et al., 1994) and metabolism (Lind, 1992). In accordance with di!erential cellular metabolism, Anderson et al. (1974) and Thomas and Rice (1975) reported increases in oxygen consumption following exposure to hydrocarbons. The aim of the study was to investigate the impact of Bass Strait crude oil and dispersed crude oil on aerobic and anaerobic enzyme activities, as a measure of "sh health. A secondary goal was to examine oxygen consumption as a physiological parameter following exposure. Changes in CCO and LDH enzyme activity were measured in the gills, liver, and muscle of Australian bass following exposure to
306 0147-6513/01 $35.00 Copyright 2001 by Academic Press All rights of reproduction in any form reserved.
METABOLIC ENZYMES IN FISH AFTER EXPOSURE TO OIL
the water accommodated fraction (WAF) of crude oil and chemically dispersed crude oil. Change in oxygen consumption in live organisms was investigated by measuring oxygen depletion from within a sealed chamber.
307
were monitored daily throughout the experiment (mean$SE, n"24). Salinity remained constant at 35$0.2 ppt, pH 8.3$0.3, conductivity at 52.6 ls/cm, dissolved oxygen at 6.4$0.2 mg/liter, and temperature at 18.3$0.63C.
MATERIALS AND METHODS
Experimental Organisms
Hydrocarbon Analyses
Eight-week-old Australian bass, Macquaria novemaculeata, were supplied by Searle Aquaculture, New South Wales, Australia, and reared in a #ow-through seawater system at the Queenscli! Marine Station, Victoria, Australia, for 4 months. M. novemaculeata were fed salmon starter pellets ad libitum. Prior to the beginning of the experiment, "sh were randomly distributed at 15 "sh per 40-liter tank and acclimated for a further 2 weeks. The seawater #owthrough rate was maintained at 150 ml/min, which ensured a 90% replacement every 4 h (Sprague, 1973). Fish weight prior to experimentation was 6.5$0.6 g (mean$SE, n"10). The seawater source was the same for rearing and experimentation.
Seawater hydrocarbon concentrations were analyzed by extracting 250 ml of stock solution twice with 25 ml dichloromethane. Sodium sulfate was added to the extracts to remove residual water. Extracts were stored at !203C until analysis. Total petroleum hydrocarbons (TPH) were analyzed by gas chromatography (GC-FID). Reagent blanks of dichloromethane and spike recoveries were also performed. TPH concentrations were analyzed in the 100% preparations rather than the experimental tanks, because the experimental concentrations were expected to be too low for accurate measurement using GC-FID.
Preparation of Stock Solutions Crude oil WAF was prepared by "lling a 20-liter glass mixing chamber with 15 liters of seawater and creating a vortex using a magnetic stirrer. Bass Strait crude oil was then added at a ratio of 1:9 (oil:seawater). The solution was capped, mixed for 20 h, and allowed to settle for 1 h to separate water and oil phases. The solution underlying the oil phases was isolated via a tap at the bottom of the bottle (Anderson et al., 1974). This solution represented the crude oil WAF. The dispersed oil WAF was prepared using the same methodology, except dispersant (Corexit 9527) was added to the #oating oil during mixing, at a ratio of 1:30 (dispersant:oil) (Gilbert, 1996). New solutions were prepared daily. In an overhead mixing tank, WAF solutions were mixed with seawater and delivered by a peristaltic pump into the respective treatments at 2% of the original stock solutions. The 2% exposure concentration was chosen, based on the 96LC values established for the di!erent toxicants. The 96LC value for dispersed crude oil was 7.15% (7.94% upper and 6.42% lower 95% con"dence interval (CI)) and 7.45% (8.26% upper and 6.71% lower 95% CI), respectively (Cohen and Nugegoda, in press). Fish were subjected to three treatments (crude oil WAF, dispersed crude oil WAF, and control) for 4 days, with two replicates per treatment, for a total of six tanks. Gills, liver, and white muscles of "ve "sh were sampled from each replicate on Day 0 and Day 4 for metabolic enzyme analysis. The remaining "ve "sh/aquarium were carried through to Day 16 in clean seawater and analyzed for bile metabolites (Cohen, in preparation). Physicochemical parameters
Metabolic Enzyme Analyses Fish were sacri"ced after being anesthetized using 3aminobenzoic acid ethyl ester and the gills, liver, and tissue were immediately sampled and stored in liquid nitrogen. CCO analyses were performed according to Pelletier et al. (1994). Tissue homogenate was prepared by homogenizing tissue in imidazole bu!er (1:10, pH 7.4), using a Ystral Model T1500 homogenizer. Samples were centrifuged at 3000 rpm for 5 min at 53C, using a Beckman GS15R centrifuge. A dosing solution was prepared that consisted of reduced cytochrome c (from horse heart, MW"12,384 g) 70 lM in imidazole bu!er (50 lM), pH 8.0. The CCO activity was read against a control composed of 900 ll dosing solution and 100 ll of 0.33% w/v potassium ferricyanide. A UV spectrophotometer (Hitachi U-2000) was used for reading absorbance at 550 nm over 2 min. Each sample was run in duplicate. Lactate dehydrogenase activity was measured according to Pelletier et al. (1994). A dosing solution consisted of 100 mM potassium phosphate bu!er, pH 7.4; 0.15 mM NADH disodium salt, and 0.8 mM pyruvic acid. Absorbance was measured at 340 nm for 2 min. Samples were run in duplicate. Enzymatic activity was reported for both enzymes in international units per milligram of protein. One unit corresponds to the transformation of 1 lmol of substrate, converted to product per minute. Protein analyses were performed on each sample using the method of Lowry et al. (1951). Oxygen Consumption Oxygen consumption was measured on Day 4, prior to sacri"cing of organisms, using a 3-liter perspex respiration
308
COHEN, NUGEGODA, AND GAGNON
chamber with an oxygen probe inserted through the cover. Organisms were placed in the chamber and allowed to acclimate for 15 min. All air bubbles were removed. To maintain a homogeneous solution the respiration chamber was placed on top of a magnetic stirrer and the water was gently mixed. Three "sh in each aquarium were randomly sampled and their respiration rate was measured individually. Oxygen consumption was monitored over a 15-min period and reported as milligrams of oxygen per gram of organism. Organisms were returned to their respective treatments following measurements. Three organisms were chosen from each treatment instead of "ve, to allow for constant monitoring, prior to "sh being sampled for metabolic enzymes. Statistical Analyses Homogeneity of variance was checked using a means versus standard deviations plot (Statistica Version 5). Log transformations were carried out on LDH in the gill, CCO and LDH in the liver, and CCO in white muscle. Two-way ANOVA (analysis of variance) was carried out on Day 0 and Day 4 data, to investigate changes in activity between the treatments. A two-way ANOVA was used to examine oxygen consumption rate between the treatments after the exposure period. A Post Hoc Test (Tukey Protected HSD) was used to compare signi"cant di!erences (P(0.05) between the treatments. RESULTS
The TPH concentrations in the stock solutions for the crude oil for WAF and dispersed oil WAF treatments were 3.4 and 19.2 mg/liter, respectively. This would result in nominal TPH concentrations of 68 and 384 lg/liter in the crude oil WAF and dispersed crude oil WAF, respectively. Spike recoveries yielded 88%. TPH background levels in seawater were found to be undetectable. Compared to the control treatments, there was no signi"cant di!erence in either CCO or LDH activity in the gills, liver, or white muscle at Day 0 in the crude oil WAF or dispersed crude oil WAF treatments (P'0.05). CCO activity was signi"cantly stimulated in the gills of "sh after the 4-day exposure to dispersed crude oil WAF, compared to the crude oil WAF and control treatments (P(0.05) (Fig. 1). There was no signi"cant di!erence in LDH activity in the gills between the treatments after the exposure period (P'0.05) (Fig. 2). The activity of cytochrome c oxidase was examined in the livers of "sh. Relative to the control treatment, CCO activity was signi"cantly higher (P(0.05) following exposure to the WAF of crude oil and dispersed crude oil (Fig. 1). LDH activity in the livers of "sh after the 4-day exposure was signi"cantly higher in the dispersed crude oil WAF treat-
FIG. 1. CCO activity (U/mg protein) (mean$SEM) in the gills, liver, and white muscle of juvenile Australian bass following waterborne exposure to the WAF of crude oil and dispersed crude oil after 4 days (gill, n"8, 7, 7; liver, n"7, 8, 7; muscle, n"8, 8, 8). An asterisk indicates that activity was signi"cantly higher than the control for that tissue.
ment (P(0.05) than in the crude oil WAF and control treatments (Fig. 2). Examination of metabolic enzyme activity in white muscle found that relative to the controls, there was no signi"cant di!erence in CCO activity (P'0.05) or LDH activity (P'0.05), following the exposure period, between the treatments. Oxygen consumption in "sh at the end of the exposure period was signi"cantly higher in the dispersed crude oil WAF treatment than in the crude oil WAF and control treatments (P(0.05) (Fig. 3). Oxygen consumption in the crude oil WAF did not signi"cantly di!er from that in the control (P'0.05).
FIG. 2. LDH activity (U/mg protein) (mean$SEM) in the gills, liver, and white muscle of juvenile Australian bass following waterborne exposure to the WAF of crude oil and dispersed crude oil after 4 days (gill, n"8, 7, 7; liver, n"7, 8, 7; muscle, n"8, 8, 8). An asterisk indicates that activity was signi"cantly higher than the control for that tissue.
METABOLIC ENZYMES IN FISH AFTER EXPOSURE TO OIL
FIG. 3. Oxygen consumption rate (mg O /g "sh) in juvenile Australian bass, following exposure to the WAF of oil and dispersed oil after 4 days (mean$SEM, n"6). An asterisk indicates that respiration in the dispersed oil treatment was signi"cantly higher (P(0.05) than that in the oil and control treatments.
DISCUSSION
From the analysis of the stock solutions, the dispersed oil WAF was found to have the highest concentration of TPH. The nominal exposure concentrations in the dispersed oil WAF treatment were more than "ve times higher than the exposure concentration in the crude oil WAF treatment. This "nding is comparable to the "ndings of previous research (Fucik, 1994). When oil is chemically dispersed it is broken up into small oil droplets that enter the water column (Gilbert, 1996), which are then made available to aquatic life. The oil droplets consist of aliphatic and aromatic hydrocarbons (Connell and Miller, 1984). The aliphatic compounds are relatively water-soluble (Anderson et al., 1974), whereas the higher molecular weight aromatic compounds are not as water-soluble and are therefore normally present at very low concentrations in the water column (Anderson et al., 1974). However, chemically dispersing crude oil allows for greater concentrations of these compounds to enter the water column (Anderson et al., 1974). The greatest CCO activity in control "sh was found in white muscle, followed by the gills and the liver. In 6-monthold largemouth bass, Micropterus salmoides, the levels of CCO in muscle were positively correlated with growth rate (Goolish and Adelman, 1987). The "sh in the present study are also small "ngerlings, undergoing rapid increases in growth. This may be the reason for the high CCO activity in white muscle. LDH activity in control "sh was also the highest in white muscle, followed by the gills and the liver. The partial pressure of oxygen drops as it passes through the gills into the respiratory cavities and is circulated until reaching the cell mitochondria (Heath, 1987). As partial pressure is related to the amount of oxygen carried by the hemoglobin, it is expected that the gills contain more oxygen than the liver, thereby prompting less anaerobic capacity. However, a build-up of lactate may be occurring in the gill epithelium,
309
due to increased oxygen consumption. LDH activity in white muscle was also approximately three times higher than in the gills. This biochemical response also complies with low partial pressure and oxygen concentrations in the muscle, prompting increased anaerobic capacity. Chambers et al. (1979) similarly found that LDH activity was highest in muscle, followed by the gill and then the liver. Exposure to the crude oil WAF caused a signi"cant stimulation in aerobic capacity in the liver. Exposure to the dispersed oil WAF treatment caused a signi"cant stimulation in aerobic activity in the gills and liver and anaerobic activity in the liver. The hepatic system is one of the most sensitive systems for detecting change, following exposure to oil. No signi"cant di!erence was detected in white muscle. It is possible that aerobic capacity was close to maximal in muscle, since young fast-growing "sh have a high metabolism (Goolish and Adelman, 1987) and, therefore, exposure to the di!erent treatments did not signi"cantly a!ect activity. This suggests that the gills and liver are more sensitive organs for detecting change in aerobic and anaerobic capacity, following exposure to crude oil and dispersed crude oil. Exposure to the dispersed oil WAF treatment caused a signi"cant increase in aerobic and anaerobic capacity, compared to the crude oil WAF treatment, which a!ected only aerobic capacity in the liver. In the dispersed oil WAF treatment the stimulation in CCO activity in gills and liver and LDH activity in the liver may be due to the high TPH concentrations present in the water column. The exposure concentration of 2% was also only three times lower than the acute toxicity value (96LC ) established for dispersed oil WAF (Cohen and Nugegoda, in press), whereas the 2% concentration in the crude oil WAF was more than 20 times less than the 96LC value established for the respective treatment (Cohen and Nugegoda, in press). LDH concentrations in the gill can be altered by direct exposure to dispersants in the water column. Gill permeability may be a!ected by the surfactants contained in the dispersants that are designed to dissolve lipids and are nonspeci"c, acting on the lipid bilayer of living cells, causing alteration of membrane permeability (Khan et al., 1986). If gill permeability is altered, there may be leakage of LDH from the tissue (Bowes and Ramos, 1994). Stimulation as well as suppression of key metabolic enzyme activity has been previously reported following exposure to hydrocarbons. Ravindran (1988) found an increase in aerobic capacity, as measured by succinate dehydrogenase and a decrease in anaerobic capacity, as measured by LDH activity in ¹ilapia mossambica from waterborne exposure to toluene after 24 h. Research conducted on Atlantic salmon, Salmo salar, has found reductions in aerobic (citrate synthase) and anaerobic (LDH) capacity following exposure to Bass Strait crude oil and dispersed crude oil (Gagnon and Holdway, 1999). The treatments utilized in the present study were 80}160 times more concentrated than concentrations
310
COHEN, NUGEGODA, AND GAGNON
utilized in the study of Gagnon and Holdway (1999), based on the percentage of dilutions chosen, which may indicate that high concentrations of hydrocarbons cause stimulation in aerobic and anaerobic capacities. Oxygen consumption was found to be signi"cantly higher in "sh exposed to the dispersed oil WAF. Increases in oxygen consumption have been found in "sh following exposure to various oils (Anderson et al., 1974). Thomas and Rice (1975) reported a linear relationship between increasing oil concentration and breathing rates in pink salmon (Oncorhynchus gorbuscha). The reason for this may be due to higher TPH concentrations causing higher stress to "sh exposed to the dispersed crude oil treatment. The present study has investigated the metabolic stress induced following exposure of "sh to the WAF of Bass Strait crude oil and dispersed crude oil. From the two remediation techniques investigated, it was demonstrated that the dispersed oil treatment induced the largest sublethal stress on exposed "sh by placing a higher concentration of total petroleum hydrocarbons into solution and causing a signi"cant increase in metabolic enzyme activities and oxygen consumption. Leaving the oil to naturally degrade on the water surface would appear to have the lesser e!ect and, therefore, may be the more appropriate remediation technique, when the oil slick is not threatening valuable resources. CONCLUSION
Increases in the activity of CCO and LDH were detected in the liver and CCO in the gills, following exposure to the dispersed crude oil WAF. There was an increase in CCO activity in the liver following exposure to crude oil WAF. Oxygen consumption increased in "sh exposed to dispersed crude oil WAF after 4 days of exposure. Therefore, the results indicate that chemically dispersing oil induces higher sublethal stress in Australian bass, exposed via the water column, than that found when the oil is allowed to naturally degrade. ACKNOWLEDGMENTS The authors thank Searle Aquaculture for providing the test organisms, the sta! at the Queenscli! Marine Station for their assistance with the project, and Professor Doug Holdway for his constructive comments. This research was supported by an RMIT National Competitive Grant Scheme Support to M.M.G.
REFERENCES Anderson, J. W., Ne!, J. M., Cox, B. A., Tatem, H. E., and Hightower, G. M. (1974). Characteristics of dispersions and water soluble extracts of crude and re"ned oils and their toxicity to estuarine crustaceans and "sh. Mar. Biol. 27, 75}88. Australian Institute of Petroleum (1996). Oceans and Oil Spills. Petroleum Topics, Australia.
Bowes, R. C., and Ramos, K. S. (1994). Assessment of cell-speci"c cytotoxic responses of the kidney to selected aromatic hydrocarbons. ¹oxicol. In
Metabolic Responses of Fish Following Exposure to Two Different Oil Spill Remediation Techniques Adam Cohen,* Dayanthi Nugegoda,* and Marthe Monique Gagnon*Department of Applied Biology and Biotechnology, RMIT, City Campus, GPO Box 2476V, Melbourne, Victoria 3001, Australia; and -School of Environmental Biology, Curtin University of Technology, Perth, Western Australia 6845, Australia Received July 24, 2000
To assess the impacts of two oil spill remediation techniques on 5sh metabolism, change in aerobic and anaerobic enzyme activities in juvenile Australian Bass, Macquaria novemaculeata, was examined. Changes in cytochrome C oxidase (CCO) and lactate dehydrogenase (LDH) activities were investigated following exposure to the crude oil water accommodated fraction (WAF) and chemically dispersed crude oil WAF. There was a signi5cant stimulation in CCO activity in the gills and livers of 5sh exposed to the WAF of Bass Strait crude oil and chemically dispersed crude oil, compared to the control treatment. In addition, LDH activity was signi5cantly stimulated in the liver of 5sh exposed to dispersed crude oil WAF, compared to the crude oil WAF. Fish exposed to the dispersed crude oil WAF treatment had signi5cantly higher oxygen consumption, as measured by oxygen depletion in a sealed chamber, than 5sh exposed to the crude oil WAF and control treatments. 2001 Academic Press Key Words: metabolic enzymes; cytochrome C oxidase; lactate dehydrogenase; crude oil; dispersed crude oil; Corexit 9527; biomarkers; oxygen consumption.
INTRODUCTION
Oil spills are caused by tankers running aground, collisions between vessels, accidents on exploration/production platforms, or at a re"nery or port area (Australian Institute of Petroleum, 1996). Spill events contribute to 12% of the total inputs of petroleum into the marine environment. O!shore oil slicks are often left to naturally disperse, while a variety of remediation techniques are available when the oil slicks threaten shorelines or valuable natural resources. Oil spill remediation options include mechanical removal, dispersing, burning, and leaving the oil to naturally degrade. Each technique results in di!erent compositions and concentrations of hydrocarbons entering the water column. Chemically dispersing crude oil places up to "ve times more total petroleum hydrocarbons (TPH) into the water column, compared to leaving the oil to naturally disperse (Fucik, 1994).
Aquatic organisms exposed to compounds such as oil may exhibit biochemical, physiological, and/or behavioral responses. In "eld studies, biomarkers of e!ects are preferred to biomarkers of exposure, because individual biological response can be extrapolated to impacts at a population level (Depledge, 1993). There has been limited research utilizing physiological parameters such as oxygen consumption, in complement with biochemical monitoring. The simultaneous measurement of both biochemical and physiological parameters could improve understanding of the impact of xenobiotics on organisms. Signi"cant increases in aerobic and decreases in anaerobic respiration have been reported following exposure to hydrocarbons (Ravindran, 1988). A key enzyme in aerobic metabolism is cytochrome C oxidase (CCO), the terminal step in the electron transport system. CCO is a mitochondrial enzyme that correlates well with actual oxygen consumption rates for di!erent tissues (Simon and Robin, 1971). Changes in carbohydrate metabolism have been reported in "sh chronically exposed to petroleum hydrocarbons (Dey et al., 1983). Lactate dehydrogenase (LDH) is found in the cellular cytoplasm and is active during glycolysis, converting pyruvate from glucose to lactic acid (Knox et al., 1994). Lactic acid is produced in "sh in larger amounts during periods of stress, exercise, and/or hypoxia (Hoar and Randall, 1970) and the LDH enzyme has been reported to increase with changes in growth rates (Pelletier et al., 1994) and metabolism (Lind, 1992). In accordance with di!erential cellular metabolism, Anderson et al. (1974) and Thomas and Rice (1975) reported increases in oxygen consumption following exposure to hydrocarbons. The aim of the study was to investigate the impact of Bass Strait crude oil and dispersed crude oil on aerobic and anaerobic enzyme activities, as a measure of "sh health. A secondary goal was to examine oxygen consumption as a physiological parameter following exposure. Changes in CCO and LDH enzyme activity were measured in the gills, liver, and muscle of Australian bass following exposure to
306 0147-6513/01 $35.00 Copyright 2001 by Academic Press All rights of reproduction in any form reserved.
METABOLIC ENZYMES IN FISH AFTER EXPOSURE TO OIL
the water accommodated fraction (WAF) of crude oil and chemically dispersed crude oil. Change in oxygen consumption in live organisms was investigated by measuring oxygen depletion from within a sealed chamber.
307
were monitored daily throughout the experiment (mean$SE, n"24). Salinity remained constant at 35$0.2 ppt, pH 8.3$0.3, conductivity at 52.6 ls/cm, dissolved oxygen at 6.4$0.2 mg/liter, and temperature at 18.3$0.63C.
MATERIALS AND METHODS
Experimental Organisms
Hydrocarbon Analyses
Eight-week-old Australian bass, Macquaria novemaculeata, were supplied by Searle Aquaculture, New South Wales, Australia, and reared in a #ow-through seawater system at the Queenscli! Marine Station, Victoria, Australia, for 4 months. M. novemaculeata were fed salmon starter pellets ad libitum. Prior to the beginning of the experiment, "sh were randomly distributed at 15 "sh per 40-liter tank and acclimated for a further 2 weeks. The seawater #owthrough rate was maintained at 150 ml/min, which ensured a 90% replacement every 4 h (Sprague, 1973). Fish weight prior to experimentation was 6.5$0.6 g (mean$SE, n"10). The seawater source was the same for rearing and experimentation.
Seawater hydrocarbon concentrations were analyzed by extracting 250 ml of stock solution twice with 25 ml dichloromethane. Sodium sulfate was added to the extracts to remove residual water. Extracts were stored at !203C until analysis. Total petroleum hydrocarbons (TPH) were analyzed by gas chromatography (GC-FID). Reagent blanks of dichloromethane and spike recoveries were also performed. TPH concentrations were analyzed in the 100% preparations rather than the experimental tanks, because the experimental concentrations were expected to be too low for accurate measurement using GC-FID.
Preparation of Stock Solutions Crude oil WAF was prepared by "lling a 20-liter glass mixing chamber with 15 liters of seawater and creating a vortex using a magnetic stirrer. Bass Strait crude oil was then added at a ratio of 1:9 (oil:seawater). The solution was capped, mixed for 20 h, and allowed to settle for 1 h to separate water and oil phases. The solution underlying the oil phases was isolated via a tap at the bottom of the bottle (Anderson et al., 1974). This solution represented the crude oil WAF. The dispersed oil WAF was prepared using the same methodology, except dispersant (Corexit 9527) was added to the #oating oil during mixing, at a ratio of 1:30 (dispersant:oil) (Gilbert, 1996). New solutions were prepared daily. In an overhead mixing tank, WAF solutions were mixed with seawater and delivered by a peristaltic pump into the respective treatments at 2% of the original stock solutions. The 2% exposure concentration was chosen, based on the 96LC values established for the di!erent toxicants. The 96LC value for dispersed crude oil was 7.15% (7.94% upper and 6.42% lower 95% con"dence interval (CI)) and 7.45% (8.26% upper and 6.71% lower 95% CI), respectively (Cohen and Nugegoda, in press). Fish were subjected to three treatments (crude oil WAF, dispersed crude oil WAF, and control) for 4 days, with two replicates per treatment, for a total of six tanks. Gills, liver, and white muscles of "ve "sh were sampled from each replicate on Day 0 and Day 4 for metabolic enzyme analysis. The remaining "ve "sh/aquarium were carried through to Day 16 in clean seawater and analyzed for bile metabolites (Cohen, in preparation). Physicochemical parameters
Metabolic Enzyme Analyses Fish were sacri"ced after being anesthetized using 3aminobenzoic acid ethyl ester and the gills, liver, and tissue were immediately sampled and stored in liquid nitrogen. CCO analyses were performed according to Pelletier et al. (1994). Tissue homogenate was prepared by homogenizing tissue in imidazole bu!er (1:10, pH 7.4), using a Ystral Model T1500 homogenizer. Samples were centrifuged at 3000 rpm for 5 min at 53C, using a Beckman GS15R centrifuge. A dosing solution was prepared that consisted of reduced cytochrome c (from horse heart, MW"12,384 g) 70 lM in imidazole bu!er (50 lM), pH 8.0. The CCO activity was read against a control composed of 900 ll dosing solution and 100 ll of 0.33% w/v potassium ferricyanide. A UV spectrophotometer (Hitachi U-2000) was used for reading absorbance at 550 nm over 2 min. Each sample was run in duplicate. Lactate dehydrogenase activity was measured according to Pelletier et al. (1994). A dosing solution consisted of 100 mM potassium phosphate bu!er, pH 7.4; 0.15 mM NADH disodium salt, and 0.8 mM pyruvic acid. Absorbance was measured at 340 nm for 2 min. Samples were run in duplicate. Enzymatic activity was reported for both enzymes in international units per milligram of protein. One unit corresponds to the transformation of 1 lmol of substrate, converted to product per minute. Protein analyses were performed on each sample using the method of Lowry et al. (1951). Oxygen Consumption Oxygen consumption was measured on Day 4, prior to sacri"cing of organisms, using a 3-liter perspex respiration
308
COHEN, NUGEGODA, AND GAGNON
chamber with an oxygen probe inserted through the cover. Organisms were placed in the chamber and allowed to acclimate for 15 min. All air bubbles were removed. To maintain a homogeneous solution the respiration chamber was placed on top of a magnetic stirrer and the water was gently mixed. Three "sh in each aquarium were randomly sampled and their respiration rate was measured individually. Oxygen consumption was monitored over a 15-min period and reported as milligrams of oxygen per gram of organism. Organisms were returned to their respective treatments following measurements. Three organisms were chosen from each treatment instead of "ve, to allow for constant monitoring, prior to "sh being sampled for metabolic enzymes. Statistical Analyses Homogeneity of variance was checked using a means versus standard deviations plot (Statistica Version 5). Log transformations were carried out on LDH in the gill, CCO and LDH in the liver, and CCO in white muscle. Two-way ANOVA (analysis of variance) was carried out on Day 0 and Day 4 data, to investigate changes in activity between the treatments. A two-way ANOVA was used to examine oxygen consumption rate between the treatments after the exposure period. A Post Hoc Test (Tukey Protected HSD) was used to compare signi"cant di!erences (P(0.05) between the treatments. RESULTS
The TPH concentrations in the stock solutions for the crude oil for WAF and dispersed oil WAF treatments were 3.4 and 19.2 mg/liter, respectively. This would result in nominal TPH concentrations of 68 and 384 lg/liter in the crude oil WAF and dispersed crude oil WAF, respectively. Spike recoveries yielded 88%. TPH background levels in seawater were found to be undetectable. Compared to the control treatments, there was no signi"cant di!erence in either CCO or LDH activity in the gills, liver, or white muscle at Day 0 in the crude oil WAF or dispersed crude oil WAF treatments (P'0.05). CCO activity was signi"cantly stimulated in the gills of "sh after the 4-day exposure to dispersed crude oil WAF, compared to the crude oil WAF and control treatments (P(0.05) (Fig. 1). There was no signi"cant di!erence in LDH activity in the gills between the treatments after the exposure period (P'0.05) (Fig. 2). The activity of cytochrome c oxidase was examined in the livers of "sh. Relative to the control treatment, CCO activity was signi"cantly higher (P(0.05) following exposure to the WAF of crude oil and dispersed crude oil (Fig. 1). LDH activity in the livers of "sh after the 4-day exposure was signi"cantly higher in the dispersed crude oil WAF treat-
FIG. 1. CCO activity (U/mg protein) (mean$SEM) in the gills, liver, and white muscle of juvenile Australian bass following waterborne exposure to the WAF of crude oil and dispersed crude oil after 4 days (gill, n"8, 7, 7; liver, n"7, 8, 7; muscle, n"8, 8, 8). An asterisk indicates that activity was signi"cantly higher than the control for that tissue.
ment (P(0.05) than in the crude oil WAF and control treatments (Fig. 2). Examination of metabolic enzyme activity in white muscle found that relative to the controls, there was no signi"cant di!erence in CCO activity (P'0.05) or LDH activity (P'0.05), following the exposure period, between the treatments. Oxygen consumption in "sh at the end of the exposure period was signi"cantly higher in the dispersed crude oil WAF treatment than in the crude oil WAF and control treatments (P(0.05) (Fig. 3). Oxygen consumption in the crude oil WAF did not signi"cantly di!er from that in the control (P'0.05).
FIG. 2. LDH activity (U/mg protein) (mean$SEM) in the gills, liver, and white muscle of juvenile Australian bass following waterborne exposure to the WAF of crude oil and dispersed crude oil after 4 days (gill, n"8, 7, 7; liver, n"7, 8, 7; muscle, n"8, 8, 8). An asterisk indicates that activity was signi"cantly higher than the control for that tissue.
METABOLIC ENZYMES IN FISH AFTER EXPOSURE TO OIL
FIG. 3. Oxygen consumption rate (mg O /g "sh) in juvenile Australian bass, following exposure to the WAF of oil and dispersed oil after 4 days (mean$SEM, n"6). An asterisk indicates that respiration in the dispersed oil treatment was signi"cantly higher (P(0.05) than that in the oil and control treatments.
DISCUSSION
From the analysis of the stock solutions, the dispersed oil WAF was found to have the highest concentration of TPH. The nominal exposure concentrations in the dispersed oil WAF treatment were more than "ve times higher than the exposure concentration in the crude oil WAF treatment. This "nding is comparable to the "ndings of previous research (Fucik, 1994). When oil is chemically dispersed it is broken up into small oil droplets that enter the water column (Gilbert, 1996), which are then made available to aquatic life. The oil droplets consist of aliphatic and aromatic hydrocarbons (Connell and Miller, 1984). The aliphatic compounds are relatively water-soluble (Anderson et al., 1974), whereas the higher molecular weight aromatic compounds are not as water-soluble and are therefore normally present at very low concentrations in the water column (Anderson et al., 1974). However, chemically dispersing crude oil allows for greater concentrations of these compounds to enter the water column (Anderson et al., 1974). The greatest CCO activity in control "sh was found in white muscle, followed by the gills and the liver. In 6-monthold largemouth bass, Micropterus salmoides, the levels of CCO in muscle were positively correlated with growth rate (Goolish and Adelman, 1987). The "sh in the present study are also small "ngerlings, undergoing rapid increases in growth. This may be the reason for the high CCO activity in white muscle. LDH activity in control "sh was also the highest in white muscle, followed by the gills and the liver. The partial pressure of oxygen drops as it passes through the gills into the respiratory cavities and is circulated until reaching the cell mitochondria (Heath, 1987). As partial pressure is related to the amount of oxygen carried by the hemoglobin, it is expected that the gills contain more oxygen than the liver, thereby prompting less anaerobic capacity. However, a build-up of lactate may be occurring in the gill epithelium,
309
due to increased oxygen consumption. LDH activity in white muscle was also approximately three times higher than in the gills. This biochemical response also complies with low partial pressure and oxygen concentrations in the muscle, prompting increased anaerobic capacity. Chambers et al. (1979) similarly found that LDH activity was highest in muscle, followed by the gill and then the liver. Exposure to the crude oil WAF caused a signi"cant stimulation in aerobic capacity in the liver. Exposure to the dispersed oil WAF treatment caused a signi"cant stimulation in aerobic activity in the gills and liver and anaerobic activity in the liver. The hepatic system is one of the most sensitive systems for detecting change, following exposure to oil. No signi"cant di!erence was detected in white muscle. It is possible that aerobic capacity was close to maximal in muscle, since young fast-growing "sh have a high metabolism (Goolish and Adelman, 1987) and, therefore, exposure to the di!erent treatments did not signi"cantly a!ect activity. This suggests that the gills and liver are more sensitive organs for detecting change in aerobic and anaerobic capacity, following exposure to crude oil and dispersed crude oil. Exposure to the dispersed oil WAF treatment caused a signi"cant increase in aerobic and anaerobic capacity, compared to the crude oil WAF treatment, which a!ected only aerobic capacity in the liver. In the dispersed oil WAF treatment the stimulation in CCO activity in gills and liver and LDH activity in the liver may be due to the high TPH concentrations present in the water column. The exposure concentration of 2% was also only three times lower than the acute toxicity value (96LC ) established for dispersed oil WAF (Cohen and Nugegoda, in press), whereas the 2% concentration in the crude oil WAF was more than 20 times less than the 96LC value established for the respective treatment (Cohen and Nugegoda, in press). LDH concentrations in the gill can be altered by direct exposure to dispersants in the water column. Gill permeability may be a!ected by the surfactants contained in the dispersants that are designed to dissolve lipids and are nonspeci"c, acting on the lipid bilayer of living cells, causing alteration of membrane permeability (Khan et al., 1986). If gill permeability is altered, there may be leakage of LDH from the tissue (Bowes and Ramos, 1994). Stimulation as well as suppression of key metabolic enzyme activity has been previously reported following exposure to hydrocarbons. Ravindran (1988) found an increase in aerobic capacity, as measured by succinate dehydrogenase and a decrease in anaerobic capacity, as measured by LDH activity in ¹ilapia mossambica from waterborne exposure to toluene after 24 h. Research conducted on Atlantic salmon, Salmo salar, has found reductions in aerobic (citrate synthase) and anaerobic (LDH) capacity following exposure to Bass Strait crude oil and dispersed crude oil (Gagnon and Holdway, 1999). The treatments utilized in the present study were 80}160 times more concentrated than concentrations
310
COHEN, NUGEGODA, AND GAGNON
utilized in the study of Gagnon and Holdway (1999), based on the percentage of dilutions chosen, which may indicate that high concentrations of hydrocarbons cause stimulation in aerobic and anaerobic capacities. Oxygen consumption was found to be signi"cantly higher in "sh exposed to the dispersed oil WAF. Increases in oxygen consumption have been found in "sh following exposure to various oils (Anderson et al., 1974). Thomas and Rice (1975) reported a linear relationship between increasing oil concentration and breathing rates in pink salmon (Oncorhynchus gorbuscha). The reason for this may be due to higher TPH concentrations causing higher stress to "sh exposed to the dispersed crude oil treatment. The present study has investigated the metabolic stress induced following exposure of "sh to the WAF of Bass Strait crude oil and dispersed crude oil. From the two remediation techniques investigated, it was demonstrated that the dispersed oil treatment induced the largest sublethal stress on exposed "sh by placing a higher concentration of total petroleum hydrocarbons into solution and causing a signi"cant increase in metabolic enzyme activities and oxygen consumption. Leaving the oil to naturally degrade on the water surface would appear to have the lesser e!ect and, therefore, may be the more appropriate remediation technique, when the oil slick is not threatening valuable resources. CONCLUSION
Increases in the activity of CCO and LDH were detected in the liver and CCO in the gills, following exposure to the dispersed crude oil WAF. There was an increase in CCO activity in the liver following exposure to crude oil WAF. Oxygen consumption increased in "sh exposed to dispersed crude oil WAF after 4 days of exposure. Therefore, the results indicate that chemically dispersing oil induces higher sublethal stress in Australian bass, exposed via the water column, than that found when the oil is allowed to naturally degrade. ACKNOWLEDGMENTS The authors thank Searle Aquaculture for providing the test organisms, the sta! at the Queenscli! Marine Station for their assistance with the project, and Professor Doug Holdway for his constructive comments. This research was supported by an RMIT National Competitive Grant Scheme Support to M.M.G.
REFERENCES Anderson, J. W., Ne!, J. M., Cox, B. A., Tatem, H. E., and Hightower, G. M. (1974). Characteristics of dispersions and water soluble extracts of crude and re"ned oils and their toxicity to estuarine crustaceans and "sh. Mar. Biol. 27, 75}88. Australian Institute of Petroleum (1996). Oceans and Oil Spills. Petroleum Topics, Australia.
Bowes, R. C., and Ramos, K. S. (1994). Assessment of cell-speci"c cytotoxic responses of the kidney to selected aromatic hydrocarbons. ¹oxicol. In