* Department of Pharmacology and Toxicology, West Virginia University, Morgantown, West Virginia 26506;
Department of Environmental and Molecular Toxicology, Oregon State University, Corvallis, Oregon 97331;
Environmental Toxicology Center, Department of Pharmacology, University of Wisconsin, Madison, Wisconsin 53705;
§ Department of Biochemistry, West Virginia University, Morgantown, West Virginia 26506;
¶ CDC/NIOSH, Health Effects Laboratory Division, Physiology Pathology Research Branch, Morgantown, West Virginia 26505; and
|| Environmental Health Sciences Center, Oregon State University, Corvallis, Oregon 97331
Received May 11, 2000; accepted June 29, 2000
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key Words: 7,12-dimethylbenz(a)anthracene; trout; liver cells; CYP1A1; DNA adducts.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Formation of the genotoxic bay region diol epoxide of DMBA in mammals proceeds through the action of the cytochrome P450 (CYP) family of enzymes. The PAH-inducible isozymes of the 1A and 1B families, as well as the phenobarbital-inducible isozymes of the 2B family have been shown to participate in mammalian metabolism of DMBA (Diamond et al., 1972; Dipple et al., 1984
; McCord et al., 1988
; Morrison et al., 1991
). Recently, Jefcoate's laboratory (Christou et al., 1994
; Pottenger and Jefcoate, 1990
; Savas et al., 1993
) has identified CYP1B1 as a prominent enzyme-metabolizing DMBA to the 3,4-diol in mammalian cells. However,
-naphthoflavone, a competitive inhibitor of mammalian CYP1A-mediated PAH metabolism, reduces DMBA-DNA adduct formation in hamster embryo cells in culture (Diamond et al., 1972
) and inhibits binding of DMBA to DNA in the skin of NIH Swiss and C57 BL mice (Dipple et al., 1984
). Studies have not established which CYP isozyme(s) is responsible for DMBA-DNA binding in trout. A CYP1B1 orthologue has recently been identified in teleosts (Godard et al., 1999
); however, the capacity of this enzyme to metabolize DMBA has not yet been reported. Phenobarbital does not induce the CYP2B-related genes in teleosts (Kleinow et al., 1990
; Miranda et al., 1990
; Stegeman et al., 1990
); however, phenobarbital does induce CYP1A1 expression, accompanied by aromatic hydrocarbon receptor transformation (Sadar et al., 1996
). Mammalian CYP1A2 evolved from a relatively recent gene duplication, and a true CYP1A2 orthologue is not present in trout (Goksoyr et al., 1991
). Trout have been shown to possess at least 2 closely related PAH-inducible CYP1A isozymes, both of which have high protein sequence homology to mammalian CYP1A1 (Berndston and Chen, 1994; Heilmann et al., 1988
; Williams and Buhler, 1984
). Neither the gene regulation nor the catalytic properties of these duplicated proteins have been fully characterized and, for the purposes of this communication, these isozymes will be referred to collectively as trout CYP1A1.
Recent studies indicate that liver microsomes from ß-naphthoflavone-treated trout produce several DMBA metabolites, predominantly an unidentified polar metabolite(s), 2-OH-DMBA, and 4-OH-DMBA, with lesser amounts of the DMBA 8,9-, 5,6-, and 3,4-dihydrodiols (Miranda et al., 1997). A purified CYP1A1, in the presence of added epoxide hydrolase produced a similar spectrum, with DMBA-8,9-dihydrodiol being the major metabolite (40% of total) and the 3,4-dihydrodiol being less than 3% of total. The objective of the present study was to test the hypothesis that CYP1A1 plays a significant role in DMBA metabolism in trout liver cells and to establish if modulation of CYP1A1 isozymes may be important in DMBA-DNA adduct formation and carcinogenicity in trout.
Removal of trout liver cell DMBA-DNA adducts was also investigated. Repair of DMBA-DNA adducts has been shown to occur at significant rates, both in mammalian target and non-target tissue, in vitro and in vivo systems (Daniel and Joyce, 1984; Janss et al., 1972
; Tay and Russo, 1981
). The susceptibility of different strains of rats to DMBA-induced tumor formation has been found to correlate less with the levels of adducts actually formed than with the length of time adducts persist (Daniel and Joyce, 1984
). Relative to mammalian cells, fish cells have been shown to have a reduced ability to repair some types of genomic DNA damage (Bailey et al., 1988
, 1995
; Walton et al., 1983
), which may increase their susceptibility to compounds that adduct to DNA. However, DMBA-DNA adduct removal in trout has not been measured.
![]() |
MATERIALS AND METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Isolation and Culture of Trout Liver Cells and Mouse Embryo Fibroblasts
Rainbow trout (Oncorhynchus mykiss) were obtained from the Bowden National Fish Hatchery, Bowden, WV, or from the Food Toxicology and Nutrition Laboratory (FTNL), Oregon State University, Corvallis, OR. Liver cell cultures were prepared and maintained at 15°C as described previously (Miller et al., 1989). Tumor studies were conducted at the FTNL, using duplicate groups of 100 animals by procedures described elsewhere (Bailey et al., 1995
). Medium in cell cultures was replaced daily with the indicated concentrations of specific compounds. Mouse embryo fibroblasts were the generous gift of Dr. A. Dipple; these cells were obtained as frozen cultures and were cultured at 37°C as described (Milner et al., 1985
).
Water-Soluble DMBA Metabolites
To investigate conversion of DMBA to water-soluble metabolites, cells were incubated with medium containing 0.1 µg/ml [3H]-DMBA for indicated times, then extracted with water, methanol, and chloroform (1:1:1), as described (Baird et al., 1978). Radioactivity in the aqueous and organic phases was analyzed by liquid scintillation counting, to determine the percentage of DMBA converted to water soluble metabolites.
To determine the proportion of DMBA metabolites conjugated with sulfates or glucuronides, aliquots of medium were removed at various times and submitted to either no treatment (control), ß-glucuronidase treatment, or sulfatase treatment. Additionally, samples were removed from cells that were pretreated for 24 h with 0.6 mM diethyl maleate and 1 mM buthionine sulfoximine, to deplete GSH levels prior to incubation with fresh medium containing 0.1 µg [3H]-DMBA/ml. One-ml aliquots of medium were mixed with either 2000 units of type B-10 bovine ß-glucuronidase (Sigma, G 0501) or 20 units of type V limpet sulfatase (Sigma, S 8629) and 40 mM D-saccharic acid 1,4 lactone dissolved in 1 ml of sodium acetate (0.5M, pH 5.2). The limpet sulfatase was utilized due to its low levels of contaminating ß-glucuronidase activity (< 2 Sigma units/mg solid), and the saccharic acid, 1,4 lactone, was added to the sulfatase enzymatic treatment to inhibit this small but significant amount of contaminating ß-glucuronidase activity. Control treatments utilized 1 ml of medium mixed with 1 ml of sodium acetate (0.5 M, pH 5.2). Additionally, 1 mg/ml of ascorbic acid was added to each of the control or enzymatic solutions in order to prevent auto-oxidation of [3H]-DMBA phenols to quinones. Each sample was incubated on a rotary shaker at 37° C for 4 h prior to extraction. To extract medium, 1 ml was added to an equal volume of ethyl acetate/acetone (2:1) containing 10 mM dithiothreitol. This mixture was extracted for 5 min and then centrifuged in a Beckman microfuge to generate aqueous and organic phases. The phase-separated materials were then submitted to scintillation-counting, and the percentage of aqueous and organic [3H]-DMBA was calculated and corrected against the zero time values for each treatment. Water-soluble DMBA conjugates not hydrolyzed by sulfatase or ß-glucuronidase were designated "unidentified" DMBA metabolites.
Organic-Soluble DMBA Metabolites
Phase-I metabolism of DMBA was investigated in 2 ways: initial studies partially characterized [3H]-DMBA-free primary metabolites (organic-soluble, unconjugated) produced by trout liver cells and mouse embryo fibroblasts by HPLC separation. Limited metabolites were detected this way (see Results section), and in subsequent studies, metabolites produced in trout liver cells were analyzed by hydrolysis of glucuronide conjugates, followed by HPLC separation and fluorometric analysis of the resultant DMBA hydroxylation derivatives (Christou et al., 1994). Free [3H]-DMBA metabolites and parent DMBA were separated by HPLC, using the conditions described by Dipple, et al. (1983). Trout liver and mouse embryo cells were incubated with medium containing 0.1 µg [3H]-DMBA/ml medium for 6 h. Media was then removed and extracted (Flowers and Miles, 1991
) with an equal volume of acetone/ethyl acetate (1:1.33). Aqueous and organic phases (containing water-soluble, conjugated, phase II metabolites and organic-soluble, phase I metabolites, respectively), were separated; MgSO4 was added to the organic phase, which was then centrifuged (2000 x g, 5 min); the supernatant was dried with nitrogen and resuspended in methanol. This sample was then passed through a 0.45-µm filter, dried under nitrogen, resuspended in methanol, and subjected to HPLC analysis. Chromatographic analysis was performed on a Waters HPLC fitted with a Perkin-Elmer 5 µm C-18 reverse phase column, as described (Dipple et al., 1983
). Reference DMBA metabolites were also chromatographed; UV absorbance of these compounds was monitored at 254 nm. Percent of non-aqueous radioactivity either co-eluting with DMBA or in separate peaks was detected and integrated by a Radiomatic Flo-One Beta Radioactive Detector. Peaks were defined as
1000 3H counts/min above background.
To identify phase-I metabolites which had been conjugated with glucuronide groups, cultured trout-liver cells were incubated with DMBA (10 µM) for 3045 min and medium was collected, digested with ß-glucuronidase, extracted, dried under nitrogen, dissolved in DMSO, then analyzed by HPLC with fluorometric detection of parent compound and metabolites, as described in detail (Christou et al., 1986). These studies used short (3045 min) exposures of cells with DMBA to minimize the production of many multiply hydroxylated metabolites, which did not co-migrate with standards, and likely represented the bulk of the polar metabolites. Additionally, these short incubation times minimize DMBA activation of the arylhydrocarbon receptor and the subsequent increase in CYP metabolic activity.
EROD Assays
CYP1A1 activity was monitored using ethoxyresorufin-O-deethylase (EROD) assays, to determine if there is a correlation between CYP1A1 activity and DMBA metabolism, DNA adduct formation, or tumorigenesis. Isolated trout liver cells were pretreated with 0.05 µg BNF/ml medium to induce CYP1A1 (Miller et al., 1993), and CYP1A1 activity was inhibited by dosing cells with 100 nM ANF; control cells received dimethylsulfoxide solvent. EROD activity was used to measure CYP1A1 activity, using a Shimadzu RF 5000 U Spectrofluorophotometer, as described by Burke and Mayer (1974); assays were conducted at 20°C. Due to the difficulty of obtaining microsomes from a small number of cultured cells, EROD assays were performed with whole-cell homogenates, rather than with microsomal proteins. Studies were conducted to ensure that EROD activity measured in whole cell homogenate was comparable to that measured in microsomes. To this end, isolated trout liver cells were pretreated with 0.05 µg/ml BNF or DMSO solvent for 48 h; cells were collected by centrifugation (1000 x g, 5 min), the cell pellet was resuspended in 0.1 M KCL; 0.1 M Tris-acetate, pH 7.4; 1 mM ethylenediaminetetraacetic acid (EDTA), and 0.1 mM phenylmethylsulfonyl fluoride, and was disrupted with a Branson sonicator, on ice. This sonicate was the cell homogenate. Microsomes were prepared by centrifuging an aliquot of the homogenate (10,000 x g) for 20 min, followed by centrifugation of the supernatant (100,000 x g) for 1 h, and the resulting pellet (microsomes) was suspended in resuspension buffer (50 mM Tris, 1 mM dithiothreitol, 1 mM EDTA, 20% glycerol). With control cells, EROD-specific activity was 14.2 pmol resorufin/min/mg homogenate protein in cell homogenates, and was 51.5 pmol resorufin/min/mg microsomal protein in microsomes. Total EROD activity measured in cell homogenates was comparable to that in microsome preparations. Total EROD activity in control cell homogenates was 102.0 pmol resorufin/min/total mg homogenate protein, and was 108.0 pmol resorufin/min/total mg microsomal protein in microsomes from control cells. BNF pretreatment similarly induced EROD-specific activity and total activity
8.2-fold in both cell homogenates and microsomes. These data demonstrate that EROD measurements using whole cell homogenates are valid, and this preparation was used in all subsequent EROD measurements.
DNA Isolation
For most studies, DNA was isolated from cells as described by Shen et al. (1991). Aliquots of DNA isolated from trout liver cells, which had been incubated with [3H]-DMBA (0.1 µg/ml medium), were counted in a liquid scintillation counter, to quantify DMBA covalently bound to DNA (pmol DMBA/mg DNA). For experiments in which DNA was hydrolyzed for HPLC analysis, DNA was isolated by cesium chloride gradient centrifugation (Lieberman and Dipple, 1972).
Characterization of DMBA-DNA Adducts
Trout liver cells or mouse embryo cells were treated with 0.1 µg/ml [3H]-DMBA for 24 h, and DNA was isolated from cells and enzymatically hydrolyzed to deoxyribonucleosides (Milner et al., 1985). DNA hydrolysates were loaded onto Sep-Pak C18 cartridges and washed with water, then with 40% methanol to remove buffer salts and unmodified nucleosides; typical DMBA-DNA adducts were eluted with methanol. Adducts eluted from Sep Pak columns with methanol were subsequently analyzed by HPLC on a Beckman Ultrasphere octadecyl silane column with a 4565% methanol gradient. In some studies, a more thorough DNA digestion procedure, particularly effective in digesting liver cell DNA (Eberhart et al., 1992
), was performed prior to Sep Pak chromatography.
To determine if DMBA-DNA adducts from trout liver cells might be conjugated to sulfate or glucuronide groups, hydrolyzed DNA was incubated for 3 h at 37°C with sulfatase or ß-glucuronidase, as previously described, prior to Sep Pak column chromatography.
[32P]-Post-labeling of DMBA-DNA Adducts
A DMBA-DNA adduct standard for post-labeling was generated by exposing fetal NIH mouse cell cultures to 50 ng DMBA/ml medium for 24 h and isolating the DNA as described (Vericat et al., 1989). Control or adducted DNA was hydrolyzed to deoxynucleoside 3`-monophosphates by incubating 1525 µg of DNA in with 10 µl of a mixture of 2 µg/µl spleen phosphodiesterase and 0.5 µg/µl of micrococcal endonuclease at 37°C for 4 h.
To enrich the adducts prior to chromatography, the DNA digest (20 µl, 1020 µg DNA) was mixed with 20 µl of 10 mM tetrabutylammonium chloride, 20 µl of 100 mM ammonium formate (pH 3.5), and 140 µl of distilled water in a 1.5-ml microcentrifuge tube. The mixture was extracted twice with one volume of water-saturated 1-butanol by vortexing for 30 s and centrifuging for 3 min in a table-top microcentrifuge. The butanol layer containing DNA adducts was transferred to a clean 1.5-ml microcentrifuge tube. The combined organic phases were back-extracted thrice with 180 µl of 1-butanol-saturated, distilled, deionized water to remove trace normal nucleotides. The butanol phase after the last extraction was placed in a methanol-rinsed 1.5-ml microcentrifuge tube and evaporated in a Savant vacuum concentrator. The adduct concentrate was dissolved in 100 µl of distilled water by vortexing, then evaporated again. The adduct residue from 1020 µg of DNA was dissolved in 15 µl of double-distilled water. To this solution was added a 10 µl aliquot (200 µCi) of a [-32P]-ATP labeling mix containing 182 µl of labeling buffer (100 mM bicine, 100 mM MgCl2, 100 mM dithiothreitol, 10 mM spermidine, pH 9.0), 32 µl of carrier-free [
-32P]-ATP (4.4 mCi), and 6 µl of polynucleotide kinase (30 U/µl). The solution was mixed using a pipetter equipped with guarded pipette tips and incubated in a water bath at 37°C for 45 min.
To remove residual labeled, normal nucleotides, [-32P]-ATP, [32Pi], and other radioactive contaminants and to resolve [32P]-adducts, 9.8 to 19.6 µg of DNA (
196 µCi) were chromatographed on a 4-directional, PEI-cellulose TLC system (Schmeiser et al., 1988
; Vericat et al., 1989
). Radioactive spots on developed chromatograms were located by autoradiography at 80°C for various periods of time, using Hyperfilm-MP in cassettes with Du Pont Lightning-plus intensifying screens. To aid in the matching of chromatograms to their images on the film, chromatograms were marked with alignment dots using trace amounts of [32P] in black ink.
[33P]-Post-labeling HPLC of Adducts
DNA (10 µg) from trout fed 1000 ppm DMBA in the diet (123-day sample) was digested with nuclease P1 and prostatic acid phosphatase, evaporated, and 5`-labeled with [-33P]-ATP and polynucleotide kinase. The labeled sample was digested to 5`-deoxyribonucleotide adducts with snake venom phosphodiesterase and analyzed by HPLC (Ralston et al., 1995
).
DNA Repair
Trout liver cells were treated with medium containing 0.1 µg/ml [3H]-DMBA or 0.1 µg/ml [3H-benzo(a)pyrene for 24 h. Medium was then removed and replaced with DMBA- and benzo(a)pyrene-free medium, respectively. Cells were collected at this time and 12, 24, and 48 h later. For all time points, DNA was isolated and adduct formation was quantitated by liquid scintillation counting, as described.
Tumor Induction
Shasta strain rainbow trout were hatched and reared in 12°C well water at the Food Toxicology and Nutrition Laboratory, Oregon State University, Corvallis, OR as previously described (Schoenhard et al., 1976). Duplicate groups of 100 trout were fed 1 of 3 diets: (1) Oregon Test Diet (OTD) alone; (2) OTD containing 800 ppm DMBA for weeks 29; or (3) OTD containing 500 ppm BNF for weeks 110, and also containing 800 ppm DMBA for weeks 29. Trout in groups 1, 2, and 3 were then returned to OTD until sacrifice at 11 months. After sacrifice, the livers, stomachs, and swim bladders of each fish were examined for gross tumors and fixed in Bouin's solution for histology. Tumors were classified according to established criteria (Hendricks et al., 1984
). Another group of 100 trout were bath-exposed to 5 ppm DMBA for 16 h and livers removed for DNA adduct determination after the exposure.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Phase-I Metabolism of DMBA and Effects of Modulating CYP1A1
The extent to which trout liver cells metabolize DMBA to organic-soluble ("free" Phase-I) metabolites was investigated and compared to mouse embryo fibroblasts. Cells were incubated with [3H]-DMBA, and organic-soluble metabolites released into medium were extracted and analyzed by HPLC. Figure 2 shows the HPLC profiles of major DMBA metabolites formed in trout liver and mouse embryo cells after incubation with DMBA for six h. HPLC profiles of [3H]-DMBA metabolites differed both quantitatively and qualitatively between the teleost and mammalian cells. Mouse embryo fibroblasts produced a profile consisting of the parent DMBA, and
8 metabolites. These 8 metabolites, formed in mouse fibroblasts, accounted for
40% of the radioactive material in the organic-soluble fraction, and DMBA-3,4-dihydrodiol comprised
10% of these metabolites. In contrast, trout liver cells produced only 2 to 3 detectable DMBA metabolites, which comprised
2% of the radioactive material in the organic-soluble fraction; DMBA-3,4-dihydrodiol was not detected.
|
To further characterize Phase-I DMBA metabolites formed in trout liver cells, cells were incubated with DMBA for 45 min, then media samples were incubated with ß-glucuronidase, prior to extraction and separation by HPLC with fluorometric detection. Table 1 shows that the majority of these DMBA metabolites co-elute with 5,6-, 8,9- and 10,11-DMBA dihydrodiols, and with DMBA, 2- or 3- or 4-phenol. Minor metabolites co-migrated with 7-OH-methyl-12-methyl-benz(a)anthracene and 12-OH- methyl-7-methyl-benz(a)anthracene. Only a very small amount of DMBA 3,4-dihydrodiol (
2% of total metabolites) was detected. In addition, polar metabolites were observed that did not migrate with any DMBA standards (not shown). Because the identity of these polar metabolites is unknown, they could not be quantitated by fluorometric detection and they were not included in the analysis in Table 1
. Pre-treating trout liver cells with BNF to induce CYP1A1 increased the rate of production of most of the identified metabolites
2-fold without substantially changing the proportion of any particular metabolite formed (Table 1
).
|
|
Influence of BNF on Hepatic CYP1A1, DMBA-DNA Adducts and Tumor Response in DMBA-Treated Trout
Based on these findings, the effect of dietary BNF on hepatic CYP1A1, DMBA-DNA adduction, and tumor induction was investigated in vivo. Duplicate groups of 100 fingerling rainbow trout were fed 1 of 3 diets (control diet, 800 ppm DMBA, or 800 ppm DMBA plus 500 ppm BNF). A fourth group was fed 500 ppm BNF alone, to quantify CYP1A1 response, but not carried onto full tumor determination, since we have observed this treatment not to be carcinogenic or toxic (Bailey et al., unpublished results). This dose of BNF has been shown repeatedly (Takahashi et al., 1995) to strongly induce trout hepatic-CYP1A1 protein and EROD activity; however, the ability of DMBA alone to induce trout CYP1A1 by a dietary tumorigenic protocol has not been previously tested. Samples were taken at the end of the DMBA-exposure period for adduct and CYP1A1 assessment, and remaining trout were fed control diet until termination 7 months later. As seen in Table 2
, DMBA, by this protocol, was weakly carcinogenic in liver, and strongly carcinogenic in stomach and swim bladder. DMBA treatment alone significantly elevated hepatic CYP1A1 levels and produced levels of DMBA-DNA adducts readily detectable by [33P]-HPLC post-labeling. While BNF alone also strongly elevated hepatic CYP1A1, combined treatment with DMBA somewhat reduced hepatic CYP1A1 from that induced by DMBA or BNF alone, but did not alter DMBA-DNA adducts or tumor response compared with animals receiving DMBA alone. BNF co-treatment strongly inhibited tumor response in stomach and swim bladder; however, DNA adducts and CYP1A1 levels were not assessed in these target tissues.
|
Very different results were seen when [3H]-DMBA-adducted DNA digests obtained from trout liver cells were chromatographed on Sep Pak columns. In contrast to adducts obtained from mouse embryo cells, 6090% of the [3H]-adducts eluted from the Sep Pak column with water, 127% of the [3H]-adducts eluted with 40% methanol, and 10% of the of the [3H]-adducts eluted with 100% methanol. Further HPLC analysis of these methanol fractions was not possible due to the low amount of radioactivity contained within them.
The apparent polarity of the DMBA-DNA adducts obtained from trout liver cells could have resulted from incomplete digestion of trout liver-cell DNA, yielding adducts with oligonucleotides rather than individual deoxyribonucleosides. To determine if this were the case, a more vigorous digestion of trout liver cell DNA was undertaken, as described (Eberhart et al., 1992). This procedure is effective in completely digesting rat liver DNA; however, it did not alter the Sep Pak elution profile of trout liver cell DNA (not shown). These results suggest that DMBA-DNA adducts obtained from trout liver cells are more polar in nature than those formed in mammalian cells.
Further confirmation that DMBA-DNA adducts formed in trout liver cells are more polar than those formed in mouse embryo cells was obtained by extracting DNA digests in a chloroform:methanol:water mixture. Ninety five percent of [3H]-DMBA-DNA adducts from trout liver cells partitioned into the aqueous phase, whereas only 10% of [3H]-DMBA-DNA adducts from mouse embryo cells partitioned into the aqueous phase. DNA digests obtained from trout liver cells were also chromatographed on PEI cellulose TLC plates with 0.8 M (NH4)2SO4. DMBA-nucleoside adducts formed in mouse embryo cells do not migrate from the origin under these conditions. However, when DNA digests from trout liver cells were chromatographed, more than 95% of the radioactivity migrated from the origin (not shown), further demonstrating that DMBA-DNA adducts formed in trout liver cells are more polar than those formed in mouse embryo cells. To test the possibility that DMBA-DNA adducts formed in trout liver cells contain sulfate or glucuronide groups, making them hydrophilic, DNA from trout liver cells, incubated with [3H]-DMBA, was isolated and incubated with sulfatase or ß-glucuronidase prior to Sep Pak column chromatography. These enzymatic treatments did not change the elution profile of DMBA-DNA adducts (not shown).
DMBA-DNA adduction was further examined both by conventional [32P]-post-labeling and by [33P]-post-labeling-HPLC. Three major adducts were identified by [32P]-post-labeling and TLC when mouse embryo cells were incubated with DMBA (Fig. 4A). These adducts were barely detectable in trout liver cells incubated with DMBA (Fig. 4C
). In contrast, when trout liver cells were exposed to DMBA-3,4-dihydrodiol, 3 prominent adducts were observed (Fig. 4B
), which were indistinguishable from those formed in mouse embryo cells incubated with the parent DMBA (Fig. 4A
). In vivo, exposure of trout to DMBA by the highly efficient and carcinogenic gill uptake procedure (Fig. 4D
) did result in detectable production in the liver of one adduct that had the same Rf as one from mouse embryo cells. Based on other studies (Vericat et al., 1989
), this adduct is tentatively assigned as the deoxyadenosine adduct from DMBA-(4S,3R)-dihydrodiol(2S,1R-epoxide). Three other unknown adducts were also observed. However, dietary exposure for 18 weeks to carcinogenic doses of DMBA was determined to elicit only trace levels of these adducts (data not shown). The DMBA-DNA adducts formed in livers by this exposure protocol were further characterized by the [33P]-HPLC post-labeling procedure and found to consist almost entirely of polar adducts (Fig. 5
) not detected by the conventional [32P]-post-labeling, thin-layer-chromatography procedures used for other samples. This confirms that most DMBA-DNA adducts formed in trout liver are more polar than those formed in mouse embryo cells.
|
|
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Examination of the rate of repair of DMBA-DNA adducts in trout liver cells indicated repair of these adducts from the total genome is undetectable in 48 hours (Fig. 6), whereas repair of DMBA-damaged DNA in mammalian cells appears to be more efficient. Tay and Russo (1981) demonstrated that rat mammary epithelial cells removed 12 to 36% of DMBA-DNA adducts in 48 h. It is unclear whether the lack of repair of DMBA-DNA adducts by trout liver cells is due to low global excision repair capacity in the teleost system or to formation of adducts that are intrinsically less repairable than those formed in mammalian cells. However, these findings are consistent with other studies indicating teleosts are less efficient in repairing bulky adducts than their mammalian counterparts (Bailey et al., 1988
).
Enzymatic Pathways to DMBA-DNA Adduction in Trout
The enzymes involved in DMBA metabolism and bioactivation in trout are not fully characterized, and may vary depending on diet and other treatment history. Recent studies indicate that purified trout CYP1A1 metabolizes DMBA to the 3,4-diol proximate carcinogen, but only as a minor metabolite (Miranda et al., 1997). BNF induction of CYP1A1 in isolated trout liver cells was shown here to stimulate metabolism of DMBA to water-soluble compounds (Fig. 1
). However, the 3,4-dihydrodiol metabolite was only a very minor metabolite formed in trout liver cells, and CYP1A1 induction did not increase the relative amount of this metabolite (Table 1
). In trout liver cells, DMBA metabolites formed by CYP1A1, such as the 3,4-diol, may be efficiently metabolized further to detoxification products or polar DNA-binding metabolites. When cells were incubated with 4 µg/ml DMBA-3,4-diol (Fig. 4
), the high concentration may have resulted in more metabolic activation to the diol epoxide. Neither CYP1A1 induction with BNF nor inhibition by ANF substantially affected DMBA-DNA adduct formation (Fig. 3
). Therefore, the present results provide no evidence that CYP1A1 isoform(s) selectively catalyze DMBA metabolism to DNA-reactive intermediates in trout liver cells. These results are in apparent contrast to studies with mammalian cells, in which ANF reduced DMBA-DNA adducts (Dipple et al., 1984
). However, ANF inhibits both CYP1A1 and 1A2 in mammalian cells, and trout appear not to possess CYP1A2 orthologues (Goksoyr et al., 1991
). Miranda et al. (1997) demonstrated that microsomes from BNF-treated trout catalyzed DMBA metabolism to DNA-binding species in vitro; however the adducts formed were not characterized, and their quantitative importance could not be evaluated. Additional studies are needed to establish which trout CYP isozyme(s) participate in formation of stable DMBA-DNA adducts.
BNF Influence on DMBA DNA Adduction and Tumorigenicity
These are among the first studies to report DMBA carcinogenicity in any fish by dietary exposure. The results show that DMBA is highly carcinogenic by this route, with stomach and swim bladder as the major target organs, and a lesser response in liver (Table 2). The results also show that co-feeding of 500 ppm BNF did not significantly elevate hepatic tumor response. There are 2 opposing mechanisms through which BNF co-exposure might have potentially altered DMBA adduction and tumorigenicity: (1) by providing CYP1A1 induction additional to that seen with DMBA alone, and (2) through catalytic inhibition of CYP1A1 or other trout CYP enzymes that may bioactivate DMBA. BNF is a potent inhibitor of trout CYP1A1 and other CYP enzymes in vitro, and 500 ppm dietary BNF strongly inhibits AFB1 adduction and tumorigenicity in trout, primarily through inhibition of CYP bioactivation (Takahashi et al., 1996
). The significant inhibition of tumor response in stomach and swim bladder may well reflect such a mechanism in these target organs, but this was not investigated directly. In liver, BNF co-treatment was found to somewhat reduce hepatic CYP1A1 activity compared with animals receiving DMBA alone, but did not alter either DMBA-DNA adduction or tumorigenicity under conditions where it does both against AFB1. In contrast, we have shown that co-treatment with Aroclor 1254, also a potent CYP1A1 inducer, strongly elevated DMBA hepatic tumor response in this model (Bailey et al., unpublished results). These results suggest that members of the trout CYP1A1 isozyme subfamily may be differentially regulated, that BNF and Aroclor 1254 induce different ratios of these isozymes, and that the isozymes have distinctly different catalytic properties for DMBA metabolism and bioactivation. In support of this, recent studies (Curtis et al., 1996
) indicate differential regulation of 2 trout CYP1A1 genes by 2,4,5,2`,4`,5`-hexachlorobiphenyl. However, 2,4,5,2`,4`,5`-hexachlorobiphenyl is a non-coplanar polychlorinated biphenyl, which is not normally considered an inducer of CYP1A1 genes. Because some commercially available preparations of this polychlorinated biphenyl may contain a furan with potent CYP1A1-inducing ability, these conclusions may be compromised. Metabolism and induction studies with the cloned trout CYP1A1 isoforms will be necessary, in order to test this hypothesis. Finally, DMBA tumorigenesis in some systems may be driven primarily through the formation of unstable DNA adducts resulting from one-electron oxidation processes (Cavalieri and Rogan, 1992
). We do not believe this mechanism of activation plays a major role in DMBA tumor induction in trout. While high levels of dietary antioxidant butylated hydroxyanisole did not reduce DMBA tumorigenicity in trout (Bailey et al., unpublished results), the profile of DMBA-mediated Ki-ras oncogenic mutations is not fully compatible with apurinic site mutagenesis (Reddy et al., 1995
), and there appear to be ample stable, polar DMBA adducts in this model to account for its mutagenicity and carcinogenicity.
In summary, the cellular interactions of DMBA in trout liver cells have been initially characterized and found to differ substantially from those occurring in mouse embryo cells, particularly in the processes of bioactivation and DNA repair. Studies to further characterize the interactions of DMBA with trout liver cells may provide novel insight into mechanisms of chemical carcinogenesis.
![]() |
ACKNOWLEDGMENTS |
---|
![]() |
NOTES |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bailey, G. S., Williams, D. E., Wilcox, J. S., Loveland, P. M., Coulombe, R. A., and Hendricks, J. D. (1988). Aflatoxin B1 carcinogenesis and its relation to DNA adduct formation and adduct persistence in sensitive and resistant salmonid fish. Carcinogenesis 9, 19191926.[Abstract]
Baird, W. M., and Dipple, A. (1977). Photosensitivity of DNA-bound 7,12-dimethylbenz(a)anthracene. Intl. J. Cancer 20, 427431.[ISI][Medline]
Baird, W. M., Chemerys, R., Chern, C. J., and Diamond, L. (1978). Formation of glucuronic acid conjugates of 7,12- phenols in 7,12- dimethylbenz(a)anthracene-treated hamster embryo cell cultures. Cancer Res. 38, 34323437.[Abstract]
Berndtson, A. K, and Chen, T. T. (1994). Two unique CYP1 genes are expressed in response to 3-methylcholanthrene treatment in rainbow trout. Arch. Biochem. Biophys. 310, 187195.[ISI][Medline]
Bigger, C. A., Sawicki, J. T., Blake, D. M., Raymond, L. G., and Dipple, A. (1983). Products of binding of 7,12-dimethylbenz(a)anthracene to DNA in mouse skin, Cancer Res. 43, 56475651.[Abstract]
Burke, M. D., and Mayer, R. T. (l974). Ethoxyresorufin: Direct fluorometric assay of a microsomal O-deethylation, which is preferentially inducible by 3-methylcholanthrene. Drug Metab. Dispos. 2, 583588.[ISI][Medline]
Carmichael, P. L., Platt, K. L., She, M. N., Lecoq, S., Oesch, F., Phillips, D. H., and Grover, P. L. (1993). Evidence for the involvement of a bis-diol-epoxide in the metabolic activation of dibenz(a,h)anthracene to DNA-binding species in mouse skin. Cancer Res. 53, 944948.[Abstract]
Cavalieri, E. L., and Rogan, E. G. (1992) The approach to understanding aromatic hydrocarbon carcinogenesis. The central role of radical cations in metabolic activation. Pharmacol.-Ther. 55, 183199.[ISI][Medline]
Christou, M., Marcus, C., and Jefcoate, C. R. (1986). Selective interactions of cytochromes P-450 with the hydroxymethyl derivatives of 7,12-dimethylbenz[a]anthracene. Carcinogenesis 7, 871877.[Abstract]
Christou, M., Savas, U., Spink, D. C., Gierthy, J. F., and Jefcoate, C. R. (1994). Co-expression of human CYP1A1 and a human analog of cytochrome P450-EF in response to 2,3,7,8-tetrachloro-dibenzo-p-dioxin in the human mammary carcinoma-derived MCF-7 cells. Carcinogenesis 15, 725732.[Abstract]
Curtis, L. R., Foster, E. P., da Costa, E. G., Vrolick, N., and Chen, T. T. (1996). 2,4,5,2`,4`,5`-hexachlorobiphenyl (HCB) induces rainbow trout hepatic aryl hydrocarbon hydroxylase (AHH) and increases expression of one of two cytochrome P450 (CYP) 1A mRNAs. Toxicologist 30, 73.
Daniel, F. B., and Joyce, N. J. (1984) 7,12-Dimethylbenz(a)anthracene-DNA adducts in Sprague-Dawley and Long-Evans female rats: The relationship of DNA adducts to mammary cancer. Carcinogenesis 5, 10211026.[Abstract]
Devanesan, P. D., RamaKrishna, N. V., Padmavathi, N. S., Higginbotham, S., Rogan, E. G., Cavalieri, E. L., Marsch, G. A., Jankowiak, R., and Small, G. J. (1993). Identification and quantitation of 7,12-dimethylbenz[a]anthracene-DNA adducts formed in mouse skin. Chem. Res. Toxicol. 6, 364371.[ISI][Medline]
Diamond, L., McFall, R., Miller, J., and Gelboin, H. V. (1972. The effects of two isomeric benzoflavones on aryl hydrocarbon hydroxylase and the toxicity and carcinogenicity of polycyclic aromatic hydrocarbons. Cancer Res. 32, 731736.[ISI][Medline]
DiGiovanni, J., Nebzydoski, A. P., and Decina, P. C., (1983). Formation of 7-hydroxy-12-methylbenz[a]anthracene-DNA adducts from 7,12-dimethylbenz[a]anthracene in mouse epidermis. Cancer Res. 43, 42214226.[Abstract]
Dipple, A., and Hayes, M. E. (1979) Differential excision of carcinogenic hydrocarbon-DNA adducts in mouse embryo cell cultures. Biochem. Biophys. Res. Commun. 91, 12251231.[ISI][Medline]
Dipple, A., Pigott, M. A., Bigger, A. H. and Blake, D. M. (1984). 7,12-Dimethylbenz(a)anthracene-DNA binding in mouse skin: response of different mouse strains and effects of various modifiers of carcinogenesis. Carcinogenesis 5, 10871090.[Abstract]
Dipple, A., Pigott, M., Moschel, R. and Constantino, N. (1983). Evidence that binding of 7,12-dimethylbenz-(a)-anthracene to DNA in mouse embryo cell cultures results in extensive substitution of both adenine and guanine residues. Cancer Res. 43, 41324135.[Abstract]
Eberhart, J., Coffing, S. L., Anderson, J. N., Narcus, C., Kaloqeris, T. J., Baird, W. M., Park, S. S., and Gelboin, H. V. (1992). The time-dependent increase in the binding of benzo(a)pyrene to DNA through (+)-anti-benzo(a)pyrene-7,8-diol-9,10-epoxide in primary rat hepatocyte cultures, results from induction of cytochrome P4501A1 by benzo(a)pyrene treatment. Carcinogenesis 13, 297301.[Abstract]
Flowers, N. L., and Miles, P. R. (1991). Alterations of pulmonary benzo(a)pyrene metabolism by reactive oxygen metabolites. Toxicology 68, 259274.[ISI][Medline]
Fong, A. T., Dashwood, R. H., Cheng, R., Mathews, C., Ford, B., Hendricks, J. D., and Bailey, G. S. (1993). Carcinogenicity, metabolism, and Ki-ras proto-oncogene activation by 7,12-dimethylbenz(a)anthracene in rainbow trout embryos. Carcinogenesis 14, 629635.[Abstract]
Fuchs, J., Mlcoch, J., Oesch, F., and Platt, K. L. (1993). Characterization of highly polar DNA adducts derived from dibenz(a,h,)anthracene (DBA), 3,4-dihydroxy-3,4-dihydro-DBA, and 3,4,10,11-tetrahydroxy-3,4,10,11-tetrahydro-DBA. Toxicol. Ind. Health 9, 503509.[ISI][Medline]
Godard, C. A., Said, M. R.,Moore, M. J., Dickerson, R.L., and Stegeman, J. J. (1999). Molecular identification of cytochrome P450 1B in liver of scup (Stenotomus chrysops) and striped dolphin (Stenella coeruleoalba) and cytochrome P450 1A1 in striped dolphin testis. The Toxicologist 48, 99.
Goksoyr, A., Andersson, T., Buhler, D. R., Stegeman, J. J., Williams, D. E., and Forlin, L. (1991). Immunochemical cross reactivity of beta-naphthoflavone-inducible cytochrome P450 (P4501A) in liver microsomes from different fish species and rat. Fish Physiol. Biochem. 9, 113.[ISI]
Heilmann, L. J., Sheen, Y. Y., Bigelow, S. W., and Nebert, D. W. (1988), Trout P4501A1 cDNA and deduced protein sequence, expression in liver, and evolutionary significance. DNA 7B, 379387.[ISI][Medline]
Hendricks, J. D., Meyers, T. R., and Shelton, D. W., (1984). Histological progression of hepatic neoplasia in rainbow trout (Salmo gairdneri). Natl. Cancer Inst. Monogr., 65, 321336.[Medline]
Janss, D. H., Moon, R. C., and Irving, C. C. (1972). The binding of 7,12-dimethylbenz(a)anthracene to mammary parenchymal DNA and protein in vivo. Cancer Res. 32, 254258.[ISI][Medline]
Joyce, N. J., and Daniel, F. B. (1982). 7,12-Dimethylbenz[a]anthracene-deoxyribonucleoside adduct formation in vivo: Evidence for the formation and binding of a mono-hydroxymethyl-DMBA metabolite to rat liver DNA. Carcinogenesis 3, 297301.[ISI][Medline]
Kirby, G. M., Stalker, M., Metcalfe, C., Kocal, T., Ferguson, H., and Hayes M. A. (1990). Expression of immunoreactive glutathione S-transferases in hepatic neoplasms induced by aflotoxin B1 or 1,2-dimethylbenz(a)anthracene in rainbow trout (Oncorhynchus mykiss) Carcinogenesis 11, 22552257.[Abstract]
Kleinow, K. M., Haasch, M. L., Williams, D. E. and Lech, J. J. (1990) A comparison of hepatic P450 induction in rat and trout (Oncorhynchus mykiss): Delineation of the site of resistance of fish to phenobarbital-type inducers. Comp. Biochem. Physiol. 96, 259270.[ISI]
Lieberman, M. W., and Dipple, A. (1972). Removal of bound carcinogen during DNA repair in non-dividing human lymphocytes. Cancer Res. 32, 18551860.[ISI][Medline]
McCord, A., Burnett, A. K., Wolf, C. R., Morrison, V., and Craft, J. A. (1988). Role of specific cytochrome P450 isoenzymes in the regio-selective metabolism of 7,12-dimethylbenz(a)anthracene in microsomes from rats treated with phenobarbital or Sudan III. Carcinogenesis 9, 14851491.[Abstract]
Miller, M. R., Blair, J. B., and Hinton, D. E. (1989). DNA-repair synthesis in isolated rainbow trout liver cells. Carcinogenesis 10, 995-1001.[Abstract]
Miller, M. R., Saito, N., Blair, J. B., and Hinton, D. E. (1993). Acetaminophen toxicity in cultured trout liver cells: II. Maintenance of cytochrome P450 1A1. Exp. Mol. Pathol. 58, 127138.[ISI][Medline]
Milner, J. A., Pigott, M. A., and Dipple, A. (1985). Selective effects of selenium selenite on 7,12-dimethylbenz(a)anthracene-DNA binding in fetal mouse cell cultures. Cancer Res. 45, 63476354.[Abstract]
Miranda, L. M., Henderson, M. C., Williams, D. E., and Buhler, D. R. (1997). In vitro metabolism of 7,12-dimethylbenz(a)anthracene by rainbow trout liver microsomes and trout P450 isoforms. Tox. Appl. Pharm. 142, 123132.[ISI][Medline]
Miranda,C. L., Wang,J.-L., Henderson,M. C., and Buhler, D. R. (1990). Immunological characterization of constitutive isozymes of cytochrome P450 from rainbow trout: Evidence for homology with phenobarbital-induced rat P450s. Biochim. Biophys. Acta. 1037, 155160.[ISI][Medline]
Morrison, V. M., Burnett, A. K., Forrester, L. M., Wolf, C. R., and Craft, J. A. (1991). The contribution of specific cytochromes P450 in the metabolism of 7,12-dimethylbenz(a)anthracene in rat and human liver microsomal membranes. Chem. Biol. Interact. 79, 179196.[ISI][Medline]
Moschel, R. C., Pigott, M. A., Costantino, N., and Dipple, A. (1983). Chromatographic and fluorescence spectroscopic studies of individual 7,12-dimethylbenz(a)anthracene-deoxyribonucleoside adducts. Carcinogenesis 4, l201l204.
Nimmo, I. A., and Clapp, J. B. (1979). A comparison of the glutathione S-transferases of trout and rat liver. Comp. Biochem. Physiol. 63B, 423427.[ISI]
Pottenger, L. H., and Jefcoate, C. R. (1990). Characterization of a novel cytochrome P450 from the transformable cell line, C3H/10T1/2. Carcinogenesis 11, 321327.[Abstract]
Ralston, S. L., Seidel, A., Luch, A., Platt, K. L., and Baird, W. M. (1995). Stereoselective activation of dibenzo[a,l]pyrene to (-)-anti (11R,12S,13S,14R)- and (+)-syn (11S, 12R, 13S, 14R)-11,12-diol-13, 14-epoxides, which bind extensively to deoxyadenosine residues of DNA in the human mammary carcinoma cell line MCF-7. Carcinogenesis 16, 28992907.[Abstract]
Reddy, A., Cheng, R., Hendricks, J., and Bailey, G. (1995). Comparative tumorigenicity and Ki-ras activation by dibenzo(a,l)pyrene and 7,12-dimethylbenzanthracene in rainbow trout. Proc. Amer. Assoc. Cancer Res. 36, 966.
Sadar, M. D., Ash, R., Sundqvist, J., Olsson, P.-E., and Andersson, T. B. (1996). Phenobarbital induction of CYP1A1 gene expression in a primary culture of rainbow trout hepatocytes. J. Biol. Chem. 271, 1763517643.
Savas, U., Christou, M., and Jefcoate, C. R. (1993). Mouse endometrium stromal cells express a polycyclic aromatic hydrocarbon-inducible cytochrome P450 that closely resembles the novel P450 in mouse embryo fibroblasts (P450EF). Carcinogenesis 14, 20132018.[Abstract]
Sawicki, J. T., Moschel, R. T., and Dipple, A. (1983). Involvement of both syn- and anti-dihydrodiol epoxides in the binding of 7,12-dimethylbenz(a)anthracene to DNA in mouse embryo cell cultures. Cancer Res. 43, 32123218.[Abstract]
Schmeiser, H., Dipple, A., Schurdak, M. E., Randerath, E., and Randerath,K., (1988). Comparison of 32P-post-labeling and high pressure liquid chromatographic analyses for 7,12-dimethybenz[a]anthracene-DNA adducts. Carcinogenesis 9, 633638.[Abstract]
Schnitz, A. R., and O'Connor, J. M. (1992). In vivo DNA/RNA adduction of 7,12-dimethylbenz(a)anthracene (DMBA) and benzo(a)pyrene (BaP) in the liver of rainbow trout (Oncorhynchus mykiss). J. Env. Path. Tox. Oncol. 11, 229233.
Schnitz, A. R., Squibb, K. S., and O'Connor, J. M. (1987). Fate of 7,12-dimethylbenz(a)anthracene in rainbow trout, Salmo gairdneri. Bull. Environ. Contam. Toxicol. 39, 2936.[ISI][Medline]
Schoenhard, G. L., Lee, D. J., Howell, S. E., Pawlowski, N. E., Libbey, L. M., and Sinnhuber, R. O. (1976). Aflatoxin B1 metabolism to aflatoxicol and derivatives lethal to Bacillus subtilis GSY 1057 by rainbow trout (Salmo gairdneri) liver. Cancer Res. 36, 20402045.[Abstract]
Shen, W., Kamendulius, L. M., Ray, S. D., and Corcoran, G. B. (1991). Acetaminophen-induced cytotoxicity in cultured mouse hepatocytes: Correlation of nuclear Ca2+ accumulation and early DNA fragmentation with cell death. Tox. Appl. Pharm. 111, 242254.[ISI][Medline]
Smolarek, T. A., Morgan, S. L., Moynihan, C. G., Lee, H., Harvey, R. G. and Baird, W. M. (1987). Metabolism and DNA adduct formation of benzo(a)pyrene and 7,12-dimethylbenz(a)anthracene in fish cell lines in culture. Carcinogenesis 8, 15011509.[Abstract]
Stegeman, J. J., Woodin, B. R., and Smolowitz, R. M. (1990). Structure, function, and regulation of cytochrome P450 forms in fish. Biochem. Soc. Trans. 18, 1921.[ISI][Medline]
Takahashi, N., Miranda, C. L., Henderson, M. C., Buhler, D. R., Williams, D. E., and Bailey, G. S. (1995). Inhibition of in vitro aflatoxin B1-DNA binding in rainbow trout by CYP1A1 inhibitors: -Naphthoflavone, ß-naphthoflavone, and trout CYP1A1 peptide antibody. Comp. Biochem. Physiol. C Pharmacol. Toxicol. Endocrinol. 110, 273280.
Takahashi, N., Williams, D. E., and Bailey, G. S. (1996). The model Ah-receptor agonist ß-naphthoflavone inhibits aflatoxin B1-DNA binding in vivo in rainbow trout at dietary levels that do not induce CYP1A1 enzymes. Carcinogenesis 17, 7987.[Abstract]
Tay, L. K., and Russo, J. (1981). Formation and removal of 7,12-dimethylbenzanthracene-nucleic acid adducts in rat mammary epithelial cells with different susceptibility to carcinogenesis. Carcinogenesis 2, 13271333.[ISI][Medline]
Vericat, J. A., Cheng, S. C., and Dipple, A. (1989). Absolute stereochemistry of the major 7,12-dimethylbenz[a]anthracene-DNA adducts formed in mouse cells. Carcinogenesis 10, 567570.[Abstract]
Walton, D. G., Acton, A. B., and Stich, H. F. (1983). DNA repair synthesis in cultured mammalian and fish cells following exposure to chemical mutagens. Mutat. Res. 124, 153161.[ISI][Medline]
Williams, D., and Buhler, D. (1984). Benzo(a)pyrene hydroxylase catalyzed by purified isozymes of cytochrome P450 from beta-naphthoflavone fed rainbow trout. Biochem. Pharmacol. 33, 37433753.[ISI][Medline]