Kinetics and effects of dichloroacetic acid in rainbow trout

Aquatic Toxicology 94 (2009) 186–194

Contents lists available at ScienceDirect

Aquatic Toxicology journal homepage: www.elsevier.com/locate/aquatox

Kinetics and effects of dichloroacetic acid in rainbow trout夽 Patrick N. Fitzsimmons ∗ , Alex D. Hoffman, Gregory J. Lien, Dean E. Hammermeister, John W. Nichols U.S. Environmental Protection Agency, Office of Research and Development, National Health and Environmental Effects Research Laboratory, Mid-Continent Ecology Division, 6201 Congdon Boulevard, Duluth, MN 55804, USA

a r t i c l e

i n f o

Article history: Received 19 December 2008 Received in revised form 29 June 2009 Accepted 1 July 2009 Keywords: Disinfection byproducts Halogenated acetic acids Dichloroacetic acid Trout Kinetics

a b s t r a c t Halogenated acetic acids (HAAs) produced by chlorine disinfection of municipal drinking water represent a potentially important class of environmental contaminants. Little is known, however, about their potential to adversely impact fish and other aquatic life. In this study we examined the kinetics and effects of dichloroacetic acid (DCA) in rainbow trout. Branchial uptake was measured in fish confined to respirometer-metabolism chambers. Branchial uptake efficiency was <5%, suggesting passive diffusion through aqueous channels in the gill epithelium. DCA concentrations in tissues following prolonged (72, 168, or 336 h) waterborne exposures were expressed as tissue:plasma concentration ratios. Concentration ratios for the kidney and muscle at 168 and 336 h were consistent with the suggestion that DCA distributes primarily to tissue water. Reduced concentration ratios for the liver, particularly at 72 h, indicated that DCA was highly metabolized by this tissue. Routes and rates of elimination were characterized by injecting chambered animals with a high (5.0 mg/kg) or low (50 ␮g/kg) bolus dose. DCA was rapidly cleared by naïve animals resulting in elimination half-lives (t1/2 ) of less than 4 h. Waterborne pre-treatment of fish with DCA increased the persistence of a subsequently injected dose. In high dose animals, pre-treatment caused a 4-fold decrease in whole-body clearance (CLB ) and corresponding increases in the area under the plasma concentration–time curve (extrapolated to infinity; AUC0→∞ ) and t1/2 . Qualitatively similar results were obtained in low dose fish, although the magnitude of the pretreatment effect (∼2.5-fold) was reduced. Renal and branchial clearance contributed little (combined, <3% of CLB ) to the elimination of DCA. Biliary elimination of DCA was also negligible. The steady-state volume of distribution (VSS ) did not vary among treatment groups and was consistent with results of the tissue distribution study. DCA had no apparent effects on respiratory physiology or acid–base balance; however, the concentration of blood lactate declined progressively during continuous waterborne exposures. A transient effect on blood lactate was also observed in bolus injection experiments. The results of this study suggest that clearance of DCA is due almost entirely to metabolism. The pathway responsible for this activity exhibits characteristics in common with those of mammalian glutathione S-transferase zeta (GST␨), including non-linear kinetics and apparent suicide inactivation by DCA. Observed effects on blood lactate are probably due to the inhibition of pyruvate dehydrogenase kinase in aerobic tissues and may require the participation of a monocarboxylase transport protein to move DCA across cell membranes. Published by Elsevier B.V.

Abbreviations: AUC0→∞ , area under the plasma concentration–time curve (extrapolated to infinity); BCA, bromochloroacetic acid; CLB , whole-body clearance; CLM , metabolic clearance; CLR , renal clearance; DCA, dichloroacetic acid; fu , unbound fraction; GST␨, glutathione S-transferase zeta; HAAs, haloacetic acids; Hct, hematocrit; i.a., intra-arterial; MBCA, methyl bromochloroacetate; MDCA, methyl dichloroacetate; MDL, method detection limit; MRT, mean residence time; MTBE, t-butyl methyl ether; PDH, pyruvate dehydrogenase; t1/2 , elimination half-life; UE , oxygen uptake efficiency; VO2 , oxygen consumption; VSS , steady-state volume of distribution; VVOL , ventilation volume. 夽 The information in this document has been funded wholly by the U.S. Environmental Protection Agency. It has been subjected to review by the National Health and Environmental Effects Research Laboratory and approved for publication. Approval does not signify that the contents reflect the views of the Agency, nor does mention of any trade names or commercial products constitute endorsement or recommendation for use. ∗ Corresponding author. Tel.: +1 218 529 5184; fax: +1 218 529 5003. E-mail address: fi[email protected] (P.N. Fitzsimmons). 0166-445X/$ – see front matter. Published by Elsevier B.V. doi:10.1016/j.aquatox.2009.07.001

P.N. Fitzsimmons et al. / Aquatic Toxicology 94 (2009) 186–194

1. Introduction

2. Materials and methods

Dichloroacetic acid (DCA) is one of the several haloacetic acids (HAAs) formed during chlorine disinfection of municipal water supplies. The HAAs represent a recognized health risk to humans and their effects on mammals have been extensively investigated (Richardson et al., 2002). Concentrations of DCA in finished drinking water typically range from 5 to 20 ␮g/L, although concentrations greater than 100 ␮g/L have been measured (Uden and Miller, 1983; Krasner et al., 1989; Williams et al., 1997). Chronic exposure to DCA causes liver cancer in rats (DeAngelo et al., 1996) and mice (Herren-Freund et al., 1987). The mechanism of hepatocarcinogenecity is unknown but does not appear to be mediated by direct genotoxic effects. DCA also causes developmental toxicity in cultured rodent embryos (Ward et al., 2000; Hunter et al., 2006). Acute exposure to DCA lowers circulating concentrations of lactate and glucose by activating the pyruvate dehydrogenase (PDH) complex (Crabb et al., 1981; Stacpoole, 1989). As such, DCA has been used as an investigational drug for the treatment of several metabolic and cardiovascular disorders including diabetes (Stacpoole and Greene, 1992; Brown and Gore, 1996), congenital lactic acidosis (Stacpoole et al., 1997), and myocardial ischemia (Bersin and Stacpoole, 1997). DCA is rapidly absorbed from the gastrointestinal tract of rodents, and maximum levels in blood are generally measured within 1–2 h (Larson and Bull, 1992; Gonzalez-Leon et al., 1999; Saghir and Schultz, 2002). Elimination is due almost entirely to biotransformation. In naïve rodents, 30–50% of an oral gavage dose is converted to CO2 within 24 h (Lin et al., 1993; Xu et al., 1995; Gonzalez-Leon et al., 1997, 1999). Urinary metabolites may account for an additional 25% of an administered dose (Larson and Bull, 1992; Lin et al., 1993; Xu et al., 1995). Renal clearance of parent DCA is typically low, ranging from 0.3 to 2% of the dose (Larson and Bull, 1992; Lin et al., 1993; Xu et al., 1995; Schultz et al., 1999). Elimination half-lives in naïve rodents range from 0.1 to 2.4 h (Larson and Bull, 1992; Schultz et al., 1999; Gonzalez-Leon et al., 1997, 1999; James et al., 1998) while those in pre-exposed animals are significantly longer, ranging from 1.1 to 10.8 h (Gonzalez-Leon et al., 1997, 1999). Metabolism of DCA occurs primarily in the liver and is mediated by glutathione S-transferase zeta (GST␨; Tong et al., 1998). Pre-exposure to DCA increases the elimination half-life of a subsequently administered dose due to suicide inactivation of GST␨ (Anderson et al., 1999). DCA and other HAAs are continuously released to the environment in municipal wastewater effluents. Little is known, however, about their potential to adversely impact fish and other aquatic life. Japanese medaka exposed to DCA in water for 4 weeks showed early signs of liver toxicity including hepatocellular vacuolation, glycogen accumulation, and karyomegaly (McHugh Law et al., 1998). Unfortunately, these observations were not accompanied by any measures of DCA uptake and accumulation. Schultz et al. (2007) investigated the kinetics of DCA and several other HAAs in Japanese medaka using an oral dosing method. This approach provided a basis for comparisons with kinetic data from rodent oral dosing studies. Environmental exposures of fish to HAAs are likely, however, to be dominated by uptake directly from water. Information regarding the absorption of HAAs across the gills is therefore needed, as is information on their subsequent distribution and elimination. The purpose of this study was to investigate the branchial uptake, tissue distribution, and elimination of DCA in rainbow trout (Oncorhynchus mykiss). Experiments were designed to provide insight into the role of biotransformation in DCA clearance, including the effect of DCA pre-exposure. Potential effects of DCA on respiratory physiology, acid–base balance, and intermediary metabolism were also assessed.

2.1. Experimental design

187

Rainbow trout were exposed to DCA using three dosing protocols. Short-term (72 h) inhalation exposures were designed to measure the absorption of DCA across the gills of fish confined to respirometer-metabolism chambers (McKim and Goeden, 1982). Additional studies with chambered animals were conducted to evaluate the routes and rates of DCA elimination following a bolus intra-arterial (i.a.) injection. Selected blood chemistry parameters were measured in each of these efforts to investigate the potential impacts of DCA on intermediary metabolism. Effects on respiratory physiology and acid–base balance were also assessed. Finally, longer term (168 and 336 h) exposures with free-swimming fish were conducted to investigate the distribution of DCA among tissues and organs. 2.2. Chemicals DCA (>99% pure), bromochloroacetic acid (97% pure, BCA), methyl dichloroacetate (>99% pure, MDCA), and t-butyl methyl ether (99.8% pure, MTBE) were purchased from Sigma–Aldrich (St. Louis, MO). Methyl bromochloroacetate (100 ␮g/mL in MTBE, MBCA) was obtained from Chem Service (West Chester, PA). Diazomethane was prepared from N-methyl-N-nitroso-ptoluenesulfonamide following the protocol given by Sigma–Aldrich (technical bulletin AL-180). 2.3. Animals and surgical preparation Trout were obtained as juveniles from Aquatic Research Organisms (Kamloops Arlee strain, Hampton, NH) and the U.S. Geological Service (Erwin strain, LaCrosse, WI), and were raised to the size required for each experiment. Fish were held in sand-filtered Lake Superior water under a natural (Duluth, MN) photoperiod and fed commercial trout chow (Silver Cup, Nelson and Sons Inc., Murray, UT). Temperature was maintained at 11 ± 1 ◦ C and dissolved oxygen at 85–100% of saturation. Additional water characteristics were: total hardness, 45–46 mg/L as CaCO3 ; alkalinity, 41–44 mg/L as CaCO3 ; pH, 7.6–7.8; total ammonia, <1 mg/L. Fish used in chambered studies were surgically prepared as previously described (McKim and Goeden, 1982). Briefly, each fish was spinally transected, fitted with a urinary catheter, and cannulated from the dorsal aorta. A latex membrane was sewn around the fish’s mouth to separate inspired (compartment A) and expired (compartment B) water. A second membrane just posterior to the pectoral fins prevented dilution of expired water. Fish were allowed to recover from surgery for 24–48 h before the start of an experiment. 2.4. Inhalation experiments 2.4.1. Chambered animals Four fish (Kamloops Arlee strain, 800–1000 g) were exposed to DCA in water for 72 h. Branchial uptake was assessed by measuring DCA concentrations in inspired and expired water. The difference in concentration between inspired and expired water provides a direct estimate of absorption efficiency across the gills. Exposures were conducted by metering a stock solution of DCA into a toxicantmixing cell where it was diluted with Lake Superior water to a target concentration of 200 mg/L. Water from the mixing cell was then supplied to compartment A of each chamber at a flow rate (500 mL/min) in excess of the fish’s respiratory requirement. Stock solutions (nominally 50 g/L) were prepared by mixing 620 mL of DCA and 720 mL of 10.0 M NaOH with Lake Superior water to obtain

188

P.N. Fitzsimmons et al. / Aquatic Toxicology 94 (2009) 186–194

a final volume of 19 L. The pH of this solution was then adjusted to 7.8 using 0.1 M NaOH. Blood, urine, inspired water, and expired water were sampled to determine DCA concentrations after 0, 2, 4, 8, 24, 30, 48, 54, and 72 h of exposure. Blood (50 ␮L) was obtained through the aortic cannula and urine (350 ␮L) was collected via the urinary catheter. Inspired and expired water (1 mL) were pipetted directly from compartments A and B, respectively. All samples were stored at 4 ◦ C and analyzed within 48 h of collection. Samples of muscle, kidney, liver, and bile were collected at the end of each test (72 h) and frozen at −20 ◦ C. Muscle was sampled from above the lateral line, anterior to the dorsal fin. The liver, kidney, and bile were collected in their entirety. Additional samples of blood (100 ␮L) were collected at each time point to measure pH, pCO2 , lactate, glucose, Na+ , Ca2+ , K+ , Cl− , HCO3 − , and osmolarity. Samples of inspired and expired water (1 mL) were also collected at each time point to measure pH, pCO2 , and HCO3 − . Hematocrits were determined at 0, 24, 48, and 72 h. 2.4.2. Free-swimming animals Eight fish (Erwin strain, 350–550 g) were exposed to DCA in water (nominally 200 mg/L) under free-swimming conditions. Exposures were conducted in an 80 L circular tank supplied with 1 L/min of water. Fish were removed at 168 or 336 h (n = 4 at each time point) and dissected to measure DCA concentrations in blood, muscle, liver, kidney, and bile. 2.5. Injection experiments The elimination of DCA from plasma was measured in a total of sixteen animals. Eight fish (Kamloops Arlee strain, 750–1000 g) were injected with a high dose of DCA (nominally 5.0 mg/kg) while another eight (Erwin strain, 800–1100 g) were given a low dose (nominally 50 ␮g/kg). Four fish from each dosing level were pretreated by exposure to DCA in water (nominally 200 mg/L) for 1 week, followed by 1 week of depuration. The other four fish had not previously been exposed to DCA. The resulting four experimental groups were subsequently referred to as “high dose – pre-treatment,” “high dose – naïve,” “low dose – pre-treatment,” and “low dose – naïve.” All fish were given a bolus injection (0.5 mL/kg) of DCA through the dorsal aortic cannula. Blood (50–400 ␮L, depending on dose and sampling time) and urine (350 ␮L, high dose animals only) were sampled at 0.25, 0.5, 1, 2, 4, 6, 8, 24, and 48 h post-injection and analyzed for DCA. Additional samples of blood (100 ␮L), inspired water (1 mL), and expired water (1 mL) were collected from high dose groups to measure selected biochemical and physiological parameters (see above). Dosing solutions were prepared by dissolving DCA in 25 mL of Cortland’s physiological saline (Wolf, 1963). The pH of this solution was then adjusted to 7.8 using 0.1 M NaOH. Branchial elimination of DCA was assessed by injecting fish (Erwin strain, 800–1000 g) with a bolus dose and then measuring the appearance of chemical in expired water. Preliminary studies of fish injected with 5 mg/kg showed that concentrations of DCA in expired water were near the limit of detection. Subsequent experiments (n = 3) were therefore performed using a bolus dose of 10 mg/kg. Blood (50 ␮L) and expired water (500 mL) were collected at 2, 4, 6, and 8 h post-injection. Water was siphoned from compartment B through a sampling port located in the wall of the respirometer-metabolism chamber. Water samples were immediately filtered through Alltech (Deerfield, IL) 0.2 ␮m nylon filters and stored at 4 ◦ C. All samples were extracted and analyzed within 48 h of collection as described below. The fractional contribution of branchial elimination to total DCA clearance was calculated by dividing the total mass of chemical eliminated during each experiment by the amount injected. The

mass of DCA eliminated by each fish was estimated by multiplying the chemical concentration in expired water, averaged across all four sampling times, by the total volume of water moved across the gills. 2.6. Control experiments Four fish (Erwin strain, 900–1050 g) were placed in respirometer-metabolism chambers and blood (100 ␮L) was sampled at 0, 0.5, 1, 2, 6, 8, 24, 30, 48, 54, and 72 h to measure selected blood chemistry parameters (see above). Samples of inspired and expired water (1 mL) were also collected to measure pH, pCO2 , and HCO3 − . 2.7. Blood/plasma partitioning, plasma binding The partitioning of DCA between blood and plasma, and the binding of DCA to plasma proteins were measured in a set of in vitro experiments. Samples of whole blood (2 mL) were spiked with 2.0 or 20.0 ␮g of DCA and allowed to equilibrate for 2 h at 4 ◦ C. An aliquot of blood (50 or 100 ␮L) was retained to measure the DCA concentration. The balance of the sample was centrifuged at 3000 × g for 10 min to obtain plasma. The resulting plasma sample was then split. A portion (50 or 100 ␮L) was collected to measure the DCA concentration while the remainder was used to evaluate plasma protein binding. Binding of DCA to plasma proteins was measured using a micropartition filtration system (10,000 Da cutoff; Microcon YM10; Millipore, Billerica, MA). Plasma samples (500 ␮L) were placed into the filter devices and centrifuged at 5000 × g for 30 min. Preliminary experiments indicated that binding of DCA to the filter membrane was minimal. The unbound fraction (fu ) of DCA in plasma was calculated as the concentration of DCA in the ultrafiltrate divided by that in plasma. 2.8. Analytical methods 2.8.1. Water and dosing solutions Bolus dosing solutions and water samples collected during the inhalation experiments were analyzed for DCA using reversephase HPLC and UV detection (Waters, Milford, MA). Samples were diluted with de-ionized water if needed (dosing solutions only) and injected directly onto the HPLC column without prior extraction. Chromatographic separation of DCA was achieved on an Aqua C18 column (125 A; 3 ␮m; 3.0 mm × 100 mm; Phenomenex, Torrance, CA). The mobile phase consisted of 100 mM sodium phosphate buffer (pH 6.8) containing 1% acetonitrile. The isocratic flow rate was 0.63 mL/min and the injection volume was 30 ␮L. DCA eluted from the column at 1.85 min and was detected at a wavelength of 204 nm. DCA was quantified using standard solutions prepared in phosphate buffer. DCA concentrations in expired water samples collected during the injection experiments were too low to be measured by UV detection. A method was therefore developed to measure DCA in expired water using GC/ECD. Samples of water (50 mL) were spiked with 100 ␮L of internal standard (BCA in MTBE, 1.0 ␮g/mL), acidified with 250 ␮L 50% H2 SO4 , and extracted 3 times with MTBE (total volume of 20 mL). The extracts were dried with anhydrous sodium sulfate and concentrated to ∼0.5 mL under nitrogen gas. Each extract was then transferred to a pre-marked GC vial insert and concentrated to a final volume of ∼175 ␮L. DCA and BCA were converted to their methyl esters (MDCA and MBCA, respectively) by adding 25 ␮L of diazomethane. Extracts of expired water were analyzed on a Hewlett-Packard 5890 GC/ECD (Palo Alto, CA). Analytes (MDCA and MBCA) were separated using a 30 m × 0.25 mm Supelco SPB-1 column with

P.N. Fitzsimmons et al. / Aquatic Toxicology 94 (2009) 186–194

a 1 ␮m bonded phase (Sigma–Aldrich). Instrumental conditions were: hydrogen carrier gas at a flow rate of 2 mL/min; nitrogen make-up gas at a flow rate of 30 mL/min; injection port temperature, 225 ◦ C; detector temperature, 300 ◦ C; initial oven temperature, 45 ◦ C; initial time, 1 min; rate, 13 ◦ C/min; final oven temperature, 140 ◦ C. Analytes were quantified using standard solutions prepared in MTBE. Extraction recoveries for DCA and BCA from spiked water samples ranged from 70 to 75%. DCA concentrations in samples of expired water were therefore corrected for recovery of internal standard (BCA). The method detection limit (MDL) for DCA in expired water was 30–45 ng/L. This was calculated for a 50 mL water sample producing a DCA peak area equal to 3 times the background interference measured in control (pre-dose water) samples. 2.8.2. Biological samples DCA concentrations in biological samples were measured using methods described by Schultz et al. (1999). Samples of blood, urine, bile, plasma, and plasma ultra-filtrate (50–350 ␮L) were diluted with de-ionized water to a volume of 350 ␮L and mixed with 25 ␮L of internal standard (BCA in MTBE, 36.0 ␮g/mL). The samples were then acidified with 25 ␮L of 50% H2 SO4 and extracted with 900 ␮L of MTBE. Aqueous and organic layers were separated by centrifuging for 10 min at 10,000 × g. Organic extracts were immediately transferred to glass GC vials and capped. DCA and BCA were then converted to their methyl esters by adding 25 ␮L of diazomethane through the rubber septum. Samples of muscle, liver, and kidney (1–2 g) were homogenized (Polytron tissue homogenizer, Brinkman Instruments, Westbury, NY) in de-ionized water for 2 min at 15,000 rpm. Muscle was mixed with water at a ratio of 1:10 (w:v) while liver and kidney were mixed at a ratio of 1:7. Samples of homogenate (350 or 500 ␮L, containing 50 mg of tissue) were then extracted as described above. The recovery of DCA and BCA from all biological samples was typically greater than 90% using this “standard” extraction procedure. Therefore, it was unnecessary to correct DCA concentrations for recovery of internal standard (BCA). The extraction procedure was modified for low dose blood samples in order to increase analytical sensitivity. Gains in sensitivity were accomplished by concentrating the sample extracts and minimizing background interference. Blood samples (100–400 ␮L) were centrifuged at 3000 × g for 5 min to separate plasma from red blood cells. The resulting plasma was measured and diluted to 350 ␮L with 0.1 M acetate buffer (pH 5.2). Samples were spiked with internal standard (25 ␮L of BCA in MTBE, 1.0 ␮g/mL), acidified with 1 drop of 50% H2 SO4 , and extracted with 350 ␮L of MTBE. Solvent extracts were concentrated under nitrogen to ∼75 ␮L in pre-marked GC vial inserts. DCA and BCA were then converted to their methyl esters by adding 25 ␮L of diazomethane diluted 1:10 in MTBE. Recovery of DCA and BCA from plasma using this “low dose” extraction procedure ranged from about 40 to 85%. In this case, DCA concentrations in low dose plasma samples were corrected for recovery of internal standard. All biological samples were analyzed by GC/ECD as described above. The MDL for DCA in biological samples collected during the inhalation and high dose injection experiments was determined by the lowest analytical standard employed for these sets of samples (0.010 ␮g/mL). Samples that produced a DCA peak area below the area of the lowest analytical standard could not be quantified. The resulting MDLs were calculated to be 13–15 ng/mL for blood, urine, and bile and 90 ng/g for muscle, kidney, and liver. The MDL for DCA in plasma samples extracted using the “low dose” extraction procedure was calculated to be 3–9 ng/mL. This was based on a DCA peak area equal to 3 times the background interference measured in control (pre-dose plasma) samples.

189

2.8.3. Biochemical and physiological parameters Blood chemistry profiles and the pH, pCO2 , and HCO3 − content of inspired and expired water were determined using an ABL 705 blood chemistry analyzer (Radiometer America Inc., Westlake, OH). All measurements were made at 37 ◦ C. Measured values for pH and pCO2 were then adjusted post-analysis to the physiological temperature of trout (11 ◦ C) using empirical relationships provided by the instrument manufacturer (Eqs. (1) and (3), respectively, in the ABL 705 reference manual, pp. 6–25). Concentrations of HCO3 − in blood, inspired water, and expired water were calculated from adjusted pH and pCO2 values (Eq. (4) in the reference manual, pp. 6–25). Ventilation volume (VVOL ), oxygen uptake efficiency (UE ), and oxygen consumption (VO2 ) were continuously monitored using an automated collection system (Carlson et al., 1989) and customized software. 2.9. Toxicokinetic analysis and data handling Comparisons of kinetic data between high and low dose injection groups were made possible by converting DCA concentrations in blood (the sampled compartment in high dose animals) to equivalent concentrations in plasma (the sampled compartment in low dose animals). This was accomplished by multiplying measured blood concentrations by the blood:plasma partitioning ratio, determined from in vitro studies to be 1.12 (see Section 3). Individual DCA plasma concentration–time profiles for injected fish were log-transformed to examine the kinetics of elimination. All plots exhibited an initial rapid drop in DCA concentration followed by a less rapid, log-linear decline. Non-compartmental analysis based on statistical moments was used to estimate mean residence time (MRT; h), whole-body clearance (CLB ; mL/h), renal clearance (CLR ; mL/h), and steady-state volume of distribution (VSS ; mL/kg) (PCNONLIN; Scientific Consulting, Inc., Apex, NC). The MRT was also used to calculate an effective elimination half-life (t1/2 , as 0.693MRT; Gibaldi and Perrier, 1982). This approach does not depend on the assignment of an explicit model structure but does assume first-order kinetics. A statistical analysis was performed on log-transformed parameter estimates assuming equal variances. Differences between parameter estimates for the different treatment groups were evaluated using the two-sample t test (p < 0.05). Differences in blood chemistry parameters between control and treated fish were evaluated at each sampling time using the twosample t test (p < 0.05). Differences in VVOL , UE , and VO2 between control and treated animals were evaluated by using the twosample t test (p < 0.05) to compare hourly means for each parameter. 3. Results 3.1. DCA kinetics 3.1.1. Branchial uptake and elimination DCA concentrations in plasma, urine, inspired water, and expired water during 72 h inhalation exposures are shown in Fig. 1. Gill uptake efficiency was low at all sampling times (1–5%) and tended to decrease at later time points. DCA concentrations in plasma increased slowly over time. Concentrations measured after 24, 48, and 72 h of exposure were 3, 5, and 12% of inspired water values, respectively. Measured concentrations of DCA in urine also increased over time but remained very low throughout each experiment (<1% of inspired water). DCA concentrations in expired water following a bolus i.a. injection (10 mg/kg) were low but detectable. Measured concentrations averaged 336 ± 9 ng/L (mean ± SE, n = 3) at 2 h post-injection and declined to 148 ± 34 ng/L by 8 h. For comparison, DCA concentrations in blood at the 2 and 8 h sampling times averaged 18.1 ± 2.7

190

P.N. Fitzsimmons et al. / Aquatic Toxicology 94 (2009) 186–194

Fig. 1. DCA concentrations measured in plasma (䊉), urine (), inspired water (), and expired water () from rainbow trout exposed to 173 ± 1.0 mg/L DCA during a 72 h inhalation exposure. Reported values are given as the mean ± SE; n = 4 trout, except for the 24, 30, and 54 h plasma values where n = 3 trout.

and 5.1 ± 1.2 ␮g/mL, respectively. For each fish, the total mass of DCA eliminated across the gills was <0.5% of the administered dose. Subsequent studies showed that most of an injected dose was eliminated by fish during the first 8 h post-injection (naïve animals; see below). 3.1.2. Tissue distribution DCA concentrations in plasma, muscle, kidney, liver, and bile following 72, 168, or 336 h of exposure are shown in Table 1. Data reported for the 72 h exposure were obtained from chambered animals while data given for the 168 and 336 h sampling times were from free-swimming fish. DCA concentrations in all tissues increased throughout 168 h but varied considerably among individual animals making the interpretation of this data difficult. The distribution of DCA is better understood when tissue concentrations are expressed as tissue:plasma concentration ratios to normalize for differences in uptake among fish (Table 1). Muscle:plasma and kidney:plasma concentration ratios increased throughout 336 h suggesting that fish had not yet achieved steadystate. The liver:plasma concentration ratio was very low at 72 h and then increased at later time points. All tissue:plasma concentration ratios at 168 and 336 h were 2–3 times higher than the fractional extra-cellular fluid volume of trout (approximately 0.19; Nichols, 1987), suggesting that DCA had been distributed in intracellular water. DCA concentrations in bile were low at all time points (<1% of plasma values). Assuming that the bile flow rate in trout is 3.0 ␮L/min/kg (Sanz et al., 1993), the mass of DCA eliminated in bile was calculated to be <0.001% of that absorbed across the gills.

Fig. 2. Log-transformed plasma elimination profiles for naïve (䊉) and DCA pretreated () trout following a bolus arterial injection of a high (5 mg/kg, A) or low (50 ␮g/kg, B) dose of DCA. Plasma was sampled at selected times over a 24 and 48 h period for the low and high dose groups, respectively. In high dose experiments, the actual doses were 4.4 mg/kg for naïve fish and 5.0 mg/kg for pre-treated fish. Data from naïve fish were normalized to a uniform dose of 5.0 mg/kg. Each data point is the sample mean ± SE; n = 3 (pre-exposed, high dose), 4 (naïve, high and low doses), or 5 (pre-exposed, low dose) trout.

3.1.3. Clearance of an injected dose A summary of kinetic parameters describing the elimination of DCA is presented in Table 2. DCA was rapidly cleared from the plasma of naïve fish following a bolus i.a. injection (Fig. 2A and B). The concentration of DCA was near analytical detection limits at 6 and 12 h post-injection in the low and high dose treatment groups, respectively. CLB rates in low dose fish were approximately 2 times greater than the rates measured in high dose animals. Estimated t1/2 values were correspondingly shorter in low dose fish.

Table 1 Distribution of DCA to plasma, bile, and selected tissues. Sample

72 h DCA concentration

Exposure water Plasma Muscle Kidney Liver Bile

173.1 (1.0) 20.51 (3.98) 4.40 (0.61) 7.07 (1.26) 0.18 (0.07) 0.03 (0.01)

168 h Tissue:plasma ratio

0.22 (0.03) 0.35 (0.04) 0.01 (0.01)

DCA concentration 172.2 (3.6) 37.87 (8.86) 14.65 (3.36) 26.78 (6.51) 14.71 (3.27) 0.10 (0.04)

336 h Tissue:plasma ratio

0.40 (0.04) 0.69 (0.07) 0.39 (0.04)

DCA concentration 171.4 (2.6) 21.74 (5.13) 12.09 (4.56) 16.46 (4.00) 9.41 (2.22) 0.13 (0.09)

Tissue:plasma ratio

0.61 (0.19) 0.74 (0.03) 0.44 (0.04)

Data reported for the 72 h exposure were obtained from fish that were exposed to DCA while confined to respirometer-metabolism chambers. Data given for the 168 and 336 h exposures are from free-swimming fish. Values are expressed as the sample means (SE) for n = 4 replicates for tissues, and n = 8–21 replicates for exposure water. Concentrations are reported in units of ␮g/mL for exposure water, plasma, bile, and urine; and ␮g/g (wet wt) for muscle, kidney, and liver.

P.N. Fitzsimmons et al. / Aquatic Toxicology 94 (2009) 186–194

191

Table 2 Kinetic parameters for naïve and pre-treated trout injected with a high (5 mg/kg) or low (50 ␮g/kg) dose of DCA. Parameter

High dose Naïve (n = 4)

Parameters calculated from plasma data only AUC0→∞ (␮g/h) 49.1 (3.9) VSS (mL/kg) 468.8 (32.7) 65.7 (4.6) VSS (% TBW)a CLB (mL/h/kg) 91.9 (7.0) t1/2 (h) 3.5 (0.1) Parameter

Low dose Pre-treatment (n = 3)

Naïve (n = 4)

Pre-treatment (n = 5)

226.8 (41.2)b 393.9 (28.1) 55.2 (3.9) 23.5 (3.6)b 13.2 (3.1)b

0.25 (0.04) 498.0 (72.1) 69.8 (10.1) 167.1 (25.1)b 2.2 (0.5)b

0.62 (0.08)c 486.0 (17.5) 68.1 (2.4) 67.5 (7.6)c 5.3 (0.8)c

High dose Naïve (n = 3)

Parameters calculated from trout that provided both plasma and urine data 98.3 (4.1) CLB (mL/h/kg) CLR (mL/h/kg) 0.4 (0.1) 97.9 (4.1) CLM (mL/h/kg)d

Pre-treatment (n = 2) 21.4 (5.1)b 0.4 (0.2) 21.0 (4.9)

Parameters were calculated from measured DCA concentrations in plasma and urine from the following sampling times: high dose – naïve, 0–8 h; high dose – pre-treatment, 0–48 h; low dose – naïve, 0–6 h; low dose – pre-treatment, 0–8 h. Kinetic parameters are reported as the mean (SE) of values for individual trout and are based on the measured amount of DCA given to each fish. a TBW = total body water. Comparisons are based on the average value (714 mL/kg) for freshwater teleosts given by Thorson (1961). b Significantly different than high dose – naïve (p < 0.05). c Significantly different than low dose – naïve (p < 0.05). d CLM values were calculated as the difference between CLB and CLR values.

Pre-treatment with DCA substantially reduced the CLB of a subsequently injected dose (Fig. 2A and B). In high dose fish, DCA could be measured in the plasma of pre-treated animals throughout 48 h post-injection. Pre-treatment with DCA caused a nearly 4-fold increase in t1/2 and a 4.5-fold increase in the area under the plasma concentration–time curve (extrapolated to infinity; AUC0→∞ ). Pretreatment with DCA also had a large impact on the elimination kinetics in low dose fish, although the magnitude of this effect was less pronounced. CLB rates for pre-treated fish were 2.5 times lower than the rates calculated for naïve animals, while estimates of t1/2 and AUC0→∞ in pre-treated animals were 2.5 times greater. In high dose animals, renal clearance (CLR ) of DCA accounted for <2% of CLB in both naïve and DCA pre-treated fish. As indicated previously, elimination by branchial and biliary routes was negligible. Assuming that the balance of CLB was due to metabolism, the metabolic clearance (CLM ) of DCA can be calculated as CLB minus CLR . Calculated in this manner, the mean ± SE of CLM values for naïve trout was 97.9 ± 4.1 mL/h/kg while that for pre-treated animals was 21.0 ± 4.9 mL/h/kg (Table 2). There were no significant differences in VSS among treatment groups. VSS estimates ranged from about 400 to 500 mL/kg, which is equivalent to 60–70% of total body water. 3.2. Blood:plasma partitioning, plasma binding Samples of whole blood were spiked with DCA at two concentrations (1 and 10 ␮g/mL). No differences were observed between treatment groups so data from both groups were combined. The measured blood:plasma concentration ratio was 0.89 ± 0.01 (mean ± SE, n = 3), indicating nearly equal distribution of DCA between red blood cells and plasma. The fu of DCA in plasma was 1.00 ± 0.02. 3.3. Biochemistry and physiology 3.3.1. Control experiments Blood chemistry and respiratory physiology data for control fish were in good agreement with previously reported values for rainbow trout (McKim et al., 1987; Barron et al., 1987; Weber, 1991). Ranges for measured parameters in untreated fish were: VVOL , 85–121 mL/min/kg; UE , 50–75%; VO2 , 37–50 mg/h/kg;

blood pH, 7.9–8.1; Hct, 21–24%; osmolarity, 295–302 mOsm; Na+ , 144–147 mM; K+ , 2.4–3.0 mM; Ca2+ , 1.3–1.4 mM; Cl− , 120–126 mM; HCO3 − , 7.8–14.6 mM; lactate, 0.8–1.0 mM; glucose, 6.8–8.3 mM; pCO2 , 3.8–4.3 mm Hg. 3.3.2. Inhalation experiments Exposure to DCA in water had no affect on the pH, pCO2 , ion composition (Na+ , K+ , Ca2+ , Cl− , HCO3 − ), osmolarity, or hematocrit of blood. Respiratory physiology parameters (VVOL , UE , or VO2 ) were similarly unaffected, as were changes in pH, pCO2 , and HCO3 − concentration (inspired vs. expired water) that are normally associated with chambered animals. Exposure to DCA in water did, however, lower circulating levels of blood lactate (Fig. 3A). Lactate concentrations declined slowly throughout each exposure resulting in a mean value that was significantly different from the control mean at the last sampling time (72 h). 3.3.3. Injection experiments The injection of trout with DCA had no effect on respiratory physiology parameters. Blood chemistry parameters were also unaffected, except for a transient decrease in blood lactate (Fig. 3B). In contrast to the slow decline in lactate concentrations observed during inhalation experiments, blood lactate decreased rapidly following the injection and then increased over the next 24 h as DCA was cleared from blood. 4. Discussion The primary goals of this investigation were to characterize the branchial uptake, tissue distribution, and elimination of DCA in rainbow trout, and to provide insight into mechanisms controlling these processes. An additional effort was made to evaluate potential effects on respiratory physiology, acid–base balance, and intermediary metabolism. The DCA concentration in water employed during inhalation exposures (nominally 200 mg/L) was somewhat lower than that shown previously to alter liver histology in Japanese medaka (500 mg/L; McHugh Law et al., 1998). Kinetic studies in mammals have tended to employ injected doses based on therapeutic treatment levels (50 mg/kg or higher). Injection studies with trout were conducted at much lower dose levels (0.05 and 5.0 mg/kg), in part to increase the environmental relevance of the

192

P.N. Fitzsimmons et al. / Aquatic Toxicology 94 (2009) 186–194

Fig. 3. Blood lactate concentrations measured in control () and DCA treated (䊉) trout. Animals represented in (A) were exposed to 173 ± 1.0 mg/L (see measured values shown in Fig. 1) in water for 72 h while those represented in (B) were given a bolus arterial injection of 5 mg/kg body weight. Each value represents the mean ± SE; n = 4 trout. Values denoted by an asterisk (*) are significantly different from values for control animals at the same sampling time.

findings and also because the activities of enzymes responsible for DCA clearance were anticipated to be lower. Because DCA exhibits high water solubility, natural exposures to fish are likely to occur primarily via uptake across the gills. In the present study, DCA was found to be poorly absorbed across the gill epithelium. This is not surprising, given that DCA is almost entirely ionized (>99.99%; pKa = 1.26) at the pH (7.8) of exposure water. Nevertheless, continuous exposures to DCA resulted in low (1–5%) but measurable rates of branchial uptake. McKim et al. (1985) used chambered trout to measure the branchial uptake of 14 organic compounds and related this information to each compound’s octanol–water partition coefficient. The lowest uptake rates (5–10%) were observed for very low (<1.0) log KOW compounds, and it was suggested that branchial absorption of small (molecular weight <100), highly polar compounds occurs primarily by diffusion through “aqueous pores” in the gill epithelium. A similar proposal, involving chemical diffusion through tight junctions between cells (i.e., paracellular transport) was advanced by Saarikoski et al. (1986) to explain measured rates of uptake for several ionized compounds in guppies (Poecilia reticulate). More recently, Erickson et al. (2006a,b) used data from studies of chambered trout to show that ionized forms of weak acids and bases can contribute to branchial uptake by maintaining high gradients of neutral molecules across epithelial membrane barriers. This mech-

anism may be sufficient to support relatively high levels of branchial absorption for some compounds despite the fact that they are highly ionized in inspired water. The pKa values for compounds studied by Erickson et al. (2006a,b) ranged from 4.74 to 8.62. The mechanism by which DCA is taken up across trout gills is unknown. Based on its very low pKa value, however, it seems unlikely that concentrations of neutral diffusing species within and adjacent to the gill epithelium would be sufficient to support branchial uptake. The molecular weight of DCA (128.9) is somewhat larger than the cutoff for aqueous diffusion given by McKim et al. (1985); nevertheless, diffusion through aqueous channels (“pores” or paracellular junctions) appears to be the most likely mechanism by which DCA crosses gill membranes. Bolus dosing experiments were conducted to determine the routes and rates of DCA elimination by trout. Measured rates of renal clearance (CLR ; 0.4 mL/h/kg) were considerably lower than reported urine flow rates for trout (2.0–5.9 mL/h/kg; Nichols et al., 1991; Curtis and Wood, 1991). Previously, Schultz et al. (1999) found that renal clearance of DCA in rats (3.1 mL/h/kg; adjusted for protein binding) was lower than the urine flow rate (8.3 mL/h/kg) and it was proposed that DCA is taken up within the kidney by tubular re-absorption. The results of the present study suggest that a similar mechanism may exist in trout. The percentage contribution of CLR to whole-body clearance (CLB ) was less than 2%, indicating that most of the DCA cleared by trout was eliminated by some other route. Elimination of DCA across the gills following an injected dose was also found to be negligible (<0.5% of CLB ). No attempt was made to measure fecal elimination of DCA. In waterborne exposures, however, the mass of DCA eliminated in bile was a small (1%) fraction of that taken up across the gills. These findings are consistent with earlier work on mammals and provide indirect evidence that biotransformation is the principal route by which trout eliminate DCA. In rodents, a substantial amount of DCA is metabolized to CO2 (up to 50%; Lin et al., 1993; Xu et al., 1995; Gonzalez-Leon et al., 1997, 1999). Additional metabolites are identical to endogenously produced compounds (e.g., glycolate, oxalate, glyoxylate; Larson and Bull, 1992; Lin et al., 1993; Xu et al., 1995). In either case, these metabolic products can be quantified only by using radiolabeled starting material. Similar studies would be required to directly demonstrate DCA metabolism in trout. Metabolic clearance of DCA in mammals is largely mediated by GST␨ (Tong et al., 1998). The activity of this enzyme has been shown to exhibit saturable kinetics resulting in an inverse nonlinear relationship between CLB and the administered dose of DCA. For example, Saghir and Schultz (2002) observed a 25-fold decrease in CLB when the dose was increased across two orders of magnitude. In the present study, a comparable increase in dose (50 ␮g/kg to 5 mg/kg) was accompanied by a 2-fold decrease in CLB . An important aspect of the GST␨ pathway in mammals is that it exhibits suicide inactivation by DCA during repeated or extended exposures (James et al., 1997; Anderson et al., 1999). This inactivation causes a reduction in GST␨ activity (and by extension CLB ) and increases the biological persistence of DCA remaining in the body or introduced during subsequent exposures. In rainbow trout, pre-exposure to DCA in water increased the biological persistence of a subsequently administered i.a. dose. This change resulted in a significant decrease in CLB and increases in AUC0→∞ and t1/2 . Recently, Schultz et al. (2007) reported that pre-treatment of Japanese medaka with DCA caused a significant decrease in the rate of elimination of a related dihaloacetic acid (bromochloroacetic acid) following dietary exposure. Taken together, these findings suggest that a GST␨-like metabolic pathway is responsible for DCA metabolism by fish.

P.N. Fitzsimmons et al. / Aquatic Toxicology 94 (2009) 186–194

The results of the present study provide an opportunity to compare CLB rates in trout and rats given the same intravascular dose of DCA. The estimated CLB for naïve rats given 5.0 mg/kg DCA was about 5.3 L/h/kg (Saghir and Schultz, 2002) while that calculated for naïve trout was about 0.1 L/h/kg. This comparison is complicated, however, by the fact that CLB in rats exceeds the hepatic blood flow rate (3.13 L/h/kg; Saghir and Schultz, 2002). Based on this finding it was suggested that DCA clearance by rats involves substantial extra-hepatic metabolism. In contrast, the estimated CLB for trout is about 25% of estimated liver blood flow (0.42 L/h/kg; based on data given by Nichols et al., 2004). It is therefore possible that the metabolism of DCA by trout occurs entirely within liver tissue. In contrast to the low rate of transport measured across the gills, results from the current study suggest that DCA can readily penetrate cell membranes within the body. Steady-state volumes of distribution (VSS ) estimated from plasma clearance profiles were equivalent to about 60–70% of total body water. In addition, blood lactate concentrations declined rapidly following exposure to an injected dose. The likely mechanism by which this effect occurs (inhibition of pyruvate dehydrogenase kinase; see below) requires DCA to cross both cell and mitochondrial membranes. In mammals, the transport of DCA across cell membranes is facilitated by a monocarboxylate transport protein. Halestrap (1975) studied the kinetics and substrate specificity of the pyruvate transport protein in rat liver mitochondria and showed that DCA competes with pyruvate, inhibiting its uptake. The transport of DCA into rat liver mitochondria was found to be largely dependent on this carrier protein and was inhibited by the monocarboxylate transport inhibitor, ␣-cyano4-hydroxycinnamate. Laberee and Milligan (1999) showed that uptake of lactate in rainbow trout white muscle is mediated, at least in part, by a low-affinity, high-capacity transport protein. Lactate uptake was partially inhibited at high concentrations of pyruvate, indicating the involvement of a monocarboxylate carrier. We are not aware of any other reports of monocarboxylate transporters in fish. Nevertheless, the similarity of our results to those obtained in studies with mammals suggests that the membrane transport of DCA in trout may be facilitated by a specific transport protein. To directly assess the distribution of DCA in trout we measured concentrations in plasma and tissues following extended inhalation exposures. DCA concentrations in plasma increased throughout 72 h chambered exposures. In longer (168 or 366 h) free-swimming exposures, plasma concentrations appeared to level off at values considerably lower than measured DCA concentrations in water. As indicated previously, DCA clearance by trout appears to be due primarily to metabolism. DCA concentrations in plasma at 168 or 366 h are likely, therefore, to reflect a dynamic steady-state that results from competing rates of branchial uptake and metabolic clearance. Assuming that DCA is distributed in both extra- and intracellular water, we may calculate theoretical steady-state tissue:plasma concentration ratios based on the ratio of the water content of a tissue (generally about 75%; Nichols et al., 1991) to that of plasma (∼90%). This corresponds to a steady-state tissue:plasma concentration ratio of about 0.85. Tissue:plasma concentration ratios measured in kidney and muscle at 336 h were in reasonably good agreement with theoretical estimates while those determined at 72 and 168 h (especially for muscle) were considerably lower. This finding appears to contradict the earlier suggestion (based on kinetic evaluations of DCA injection data as well as observed effects on lactate) that DCA penetrates cell membranes. As indicated previously, however, fitted VSS values from bolus injection studies were only 60–70% of total body water. Moreover, fish exposed throughout 72 h were unlikely to have achieved an internal steady-state condition since DCA concentrations in blood were increasing throughout this time period.

193

In contrast to measured concentrations of DCA in muscle or kidney, DCA concentrations in liver after 72 h of inhalation exposure were extremely low, resulting in low tissue:plasma concentration ratios. Liver:plasma ratios increased at later time points but remained lower than the values exhibited by muscle or kidney. The low liver:plasma ratio measured at 72 h suggests that DCA was being actively metabolized within this tissue. Higher ratios at later time points may have been due to suicide inactivation of the enzyme responsible for DCA clearance, which would have allowed DCA concentrations to increase. The fact that liver:plasma ratios remained lower than muscle:plasma or kidney:plasma values at later time points suggests, however, that this inactivation was incomplete. Alternatively, DCA may be metabolized by more than one pathway, at least one of which is not subject to suicide inactivation. In mammals, high doses of DCA (typically 50 mg/kg or higher) lower circulating levels of lactate and glucose (Crabb et al., 1981; Stacpoole, 1989). The mechanism by which this effect occurs involves the inhibition of pyruvate dehydrogenase kinase, which causes pyruvate dehydrogenase to remain in its unphosphorylated, catalytically active state. This results in a shift in cellular metabolism away from anaerobic utilization of pyruvate (forming lactate) and toward the production of acetyl coenzyme A. Under normal conditions, blood lactate is taken up by the liver and synthesized into glucose. This glucose is released to the bloodstream and taken up by working muscle. Anaerobic metabolism of glucose by muscle tissue then completes what is known as the Cori cycle. By reducing lactate concentrations in blood, DCA effectively short-circuits this cycle. Additional reductions in blood glucose, particularly in diabetics, arise because of a systemic shift toward carbohydrate metabolism. In the present study, a relatively low dose of DCA (5 mg/kg) caused a transient decrease in blood lactate concentration. A progressive decline in blood lactate was also observed during continuous waterborne exposures to 200 mg/L DCA. While it is reasonable to suggest that DCA reduces blood lactate in trout by the same general mechanism as in mammals, the precise nature of these changes and their consequences for the animal remain unclear. In fish, the Cori cycle plays little role in the metabolic fate of lactate (Milligan, 1996). Instead, most (80–85%) of the lactate produced by white muscle is retained within the muscle and used for glycogen re-synthesis (Milligan and Wood, 1986; Turner et al., 1983). Lactate that does enter the general circulation is primarily consumed as an energy substrate for aerobic metabolism. Complicating matters further, experimental procedures used in this study (including transection of the spinal cord) have the effect of paralyzing the major white muscle groups. In previous studies with trout, circulating concentrations of lactate were shown to increase by a factor of 10 following exhaustive exercise (Milligan and Wood, 1987; Pagnotta et al., 1994). Because white muscle does not possess an appreciable ability to aerobically metabolize pyruvate, DCA may have little or no impact on these exercise-related changes in blood lactate. The source of lactate measured under the “resting” conditions of this study is less clear, and it is possible that much of it originates from low levels of anaerobic metabolism in largely aerobic tissues. Additional research, perhaps involving exercised animals, is required to determine where DCA impacts on cellular metabolism occur in fish and whether these effects are toxicologically important. To summarize, HAAs produced by chlorine disinfection of municipal drinking water represent a potentially important class of environmental contaminants. In this investigation we examined the kinetics and effects of an important HAA (DCA) in rainbow trout. The efficiency with which DCA is taken up across the gills

194

P.N. Fitzsimmons et al. / Aquatic Toxicology 94 (2009) 186–194

is low. Once it enters the body, however, its fate and effects are highly reminiscent of earlier findings in studies with mammals. These observations suggest that DCA is metabolized by a GST␨like enzyme which exhibits suicide inactivation during prolonged exposures, and that high levels of DCA lower blood lactate through a mechanism that involves inhibition of pyruvate dehydrogenase kinase. Acknowledgements We thank Dr. Rodney Johnson and Dr. Anthony DeAngelo for critical reviews of this manuscript. In addition, we thank Dr. Irvin Schultz for helpful insights regarding the analysis and interpretation of kinetic data sets. References Anderson, W.B., Board, P.G., Gargano, B., Anders, M.W., 1999. Inactivation of glutathione transferase zeta by dichloroacetic acid and other fluorine-lacking ␣-haloalkanoic acids. Chem. Res. Toxicol. 12, 1144–1149. Barron, M.G., Tarr, B.D., Hayton, W.L., 1987. Temperature-dependence of cardiac output and regional blood flow in rainbow trout, Salmo gairdneri Richardson. J. Fish Biol. 31, 735–744. Bersin, R.M., Stacpoole, P.W., 1997. Dichloroacetate as metabolic therapy for myocardial ischemia and failure. Am. Heart J. 134, 841–855. Brown, J.A., Gore, D.C., 1996. In vivo metabolic response of glucose to dichloroacetate in humans. J. Surg. Res. 61, 391–394. Carlson, R.W., Lien, G.W., Holmen, B.A., 1989. An Automated Monitoring System for Fish Physiology and Toxicology. EPA/600/3-89/011. U.S. Environmental Protection Agency, Duluth, MN. Crabb, D.W., Yount, E.A., Harris, R.A., 1981. The metabolic effects of dichloroacetate. Metabolism 30, 1024–1039. Curtis, J.B., Wood, C.M., 1991. The function of the urinary bladder in vivo in the freshwater rainbow trout. J. Exp. Biol. 155, 567–583. DeAngelo, A.B., Daniel, F.B., Most, B.M., Olson, G.R., 1996. The carcinogenicity of dichloroacetic acid in the male fischer 344 rat. Toxicology 114, 207–221. Erickson, R.J., McKim, J.M., Lien, G.J., Hoffman, A.D., Batterman, S.L., 2006a. Uptake and elimination of ionizable organic chemicals at fish gills. I. Model formulation, parameterization, and behavior. Environ. Toxicol. Chem. 25, 1512– 1521. Erickson, R.J., McKim, J.M., Lien, G.J., Hoffman, A.D., Batterman, S.L., 2006b. Uptake and elimination of ionizable organic chemicals at fish gills. II. Observed and predicted effects of pH, alkalinity, and chemical properties. Environ. Toxicol. Chem. 25, 1522–1532. Gibaldi, M., Perrier, D., 1982. Pharmacokinetics, 2nd ed. Marcel Dekker, New York, NY, pp. 409–417. Gonzalez-Leon, A., Schultz, I.R., Xu, G., Bull, R.J., 1997. Pharmacokinetics and metabolism of dichloroacetate in the F344 rat after prior administration in drinking water. Toxicol. Appl. Pharmacol. 146, 189–195. Gonzalez-Leon, A., Merdink, J.L., Bull, R.J., Schultz, I.R., 1999. Effect of pre-treatment with dichloroacetic or trichloroacetic acid in drinking water on the pharmacokinetics of a subsequent challenge dose in B6C3F1 mice. Chem. Biol. Interact. 123, 239–253. Halestrap, A.P., 1975. The mitochondrial pyruvate carrier. Kinetics and specificity for substrates and inhibitors. Biochem. J. 148, 85–96. Herren-Freund, S.L., Pereira, M.A., Khoury, M.D., Olson, G., 1987. The carcinogenicity of trichloroethylene and its metabolites, trichloroacetic acid and dichloroacetic acid, in mouse liver. Toxicol. Appl. Pharmacol. 90, 183–189. Hunter III, S.E., Blanton, M.R., Rogers, E.H., Mole, M.L., Andrews, J., Chernoff, N., 2006. Short-term exposures to dihaloacetic acids produce dysmorphogenesis in mouse conceptuses in vitro. Reprod. Toxicol. 22, 443–448. James, M.O., Cornett, R., Yan, Z., Henderson, G.N., Stacpoole, P.W., 1997. Glutathionedependent conversion to glyoxylate, a major pathway of dichloroacetate biotransformation in hepatic cytosol from humans and rats, is reduced in dichloroacetate-treated rats. Drug Metab. Dispos. 25, 1223–1227. James, M.O., Yan, Z., Cornett, R., Jayanti, R.C.V.M., Henderson, G.N., Davydova, N., Katovich, M.J., Pollock, B., Stacpoole, P.W., 1998. Pharmacokinetics and metabolism of [14 C]dichloroacetate in male Sprague–Dawley rats. Drug Metab. Dispos. 26, 1134–1143. Krasner, S.W., McGuire, M.J., Jacangelo, J.G., Patania, N.L., Reagan, K.M., Aieta, E.M., 1989. The occurrence of disinfection by-products in US drinking water. J. Am. Water Works Assoc. 81, 41–53. Laberee, K., Milligan, C.L., 1999. Lactate transport across sarcolemmal vesicles isolated from rainbow trout white muscle. J. Exp. Biol. 202, 2167–2175.

Larson, J.L., Bull, R.J., 1992. Metabolism and lipoperoxidative activity of trichloroacetate and dichloroacetate in rats and mice. Toxicol. Appl. Pharmacol. 115, 268–277. Lin, E.L.C., Mattox, J.K., Daniel, F.B., 1993. Tissue distribution, excretion and urinary metabolites of dichloroacetic acid in the male fischer 344 rat. J. Toxicol. Environ. Health 38, 19–32. McHugh Law, J., Lopez, L., DeAngelo, A.B., 1998. Hepatotoxicity of the drinking water disinfection by-product, dichloroacetic acid, in the medaka small fish model. Toxicol. Lett. 94, 19–27. McKim, J.M., Goeden, H.M., 1982. A direct measure of the uptake efficiency of a xenobiotic chemical across the gills of brook trout (Salvelinus fontinalis) under normoxic and hypoxic conditions. Comp. Biochem. Physiol. 72C, 65–74. McKim, J., Schmieder, P., Veith, G., 1985. Absorption dynamics of organic chemical transport across trout gills as related to octanol–water partition coefficient. Toxicol. Appl. Pharmacol. 77, 1–10. McKim, J.M., Schmieder, P.K., Carlson, R.W., Hunt, E.P., Niemi, G.J., 1987. Use of respiratory–cardiovascular responses of rainbow trout (Salmo gairdneri) in identifying acute toxicity syndromes in fish: Part 1. Pentachlorophenol, 2,4dinitrophenol, tricaine methanesulfonate and 1-octanol. Environ. Toxicol. Chem. 6, 295–312. Milligan, C.L., 1996. Metabolic recovery from exhaustive exercise in rainbow trout. Comp. Biochem. Physiol. 113A, 51–60. Milligan, C.L., Wood, C.M., 1986. Tissue intracellular acid–base status and the fate of lactate after exhaustive exercise in the rainbow trout. J. Exp. Biol. 123, 123–144. Milligan, C.L., Wood, C.M., 1987. Regulation of blood oxygen transport and red cell pHi after exhaustive activity in rainbow trout (Salmo gairdneri) and starry flounder (Platichthys stellatus). J. Exp. Biol. 133, 263–282. Nichols, D.J., 1987. Fluid volumes in rainbow trout, Salmo gairdneri: application of compartmental analysis. Comp. Biochem. Physiol. 87A, 703–709. Nichols, J.W., McKim, J.M., Lien, G.J., Hoffman, A.D., Bertelsen, S.L., 1991. Physiologically based toxicokinetic modeling of three waterborne chloroethanes in rainbow trout (Oncorhynchus mykiss). Toxicol. Appl. Pharmacol. 110, 374–389. Nichols, J.W., Fitzsimmons, P.N., Whiteman, F.W., Dawson, T.D., Babeu, L., Juenemann, J., 2004. A physiologically based toxicokinetic model for dietary uptake of hydrophobic organic compounds by fish. I. Feeding studies with 2,2 ,5,5 tetrachlorobiphenyl. Toxicol. Sci. 77, 206–218. Pagnotta, A., Brooks, L., Milligan, L., 1994. The potential regulatory roles of cortisol in recovery from exhaustive exercise in rainbow trout. Can. J. Zool. 72, 2136–2146. Richardson, S.D., Simmons, J.E., Rice, G., 2002. Disinfection byproducts: the next generation. Environ. Sci. Technol. 36, 198A–205A. Saarikoski, J., Lindstrom, R., Tyynelä, M., Viluksela, M., 1986. Factors affecting the absorption of phenolics and carboxylic acids in the guppy (Poecilia reticulate). Ecotoxicol. Environ. Saf. 11, 158–173. Saghir, S.A., Schultz, I.R., 2002. Low-dose pharmacokinetics and oral bioavailability of dichloroacetate in naïve and GST␨-depleted rats. Environ. Health Perspect. 110, 757–763. Sanz, A., Cardenete, G., Hidalgo, F., Garcia-Gallego, M., 1993. Study of biliary secretion in trout (Oncorhynchus mykiss) by cannulation of bile duct. Comp. Biochem. Physiol. 104A, 525–529. Schultz, I.R., Merdink, J.L., Gonzalez-Leon, A., Bull, R.J., 1999. Comparative toxicokinetics of chlorinated and brominated haloacetates in F344 rats. Toxicol. Appl. Pharmacol. 158, 103–114. Schultz, I.R., Reed, S., Pratt, A., Skillman, A.D., 2007. Quantitative oral dosing of water soluble and lipophilic contaminants in the Japanese medaka (Oryzias latipes). Comp. Biochem. Physiol. 145C, 86–95. Stacpoole, P.W., 1989. The pharmacology of dichloroacetate. Metabolism 38, 1124–1144. Stacpoole, P.W., Greene, Y.J., 1992. Dichloroacetate. Diabetes Care 15, 785–791. Stacpoole, P.W., Barnes, C.L., Hurbanis, M.D., Cannon, S.L., Kerr, D.S., 1997. Treatment of congenital lactic acidosis with dichloroacetate. Arch. Dis. Child. 77, 535–541. Thorson, T., 1961. The partitioning of body water in osteichthyes: phylogenetic and ecological implications in aquatic vertebrates. Biol. Bull. 120, 238–254. Tong, Z., Board, P.G., Anders, M.W., 1998. Glutathione transferase zeta-catalyzed biotransformation of dichloroacetic acid and other ␣-haloacids. Chem. Res. Toxicol. 11, 1332–1338. Turner, J.D., Wood, C.M., Clark, D., 1983. Lactate and proton dynamics in the rainbow trout (Salmo gairdneri). J. Exp. Biol. 104, 247–268. Uden, P.C., Miller, J.W., 1983. Chlorinated acids and chloral in drinking water. J. Am. Water Works Assoc. 75, 524–527. Ward, K.W., Rogers, E.H., Hunter III, S.E., 2000. Comparative pathogenesis of haloacetic acid and protein kinase inhibitor embryotoxicity in mouse whole embryo culture. Toxicol. Sci. 53, 118–126. Williams, D.T., LeBel, G.L., Benoit, F.M., 1997. Disinfection by-products in Canadian drinking water. Chemosphere 34 (2), 299–316. Weber, J.M., 1991. Effect of endurance swimming on the lactate kinetics of rainbow trout. J. Exp. Biol. 158, 463–476. Wolf, K., 1963. Physiological salines for fresh-water teleosts. Prog. Fish Cult. 25, 135–140. Xu, G., Stevens, D.K., Bull, R.J., 1995. Metabolism of bromodichloroacetate in B6C3F1 mice. Drug Metab. Dispos. 23, 1412–1416.