TOXICOLOGY

AND APPLIED

PHARMACOLOGY

117,65-74

( 1992)

Dietary 2,3,7,8-Tetrachlorodibenzofuran in Rainbow Trout: Accumulation, Disposition, and Hepatic Mixed-Function Oxidase Enzyme Induction’ DEREK C. G. MUIR,~ ALVIN

L. YARECHEWSKI,

DONALD

A. METNER,

AND W. LYLE

LOCKHART

Department of Fisheries and Oceans, Freshwater Institute, 501 University Crescent, Winnipeg, Canada R3T 2N6 Received February 19, 1992; accepted July 27, 1992

haust), and chlorine-bleached kraft pulp and paper mill effluents (Fielder et al., 1990). In Great Lakes lake trout (Salv&us namaycush), mean TCDF concentrations (whole fish basis) ranged from 11.3 to 42.3 ng kg-’ (DeVault et al., 1989) but elevated levels of TCDF (up to 400 ng kg-‘) have been found in bottom-feeding fish below pulp mills using chlorine bleaching (Whittle et al., 1990; USEPA, 199 1). Food chain transfer has been recognized as the most important pathway of accumulation of hydrophobic organochlorine pollutants such as PCBs (Thomann and Connolly, 1984) and 2,3,7,8-TCDD (Cook et al., 1990; Endicott et al., 1990) in piscivorous fish. For very hydrophobic compounds the assimilation efficiency (a) from ingested food and depuration rates (Q are critical pharmacokinetic parameters in the prediction of food chain transfer (Thomann, 1989). There have been relatively few measurements of cxand kD for TCDF in fish (Opperhuizen and Sijm, 1990) and the limited published data set shows a wide range of values. Kuehl et al. ( 1986) found that field-exposed carp (Cyprinius carpio), held in the laboratory, had depurated 82% of TCDF residues in 336 days, while studies with rainbow trout fry (Oncorhynchus mykiss) and guppies (Poecilia reticulata) found kD’s ranging from 0.062 to 2.6 day-’ corresponding to t1,2’sof 11 to < 1 day (Opperhuizen and Sijm, 1990; Merhle et al., 1988). Catfish (Ictalurus punctatus) and largemouth bass (Micropterus salmoides) exposed to TCDF in pulp mill effluent in experimental streams had tI12’s of 11 and 58 days, respectively (NCASI, 1991). Loonen et al. (1992) reported an Q of 3.6% for TCDF in guppies, while studies with 2,3,4,7,8-pentachloro-dibenzofuran (PnCDF) yielded (Yvalues of 41 to 44% following 30-day dietary exposures (Muir et al., 1990). The range of tt12’s reported for TCDF and other 2,3,7,8substituted PCDFs has led some authors to question whether first-order rate processes (i.e., independent of initial concentration in the organism and in the food or water) can be applied to these compounds (Opperhuizen and Sijm, 1990). Opperhuizen and Schrap (1988) have observed deviations from the first-order kinetics rate model (Bruggeman et al., 198 1) in the uptake of PCB congeners from food by guppies. An objective of this study was to test the assumptions of the

Dietary 2,3,7,8-Tetrachlorodibenzofuran in Rainbow Trout: Accumulation, Disposition, and Hepatic Mixed-Function Oxidase Enzyme Induction. MUIR, D. C. G., YARECHEWSKI, A. L., METNER, D. A., AND LOCKHART, W. L. (1992). Tuxicof. Apgl. Pharmacol. 117,65-74.

Juvenile rainbow trout (Oncorhynchus mykiss) were exposed to dietary 2,3,7,8-[3H]tetrachlorodibenzofuran (TCDF) (0.36 to 42.8 ng g-‘) and accumulation, tissue distribution, biotransformation, and hepatic monooxygenase enzyme (MO) induction were studied. The assimilation efficiency of TCDF ranged from 49 to 62% in 30-day exposures and was independent of the TCDF level in the diet. Depuration half-lives (whole body) of TCDF following 30-day exposure ranged from 40 to 77 days and were significantly more rapid in fish exposed to 42.8 ng g-l. Liver somatic index (LSI) and rate of increase in liver weight were elevated in fish exposed to 42.8 ng g-r TCDF compared to controls. Exposure to 9.2 ng g-r TCDF in the diet for 140 days also resulted in higher LSI values, as well as increased mortality (16%), but had no significant effects on growth. [3H]TCDF was found mainly in the carcass (63-74%) and GI tract (18-31%), with lesser amounts in liver (0.6-2.3%) during the 140-day exposure, primarily (>98%) in the form of the parent compound. Radioactivity in bile was found mainly as a single polar transformation product by reverse-phase HPLC. Glucuronidase hydrolysis yielded a product with the retention time expected of hydroxylated TCDF, suggesting the presence of a glucuronide conjugate. MO enzyme induction measured by ethoxyresorufin0-deethylase (EROD) activity in liver (postmitochondrial supernatant) was 137.5 and 15 times higher than that in control fish after 30 days dietary exposure to 42.8 and 9.2 ng g-‘, respectively. EROD activities were correlated with TCDF concentrations in liver (R* = 0.59, N = 45). 0 1992 Academic press, h.

The major polychlorinated dibenzofuran (PCDF) congener found in most freshwater and marine fish is 2,3,7,8tetrachlorodibenzofuran (TCDF) (Rappe et al., 1987; Devault et al., 1989). Sources of TCDF include PCB fluids, combustion processes (waste incineration, automobile ex’ Presented in part at the I1 th Annual Meeting of the Society of Environmental Toxicology and Chemistry, Arlington, VA, November 1990. ’ To whom correspondence should be addressed. 65

0041-008X/92 $5.00 Copyright 0 1992 by Academic Press, Inc. All rights of reproduction in any form reserved.

66

MUIR

first-order kinetic model by examining assimilation and depuration over a wide range of TCDF concentrations in food. A second objective of this study was to examine the relationship between TCDF concentrations and induction of cytochrome P450 hepatic monooxygenase (MO) activity in rainbow trout over a range of tissue concentrations and exposure times. A single intraperitoneal dose of 2,3,7,8-TCDF at 3.1 pg kg-’ produced a strong, sustained induction of P450E mRNA (P450 IA 1) and ethoxyresorufin-O-deethylase (EROD) activity in scup (Stenotomus chrvsops) (Hahn et al., 1989). We have previously observed sustained EROD activity, following dietary exposure to PnCDF, at whole-body concentrations of about 1 ng g-’ (Muir et al., 1990). METHODS Chemicals ‘H-labeled 2,3,7&TCDF having specific activity of 140 dpm pg-’ (7.14 X 10” Bq/mmol) was obtained from Chemsyn Science Laboratories (Lenexa KS). When received the TCDF standard was 75% radiochemically pure. Analysis of the standard by GC-mass spectrometry on a 30-m DE5 capillary column indicated that the sole impurity was a trichlorodibenzofuran, either 2,3,7- or 2,3,8-congener. Purification was carried out by reverse-phase HPLC (Nova Pak C 18, 15 cm X 3.9 mm i.d. column, Waters Associates, Milford, MA) using an isocratic solvent system of methanokwater (9:l) at 1.5 ml min-‘. The final purified product was 97.5% radiochemically pure with the remaining impurity consisting of the trichlorocongener. Fish and Food Preparation Juvenile rainbow trout (2-3 g initial weight), supplied by the Freshwater Institute hatchery, were acclimated for 2 weeks in uv-dechlorinated, carbonfiltered city tap water (10°C). All fish were fed commercial fish food (Martin’s Feed Mills. Ltd., Elmira, ON, Canada) consisting of 4 1% protein, 15% lipid, and 3% fiber. This food was found to contain 33 ng g-’ total PCB congeners and sub-ng g-’ levels of mono-ortho-substituted PCBs (Muir et al,, 1990). [3H]TCDF was added to food by suspending the food particles in a hexane solution (50 ml) containing graded concentrations of TCDF and slowly evaporating to dryness under reduced pressure in a rotary evaporator. The food was allowed to air-dry (24 hr) and was stored at 10°C in sealed glass jars. Food fed to control fish was treated in the same manner but without radiolabeled compound. Final concentrations of the TCDF congener in the spiked food were determined by extraction and analysis as described below for fish tissue. Uptake and Depuration Studies The design of dietary exposure studies with TCDF was identical to that used previously for PnCDF (Muir et al., 1990). The methods used are described briefly: (a) Uptake/depuration study. Trout (38; initial wt, 1-2 g) were held in fiberglass aquaria in a continuous flow of water (1 liter min-i) and fed TCDFtreated food (adjusted at each sampling interval) at 0.02 g ag fish-’ . day-’ for 30 days. Six treatments (0.361, 2.90, 3.60, 7.20, and 42.8 ng TCDF g-’ food plus untreated control) were used. Three fish per treatment were sampled after 5, 10, 20, and 30 days. After 30 days the remaining fish were fed clean food at approximately 0.02 gag fish-’ . day-’ for up to 180 days to study elimination of radioactivity associated with 2,3,7,8-TCDF. At each sampling time livers were removed and frozen immediately on dry ice. The frozen livers were held at -60 to -80°C until analysis (less than 3 months) for MO activity. The rest of the fish was frozen at -20°C until analysis for [‘HITCDF.

ET AL. (b) Uptake/disposition study. A second study examined uptake of TCDF by rainbow trout (21 fish per treatment) over a 14Oday period at 0.910 and 9.20 ng TCDF g-’ food. Two groups of fish fed untreated food served as controls. The feeding rate was 0.0 15 g. g fish-’ . day-‘. Fish were sampled in triplicate after 0, 10, 30, 50, 70, 100, and 140 days. In addition to liver, the gastrointestinal tract and bile were removed from each fish (all samples from the 9.2 ng g-’ exposure and at 100 and 140 days for the 0.9 1 ng g-’ exposure) for separate analysis. Sample Analysis Whole fish, GI tract, livers (subsamples not used for MO assays),and bile samples were weighed and lyophilized. Extractable radioactivity was determined by homogenizing the dried samples (except bile) with toluene and assaying portions of the extract by liquid scintillation counting (LSC). Food samples (0.5 g) were extracted by the same method. Unextractable ‘H in carcass, GI tract, and liver was determined by combusting a portion of the extracted tissue, isolated from the extract by filtration (Whatman No. 1 paper), on a Packard Model D306 oxidizer (Packard Instrument Co., Downers Grove, IL). ‘Hz0 was diluted with Monophase scintillation fluor (Packard Instrument Co.) prior to LSC. Toluene-extractable lipids in fish and food were determined by air-drying portions of the toluene extract to constant weight. Concentrations of TCDF in toluene extracts were determined by reversephase HPLC using an isocratic solvent system of methanokwater (9: 1). All sample extracts from the highest dietary concentration (42.8 ng g-r) were analyzed directly without additional sample cleanup. [“HITCDF was detected by collecting the eluate in 0.5-min fractions and assaying by L-SC.Bile samples were suspended in water and extracted with an equal volume of dichloromethane (DCM). Portions of the aqueous and DCM phases were assayed by LSC for total radioactivity and by reverse-phase HPLC for TCDF. The aqueous phase was evaporated to about 0.5 ml, diluted with 0.05 M phosphate buffer (pH 7.8), and incubated with glucuronidase (Sigma Chemical Co., St. Louis, MO) for 4 hr at 37“C. Following the incubation the mixture was reextracted with DCM and assayed by LSC and by reverse-phase HPLC. MO Assays Analysis of liver samples for MO enzyme activity was carried out with postmitochondrial supematants as described previously (Muir et al., 1990). The small size of the livers throughout most of the experiment, and the need to assay a subsample for TCDF concentration, precluded preparation of microsomal fractions. Livers were subsampled for TCDF analysis if they weighed more than 0.3 g. In brief, samples (0.05-0.3 g) were homogenized in 0.5 to 2.0 ml Hepes-KCl(O.02 M Hepes, 0.15 M KCl, pH 7.5), depending on sample size, and homogenates were centrifuged for 20 min (15,600g). All preparative steps were done in a cold room at 2°C. The supernatants were immediately analyzed for EROD activity using the method of Pohl and Fouts (1980) with several modifications (Muir et al., 1990). The reaction was started by the addition of 10 ~1ofethoxyresorufin in dimethyl sulfoxide (0.04 mg . ml-‘). The samples were incubated for precisely 2 min in a water bath at 25’C and then the reaction was stopped by addition of 2.5 ml of methanol. The samples were centrifuged at 24,000g to pellet the precipitated protein, and the amount of resorufin in the supernatant was determined spectrofluorometrically using an excitation wavelength of 530 nm and an emission wavelength of 585 nm. Protein was determined using the Lowry method as modified by Markwell et al. (198 1). Data Analysis Growth rates were determined by fitting all fish and liver weight data to an exponential model (In weight versus time). TCDF concentrations were corrected for growth dilution and were lipid normalized for calculation of pharmacokinetic parameters. The best value of assimilation (a) was calculated by fitting the concentration data to the integrated form of the kinetic rate equation for constant dietary exposure (Bruggeman et al.. 1981) using the

DIETARY

ACCUMULATION

AND MFO INDUCTION

iterative nonlinear regression NLIN (SAS, 1986). Depuration rate constants (kr,) were obtained by fitting data from the deputation phase of the experiment to a first-order decay curve (In concentration versus time). For the uptake study, TCDF levels in carcass, liver, and GI tract were combined to yield a “whole fish” value, and cxand kD were estimated by NLIN with two unknown parameters. The equilibrium biomagnification factor (BMF) was calculated from the equation BMF = oi vF/k, (Bruggeman et al., 198 I), where F is the feeding rate. BMF is equivalent to the BAF (bioaccumulation factor) reported in our previous dietary assimilation studies (Muir et al., 1990). Differences between growth rate constants and elimination rates among treatments were examined by testing the homogeneity of slopes in an analysis of covariance. The Student t test was used to compare pairs of elimination and growth rate constants at the p < 0.05 level of significance (Steel and Torrie, 1980).

RESULTS

Effects on Growth and Liver Somatic Index Differences were observed in growth rates of whole fish between the highest concentration exposure and most of the other exposures during the uptake/depuration study (Table 1). Analysis of covariance with the fish weights showed significant weight-time interactions. Growth of trout exposed for 30 days to 42.8 ng g-’ TCDF was significantly slower (0.0114 day-‘) than that at three of the five lower concentration exposures (0.014-0.015 day-‘) but differed from

67

BY 2,3,7,8-TCDF

growth of controls (0.0133 day-‘) only at p < 0.1. There were no significant differences in whole fish growth rates among the lower treatment concentrations. Growth rates of rainbow trout during the 140-day uptake study did not differ (Table 1) significantly. Growth was about twofold lower than that in the uptakefdepuration experiment because of the lower feeding rate. The increase in liver weight in fish exposed to 42.8 ng g-’ TCDF (0.0 17 1 day-‘) was more rapid than in controls (0.0125 day-‘) and all TCDF treatments, except the lowest concentration (0.0153 day-‘) (Table 1). Mean liver somatic index (LSI) in the 30-day exposure to 42.8 ng g-’ TCDF was higher after 72 days (1.83 + 0.43) than in controls and low concentration exposures (1.2-l .3) (Table 1). LSI values for the low concentration exposures (0.361-7.20 ng g-‘) did not differ significantly from control values during the depuration phase. In the 140-day uptake study, LSI values at the higher concentration exposure ( 1.95 f 0.97) were higher than those in controls (1.03 + 0.23) after 72 days (Table I), but liver weights did not increase more rapidly in treated than in control fish (Table 1). Although not quantified, the fish in the 140-day exposure to 9.2 ng g-’ TCDF appeared to consume less food during the 50- to 140-day period. Mortality in this treatment was 16% compared to 8% for fish on control diets

TABLE 1

Growth Rates, Lipid Content, Liver Somatic Index, and Cumulative Mortality of Juvenile Rainbow Trout Exposed to Dietary 2,3,7,8-TCDF ,~ Duration Concn in food (ng P-‘1

Uptake (days)

Growth rate constants’

Depuration (days)

Whole fish (day-’ x 10’)

Liver (day-’ X 103)

Uptake/depuration Ctrl 1 0.361 2.90 3.60 7.20 42.8

30 30 30 30 30 30

180 180 140 180 180 180

13.3 + 0.8 (0.89) 14.8 -c 0.7 (0.92) 14.0 + 0.9 (0.88) 14.0 t 0.8 (0.90) 15.0 f 0.8 (0.9 1) 11.4 + 1.o (0.79)

Lipidb (%)

LSI’ (W)

%d Mortality

study

12.5 f 15.3 f 12.6 + 12.4 k 13.1 + 17.1 r+

1.2 (0.78) 1.3 (0.83) 1.8 (0.68) 1.4 (0.74) 1.1 (0.82) I.1 (0.91)

-e 10.0 -t 9.0 f 10.5 5 10.9 rf 9.9 f

1.7 1.6 2.0 1.0 1.7

1.29 f 1.27 ? I .04 f 1.49 + 1.19 + 1.83 +

0.13 0.17 0.32 0.33 0.18 0.43

0 0 0 0 3 0

1.8 (0.64) 1.7 (0.50) 1.9 (0.57) 2.1 (0.62)

5.3 2 5.3 + 5.2 + 5.1 f

1.3 0.3 1.2 1.3

1.20 ck 0.10 0.87 f 0.15 1.37 + 0.02 1.95 iT 0.97

0 0 8 16

Uptake study Ctrl 2 Ctrl 3 0.910 9.20

140 140 140 140

-

7.6 6.3 6.0 7.0

+ + + +

0.8 (0.83) 0.9 (0.71) I .O (0.65) 0.8 (0.83)

9.4 6.7 9.0 10.9

+ t * f

a Rate constants * standard error of the regression coefficient, for the model In wt = a + b (time), where a = intercept and b = rate constant. For the uptake/depuration study N = 38 (O-2 10 days) for whole fish and N = 32 (0- 130 days) for liver. For the uptake study, N = 2 1 for whole fish and N = 18 for liver. Coefficient of variation for the model is shown in parentheses. b Mean percentage lipid in fish from ah sampling dates within each exposure. ’ Mean LSI at 72 days for the uptake/depuration study and 70 days for the uptake study. d Cumulative mortality expressed as percentage of the number of fish exposed on Day 0. e Percentage lipid not determined.

68

MUIR

and 0 to 3% in the uptake/depuration exposures (Table 2). Fin erosion was observed in all the dead fish from the 9.2 ng g-’ exposure. Average (toluene-extractable) lipid in the juvenile trout ranged from 8 to 12% in the uptake/depuration study and from 4.5 to 6.8% in the 140-day uptake study (Table 1). There were no differences between mean percentage lipid in treated versus control fish and no trends over time. Assimilation and Depuration of TCDF Accumulation of 2,3,7,8-[3H]TCDF by juvenile rainbow trout during a 30-day dietary exposure, and 180-day depuration, is illustrated in Fig. 1A for three of the five treatment concentrations. Uptake of TCDF did not reach a steady state by 30 days and depuration proceeded slowly with t,,*‘s ranging from 40 to 77 days (Table 2). After 30 days exposure TCDF concentrations ranged from 0.095 ng g-l in the 0.36 1 ng g-’ treatment to 11.0 + 0.97 ng g-i in the 42.8 ng g-’ exposure (Fig. 1A). During the 140-day dietary exposure, apparent steady-state concentrations of about 2 ng g-’ were reached after about 70 days at the high treatment level (9.2 ng g-‘) although a single fish remaining at 140 days had 2.8 ng g-’ (Fig. 1B). Feeding at 0.9 1 ng g-’ for 140 days did not result in steady-state concentrations (Fig. 1B). The results in Fig. 1 are based on toluene-extractable radioactivity in whole fish corrected for growth dilution. HPLC analysis indicated that >98% of the toluene-extractable radiolabel was in the form of the parent compound throughout the uptake and depuration phases. Furthermore >98% of the total radioactivity in carcass and liver samples was extractable with toluene (Table 3), indicating little incorporation or exchange of the tritium label. Therefore the results in Fig. 1 can be considered indicative of the behavior of 2,3,7,8-TCDF congener. Elimination of TCDF followed a single first-order decay curve at all concentrations (Fig. 1A and Table 2). An analysis of covariance for the depuration data from all five treatments had a significant time-treatment interaction (p < O.Ol), indicating differences in k. values. Paired comparisons of the slopes (rate constants) of the depuration curves indicated that elimination following the 42.8 ng g-i TCDF treatment (0.0174 day-i) was significantly more rapid than at 0.361, 3.60, and 7.20 ng g-’ exposures (0.0090-0.0 100 day-‘). There were no differences in k. among the lower concentration exposures. Depuration rates for TCDF during the 140-day uptake study, calculated by iterative nonlinear regression, were within the range observed for actual depuration following 30-day exposure and did not differ significantly (Table 2). The k,‘s give rise to a tllZ of TCDF of about 50 days. Assimilation efficiencies of TCDF ranged from 48.8 to 62.4% and BMFs from 0.9 1 to 1.9 1 in the uptake/depuration study (Table 1). Unlike the results for depuration and growth there were no trends in LY with exposure concentration.

ET

AL.

However, a lower BMF was observed with the highest 30day exposure. In the 140-day uptake study, LYof 33.3% and BMF of 0.36 for the 140-day exposure at 9.2 ng g-’ were lower than those of other treatments. Tissue Distribution

and Metabolism

The tissue distribution of TCDF in rainbow trout over time during the 140-day exposure is shown in Table 3. The majority of toluene-extractable radioactivity (64-74%) was found in the carcass (whole fish with GI tract removed). The GI tract accounted for 18 to 3 1% of the radiolabel and liver from 0.6 to 2.3%. Unextractable radioactivity, determined by combustion of toluene-extracted tissue, ranged from 2.6 to 4.8% in the carcass, from 0.3 to 2.3% in the GI tract, and from 0.05 to 0.20% in the liver. HPLC analysis of the toluene extracts of the fish carcass indicated that >98% of the radiolabel in this tissue was in the form of TCDF at all sampling times (data not shown). Most samples had two small additional peaks with relative retention times (RRT) to TCDF of 0.30 and 0.80. Maximum proportions of these peaks, representing 0.3 and 0.2% of total radiolabel in the extracts of whole fish, were observed after 30 days exposure. The peak with RRT of 0.80 had a retention time identical to that of the trichlorofuran congener in the standard. In bile samples, 9 to 78% of the radioactivity was extractable with DCM and the remainder was associated with the aqueous phase (Table 3). Although the combined aqueous and DCM phases represented only 0.05 to 1.4% of the body burden, concentrations were high (ranging from 2.3 to 4.3 ng g-’ expressed as equivalents of TCDF) because of the small sample volume. HPLC analysis indicated that most of the DCM-extractable radioactivity was in the form of two or more poorly resolved polar peaks (Fig. 2) with RRTs of 0.14 to 0.28 (Fig. 2C). The aqueous phase contained a single major peak with RRT of 0.18 and a minor component at 0.32 (Fig. 2B). Glucuronidase hydrolysis of the aqueous phase yielded a more nonpolar product (RRT = 0.32; Fig. 2D) with the RRT expected of a hydroxylated TCDF or TCDD under these HPLC conditions (Muir et al., 1985). The results suggested that the polar product in the aqueous phase was a glucuronide conjugate. Hepatic Mixed-Function

Oxidase Enzyme Activity

EROD activity in postmitochondrial supernatants of rainbow trout liver was strongly induced by dietary exposure to 42.8 ng g-i TCDF for 30 days (Fig. 3 and Table 2). At 30 days, EROD was 137.5 and 82.5 times higher than in controls and the 7.2 ng g-’ exposure, respectively. During the depuration phase, EROD activities declined rapidly (tl,* = 16 days) to levels similar to those of control fish after 100 days. Mean EROD activities in fish exposed for 30 days to 0.361 to 7.2 ng g-’ of TCDF were not significantly different from control values.

DIETARY

ACCUMULATION

AND MFO INDUCTION

69

BY 2,3,7,8-TCDF

TABLE 2 Bioaccumulation Parameters for 2,3,7,8-TCDF in Juvenile Rainbow Trout during Dietary Exposures Duration Concn in food (ng g-‘1

Uptake (days)

Depuration (days)

Depuration’ rate constant (day-’ X 103) Uptake/depuration

0.361 2.90 3.60 7.20 42.8

30 30 30 30 30

180 140 180 180 180

9.5 12.6 9.0 10.0 17.4

+ 2 + + +

BMF

Assimilation efficiencyd @)

5 6 5 4 6

1.91 1.61 1.74 1.77 0.91

55.9 62.4 48.2 54.5 48.8

47+ 19 54+31

0.96 0.36

54.7 + 10.0 33.3 +_ 8.8

hb (days)

study

0.7 (0.90) 1.5 (0.80) 0.6 (0.89) 0.5 (0.94) 2.6 (0.66)

73* 55* 77f 70+ 40+

k rf: + + f

3.5 2.7 2.8 3.4 1.2

Uptake study 0.910 9.20

140 140

-

14.6 + 5.9 12.9 xi 7.5

n Elimination rate constant (kD) for the model In concentration (lipid wt basis) = a + b (time) for elimination of toluene-extractable radioactivity for 180 days of depuration (coefficients of variation for the model are shown in parentheses). For the uptake study, kD was estimated (along with a) by twoparameter nonlinear regression. b Half-life calculated from the equation tu2 = 0.693/k,. c Biomagnification factor calculated from the equation BMF = aF/kD, where (Y is the assimilation efficiency determined by nonlinear regression of uptake data; F is the feeding rate on a lipid basis, and kD is the elimination rate. BMF is equivalent to the BAF (bioaccumulation factor) reported in our previous dietary assimilation studies. d Assimilation efficiency calculated by fitting the data to the integrated form of the kinetic rate equation for constant dietary exposure using iterative nonlinear regression: C,, = (aFCf&kD)[ 1 - exp(-k, t)] where C 6Jhis the concentration in fish (lipid basis); C roodis the concentration in food; and t is the time (days) of uptake.

During the 140-day uptake study EROD activities in fish sistent with results for other 2,3,7,8+ubstituted chlorinated exposed to 9.2 ng g-’ were 15 times higher than those of dibenzo-pdioxins (PCDDs) and PCDFs in rainbow trout controls after 30 days and remained at approximately the (Muir et al., 1992). Values of (y for TCDF were independent same levels between Days 30 and 100. In the lower concen- of dietary exposure level in the 30-day uptake study. Loonen tration exposure, EROD activity was about twofold higher et al. (1992) reported an Q of only 3.6% for TCDF in guppies than that of controls between Days 30 and 140 but the dif- during a 180-day feeding study with a mixture of 2,3,7,8ference was not statistically significant at any sampling time. substituted PCDDs and PCDFs. The dietary concentrations EROD activities were significantly correlated (p < 0.01) of the TCDF were from 1.8 to 18.7 times higher than those with TCDF concentrations in whole fish and in liver (Figs. used in the present study (140-day uptake) and may have 4A and 4B). The strongest relationship between EROD and been in the range where deviations from first-order kinetics TCDF concentration was observed for liver during the 140- occurred. Opperhuizen and Schrap ( 1990) found that values day uptake study (EROD = 0.002 (kO.02) + 0.091 of (Y for hexa- and octachlorobiphenyl biphenyl congeners (+0.012). [TCDF], R* = 0.59), where [TCDF] is the TCDF were independent of concentration over a 20-fold range (from concentration (wet wt). Results for TCDF in whole fish from 7.1- 150 gg g-’ in food) but were lower at higher concentrathe 140-day study were weakly correlated with those for tions (550 and 1400 pg g-l) where high mortality and beEROD (R* = 0.29, N = 45) but there was a stronger rela- havioral changes occurred. A decline in (Y with increasing tionship when these data were combined with the uptake/ concentrations of nonmetabolizable organic chemicals in the depuration results from the high concentration exposure (R* fish and prey, and with increased duration of exposure, is = 0.36) (Fig. 4A). There was considerable variation in EROD predicted by the thermodynamic-based model of Barber et activities among individual fish with the same TCDF con- al. (199 1) which assumes a diffusive exchange of chemical centrations in liver. across the intestinal wall. Results of the present study provide only limited support for this prediction. Values of cx were lower for the 140-day exposure at 9.2 ng g-i than in all other DISCUSSION treatments, but estimated values of cr in the 0.910 ng g-i The assimilation efficiencies of 48 to 62% for TCDF that exposure were within the range observed in the uptake/depwere observed in this study are similar to those found for uration study (Table 1). The lower (Yin the 140day exposure 2,3,4,7,8-PnCDF (41-44%) (Muir et al., 1990) and are con- to 9.2 ng g-’ TCDF may be due to the reduced feeding rate

70

MUIR ET AL.

10 A ,@...“...

$,:. .x., ! ‘,.4$0... ..ng .+.." '.....,,

f” 3.0

t

d

j

1

2.0 1.0 1

0.91 ng g-’

0.50

0.05

0.02

1

-

' i 0

I 40

' 20

I 60

I 00

I 100

I 120

I 140

Time (days) FIG. 1. Accumulation and depuration of 2,3,7&TCDF by juvenile rainbow trout (whole fish minus the liver) during (A) a 30&y dietary exposure to 0.36, 7.2, or 42.8 ng g-’ concentrations followed by a 180-day depuration period and (B) a 140-day dietary exposure to 0.91 and 9.2 ng g-r concentrations. Each point represents mean toluene-extractable radioactivity from three fish corrected for growth dilution. Vertical lines indicate one standard deviation.

of the trout that was observed during the 50- to 140-day period. The calculation of equilibrium values of (Yand BMF depend on feeding rate (F) as well as TCDF concentrations in the fish (Bruggeman et al., 198 1). A precise value of feeding rate could not be determined but the nominal rate of 0.0 15 g. g fish-’ . day-‘, which was used for calculation of (Yin the 9.2 ng gg’ treatment, was probably too high. Knowledge of feeding rate is thus critical to precise estimation of (Y. Whole-body depuration t1,2’s for TCDF of 40 to 77 days in the present study are much greater than those estimated by Merhle et al. ( 1988) who reported a tl12 of 0.3 and 3 days for TCDF in rainbow trout fry, following aqueous exposure. Fish size and lipid content may explain in part the wide ranges in observed depuration rates; the initial weights of the trout used in the present work were 10 times higher, and lipid content about twofold greater, than those used by Merhle et al. ( 1988). Cook et al. (199 1) found t1,2’sof 2,3,7,8TCDD of 320 days for carp averaging 1.5 kg and 19% lipid content and 63 days for carp with average weights of 15 g and 9% lipid. The t1,2’s of 147 days for TCDF in 1.5-kg carp (Kuehl et al., 1986) and 11 days in guppies (0.1 g) following aqueous exposure (Opperhuizen and Sijm, 1990) are in general agreement with this study, assuming that the allometric relationships reported for TCDD by Cook et al. ( 199 1) also apply to TCDF. Depuration rates of TCDF by rainbow trout were independent of dietary exposure level over a 20-fold range (0.36 l7.20 ng g-‘) and were well described by a single first-order kinetic rate equation. Elimination of other di- to heptachlorosubstituted PCDDs and PCDFs is also generally well described by simple first-order equations although there have been few tests of the model assumptions (Opperhuizen and Sijm, 1990; Muir et al., 1992; Gobas and Schrap, 1990). The

TABLE 3 Distribution of 2,3,7,8-[3H]TCDF in Carcass, GI Tract, Bile, and Liver of Rainbow Trout during a 140-Day Dietary Exposure Period Percentage in each tissue Carcass Treatment

Day

0.910 ngg-I

100

9.20 ngg-’

140 :: 70 100 140'

Extract 69.1 68.0 63.5 14.4 67.2 62.1

f 7.8 f 16.6 IT 8.7 f 4.0 + 5.4 f 5.9 69.0

GI tract Unextract 3.4 8.3 2.6 4.8 2.6 4.3

+ 2.4 + 11.4 + 1.7 f 2.1 + 1.2 + 0.89 5.1

Extract 25.8 t 9.0 22.6 + 6.9 30.6 +I 10.3 17.5 + 5.1 28.2 k 5.5 30.6 k 5.7 22.7

a Mean of two bile samples from fish collected on Days 78 and 109. b Dash indicates no sample available. ’ Mean of two samples from Days 112 and 140.

Bile Unextract 0.52 t 0.05 0.33 + 0.25 0.78 k 0.14 0.59 z!c 0.14 0.74*0.11 1.3 AI 0.56 0.91

Extract 0.08" 0.06'

0.11 kO.10 0.47 + 0.29 0.03 + 0.01 0.21 + 0.20 0.12

Liver Unextract 0.19 0.06 0.09 + 0.07 0.13 f 0.12 0.02 k 0.02 0.16 f 0.12 1.27

Extract 0.60 0.75 2.3 2.0 1.1 1.2

2 0.27 + 0.33 5 0.39 f 0.93 + 0.64 + 0.46 0.78

&extract -b 0.05 k 0.03 0.09 k 0.04 0.14 f 0.08 0.20 * 0.03 0.12

DIETARY

60

ACCUMULATION

A. TCDF standard ,, ,. ,...,....................

,,

AND MFO INDUCTION

t iii”.

,100 o C. DCM extract, 100 d g 60 _,,,,,,,..,.......,,...,..........,.....,....,,,,,,,,,,,,,,,,,, .. , i ;~~~~,l:~~l

B. Aauacus

71

BY 2,3,7,&TCDF

Dhase. 100 d

. D. DCM extract after glucuronidase .._ -................................

1

hydrolysis

_

~~~~~~~~~~

40 20

.,.,., 1 2

3

4’ 5

6 7

6 9 10 11 12 13

Retention time (minutes) FIG. 2. Reverse-phase HPLC chromatograms of (A) 2,3,7,8-TCDF standard, (B) aqueous phase of bile (following dichloromethane extraction) from rainbow trout exposed to 9.2 ng g-l TCDF for 100 days, (C) dichloromethane extract of the bile sample, and (D) aqueous phase following treatment with glucuronidase.

results reported here for TCDF confirm that the assumption of first-order depuration kinetics is reasonable for juvenile rainbow trout where no effects on growth are observed. More rapid depuration of TCDF, following a 30-day exposure to 42.8 ng g-’ was associated with elevated EROD activity and maximum whole-body concentrations of 11 ng g-‘. Elevated EROD activity in the 140-day uptake study was also associated with a higher proportion of radiolabel in the bile. These results suggest that biotransformation of TCDF to polar products catalyzed by MO enzymes was elevated at higher exposure concentrations, resulting in greater elimination of TCDF. Degradation products were not measured in the 30-day uptake study (except in toluene extracts where they were barely detectable). However, the higher proportion of radioactivity in the bile of fish exposed to 9.2 ng g-‘, than at 0.9 1 ng g-’ for 140 days (Table 3), and elevated cytochrome P450-dependent MO activity in liver provide indirect evidence for increased oxidative activity. Other studies in which P450-dependent monooxygenases were induced in fishes prior to or during exposure to chemicals have also shown enhanced excretion of metabolites (Lech and Bend, 1980). The biliary excretion of trichlorobenzene was enhanced sixfold in induced trout (Lech and Bend, 1980), while benzo[a]pyrene was more rapidly oxidized and eliminated in a variety of fishes pretreated with 3-methylcholanthrene (James et al., 1988). Readily metabolized PCDD congeners, 1,2,3,7-TCDD and 1,2,3,4,7-PnCDD, were eliminated more slowly in viva by rainbow trout when oxidative enzymes were inhibited with piperonylbutoxide (PBO) and unextractable radioactivity was lower than that in non-PBOtreated fish (Sijm et al., 1990).

There is evidence in recent literature that the extent to which metabolism and excretion of PCDD/Fs can be enhanced in fishes and mammals is related to molecular structure. There was little effect of PBO exposure on the elimination of 2 33 94 >7 98-PnCDF during a 2 1-day period (Sijm et al., 1990) and juvenile rainbow trout with elevated EROD activity had kD values only slightly greater than those with low induction (Muir et al., 1990). But biliary excretion of 2 , 3 ,4 77 98-PnCDF increased twofold in rats pretreated with the chemical (Brewster and Birnbaum, 1987). In vivostudies with rats indicated that biliary excretion of TCDD metabolites was not enhanced by pretreatment with TCDD despite elevation of hepatic microsomal EROD activity (Kedderis et al., 1991). Growth of juvenile rainbow trout was reduced when TCDF levels reached 11 ng g-’ following a 30-day exposure. This threshold dose for growth reduction is similar to that observed by Merhle et al. ( 1988), who found reduced growth of rainbow trout fry when body burdens of TCDF were about 10 ng g-’ following a 28-day aqueous exposure. TCDF concentrations associated with elevated LSI values ranged from 2 to 11 ng g-’ depending on the exposure time. These levels were similar to the concentrations of TCDD found to increase LSI in 25- to 55-g trout after a single ip injection (Van der Weiden et al., 1990). The tissue distribution of TCDF in the present study was similar to that observed for 2 33 74 37 78-PnCDF in rainbow trout 2 1 days after a single oral dose where radioactivity in muscle/ skin and liver represented about 84 and 2.7%, respectively, of the body burden (Sijm et al., 1990). The distribution and biotransformation of TCDD in juvenile rainbow trout during

72

MUIR ET AL.

ratio of 15 was observed for relative induction of EROD by PnCDF and TCDF in rat liver microsomes in vitro (Mason et al., 1985). In comparison, the threshold for induction of liver microsomal EROD by Aroclor 1254 in juvenile rainbow trout was approximately 250 ng g-’ in muscle and 3 16 ng gg’ in liver following a single ip dose (Melancon et al., 1989). A stronger relationship between EROD and TCDF concentration might have been obtained if liver microsomes rather than postmitochondrial supernatants were used. This was not feasible in the present study because of the small size of the liver samples. The rate of decline in EROD activity following a 30-day exposure to PnCDF was also much slower (L,,~ = 113 days) than that observed for TCDF (tllZ = 16 days) although the whole-body t ,,2‘s of the congeners were similar. The sustained induction by PnCDF reflects the greater affinity for the Ah receptor and possibly a slower rate of oxidation of this congener in comparison with TCDF. In summary, assimilation of 2,3,7,8-TCDF during 30-day dietary exposures was independent of concentration over a lOO-fold range. In contrast, depuration rates were indepen-

0

20

40

60

80

100

120

140

Time (days) FIG. 3. EROD activity in livers (postmitochondrial supematants) of juvenile rainbow trout during (A) a 30-day dietary exposure to 0.36,7.2, or 42.8 ng g-’ concentrations followed by a loo-day depuration period and (B) a 140day dietary exposure to 0.91 and 9.2 ng g-’ concentrations. Two other 30-day exposure concentrations are not shown for clarity (see Table 1). EROD results for samples collected after I80 and 2 10 days (uptake/depuration study) are not shown because levels in treated and control fish did not differ. Vertical lines indicate one standard deviation; upper and lower lines for EROD in control and 7.2 ng g-’ exposures are omitted for clarity.

;; o.05/..~...1_.--.‘.... h 0.03

dietary exposure (about 70% in carcass, skin, gills, and muscle and about 2% in liver; Kleeman et al., 1986) were similar to those of TCDF. Kleeman et al. (1986) found >98% of [3H]TCDD in the form of parent compound except bile where three unidentified polar metabolites were observed. Oxidation and glucuronide conjugation seem to be common pathways for both PCDDs and PCDFs in rainbow trout. However, TCDD and TCDF appear to have low rates of biotransformation, judging by the small proportion of polar products in the bile, unlike non-2,3,7,8-substituted PCDDs which have high rates of conversion to polar unextractable products (Sijm et al., 1990; Muir et al., 1985). The EROD/TCDF relationship suggests a threshold concentration of about 0.2 and 1 ng g-r TCDF, in liver and whole fish, respectively, for induction of EROD (in postmitochondrial supernatants). 2,3,4,7,8-PnCDF was a more potent inducer of EROD activity by about fivefold under similar dietary exposure conditions (Muir et al., 1990). A

CJ) 0.12 E 5 E g

E

0.30

1.0

3.0

10.0

B

0.1

EROD = 0.002 + 0.091 [TCDF] I+= 0.59..! N=45

0.09

0

0.10

..j

i 0.06

Log TCDF concentration

(ng S’)

FIG. 4. EROD activity plotted against TCDF concentrations in whole fish (A) or liver (B). Results for whole fish are from fish exposed for up to 140 days to 0.91 and 9.2 ng g-’ TCDF, combined with those exposed for 30 days to 42.8 ng g-’ followed by a 180-day depuration phase. Results for liver are from the 140-day study only.

DIETARY

ACCUMULATION

AND

dent of concentration only over a 20-fold range and were significantly more rapid at the highest exposure concentration (42.8 ng g-r for 30 days). More rapid depuration was associated with higher EROD activity, lower growth rates, and elevated LSI in fish exposed to 42.8 ng g-’ for 30 days. Elevated EROD activity was also observed along with increased mortality and fin erosion, and elevated LSI during a 140day dietary exposure to 9.2 ng g-’ TCDF. The elevated EROD and higher radioactivity in bile of induced fish suggested that the rainbow trout responded to TCDF exposure by increasing liver oxidative activity. The trout were able to transform TCDF to polar products, with the major one being a glucuronide conjugate. Only 1% or less of the body burden was in the form of detectable degradation products, suggesting that little of the parent compound was metabolized and that polar products were rapidly eliminated. These results provide, for the first time, detailed pharmacokinetic data which may be useful for modeling the bioaccumulation of TCDF in aquatic biota. The BMF values of 0.9 to 1.9 obtained in this study confirm the high bioaccumulation potential of TCDF in salmonid fishes which is apparent from monitoring studies. ACKNOWLEDGMENTS This work was supported in part by a Strategic Grant from the Natural Sciences and Engineering Research Council of Canada. We thank S. Brown (Fisheries and Oceans, Winnipeg, MB), A. J. Niimi (Fisheries and Oceans, Burlington, ON), and P. V. Hodson (Fisheries and Oceans, Mont Joli, QU) for helpful reviews of the manuscript. We are grateful to M. Simon and R. J. Norstrom (Canadian Wildlife Service, Ottawa, ON KlA OH7) for GCMS analysis of the standard.

REFERENCES Barber, M. C., Suarez, L. A., and Lassiter, R. R. ( 199 1). Modelling bioaccumulation of organic pollutant in fish with an application to PCBs in Lake Ontario salmonids. Can. J. Fish. Aquat. Sci. 48, 318-337. Brewster, D. W., and Birnbaum, L. S. (1987). Disposition and excretion of 2 13, 4, 7 78pentachlorodibenzofuran in the rat. Toxicol. Appl. Pharmacol. 90,243-252. Bruggeman, W. A., Martron, L. B. J. M., Kooiman, D., and Hutzinger, 0. ( 198 I). Accumulation and elimination kinetics of di-, tri- and tetrachlorobiphenyls by goldfish after dietary and aqueous exposure. Chemosphere lo,81 1-832. Cook, P. M., Batterman, A. R., Butterworth, B. C., Lodge, K. B., and Kohlbry, S. W. (1990). Laboratory study of TCDD bioaccumulation by lake trout from Lake Ontario sediments, food chain and water. In Lake Ontario TCDD Bioaccumulation Study. Final Report. U.S. Environmental Protection Agency, Duluth, MN. Cook, P. M., Walker, M. K., Kuehl, D. W., and Peterson, R. E. (1991). Bioaccumulation and toxicity of 2,3,7,8-tetrachlorodibenzo-pdioxins and related compounds in aquatic ecosystems. In Biological Basis for Risk Assessment of Dioxins and Related Compounds (M. A. Gallo, R. J. Scheuplein, and C. A. Vander Heijden, Eds.), Banbury Report, Vol. 35, pp. 143- 167. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Devault, D., Dunn, W., Bergqvist, P-A., Wiberg, K., and Rappe, C. (1989). Polychlorinated dibenzofurans and polychlorinated dibenzo-pdioxins in

MFO INDUCTION

BY 2,3.7,8-TCDF

73

Great Lakes fish: A baseline and interlake comparison. Environ. Toxicol. Chem. 8, 1013-1022. Endicott, D., Richardson, W. R., and Ditoro, D. M. (1990). Lake Ontario TCDD modeling report. In Lake Ontario TCDD Bioaccumulation Study. Final Report. U.S. Environmental Protection Agency, Duluth, MN. Fiedler, H., Hutzinger, O., and Timms, C. W. (1990). Dioxins: Sources of environmental load and human exposure. Toxicol. Environ. Chem. 29, 157-234. Gobas, F. A. P. C., and Schrap, S. M. (1990). Bioaccumulation of some polychlorinated dibenzo-p-dioxins and octachlorodibenzofuran in the guppy (Poecilia reticulata). Chemosphere 20,495-5 12. Hahn, M. E., Woodin, B. R., and Stegeman, J. J. (1989). Induction of cytochrome P450E (P450IAl) by 2,3,7,8-tetrachlorodibenzofuran (2,3,7.8TCDF) in the marine fish scup (Stenotomus chrysops). Mar. Environ. Res. 28, 61-65. James, M. O., Heard, C. S., and Hawkins, W. E. (1988). Effect of 3-methylchloanthrene on monooxygenase, epoxide hydrolase and glutathione-stransferase in small estuarine and freshwater fish. Aquat. Toxicol. 12, l15. Kedderis, L. B., Diliberto, J. J., Linko, P., Goldstein, J. A., and Birnbaum, L. S. (199 1). Disposition of 2,3,7,8-tetrabromodibenzo-p-dioxin and 2,3,7,8-tetrachlorodibenzo-p-dioxin in the rat: Biliary excretion and induction of cytochromes CYPl Al and CYPlA2. Toxicol. Appl. Pharmacol. 111,

163-172.

Kleeman, J. M., Olson, J. R., Chen, S. M., and Peterson, R. E. (1986). Metabolism and disposition of 2,3,7,8-tetrachlorodibenzo-p-dioxin in rainbow trout. Toxicol. Appl. Pharmacol. 83, 39 l-401. Kuehl, D. W., Cook, P. M., and Batterman, A. R. (1986). Uptake and depuration studies of PCDDs and PCDFs in freshwater fish. Chemosphere 15,2023-2026.

Lech, J. J., and Bend, J. R. (1980). Relationship between biotransformation and the toxicity and fate of xenobiotic chemicals in fish. Environ. Health Perspect. 34, 115- I3 1. Loonen, H., Parsons, J. R., and Govers, H. A. J. (1992). Relative contribution of water and food to the bioaccumulation of PCDDs and PCDFs in guppies. Chemosphere, in press. Markwell, M. A. K., Haas, S. M., Tolbert, N. E., and Bieber, L. L. (1981). Protein determination in membrane and lipoprotein samples: Manual and automated procedures. Methods Enzymol. 72,296-393. Mason, G.. Sawyer, T., Keys, B., Bandiera, S., Ronkes, M., Piskorska-Pliszcynska, J., Zmudzka, B., and Safe, S. (1985). Polychlorinated dibenzofurans (PCDFs): Correlation between in vivo and in vitro structure-activity relationships. Toxicology 37, 1- 12. Melancon, M. J., Tumquist, K. A., and Lech, J. J. (1989). Relation of hepatic microsomal monooxygenase activity to tissue PCBs in rainbow trout (Salmo Gairdneri) injected with “‘C-PCBs. Environ. Toxicol. Chem. 8, 777-782. Merhle, P. M., Buckler, D. R., Little, E. E., Smith, L. M., Petty, J. D.. Peterman, P. H., Stalling, D. L., De Graeve, G. M., Coyle, J. J., and Adams, W. J. (1988). Toxicity and bioconcentration of 2,3,7,8-tetrachlorodibenzo-pdioxin and 2,3,7,8-tetrachlorodibenzofuran in rainbow trout. Environ. Toxicol. Chem. 7,47-62. Muir, D. C. G., Fairchild, W. L., Yarechewski, A. L., and Whittle, D. M. ( 1992). Derivation of bioaccumulation parameters and application of food chain models for chlorinated dioxins and furans. In Chemical Dynamics in Freshwater Ecosystems (F. A. P. C. Gobas and F. MacCorquadale, Eds.), pp. 185-208. Lewis Publishers, Ann Arbor MI. Muir, D. C. G., Yarechewski, A. L., Metner, D. A., Lockhart, W. L., Webster, G. R. B., and Friesen, K. J. (1990). Dietary accumulation and sustained hepatic mixed function oxidase enzyme induction by 2,3,4,7,8pentachlorodibenzofuran in rainbow trout. Environ. Toxicol. Chem. 9, 14651474.

74

MUIR

Muir, D. C. G., Yarechewski, A. L., and Webster, G. R. B. (1985). Bioconcentration of four chlorinated dioxins by Rainbow trout and Fathead minnows. In Aquatic Toxicology and Hazard Assessment: Eighth Symposium, ASTMSTP 891 (R. C. Bahner and D. J. Hansen, Eds.), pp. 440454. Amer. Sot. Test. Mater., Philadelphia, PA. NCASI. 199 1. Observations on the Bioaccumulation of 2,3,7&TCDD and 2.3.7,8-TCDF in Channel Catjish and Largemouth Bass and Their Survival or Growth during Exposure to Biologically Treated Bleached Kraji Mill Efluent in Experimental Streams. Technical Bull. No. 61 I. National Council of the Paper Industry for Air and Stream Improvement, Inc., New York, NY. Opperhuizen, A., and Schrap, S. M. (1988). Uptake efficiencies of two polychlorobiphenyls in fish after dietary exposure to five different concentrations. Chemosphere 17, 253-262. Opperhuizen, A., and Sijm, D. T. H. M. (1990). Bioaccumulation and biotransformation of polychlorinated dibenzo-p-dioxins and dibenzofurans in fish. Environ. Toxicol. Chem. 9, 115-186. Pohl, R. J., and Fouts, J. R. (1980). A rapid method for assaying the metabolism of 7-ethoxyresorufin by microsomal cellular fractions. Anal. Biochem. 107, 150-155. Rappe, C., Andersson, R., Bergovist, P-A., Brohede, C., Hansson, M., Kjeller, L-O., LindstGm, G., Marklund, S., Nygren, M., Swanson, S. E., Tysklind, M., and Wiberg, K. (1987). Overview on environmental fate of chlorinated dioxins and dibenzofurans: Sources, levels and isomeric pattern in various matrices. Chemosphere 16, 1603-1618.

ET AL. SAS (1986). SAS System for Linear Models, 2nd ed. SAS Institute, Gary. NC. Sijm, D. T. H. M.. Yarechewski, A. L., Muir, D. C. G., Webster, G. R. B., Seinen, W.. and Opperhuizen, A. (1990). Biotransformation and tissue distribution of 1,2,3,7-tetrachlorodibenzo-p-dioxin, 1,2,3,4,7-pentachlorodibenzo-p-dioxin, and 2,3,4.7,8-pentachlorodibenzofuran by rainbow trout. Chemosphere 21, 845-866. Steel, R. G. D., and Torrie. J. H. (1980). Principles and Procedures ofStatistics: A biometrical Approach, 2nd ed. McGraw-Hill, Inc., NY. Thomann, R. V. (1989). Bioaccumulation model of organic chemical distribution in aquatic food chains. Environ. Sci. Technoi. 23, 699-707. Thomann, R. V., and Connolly, J. P. (1984). Model of PCB in the Lake Michigan lake trout food chain. Environ. Sci. Technol. 18, 65-7 1. U.S. Environmental Protection Agency (USEPA) (199 I). Bioaccumulation of Selected Pollutants in Fish-A National Study. EPA 506/6-90/OOla. Vol. 1. U.S. EPA, Office of Water Regulation and Standards, Washington DC. Van Der Weiden. M. E. J., Van Der Kolk, J., Penninks, A., Seinen, W., and Van Den Berg. M. (1990). A dose/response study with 2.3,7.8-TCDD in the rainbow trout (Oncorhynchus mykiss). Chemosphere 20, tO531058. Whittle, M., Sergeant, D. B., Huestis, S., and Hyatt, W. H. (1990). The Occurrence of Dioxin and Furan Isomers in Fish and She&h Collected near Bleached Kraft Mills in Canada. Presented at the 10th International Symposium on Chlorinated Dioxins and Related Compounds, Bayreuth, Germany, September 1990.