Dieldrin Stimulates Biliary Excretion of 14C-Benzo[a]pyrene Polar Metabolites but Does Not Change the Biliary Metabolite Profile in Rainbow Trout (Oncorhyncus mykiss)

Melanie L. Barnhill, Marie V. M. Rosemond and Lawrence R. Curtis1

Department of Environmental and Molecular Toxicology, Oregon State University, Corvallis, Oregon 97331

Received April 17, 2003; accepted June 30, 2003


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Activities of hepatic microsomal and cytosolic epoxide hydrolases, accumulation of dieldrin in liver, and in vivo metabolism and disposition of the polycyclic aromatic hydrocarbon (PAH), benzo[a]pyrene (BP), were examined in rainbow trout pretreated with dieldrin, a chlorinated cyclodiene insecticide. Rainbow trout were fed 0.3 mg dieldrin/kg/day for 9 weeks and the same dose of dieldrin for 9 weeks, followed by 3 weeks on control diet (12 weeks). Fish then received an intraperitoneal (ip) challenge dose of 14C-BP (10 µmol/kg). Dieldrin pretreatment significantly elevated the concentration of 14C-BP in bile (142% and 200% at 9 and 12 weeks, respectively), but not liver or fat. Extraction of bile subsamples confirmed dieldrin pretreatment significantly stimulated total biliary excretion of14C-BP polar metabolites (244% and 221% at week 9 and 12, respectively). The complex metabolism of BP characterized the in vivo state of the CYP system, UDP-glucuronyltransferases, and sulfotransferases. Bile was extracted and then hydrolyzed by ß-glucuronidase and arylsulfatase to regenerate BP metabolites conjugated by phase II enzymes. Evaluation of biliary polar metabolite profiles of 14C-BP revealed no significant differences between control and dieldrin-fed fish. There was no selective enhancement of any particular metabolite, or formation of a novel metabolite with dieldrin pretreatment. This research confirmed that enhanced biliary excretion, following chronic dieldrin exposure, was not explained by induction of xenobiotic metabolizing enzymes. The results are consistent with induction of hepatic intracellular trafficking proteins in dieldrin-fed fish.

Key Words: benzo[a]pyrene; biliary excretion; dieldrin; in vivo metabolism; rainbow trout.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Dieldrin, a chlorinated cyclodiene insecticide, is a nonplanar highly persistent molecule that tends to bioaccumulate in fish, wildlife, and humans. There is a U.S. Environmental Protection Agency (U.S. EPA) ban on all uses of this insecticide within the United States (1971Go, 1990Go). However, dieldrin still occurs in all environmental media (U.S. Department of the Interior, 1999Go), and the U.S. EPA (2001)Go lists it as a priority level-1 pollutant that is persistent, bioaccumulative, and toxic (PBT). Polycyclic aromatic hydrocarbons (PAHs) also occur widely in the environment (Cerniglia and Heitkamp, 1989Go; McElroy et al., 1989Go), and the U.S. EPA (2001)Go lists benzo[a]pyrene (BP) as a PBT as well. Understanding interactions between organochlorines (OCs), like dieldrin, and PAHs is important because they occur in the environment as complex mixtures.

Dieldrin pretreatment of rainbow trout altered tissue distribution of a 14C-dieldrin challenge dose and stimulated biliary excretion of polar metabolites. Bioaccumulation of dieldrin in rainbow trout was measured following waterborne and dietary exposures for 16 weeks (Shubat and Curtis, 1986Go). When normalized for lipid content, whole-body dieldrin concentrations increased through 8 weeks. However, at week 16 whole-body concentrations decreased unexpectedly to concentrations observed in trout exposed for 2 weeks. Subsequent work confirmed this phenomenon and demonstrated that altered disposition, following dieldrin pretreatment, was not specific to a subsequent dose of dieldrin, but also occurred with the PAH 7,12-dimethylbenz[a]anthracene (DMBA) (Donohoe et al., 1998Go; Gilroy et al., 1993Go). Feeding rainbow trout 0.3 or 0.4 mg dieldrin/kg/day for 9–12 weeks stimulated the biliary excretion of a subsequent dose of 14C-dieldrin by 500% and 3H-DMBA by 240%. The same exposure also significantly increased the disposition of 14C-dieldrin to liver and mesenteric fat and elevated the levels of 3H-DMBA in liver. In another study, pretreatment (0.4 mg dieldrin/kg/day in the diet for 10–12 weeks) stimulated a two-fold increase in the uptake of 14C-dieldrin by precision-cut liver slices, as well as an increased uptake and efflux of 3H-DMBA (Gilroy et al., 1996Go).

Dieldrin and other nonplanar OCs interact with the mammalian consititutive androstane receptor (CAR) and the pregnane X receptor (PXR) to induce cytochrome P450 (CYP) 2B or CYP3A family enzymes, respectively (Goodwin et al., 2002Go; Sueyoshi and Negishi, 2001Go). However, dieldrin-stimulated biliary excretion of the 14C-dieldrin or 3H-DMBA challenge dose was unexpected because fish are refractory to CYP by nonplanar OCs (Vodicnik et al., 1981Go). In vitro work with hepatic microsomes, from control and dieldrin-pretreated fish, detected no induction of the CYP system or conjugative enzyme activities. Total CYPs, exthoxyresorufin-O-deethylase (EROD), and pentoxyresorufin-O-deethylase (PROD) activities were not significantly different (Gilroy et al., 1993Go). There were also no differences in glutathione S-transferase (GST) or UDP glucuronosyltransferase (UDPGT) activities (Gilroy et al., 1993Go). Substrates well characterized in mammals were used. Isoform-specific substrate selectivities for rainbow trout conjugative enzymes were unknown. Western blot analysis revealed no changes in six hepatic CYP isozymes (Gilroy et al., 1996Go). In addition, hepatic microsomes from control and dieldrin-pretreated fish contained equivalent aryl hydrocarbon hydroxylase (AHH) activities toward 14C-BP and 3H-DMBA (Gilroy et al., 1996Go).

Increased performance of cytosolic binding proteins, putatively involved in intracellular trafficking to sites of metabolism, and induction of proteins involved in the excretion of xenobiotics into bile were also examined (Curtis et al., 2000Go). Hepatic cytosolic binding of 3H-DMBA doubled in dieldrin-fed trout after 10 weeks of exposure. However, immunohistochemistry of liver revealed no changes in multidrug resistance (i.e., adenosine binding cassette) proteins.

This research examined in vivo metabolism and disposition of BP in rainbow trout pretreated with dieldrin. BP was selected for three reasons. First, previous research demonstrated increased biliary excretion of DMBA in dieldrin-fed fish (Donohoe et al., 1998Go). This study assessed whether a similar increase occurred with a subsequent dose of BP. Second, BP is the ligand being used in ongoing protein studies to identify the mechanism by which dieldrin pretreatment enhances biliary excretion. Third, researchers have extensively characterized the complex metabolism of BP. Therefore, this compound was used to observe the in vivo state of the CYP system, UDP-glucuronyltransferases, and sulfotransferases. The biliary polar metabolite profile of 14C-BP in control and dieldrin-fed fish was compared to assess preferential stimulation of a pathway not identified using in vitro methods. Conversion of dieldrin to the diol was a potential biotransformation pathway. Therefore, epoxide hydrolase activities in hepatic cytosol and microsomes were evaluated. The hepatic concentration of dieldrin was determined, so altered BP disposition was interpreted in context of target organ concentration.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Chemicals.
Dieldrin was purchased from AccuStandard, Inc. (New Haven, CT; 99% pure). 3H-Cis-stilbene oxide and 3H-trans-stilbene oxide were gifts from Dr. Bruce Hammock (University of California, Davis; 1 mCi/mmol). Bicinchoninic acid (BCA) protein assay reagents were purchased from Sigma Chemical Company (St Louis, MO). Unlabeled BP was purchased from Sigma Chemical Co. (St. Louis, MO; 97% pure). Radiolabeled 14C-BP was obtained from ChemSyn Laboratories (Lenexa, KS; 51.6 mCi/mmol, >98% purity by TLC). The enzymes ß-glucuronidase (type H-3 from Helix pomatia) and arylsulfatase (type V from keyhole limpets, Patella vulgata; ß-glucuronidase activity < 2 Sigma units per mg solid) were purchased from Sigma Chemical Co. (St. Louis, MO). The following standards were obtained from the National Cancer Institute (NCI) Chemical Carcinogen Reference Standard Repository (distributed by Midwest Research Institute, Kansas City, MO): 3-hydroxybenzo[a]pyrene (=98% pure by high-performance liquid chromatography, or HPLC), benzo[a]pyrene-3-sulfate potassium salt (>99% pure by HPLC), and 3-benzo(a)pyrenyl ß-D-glucopyranosiduronic acid (99% pure by HPLC). Amersham Corporation (Arlington Heights, IL) supplied NCS II Tissue Solubilizer. All additional material and reagents were purchased from Sigma Chemical Co. (St. Louis, MO) or Fisher Scientific (Pittsburgh, PA).

Animals.
Shasta strain rainbow trout (Oncorhynchus mykiss) were provided by the Marine/Freshwater Biomedical Center at Oregon State University (Corvallis, OR). Fish were held in continuous-flow (approximately 6 l/min; 13°C) circular tanks (88.9 l; 80 fish/tank) and kept on a 15-h light/9-h dark photoperiod.

Dieldrin exposure.
Fish (~2 g) were fed a growth ration (4% dry weight diet/dry weight fish) of Oregon Test Diet (Sinnhuber et al., 1977Go) with or without dieldrin (15 ppm; 0.324 mg dieldrin/kg/day) for 9 weeks. Due to signs of toxicity, fish were fed a maintenance ration (2% dry weight diet/dry weight fish) without dieldrin for the remaining 3 weeks.

Liver preparation.
Following 3, 6, 9, and 12 weeks of dieldrin pretreatment, fish were euthanized (200 mg/l tricane methane sulfonate), weighed, and livers excised. Liver is approximately 1% of the body weight in rainbow trout. To obtain sufficient mass for analysis it was necessary to composite tissue. Pooled intact liver samples were frozen at -20°C for dieldrin residue analysis (samples were pooled into 3 groups for each time-treatment combination to yield 0.5–1.0 g of tissue per group; 24 groups total). Remaining livers were analyzed for microsomal and cytosolic epoxide hydrolase (mEH and cEH) activities. Samples were pooled as described above, homogenized in 3 volumes of buffer (10 mM potassium phosphate, 0.15 M potassium chloride [KCL], 1 mM EDTA, 0.1 mM butylhydroxytoluene [BHT], 0.1 mM phenylmethanesulfonyl fluoride [PMSF], pH 7.5) and centrifuged (10,000 x g for 23 min, followed by a second centrifugation at 100,000 x g for 90 min). Supernatant was collected as the cytosolic fraction, and the microsomal pellet was resuspended in buffer (0.1 M potassium phosphate, 1 mM EDTA, 0.1 mM PMSF, 20% glycerol, pH 7.4). Microsomes and cytosol were frozen at -80°C until use.

Dieldrin residue analysis.
Livers were ground in a glass mortar with 5 g sodium sulfate. The mixture was poured into a chromatographic column, filled with sodium sulfate and dichloromethane : hexane (1:1). Sodium sulfate (an additional 2 g) was added to the mortar and then the column to remove residual sample. Eluates were evaporated to 1 ml, and solvent exchange (dichloromethane/hexane) eliminated dichloromethane. The volume was adjusted to 10 ml with hexane and diluted by a factor of 10. Two-microliter samples were analyzed by gas chromatography with electron capture detection (Varian Star 3400 Cx, Autosampler 8200), using a ramped temperature program (initial temperature 250°C, column temperature 150°C for 2 min, 255°C for 14 min, and 270°C for 15 min).

Depletion of glutathione by dialysis and enzyme assays.
Reduced and oxidized glutathione (GSH and GSSG) were depleted by dialysis to prevent conjugation of stilbene oxides in the microsomal and cytosolic epoxide hydrolase assays. Hepatic cytosol and microsomes (0.5 ml) were thawed and dialyzed for 2 h at 4°C in 1 l of a 10 mM potassium phosphate buffer (150 mM KCL, 1 mM EDTA, 0.1 mM BHT, pH 7.5). Efficiency was determined by paired analyses of dialyzed and nondialyzed samples. Glutathione concentrations were measured by HPLC with ultraviolet (UV) detection and 2,4 dinitrofluorobenzene (FDNB) (Reed and Will, 1999Go). Samples were derivatized, with initial formation of S-carboxymethyl derivatives of free thiols by a reaction with iodoacetic acid. The primary amines were converted to 2,4 dinitrophenyl (DNP) derivatives, after a second derivatization with Sanger’s reagent and FDNB.

Following dialysis, protein concentrations were determined using the BCA assay (bovine serum albumin served as the standard). Preliminary enzymatic assays at different pHs (6.5, 7.0, 7.5, 8.0, 8.5, 9.0, and 9.5) and temperatures (4, 15, 23, and 30°C) optimized conditions for subsequent work. 3H-trans-stilbene oxide and 3H-cis-stilbene oxide were used as substrates for cEH and mEH, respectively. Activities of mEH and cEH were measured radiometrically (in preparations from weeks 3, 6, 9, and 12 at optimum pH and temperature) utilizing differential partitioning of the epoxide into dodecane and the diol metabolites into the aqueous phase (Gill et al., 1983Go). Thin layer chromatography (TLC) assessed partitioning of stilbene oxide, stilbene diol, and the absence of GSH conjugates of stilbene oxide (Gill et al., 1983Go). Diol metabolites in the remaining aqueous phase were removed following extraction with hexanol. The organic phase was concentrated under nitrogen, and 25 µl was counted for diols by liquid scintillation counting (LSC). Fifty microliters were applied to silicate prelayered plates developed in toluene : n-propanol (19:1) and dried overnight. Bands of silica-gel plates (1 cm) were scraped, added to LSC cocktail (1 ml), and counted by LSC. The presence of mercapturate conjugates was assessed by LSC of the remaining aqueous phase.

14C-BP exposure.
Tissue distribution of 14C-BP was examined in fish fed control (n = 9) or 15 ppm dieldrin (n = 9) for 9 weeks, fish fed 15 ppm dieldrin for 9 weeks followed by control diet for 3 weeks (n = 9), and fish fed control diet for 12 weeks (n = 9). Fish were transferred to polypropylene buckets containing clean well-water (static conditions) and submersed charcoal filters. Fish were not fed for 24 h prior to intraperitoneal (ip) injection with 10 µmol 14C-BP/kg in menhaden oil (10 ml/kg). Fish were killed 24 h after the ip injection and gallbladder/bile, liver, and dissectible visceral fat were removed.

Tissue preparation.
Gallbladder/bile was stored in amber microcentrifuge tubes without solvent to minimize spontaneous oxidation to quinone metabolites. Bile subsamples were removed and the remainder transferred to -200C for subsequent metabolite analysis. Polar and nonpolar metabolites in the subsamples were separated using a methanol/water-chloroform extraction system (Bligh and Dyer, 1959Go). The aqueous and chloroform fractions were evaporated, Cytoscint ESTM added, and radioactivity analyzed by LSC. Liver and fat tissues were digested with NCS II Tissue Solubilizer at 40°C for 48 h. Cytoscint ESTM was added and radioactivity analyzed by LSC.

HPLC analysis of biliary metabolites.
Gallbladder/bile samples frozen for metabolite analysis were extracted using the method described by Willet et al.(2000), with the following modifications. Samples were combined with 1 ml buffer (potassium phosphate with 1.0% (w/v) bovine serum albumin, pH 6.8 at 37°C) and extracted with 2 ml ethyl acetate to isolate parent 14C-BP and unconjugated oxidized metabolites. The aqueous phase was incubated with ß-glucuronidase for 6 h at 37°C (1000 units; 8.4 µl) and extracted with 2 ml ethyl acetate to isolate cleaved glucuronide conjugates. The aqueous phase was then adjusted from pH 6.8 to 5 (using HCl) and incubated with 19 units arylsulfatase (dissolved in sodium chloride immediately before use; negligible ß-glucuronidase activity) for 6 h at 37°C. Cleaved sulfate conjugates were extracted with 2 ml ethyl acetate. Organic fractions (50 µl duplicates) were analyzed for radioactivity by LSC, and the remainder was stored at -20°C for HPLC analysis. The remaining aqueous phase (50 µl duplicates) was also analyzed by LSC for residual polar material (e.g., glutathione, sulfate, or glucuronide conjugates refractory to cleavage). Blanks and standards were processed with samples for HPLC analysis.

The organic fractions, containing BP and unconjugated metabolites, were transferred to amber microcentrifuge tubes. Ethyl acetate was evaporated to dryness under a nitrogen stream, and residue was resuspended in 50 µl methanol (Varanasi et al., 1982Go). Twenty microliters were injected onto a C-18 reverse-phase HPLC column (Vydac 218TP54, Vydac, Hesperia, CA) and the remainder stored at -20°C. A two-step gradient was used to separate BP and its metabolites, as described by Willet et al.(2000) with the following modifications: methanol : water : acetic acid (50:49.5:0.5) to methanol : water (83.5:16.5) in 30 min to 100% methanol in 60 min. Final conditions were maintained for an additional 5 min before the gradient was returned to initial settings. The flow rate was lowered (0.4 ml/min) during sample injection and then held at 1 ml/min.

Eluent from the column was analyzed by fluorescence spectrophotometry; however, radioactivity provided a more sensitive method of detection and was used for data analysis (Nishimoto et al., 1992Go; Varanasi et al., 1986Go). HPLC fractions were collected at 3-min intervals using a fraction collector (FC 203, Gilson Inc., Middleton, WI). Duplicates (300 µl) from each fraction were pipetted into Deep-Well LumaPlateTM microplates that contained solid scintillator (Bornsen, 2000Go). The plates were evaporated, sealed, and placed in the TopCount® microplate scintillation and luminescence counter (Packard Instrument Company, Meriden, CT) (Bornsen, 2000Go).

Statistical analysis.
Statgraphics Plus 5.0 was used for all statistical analyses. Two-way analysis of variance compared multiple means and determined significant time or treatment effects. A two-sample comparison, or t-test, compared two means for significant treatment effects. Significance was determined using a 95.0% confidence level (p < 0.05). Nonparametric alternatives were used when assumptions of normality or equality of variance were not met.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Animals
Dieldrin-fed fish exhibited signs of toxicity between 6 and 9 weeks of treatment. The percent cumulative mortality was fairly low (2.1% in controls and 3.8% in dieldrin-fed fish over 12 weeks); however, the body weight of dieldrin-fed fish was 75% of control fish at week 9 (Fig. 1Go). Although the change in body weight was not statistically significant, all fish were fed a control diet after 9 weeks. After an additional 3 weeks body weights for control and dieldrin-fed fish were equivalent (Fig. 1Go).



View larger version (12K):
[in this window]
[in a new window]
 
FIG. 1. Body weight of rainbow trout (initial weight ~2 g) fed control or dieldrin (0.324 mg dieldrin/kg/day) diets for 9 weeks. The body weight of dieldrin-fed fish was 75% of control fish after 9 weeks ({downarrow}). Therefore, all fish were fed control diet for an additional 3 weeks. Values are means ± SE (weeks 3 and 6: n = 9; week 9: n = 18; week 12: n = 9).

 
Dieldrin Residue Analysis
The concentration of dieldrin in liver increased with treatment time and was significantly higher at week 9 (Fig. 2Go). The estimated daily dose of dieldrin up to 9 weeks was 0.324 mg dieldrin/kg body weight (15 ppm). Three weeks after dieldrin-fed fish resumed control diet the concentration in liver decreased markedly (Fig. 2Go). Dieldrin was not detected in the livers of rainbow trout fed a control diet. Mean recoveries were 84% (n = 4).



View larger version (34K):
[in this window]
[in a new window]
 
FIG. 2. Mean dieldrin concentration in liver from rainbow trout fed control or dieldrin (0.324 mg dieldrin/kg/day) diets for 9 weeks. All fish were fed control diet for an additional 3 weeks due to signs of dieldrin toxicity. Dieldrin concentration increased at week 9 but decreased markedly after dieldrin was removed from the diet. Dieldrin was not detected in livers from control fish. Values are means ± SE. *Significantly different from control, p < 0.05.

 
Depletion of Glutathione by Dialysis and Enzyme Assays
Glutathione was efficiently depleted by dialysis. Glutathione levels in cytosol were reduced 77.4% from 2.12 ± 0.68 to 0.48 ± 0.10 nmol GSH/mg protein. Glutathione levels in microsomes were reduced 60.4% from 1.26 ± 0.05 to 0.50 ± 0.03 nmol GSH/mg protein. Sequential extraction of incubation media for mEH and cEH from control and dieldrin-fed fish (9 week samples, n = 3 per condition) validated specificity of the enzymatic assays (Gill et al., 1983Go). TLC of hexanol extract concentrates yielded a single spot 7–9 cm from the origin (data not shown). The final aqueous phase contained only trace amounts of radioactivity, indicating negligible formation of mercapturates and hence insignificant glutathione conjugation.

Preliminary assays that verified optimal pHs and temperatures (data not shown) for mEH (pH 8 to 8.5 and 23°C) and cEH (pH 6.5 to 7.5 and 23°C) yielded results similar to previous work for rainbow trout (Lauren et al., 1989Go). Differential partitioning of the epoxide into dodecane yielded 90% of the radioactivity; therefore, the enzyme was not substrate limited. The activity of mEH, the form of the enzyme active toward arene oxides (Wixtrom and Hammock, 1984Go), was unaffected by dieldrin (Fig. 3Go). The activity of cEH was also unaffected in dieldrin-fed fish (overall grand mean 3.91 ± 0.28 nmol/min/mg protein).



View larger version (12K):
[in this window]
[in a new window]
 
FIG. 3. Microsomal epoxide hydrolase (mEH) activity toward 3H-cis-stilbene oxide in rainbow trout fed control or dieldrin (0.324 mg dieldrin/kg/day) diets for 9 weeks. All fish were fed control diet for an additional 3 weeks due to signs of dieldrin toxicity. Values are means ± SE. There were no significant time or treatment effects.

 
14C-BP Disposition
Two-way analysis of variance revealed no significant interaction between time and treatment on the tissue concentrations of 14C-BP in liver or fat (Fig. 4Go). There was a significant difference in the mean concentrations of 14C-BP between weeks 9 and 12, averaged over treatment. After 12 weeks there was a significantly higher liver concentration (nmol/g) of 14C-BP, in both control and dieldrin-fed fish, than after 9 weeks (Fig. 4AGo). However, the concentration of 14C-BP in visceral fat was significantly higher in control and dieldrin-fed fish at 9 weeks compared to 12 weeks (Fig. 4BGo).



View larger version (24K):
[in this window]
[in a new window]
 
FIG. 4. Tissue concentrations of 14C-benzo[a]pyrene in liver (A) and fat (B) following dieldrin pretreatment for 9 weeks. Rainbow trout (initial weight ~2 g) were fed control or dieldrin (0.324 mg dieldrin/kg/day) diets for 9 weeks. Dieldrin was removed from the diet at this time due to overt toxicity. At weeks 9 and 12 trout were injected (ip) with 10 µmol 14C-BP/kg. Fish were sacrificed 24 h later, and gallbladder/bile, liver, and visceral fat were removed for analysis. Values are means ± SE (week 9: control n = 8, treated n = 5; week 12: control n = 9, treated n = 8). {dagger}Significant time effect: significantly different from week 9 or week 12, p < 0.05.

 
There was no significant interaction between time and treatment on the tissue concentration or total biliary excretion of 14C-BP in bile. However, there was a significant treatment effect. The mean concentration of 14C-BP in bile, averaged over time, was significantly different between control and dieldrin-fed fish. Dieldrin-fed fish had a significantly higher concentration (nmol/g) of 14C-BP in bile compared to control fish (Fig. 5AGo). Stimulation of total biliary excretion (total nmol) was more pronounced compared to the concentration of 14C-BP in bile (Figs. 5A and 5BGo). On average, polar metabolites of 14C-BP in dieldrin-fed fish were significantly elevated (2.4-fold higher at 9 weeks and 2.2-fold higher at 12 weeks) compared to control fish (Fig. 5BGo). No difference was observed in nonpolar material between control and dieldrin-fed trout (Fig. 5BGo).



View larger version (24K):
[in this window]
[in a new window]
 
FIG. 5. Concentration of 14C-benzo[a]pyrene in bile (A) and total biliary excretion (B) following dieldrin pretreatment for 9 weeks. Rainbow trout (initial weight ~2 g) were fed control or dieldrin (0.324 mg dieldrin/kg/day) diets for 9 weeks. Dieldrin was removed from the diet at this time due to overt toxicity. At week 9 and 12 trout were injected (ip) with 10 µmol 14C-BP/kg. Fish were sacrificed 24 h later, and gallbladder/bile, liver, and visceral fat were removed for analysis. Values are means ± SE (week 9: control n = 5, treated n = 3; week 12: control n = 9, treated n = 8). *Significant treatment effect: significantly different from control, p < 0.05.

 
Biliary 14C-BP Polar Metabolite Profile
The biliary metabolite profile of 14C-BP was not significantly different between control and dieldrin-fed fish (Figs. 6Go, 7Go, and 8Go). Induction of a novel biotransformation pathway was not detected in treated fish (e.g., a unique metabolite was not formed). In addition, there was no evidence that an enzyme expressed at a low level constitutively was being induced (e.g., elevated levels of specific metabolites in pretreated fish were not observed). Although there was no selective enhancement of any particular metabolite with dieldrin pretreatment, there was a general increase in many of the biliary metabolites from dieldrin-fed fish (Figs. 6Go, 7Go, and 8Go). The efficiency of hydrolysis for the glucuronide conjugate of 3-hydroxy was 91% ± 4.7; however, the sulfate conjugate of 3-hydroxy was resistant to hydrolysis.



View larger version (39K):
[in this window]
[in a new window]
 
FIG. 6. 14C-Benzo[a]pyrene parent and unconjugated oxidized metabolite profile in gallbladder/bile at week 9 (A) and week 12 (B), following dieldrin pretreatment in rainbow trout (initial weight ~2 g) for 9 weeks. Bile was extracted with ethyl acetate and run on reverse-phase HPLC to separate the metabolites. HPLC fractions were collected at 3-min intervals, and 300-µl duplicates of each fraction were placed in the TopCount system to analyze radioactivity. Values are mean cpms per fraction. Of the total radioactivity in the major peaks the percentage of parent 14C-BP (~63 min) was 64% (control) and 77% (treated) at 9 weeks (A) and 75% (control) and 66% (treated) at 12 weeks (B).

 


View larger version (41K):
[in this window]
[in a new window]
 
FIG. 7. 14C-Benzo[a]pyrene metabolite profile of cleaved glucuronide conjugates in gallbladder/bile at week 9 (A) and week 12 (B), following dieldrin pretreatment in rainbow trout (initial weight ~2 g) for 9 weeks. Metabolites remaining in the aqueous phase, following the first extraction with ethyl acetate, were hydrolyzed with ß-glucuronidase to regenerate phase-1 metabolites prior to their conjugation. Cleaved glucuronide conjugates were extracted with ethyl acetate and run on reverse-phase HPLC to separate the metabolites. HPLC fractions were collected at 3-min intervals, and 300-µl duplicates of each fraction were placed in the TopCount system to analyze radioactivity. Values are mean cpms per fraction. Of the total radioactivity in the major peaks, the parent 14C-BP (~63 min) not removed in the initial extraction was 19% (control and treated) at 9 weeks (A) and 15% (control) and 11% (treated) at 12 weeks (B). The 3-hydroxy peak (~46 min) was confirmed using a standard.

 


View larger version (44K):
[in this window]
[in a new window]
 
FIG. 8. 14C-Benzo[a]pyrene metabolite profile of cleaved sulfate conjugates in gallbladder/bile at week 9 (A) and week 12 (B), following dieldrin pretreatment in rainbow trout (initial weight ~2 g) for 9 weeks. Metabolites remaining in the aqueous phase, following the second extraction with ethyl acetate, were hydrolyzed with arylsulfatase to regenerate phase-1 metabolites prior to their conjugation. Cleaved sulfate conjugates were extracted with ethyl acetate and run on reverse-phase HPLC to separate the metabolites. HPLC fractions were collected at 3-min intervals, and 300-µl duplicates of each fraction were placed in the TopCount system to analyze radioactivity. Values are mean cpms per fraction.

 
Polar Metabolites of 14C-BP Excreted into Bile
Only a small percentage of 14C-BP in bile subsamples appeared in the organic phase (Fig. 6Go). For that reason, parent 14C-BP on HPLC chromatograms was assumed to be contamination on the outside of the gallbladder, as a result of the ip injection (e.g., BP that was not processed by the liver and excreted into bile). Therefore, biliary system performance (e.g., the amount transported from liver to bile) was estimated after parent compound was subtracted from the total recovered. This yielded the total amount (nmol) of 14C-BP polar metabolites excreted into bile (Fig. 9Go).



View larger version (29K):
[in this window]
[in a new window]
 
FIG. 9. Polar metabolites of 14C-benzo[a]pyrene excreted into bile, following dieldrin pretreatment in rainbow trout (initial weight ~2 g) for 9 weeks: phase 1 metabolites (A), cleaved glucuronide conjugates (B), cleaved sulfate conjugates (C), and residual polar material (D). Values are means ± SE (week 9: control n = 3, treated n = 4; week 12: control and treated n = 3). There were no significant time or treatment effects.

 
Two-way analysis of variance revealed no significant interaction between time and treatment on biliary excretion of polar metabolites excreted into bile (Figs. 9A–9DGo). At week 12, the amount in dieldrin-fed fish was elevated for unconjugated oxidized metabolites (Fig. 9AGo), cleaved glucuronide conjugates (Fig. 9BGo), cleaved sulfate conjugates (Fig. 9CGo), and residual polar material (Fig. 9DGo). However, the amounts in dieldrin-fed fish were not significantly different from controls.

The nmol of 14C-BP polar metabolites in each fraction were divided by the total nmol recovered in bile for each fish. This yielded the percent radioactivity in each fraction containing phase 1 metabolites, cleaved glucuronide conjugates, cleaved sulfate conjugates, and residual polar material. Statistical analysis revealed no significant time or treatment effects. Therefore, the averages of control and treated fish at weeks 9 and 12 were pooled into one average. Of the total radioactivity in bile, the majority was detected as cleaved glucuronide conjugates (24%). Only a small proportion of free metabolites (6%) and cleaved sulfate conjugates (2%) were recovered. Parent compound, assumed to be contamination on the outside of the gallbladder as a result of the ip injection, represented 25% of the total radioactivity.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
During a 16-week study of dieldrin bioaccumulation in rainbow trout lipid-normalized, whole body dieldrin concentrations increased with time up to 8 weeks (Shubat and Curtis, 1986Go). However, the dieldrin concentrations measured at 8 weeks unexpectedly decreased by 50% after 16 weeks of exposure. In vivo experiments confirmed prolonged dieldrin treatment of rainbow trout markedly altered disposition of a subsequent dose of 14C-dieldrin (Gilroy et al., 1993Go). Feeding rainbow trout 0.4 mg dieldrin/kg/day for 10 or 12 weeks stimulated the biliary excretion of a subsequent dose of 14C-dieldrin by 500% and increased the disposition of 14C-dieldrin to liver (200%) and mesenteric fat (500% at 10 weeks, 1200% at 12 weeks). Tissue distribution of the 14C-dieldrin dose was the same in fish fed dieldrin for 2, 4, 6, or 8 weeks and time-matched controls. In vitro accumulation of 14C-dieldrin doubled in liver slices from rainbow trout fed 0.3 mg dieldrin/kg/day for 10 weeks (Gilroy et al., 1996Go). There was no difference between total liver lipid in control and treated fish.

Approximately 90% of total 14C-dieldrin equivalents in bile of control and dieldrin-fed fish were polar metabolites (Gilroy et al., 1993Go). Induction of hepatic xenobiotic metabolizing enzymes was investigated as the basis for stimulated biliary excretion of 14C-dieldrin. In vitro work demonstrated no induction of the CYP system or increased activity of several conjugating enzymes (Gilroy et al., 1996Go). Hepatic microsomal metabolism of a variety of substrates selective for different CYP isoforms and Western blotting for 6 CYP isoforms revealed no differences between control and dieldrin-fed fish. Hepatic glucuronyl transferase and GSH transferase activities were also the same in control and dieldrin-fed fish. As an epoxide, dieldrin was a potential substrate of mEH (Wixtrom and Hammock, 1984Go). However, neither hepatic mEH, nor cEH activities increased in dieldrin-fed fish (Fig. 3Go).

Uptake and efflux of 3H-DMBA was significantly higher in liver slices from dieldrin-fed rainbow trout than those from controls (Gilroy et al., 1996Go). The DMBA binding capacity of hepatic cytosol from dieldrin-fed fish was twice that of controls (Curtis, 2000Go). In addition, biliary excretion of DMBA approximately doubled in dieldrin-fed rainbow trout (Donohoe et al., 1998Go). Therefore, altered hepatic elimination of a lipophilic xenobiotic, induced by prolonged dieldrin treatment, was not specific to that cyclodiene. A significant interaction occurred with the PAH, DMBA.

The present research examined the effects of dieldrin pretreatment on the metabolism and disposition of another PAH, BP. This substrate, which undergoes complex metabolism, characterized the in vivo state of the CYP system, UDP-glucuronyltransferases, and sulfotransferases. This assessed alteration of a particular CYP isoform(s) for which selective substrates or antibodies were not developed.

In the current study fish were fed 0.3 mg dieldrin/kg/day for 3, 6, or 9 weeks. The dieldrin-fed rainbow trout exhibited some, although not statistically significant, signs of toxicity after 9 weeks of treatment (Fig. 1Go). Some enlargement and vacuolization of hepatocytes was observed in rainbow trout fed 0.3 mg dieldrin/kg/day for 10 weeks (Curtis et al., 2000Go). These results indicated 0.3 mg dieldrin/kg/day approximated a threshold for subchronic toxicity in rainbow trout. There were no signs of toxicity in fish fed control diet for 3 weeks after 9 weeks of dieldrin treatment (Fig. 1Go). The peak hepatic dieldrin concentration of 1.2 ug/g (3.2 nmol/g) at 9 weeks decreased to 0.4 ug/g (1.0 nmol/g) after 3 weeks on control diet (Fig. 2Go). Rapid hepatic elimination of dieldrin during 3 weeks on control diet was consistent with stimulated biliary excretion of this cyclodiene, reported after prolonged treatment (Gilroy et al., 1993Go).

There were no statistically significant differences in 14C-BP concentrations in liver or fat associated with dieldrin pretreatment (Fig. 4Go). This was true for fish fed dieldrin for 9 weeks and for fish fed dieldrin 9 weeks, followed by control diet for 3 weeks (12 weeks). The liver concentration of 14C-BP was significantly higher after 12 weeks in control and dieldrin-fed fish, than after 9 weeks (Fig. 4AGo). However, the 14C-BP concentration in visceral fat was significantly higher in control and dieldrin-fed fish at 9 weeks, compared to 12 weeks (Fig. 4BGo). Triacylglycerols are incorporated into lipoproteins by the liver for secretion into the circulation and transportation to various tissues. The liver is the main regulator of lipid homeostasis and incorporates BP into lipoproteins (Aarstad et al., 1987Go). A change in ration between weeks 9 and 12 (4% growth ration vs. 2% maintenance ration, respectively) probably influenced 14C-BP tissue distribution. Energy storage as fat was favored during the first 9 weeks at high ration. When fish were placed on a lower ration, synthesis of fat probably decreased by week 12. Therefore, lipoprotein export by the liver to transport fat to adipose tissue probably decreased. As a result, accumulation of 14C-BP by liver increased, while distribution to fat decreased.

Consistent with earlier work using DMBA (Donohoe et al., 1998Go), dieldrin pretreatment for 9 weeks stimulated biliary excretion of 14C-BP (Fig. 5Go). Stimulated biliary excretion of 14C-BP persisted in fish fed control diet for 3 weeks after 9 weeks of dieldrin treatment (12 weeks). This occurred despite markedly reduced hepatic dieldrin concentrations after 3 weeks on control diet (Fig. 2Go). Dieldrin pretreatment significantly elevated the concentration of 14C-BP in bile (142% and 200% at 9 and 12 weeks, respectively) (Fig. 5AGo). Extraction of bile subsamples with methanol/water-chloroform confirmed dieldrin pretreatment significantly stimulated total biliary excretion of 14C-BP polar metabolites (244% and 221% at week 9 and 12, respectively; Fig. 5BGo).

Evaluation of biliary polar metabolite profiles of 14C-BP revealed no significant differences between control and dieldrin-fed fish (Figs. 6Go, 7Go, and 8Go). There was no selective enhancement of any particular metabolite or formation of a novel metabolite with dieldrin pretreatment. Many of the biliary metabolites increased in dieldrin-fed fish, which suggested no particular CYP was altered. Of the total radioactivity in bile, the majority was detected as cleaved glucuronide conjugates (24%), while only a small proportion of free metabolites (6%) and cleaved sulfate conjugates (2%) were recovered. Recovery of parent compound (25%) was assumed to be contamination on the outside of the gallbladder, as a result of the ip injection.

In agreement with other studies (Nishimoto et al., 1992Go; Varanasi et al., 1982Go, 1986Go; Willett et al., 2000Go), glucuronide conjugates predominated in gallbladder bile, while only a small proportion of sulfate conjugates were recovered (Figs. 7Go and 8Go). There may be several reasons why detection of sulfate conjugates is low. Lower water solubility and recognition by organic anion transport systems for sulfate compared to glucuronide conjugates decreased their elimination (Parkinson, 1996Go). Another potential explanation was directly related to sulfation reactions, catalyzed by sulfotransferases. Transfer of the sulfate group, from the cofactor 3'-phosphoadenosine 5'-phosphosulfate, to hydroxyl groups on PAHs introduced a good leaving group, which formed a reactive, electrophilic carbocation species (Parkinson, 1996Go). Instability of sulfate conjugates probably contributed to the small percentage recovered in bile.

Approximately 43% of the total radioactivity remained in the final aqueous phase. This residual polar material contained water-soluble conjugates including glutathione and glucuronide, or sulfate conjugates not hydrolyzed during the ß-glucuronidase or arylsulfatase reactions. Standards were chosen based on earlier studies that demonstrated 3-hydroxy is one of the primary BP metabolites formed, which was consistent with our results (Nishimoto et al., 1992Go; Varanasi et al., 1982Go; Willett et al., 2000Go).

Polar metabolites of 14C-BP (unconjugated oxidized metabolites, cleaved glucuronide conjugates, cleaved sulfate conjugates, or residual polar material) excreted into bile were not significantly increased in dieldrin-fed fish (Fig. 9Go). At week 12, all polar metabolites were elevated in dieldrin-fed fish. However, this response was not apparent at week 9. Oxidation products created during storage and sample processing explained why this data detected insignificantly enhanced biliary excretion of 14C-BP polar metabolites. Analysis of bile subsamples, immediately following each experiment, provided a better estimate of total biliary excretion than reconstruction of data following the metabolite analysis.

In conclusion, chronic dieldrin exposure stimulated biliary excretion in rainbow trout given subsequent doses of 14C-dieldrin, 3H-DMBA, and 14C-BP. This study confirmed that the interaction was not explained by induction of xenobiotic metabolizing enzymes. Therefore, the mechanism by which dieldrin pretreatment enhances biliary excretion of lipophilic compounds requires further examination. Biliary excretion of xenobiotics is complex and involves hepatic processes other than metabolism (Donohoe et al., 1998Go): uptake across the plasma membrane, intracellular trafficking to sites of metabolism and elimination, and excretion into bile (Curtis et al., 2000Go; Donohoe et al., 1998Go; Gilroy et al., 1993Go, 1996Go). Immunohistochemistry with several antibodies to adenosine binding cassette proteins revealed no increased hepatic content of these proteins in sinusoidal or canalicular plasma membrane domains of dieldrin-fed fish (Curtis et al., 2000Go). Alternative approaches to assessing involvement of this protein family are warranted. Specific binding for 3H-DMBA doubled in hepatic cytosol from dieldrin-fed fish (Curtis et al., 2000Go). The current hypothesis is that prolonged dieldrin treatment increases capacity of hepatic intracellular trafficking proteins (Curtis et al., 2000Go; Donohoe et al., 1998Go; Gilroy et al., 1993Go, 1996Go). Ongoing research focuses on BP binding proteins.


    ACKNOWLEDGMENTS
 
The Oregon Agricultural Experiment Station and the National Institute of Health (P30 ES03850) supported this research.


    NOTES
 
1 To whom correspondence should be addressed at Department of Environmental and Molecular Toxicology, ALS 1007, Oregon State University, Corvallis, OR 97331. Fax: (541) 737-0497. E-mail: larry.curtis{at}orst.edu. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Aarstad, K., Toftgard, R., and Nilsen, O. G. (1987). A comparison of the binding and distribution of benzo[a]pyrene in human and rat serum. Toxicology 47, 235–245.[CrossRef][ISI][Medline]

Bligh, E. G., and Dyer, W. J. (1959). A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37, 911–917.[ISI]

Bornsen, K. O. (2000). Using the TopCount microplate scintillation and luminescence counter and deep-well LumaPlate microplates in combination with micro separation techniques for metabolic studies, pp. 1–6. Packard Instrument Company, Meriden, CT.

Cerniglia, C. E., and Heitkamp, M. A. (1989). Microbial degradation of polycyclic aromatic hydrocarbons (PAH) in the aquatic environment. In Metabolism of Polycyclic Aromatic Hydrocarbons in the Aquatic Environment (U. Varanasi, Ed.), pp. 42–68. CRC Press, Boca Raton, FL.

Curtis, L. R., Hemmer, M. J., and Courtney, L. A. (2000). Dieldrin induces cytosolic 3H-7,12-dimethylbenz[a]anthracene binding but not multidrug resistance proteins in rainbow trout liver. J. Toxicol. Environ. Health 60, 275–289.[CrossRef][ISI]

Donohoe, R. M., Zhang, Q., Siddens, L. K., Carpenter, H. M., Hendricks, J. D., and Curtis, L. R. (1998). Modulation of 7,12-dimethylbenz[a]anthracene disposition and hepatocarcinogenesis by dieldrin and chlordecone in rainbow trout. J. Toxicol. Environ. Health 54, 227–242.[CrossRef][ISI]

Gill, S. S., Ota, K., and Hammock, B. D. (1983). Radiometric assays for mammalian epoxide hydrolases and glutathione S-transferase. Anal. Biochem. 131, 273–282.[ISI][Medline]

Gilroy, D. J., Carpenter, H. M., Siddens, L. K., and Curtis, L. R. (1993). Chronic dieldrin exposure increases hepatic disposition and biliary excretion of 14C-dieldrin in rainbow trout. Fundam. Appl. Toxicol. 20, 295–301.[CrossRef][ISI][Medline]

Gilroy, D. J., Miranda, C. L., Siddens, L. K., Zhang, Q., Buhler, D. R., and Curtis, L. R. (1996). Dieldrin pretreatment alters 14C-dieldrin and 3H-7,12-dimethylbenz[a]anthracene uptake in rainbow trout liver slices. Fundam. Appl. Toxicol. 30, 187–193.[CrossRef][ISI][Medline]

Goodwin, B., Redinbo, M. R., and Kliewer, S. A. (2002). Regulation of CYP3A gene transcription by the pregnane X receptor. Annu. Rev. Pharmacol. Toxicol. 42, 1–23.[CrossRef][ISI][Medline]

Lauren, D. J., Halarnkar, P. P., Hammock, B. D., and Hinton, D. E. (1989). Microsomal and cytosolic epoxide hydrolase and glutathione S-transferase activities in the gill, liver, and kidney of the rainbow trout, Salmo gairdneri. Biochem. Pharmacol. 38, 881–887.[CrossRef][ISI][Medline]

McElroy, A. E., Farrington, J. W., and Teal, J. M. (1989). Bioavailability of polycyclic aromatic hydrocarbons in the aquatic environment. In Metabolism of Polycyclic Aromatic Hydrocarbons in the Aquatic Environment (U. Varanasi, Ed.), pp. 1–39. CRC Press, Boca Raton, FL.

Nishimoto, M., Yanagida, G. K., Stein, J. E., Baird, W. M., and Varanasi, U. (1992). The metabolism of benzo(a)pyrene by English sole (Parophrys vetulus): Comparison between isolated hepatocytes in vitro and liver in vivo. Xenobiotica 22, 949–961.[ISI][Medline]

Parkinson, A. (1996). Biotransformation of xenobiotics. In Casarett & Doull's Toxicology. The Basic Science of Poisons (C. D. Klaassen, M. O. Amdur, and J. Doull, Eds.), pp. 113–186. McGraw-Hill, New York.

Reed, D. J., and Will, Y. (1999). The glutathione pathway. In Current Protocols in Toxicology (M. Maines, L. G. Costa, D. J. Reed, S. Sassa, and I. G. Sipes, Eds.), 6.0.1–6.0.2. Wiley & Sons, New York.

Shubat, P. J., and Curtis, L. R. (1986). Ration and toxicant pre-exposure influence dieldrin accumulation by rainbow trout. Environ. Toxicol. Chem. 5, 69–77.[ISI]

Sinnhuber, R. O., Hendricks, J. D., Wales, J. H., and Putnam, G. B. (1977). Neoplasms in rainbow trout, a sensitive animal model for environmental carcinogens. Ann. N.Y. Acad. Sci. 298, 389–408.

Sueyoshi, T., and Negishi, M. (2001). Phenobarbital response elements of cytochrome P450 genes and nuclear receptors. Annu. Rev. Pharmacol. Toxicol. 41, 123–143.[CrossRef][ISI][Medline]

U.S. Department of the Interior, U.S. Geological Survey (1999). The Quality of Our Nation’s Waters—Nutrients and Pesticides, 82p. U.S. Geological Survey Circular 1225. Government Printing Office, Washington, DC.

U.S. Environmental Protection Agency (U.S. EPA) (1971). Cancellation of registration under the FIFRA of products containing aldrin or dieldrin, PR Notice 71-4. U.S. Government Printing Office, Washington, DC.

U.S. Environmental Protection Agency (1990). Suspended, cancelled, and restricted use pesticides, EPA-20-T-1002. U.S. Government Printing Office, Washington DC.

U.S. Environmental Protection Agency, Office of Pollution, Prevention, and Toxics (2001). 2000 PBT Program Accomplishments. U.S. Government Printing Office, Washington, DC.

Varanasi, U., Nishimoto, M., Reichert, W. L., and Eberhart, B. L. (1986). Comparative metabolism of benzo(a)pyrene and covalent binding to hepatic DNA in English sole, starry flounder, and rat. Cancer Res. 46, 3817–3824.[Abstract]

Varanasi, U., Stein, J. E., Nishimoto, M., and Hom, T. (1982). Benzo[a]pyrene metabolites in liver, muscle, gonads, and bile of adult English sole (Parophrys vetulus). In Polynuclear Aromatic Hydrocarbons Seventh International Symposium on Formation, Metabolism and Measurement (M. W. Cooke and A. J. Dennis, Eds.), pp. 1221–1234. Battelle Press, Columbus OH.

Vodicnik, M. J., Elcombe, C. R., and Lech, J. J. (1981). The effects of various types of inducing agents on hepatic microsomal monooxygenase activity in rainbow trout. Toxicol. Appl. Pharmacol. 59, 364–374.[ISI][Medline]

Willett, K. L., Gardinali, P. R., Lienesch, L. A., and Di Guilio, R. T. (2000). Comparative metabolism and excretion of benzo[a]pyrene in 2 species of ictalurid catfish. Toxicol. Sci. 58, 68–76.[Abstract/Free Full Text]

Wixtrom, R. N., and Hammock, B. D. (1984). Membrane-bound and soluble-fraction epoxide hydrolases: Methodological aspects. In Biochemical Pharmacology and Toxicology (D. Zakim and D. A. Vessey, Eds.), pp. 1–93. Wiley & Sons, New York.





This Article
Abstract
FREE Full Text (PDF)
All Versions of this Article:
75/2/249    most recent
kfg192v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Disclaimer
Request Permissions
Google Scholar
Articles by Barnhill, M. L.
Articles by Curtis, L. R.
PubMed
PubMed Citation
Articles by Barnhill, M. L.
Articles by Curtis, L. R.