Antiestrogenicity of β-naphthoflavone and PAHs in cultured rainbow trout hepatocytes: evidence for a role of the arylhydrocarbon receptor

Aquatic Toxicology 51 (2000) 79 – 92 www.elsevier.com/locate/aquatox

Antiestrogenicity of b-naphthoflavone and PAHs in cultured rainbow trout hepatocytes: evidence for a role of the arylhydrocarbon receptor Jose´ Marı´a Navas, Helmut Segner * Umweltforschungszentrum Leipzig-Halle, Sektion Chemische O8 kotoxikologie, Permoserstrasse 15, D-04318 Leipzig, Germany Received 15 September 1999; received in revised form 15 February 2000; accepted 17 February 2000

Abstract The aims of the present study were to assess, (1) if polyaromatic hydrocarbons (PAHs) are able to inhibit estradiol-regulated vitellogenin synthesis in fish; and (2) if this antiestrogenic activity is mediated through the binding of PAHs to the arylhydrocarbon receptor (AhR). Cultured liver cells of rainbow trout, Oncorhynchus mykiss, were co-exposed to PAHs and 17b-estradiol (E2), and the resulting effects on induction of AhR-regulated 7-ethoxyresorufin-O-deethylase (EROD) activity and on E2-regulated vitellogenesis were investigated. The following test compounds were compared: the PAH 3-methylcholanthrene (3MC), which is a strong EROD inducer, the PAH anthracene (ANT), which is not an inducer of EROD activity, and the model EROD inducer, b-naphthoflavone (bNF). 3MC and bNF led to significant decreases of E2-triggered hepatocellular VTG synthesis, whereas ANT exerted no antiestrogenic activity. The rank order of the antiestrogenic activity of the test substances agreed with their EROD-inducing potency suggesting that their antiestrogenicity might be mediated through the AhR. Further evidence for this assumption comes from the observation that inhibitors such as a-naphthoflavone which interferes with ligand–AhR binding, and 8-methoxypsoralen (8MP), which prevents binding of the occupied AhR to responsive DNA elements, clearly reduced the antiestrogenic effects of the xenobiotics. Furthermore, from the comparison of estradiol concentrations in media of liver cells exposed to the CYP 1A-inducing agents and in media of control cells it is unlikely that the observed antiestrogenic effects were caused by an enhanced E2 catabolism. In conclusion, the results from this study indicate that, (1) AhR-binding PAHs possess an antiestrogenic activity; and (2) that the antiestrogenic activity is mediated through the AhR. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Antiestrogenicity; Vitellogenin; Hepatocytes; Polynuclear aromatic hydrocarbons (PAHs); Arylhydrocarbon receptor; Cytochrome P450 1A

Abbre6iations: AhR, arylhydrocarbon receptor; Ant, anthracene; CYP 1A, cytochrome P450 1A; ER, estrogen receptor; ERE, estrogen response elements; EROD, 7-ethoxyresorufin-O-deethylase; E2, 17b-estradiol; PAH, polyaromatic hydrocarbons; PCBs, polychlorinated biphenyls; TCDD, 2,3,7,8-tetrachlorodibenzo-p-dioxine; VTG, vitellogenin; XRE, xenobiotic response elements; aNF, a-naphthoflavone; bNF, b-naphthoflavone; 3MC, 3-methylcholanthrene. * Corresponding author. Tel.: + 49-341-2352329; fax: +49-341-2352401. E-mail address: [email protected] (H. Segner). 0166-445X/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 4 5 X ( 0 0 ) 0 0 1 0 0 - 4

80

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

1. Introduction A variety of natural and man-made chemicals can interact with the endocrine systems of humans and wildlife (Colborn and Clement, 1992; Kavlock et al., 1996), inducing reversible or irreversible alterations of the hormone metabolism and hormone-regulated processes. These compounds can, for instance, interfere with hormone action, alter the functioning of signal transduction pathways or disturb the complex control of sexual differentiation. Estrogens are a major group of steroid hormones whose primary role is the control of reproduction in vertebrates. The female reproductive hormone, 17b-estradiol (E2) influences cellular functions by binding to the estrogen receptor (ER) protein, which is a ligand-activated transcription factor belonging to the nuclear receptor superfamily (reviewed by Tsai and O’Malley, 1994). After binding of its cognate hormonal ligand, ER undergoes an activation process, which involves conformational changes and formation of a homodimer. The receptor – ligand complex interacts with cis-acting DNA elements, so-called estrogen response elements (EREs), in the vicinity of estrogen responsive genes and activates transcription. Also a number of man-made chemicals can bind to the ER and exert an estrogen-like activity through direct interaction with the ER. The stimulation of the ER pathway by non-physiological ligands, the so-called xenoestrogens, potentially leads to a disturbance of estrogen-regulated cellular and physiological processes. As a consequence, normal sexual development and reproduction of fish, wildlife and humans could be disrupted (Colborn and Clement, 1992; Sharpe and Skakkebaek, 1993; Guillette et al., 1994; Facemire et al., 1995; Sumpter et al., 1996). Whereas much research work has been done to date on xenobiotics with estrogen-like activity, less attention has been given to substances showing anti-estrogenic activity, i.e. compounds that antagonize or inhibit estrogen-dependent processes in the target tissues. Particularly for xenobiotics such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and polychlorinated biphenyls (PCBs),

which bind to the arylhydrocarbon receptor (AhR) and thereby induce the biotransformation enzyme cytochrome P450 1A (CYP 1A), antiestrogenic activities have been observed (Safe, 1994; Navas and Segner, 1998). In teleost fish, both in vivo- and in vitro-studies have described that exposure to TCDD or PCBs could be associated with reduced vitellogenin (VTG) synthesis or impaired gonad development (Chen et al., 1985; Thomas, 1989; Wannemacher et al., 1992; Anderson et al., 1996a,b; Smeets, 1999). Polyaromatic hydrocarbons (PAHs) are a class of toxic organic chemicals comprising hundreds of individual compounds, with many of them being AhR ligands. They are released into the environment primarily from combustion processes, oil spills or industrial processes using petroleum components. The principal toxic risk currently associated with PAHs is cancer, which arises as a result of the endogenous biotransformation of PAHs to reactive metabolites. Because PAHs are ligands to the AhR, they may constitute another class of antiestrogenic chemicals, as described previously, for dioxins and PCBs. In fact, a number of in vivo- and in vitro-investigations with rodents have provided evidence that PAHs act as antiestrogens in mammals (Safe, 1994; Navas and Segner, 1998). For teleosts, results from the few studies having addressed this problem point to an antiestrogenic activity of PAHs in this vertebrate group as well (Anderson et al., 1996a,b; Nicolas, 1999). However, as pointed out by Nicolas (1999), there is a lack of agreement between the results of the available reports on PAH antiestrogenicity in fish indicating a need for more, mechanistically oriented research on this topic. The present study has two main objectives, (1) to assess the antiestrogenic activity of two PAHs, 3-methylcholanthrene (3MC) and anthracene (ANT), and the model compound, b-naphthoflavone (bNF); and (2) to investigate if the antiestrogenic activity of PAHs in trout is mediated through the AhR. As experimental model, we use cultured liver cells from rainbow trout, Oncorhynchus mykiss, co-exposed to estradiol and PAHs. The three test compounds, bNF, 3MC and ANT have been selected because of their different AhR binding affinities. The antiestrogenic activity

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

of the chemicals is derived from their ability to inhibit the E2-induced hepatocellular synthesis of the egg yolk precursor, VTG.

2. Material and methods

2.1. Animals Sexually immature male and female rainbow trout (250–350 g weight) from a local trout farm were maintained in 200-l steel tanks in the facilities of the Umweltforschungszentrum, Leipzig (Centre for Environmental Research, Leipzig), Germany. Gonadosomatic indices (gonad weight/ body weight × 100) ranged between 0.5 and 1%. Fish were held under artificial 12 h light/12 h dark photoperiod in constant flow aerated water. Water temperature was maintained at 14 – 16°C.

2.2. Hepatocytes isolation and culture Hepatocytes were isolated following a two-step perfusion technique as described by Scholz et al. (1997). As culture medium, modified M199 medium (Sigma, USA) supplemented with 2mM glutamine, 10 U/ml penicillin and 10 mg/ml streptomycin was used. Media was sterilized by a 0.22 mm filter. Cells were plated in 24-well Falcon primary culture plates (Becton Dickinson, Oxnard, CA) precoated with Matrigel (0.1 mg protein/ml). Every well received 400 ml of the final suspension of cells (density, 2× 105 cells per cm2). The cells used for the extractions of microsomes and analysis of CYP 1A1 were plated on 60 mm diameter Falcon polystyrene tissue culture dishes and every plate received 5 ml of the final suspension of the cells. During the first 24 h of treatment the medium was supplemented with 5% fetal bovine serum (Sigma, USA) to enhance the attachment of the cells. During exposure to the test compounds serum free medium was used. The plates were maintained at 15°C and 80% humidity. 24 h after plating, half of the medium was substituted with fresh medium containing the desired concentrations of xenobiotics and 17b-estradiol (E2). Control wells received the solvent only (carrier control). Treatments were applied in du-

81

plicate. The medium was renewed in each 24 h. Ninety-six hours after plating the medium was removed, aliquoted, and stored at − 80°C until analysis of lactate dehydrogenase (LDH) activity and VTG in medium. The cells were briefly washed with phosphate-buffered saline (PBS), pH 7.5 and the plates were frozen by using liquid nitrogen. They were maintained at − 80°C until analysis of 7-ethoxyresorufin-O-deethylase (EROD) and protein.

2.3. Xenobiotic treatments bNF, 3MC, ANT, a-naphthoflavone (aNF) and 8-methoxypsoralen (8MP) were purchased from Sigma (USA). Chemicals were dissolved in dimethyl sulfoxide (DMSO, Merck, Germany) and added to the culture media of hepatocytes. E2 (Serva, Germany) was used as positive control (it will be designated E2-control in the text). E2 was diluted in ethanol and added to the media to achieve the final desired concentrations. Final concentration of the solvents in the assay was 0.1%. Controls received the solvents only (carrier control). For the test compounds, serial dilutions were assayed.

2.4. Cell 6iability assay Viability of the cells exposed to the above compounds was assessed by measuring the release of LDH into the extracellular medium following the method described by Scholz and Segner (1998). Activity of LDH in the medium was determined following the oxidation of NADH in a spectrophotometer at 340 nm. Cell appearance and morphology were routinely observed using an inverted microscope.

2.5. Vitellogenin assay Vitellogenin was analyzed in cell culture medium by using a non-competitive ELISA as previously described by Schrag et al. (1998). A polyclonal rabbit anti-trout VTG antibody prepared in our laboratory againts purified vitellogenin of trout plasma was used (Navas et al., in preparation). Dilutions of the samples (four to

82

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

eightfold) and of the standard were performed in M199 medium. The diluted samples and standards (100 ml) were incubated overnight at 4°C in individual wells of sealed MicroTest III Falcon flexible assay plates (Becton Dickinson, Oxnard, CA). After coating, wells were washed three times with PBS pH 7.4 (PBS, 137 mM NaCl, 2 mM KCl, 10 mM Na2HPO4, 1 mM KH2PO4) containing 1% Tween 20 (PBST). To reduce background, wells were blocked with 200 ml of 1% bovine serum albumin (BSA, Serva, Germany) diluted in PBS, for 1.5 h. All the incubation steps were carried out at room temperature. Again, wells were washed as above. Polyclonal rabbit antitrout VTG antibody was diluted to 1/20 000 and 100 ml per well were applied for 1.5 h. After washing, plates were incubated for 1.5 h with 100 ml per well of monoclonal anti-rabbit IgG antibody coupled with horseradish peroxidase (diluted 1/2000 in PBS). Plates were washed and the peroxidase activity was visualized with 200 ml per well of an ABTS (2,2%-azino-bis(3-ethylbenz-thiazoline-6-sulfonic acid)) solution containing 50 mg ABTS, 200 ml H2O2 and 200 ml acetate buffer, (100 mM C2H3NaO2, 40 mM Na2HPO4, pH corrected to 4.2 with acetic acid). After 30 min reaction, the absorbance was read at 405 nm with a SLT (Crailsheim, Germany) microtiter plate reader.

2.6. EROD and protein assays EROD activity was measured to estimate the catalytic activity of CYP 1A. Measurements of EROD activity and total protein content of the wells were analysed following the method described by Kennedy et al. (1995). Plates were removed from the freezer and hepatocytes were allowed to thaw at room temperature for 10 min. Phosphate buffered saline (PBS, 250 ml, pH 7.8, 150 mM KCl, 80 mM Na2HPO4, 20 mM KH2PO4) were added to the wells that contained the hepatocytes. Ethoxyresorufin (Sigma, USA) in methanol was diluted with PBS to yield a concentration of 35 mM. This solution (50 ml) was added to the wells (final concentration in the wells, 6 mM). Different concentrations of bovine serum albumine (BSA) and of resorufin were added to

empty wells and served as standards. BSA was prepared in sodium phosphate buffer and resorufin (Sigma, USA) standards were prepared by diluting methanol solution of resorufin in PBS. The plates were maintained at room temperature for a 10-min preincubation period. To start EROD reactions, 50 ml of a 13.4 mM solution of NADPH (Serva, Germany) in PBS were added to the wells. Plates were incubated for 7 min at 25°C and reaction was stopped with 300 ml of acetonitrile (Merck, Germany) that contained fluorescamine (Sigma, USA) at a concentration of 150 mg/ml. After 15 min (to allow maximal and stable fluorescence) plates were placed into the fluorescence plate reader and scanned for resorufin (excitation, 530 nm; emission, 590 nm) and for total proteins (excitation, 355 nm; emission, 460 nm). The concentration of VTG detected in the medium and the EROD activity observed in the cells were normalized to the quantity of cell protein present in the corresponding well.

2.7. CYP 1A ELISA The CYP 1A content of the hepatocytes was measured by a non competitive ELISA as described by Scholz et al. (1997). Microsomes were prepared from the trout liver cell cultures. Approximately 1× 106 cells were used to prepare the microsomes. The microsome samples were diluted to 10 mg/ml protein in coating buffer (50 mM Na-bicarbonate, pH 9.5) and the microtitre plates were coated with 100 ml of these solutions.

2.8. E2 ELISA Analysis of E2 in medium was performed by using a 17b-estradiol ELISA kit (r-biopharm, Germany). This is a competitive enzyme immunoassay based on the competition between the E2 from samples or standards and an E2-enzyme conjugate for the same antibody.

2.9. Statistics Every treatment was repeated in four to five independent experiments, with cells from different

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

fish (except for 8MP treatments, with n = 3 only, and measurements of E2 in medium of cells treated with 1 mM E2 and 12.5 mM bNF, with n= 1 only). In each individual experiment, every treatment was done in duplicate or triplicate. Mean values obtained for every fish were compared. Significant differences between means of various treatment groups were determined by ANOVA (P B 0.05) and means were contrasted using the Dunnett test. If the xenobiotic was applied together with E2, the group receiving E2 alone was considered as control group. If the xenobiotic was applied alone, the results were compared with those of the cells receiving the carrier (control). A regression model equation (sigmoid y = {(max− min)/[1 + (x/EC50)b)] +min} was fitted

83

to the EROD activity and VTG concentration data to estimate the concentration–response function and to calculate the EC50 values.

3. Results

3.1. Cell 6iability Alterations of liver cell viability during in vitro incubation was estimated from cellular LDH release into the medium. In control treatments, LDH leakage ranged between 13 and 20% of total LDH. When cells were exposed to E2 and/or PAHs, values of LDH release in no case were significantly increased, compared with the controls.

3.2. VTG and EROD acti6ity in the cells treated with ANT, 3MC or bNF

Fig. 1. Effect of different concentrations of anthracene (mM) on (A) EROD activity and on (B) VTG production by rainbow trout hepatocytes maintained in culture for 72 h.

The antiestrogenic activity of the test compounds was determined by their ability to decrease or inhibit the E2-induced hepatocellular synthesis of VTG. To this end, cells were pre-exposed for 24 h to a dilution series of the test compound alone and then for another 48 h to 1 mM E2 plus the different concentrations of xenobiotic. The interaction of the test compounds with the AhR was assessed from their capacity to induce the CYP 1A-catalyzed EROD activity. ANT alone, in the absence of E2, did not alter the cellular EROD activity (Fig. 1A) or cellular VTG production when compared with controls (Fig. 1B). When ANT was applied together with E2, no elevation of the EROD activity of the cells was observed (Fig. 2A). The highest concentration of ANT used (6.25 mM) provoked a significant reduction (P B0.05) of the EROD activity with respect to the control, without compromising cell viability. Simultaneously, ANT had no effects on the E2 dependent VTG synthesis (Fig. 2B). Exposure of trout hepatocytes to 3MC, in the absence of E2, induced an increase of EROD activity (Fig. 3A), being significantly different (PB 0.05) from controls at the concentrations of 0.39–1.56 mM 3MC. The EC50 value was 0.042 mM 3MC. The EROD activity decreased at the

84

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

mM bNF. A reduction of the EROD activity occurred at the highest concentration of bNF (12.5 mM). bNF alone did not induce or suppress basal VTG synthesis (Fig. 5B). When the cells were co-treated with E2 and bNF (Fig. 6), increments of EROD activity were similar to those detected after treatment with bNF alone (Fig. 6A). At the same time, bNF concentrations of 0.78 mM or higher resulted in a significant (PB 0.05) inhibition of E2-dependent VTG synthesis (Fig. 6B). EC50 values for EROD induction and VTG inhibition were 0.0138 and 0.805 mM bNF, respectively. In summing up, the test agent with the highest induction potency for EROD, bNF, was also the

Fig. 2. Effect of different concentrations of anthracene (mM) administered in combination with 1 mM E2 on (A) EROD activity and on (B) VTG production by rainbow trout hepatocytes. Asterisks indicate significant differences (PB 0.05) with respect to the E2-control.

highest concentration of 3MC used (6.25 mM). 3MC did not induce or suppress basal VTG production (Fig. 3B). EROD activity in trout hepatocytes was induced by 3MC also in the presence of E2, with an EC50 value of 0.054 mM 3MC (Fig. 4A). The induction was significant compared with controls for concentrations of 0.19 mM and higher. The E2-dependent hepatocellular VTG synthesis decreased with increasing concentrations of 3MC; the inhibitory effect had an EC50 value of 4.75 mM 3MC (Fig. 4B). Treatment of the cells with bNF led to an increase of EROD activity (Fig. 5A), which was significantly different from control for 0.024 – 6.25

Fig. 3. Effect of different concentrations of 3MC (mM) on (A) EROD activity; and (B) VTG production by rainbow trout hepatocytes maintained in culture for 72 h. Asterisks indicate significant differences with respect to the carrier-control (P B 0.05).

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

85

8MP, which inhibits the binding of the receptor– ligand complex to the DNA (Jeong et al., 1995). As observed in the experiments described above, exposure of trout liver cells to bNF together with 1 mM E2 resulted in a significant, dose-dependent induction of EROD activity (Fig. 7A), and in a significant, dose-dependent reduction of the E2-dependent VTG production (Fig. 7B). The addition of 6.25 mM aNF led to significantly lower EROD levels than with bNF alone, whereas the inhibitory effect of bNF on cellular VTG production was not altered by the aNF treatment (Fig. 7C and D). When aNF was added at a concentration of 12.5 mM, however, the inhibitory effect of bNF on cellular VTG production disappeared (Fig. 7E and F).

Fig. 4. Effect of different concentrations of 3MC administered in combination with 1 mM E2 on (A) EROD activity; and (B) VTG production by rainbow trout hepatocytes. Asterisks indicate significant differences with respect to the E2-control (PB 0.05).

strongest inhibitor of VTG synthesis. The less potent inductor of EROD activity, 3MC, was also the less potent inhibitor of VTG synthesis. Finally, ANT, which did not induce EROD activity, remained without effect on VTG production.

3.3. VTG, EROD acti6ity, and CYP 1A1 protein in the cells co-treated with bNF and aNF In order to evaluate if the observed antiestrogenic effects of PAHs on the E2-induced hepatocyte VTG synthesis might be mediated through the AhR, we used two inhibitors, aNF, which is an AhR antagonist (Merchant et al., 1993), and

Fig. 5. Effect of different concentrations of b-NF (mM) on (A) EROD activity; and (B) VTG production by rainbow trout hepatocytes. Asterisks indicate significant differences with respect to the carrier-control (PB 0.05).

86

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

3.4. VTG and EROD acti6ity in the cells treated with bNF and 8MP bNF resulted in a significant elevation of the hepatocellular EROD activity at all three concentrations applied (1.56, 3.12, and 6.25 mM) (Fig. 9A), while the two highest concentrations of bNF (3.12 and 6.25 mM) significantly reduced E2-induced VTG synthesis (Fig. 9B). When 8MP was added together with bNF, an inhibition of the bNF induced EROD activity was observed at both concentrations of 8MP used (10 and 50 mM) (Fig. 9C and E). In addition, 8MP abolished the antiestrogenic effect of bNF, i.e. VTG levels of bNF-treated cells were not significantly reduced when compared with VTG production of cells exposed to E2 only (Fig. 9D and F).

3.5. E2 content in the medium

Fig. 6. Effect of different concentrations of b-NF (mM) administered in combination with 1 mM E2 on (A) EROD activity; and (B) VTG production by rainbow trout hepatocytes. Asterisks indicate significant differences with respect to the E2-control (P B0.05).

In order to evaluate if enhanced CYP 1A expression as it results from exposure of the cells to PAHs may enhance E2 catabolism we analyzed the E2 levels in the medium of cells treated with E2 only or with E2 together with bNF. After the treatment period, E2 medium concentrations from cultures receiving E2 only were 1629 55 pmol/ml medium, whereas the E2 levels in culture medium of cells treated with 6.25 or 12.5 mM bNF were 1929 32 and 180918 pg/ml, respectively.

4. Discussion In order to prove if the observed aNF effects were mediated through the AhR, instead of being the result of a competitive inhibition of EROD catalytic activity by aNF, we measured the microsomal CYP 1A1 protein content in hepatocytes treated with bNF or with a combination of bNF and aNF (Fig. 8). Exposure to 6.25 or 12.5 mM bNF induced a statistically significant (P B 0.05) increase of the microsomal CYP 1A1 protein compared with cells treated with carrier or with E2 only. Addition of 6.25 mM aNF to bNFtreated cells prevented the induction of CYP 1A protein.

In the present work, the antiestrogenicity of PAHs has been assessed by their ability to inhibit the E2-induced production of VTG in rainbow trout hepatocytes in vitro. The results demonstrate that the CYP 1A-inducing PAH, 3MC, and the CYP 1A model inducer, bNF, modulate E2regulated VTG synthesis in trout liver cells, whereas the non-inducing PAH, ANT, has no effect on VTG. The antiestrogenic activity of the test agents in the trout system could be explained by several mechanisms. As one possible explanation, the decrease of cellular VTG synthesis in response to xenobiotic exposure could be caused by cytotoxic

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

effects of the chemicals. However, the inhibitory action on VTG secretion of trout liver cells occurred at concentrations that were not cytotoxic, as evident from LDH leakage and therefore, this explanation is unlikely. Accordingly, suppression of VTG synthesis in trout or carp hepatocytes by PCBs were found to occur at non-cytotoxic (Anderson et al., 1996a; Smeets, 1999). A number of substances, e.g. tamoxifen, act as antiestrogens by means of antagonistic binding to the ER (MacGregor and Jordan, 1998). These competitive inhibitors of the binding of E2 to its cognate receptor either form a receptor complex that is converted incompletely to the fully activated form or they cause receptor destruction.

87

For mammals, it has been shown that, although PAHs are no ligands for the ER, metabolites of PAHs are able to bind competitively to the ER and to mediate antiestrogenic effects through this pathway (Ebright et al., 1986; Arcaro et al., 1999). In teleost fish, PAHs have no binding affinity for the ER (Thomas and Smith, 1993), but currently no information is available with respect to binding of PAH metabolites to teleostean ER. The antiestrogenicity of PAHs in mammals has been interpreted as a result of the PAH binding to the AhR. The results of the present study would agree with such a hypothesis, (1) Only the CYP 1A-inducing, i.e. AhR-binding, agents led to an antivitellogenic effect. Similarly, in MCF-7 cells

Fig. 7. Effect of aNF on EROD induction and the antiestrogenic activity of bNF. bNF led to an increase of EROD (A); and a reduction of the E2-dependent vitellogenin production (B). At a concentration of 6.25 mM, aNF reduced the EROD induction by bNF (C) but not the antivitellogenic effect of bNF. At 12.5 mM, aNF inhibited the EROD induction (E) and the antiestrogenicity of bNF (F). Asterisks indicate significant differences with respect to the E2-control (P B 0.05).

88

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

Fig. 8. Effect of aNF on the induction of CYP 1A1 protein by bNF. A significant induction of CYP 1A1 protein was observed after treatment of the cells with 6.25 or 12.5 mM bNF. The simultaneous treatment of the cells with aNF (12.5 or 25 mM) resulted in a reduction of CYP 1A1 protein compared with cells treated with bNF only.

the rank order of PAHs as inducers of CYP 1 was directly related to their antiestrogenic potency (Chaloupka et al., 1992). The same observation applied for another group of AhR ligands, the polychlorinated dibenzo-p-dioxins (PCDD) (Gierthy et al., 1993; Krishnan and Safe, 1993; Zacharewski et al., 1994). In cultured hepatocytes of rainbow trout and carp, the antivitellogenic activity of dioxins and PCBs was essentially in parallel to their EROD-inducing potency (Anderson et al., 1996a; Smeets, 1999); (2) The antiestrogenic efefct of bNF could be abolished by the AhR antagonist, aNF, and by methoxypsoralen, which inhibits the binding of the AhR – ligand complex to the xenobiotic responsive elements (XRE) of the DNA. Although aNF may inhibit EROD activity by competitive inhibition (Koley et al., 1997; Smeets, 1999), it also acts as an AhR antagonist (Gasiewicz and Rucci, 1991; Merchant et al., 1993; Wang et al., 1993). In our study, the results from the CYP 1A ELISA support the view that the EROD inhibition after aNF treatment was not due to substrate competitition but to a receptor antagonistic effect of aNF. This interpretation is further supported from the methoxypsoralen results which indicate that antiestrogenicity of bNF requires binding of the AhR – ligand complex to the DNA.

Several mechanisms of AhR-mediated antiestrogenicity have been proposed (Safe et al., 1991; Safe, 1994; Navas and Segner, 1998). One possibility is an enhanced metabolism of E2 due to the AhR-mediated induction of CYP 1A activity. Increased rates of E2 metabolism as a result of exposure to AhR ligands such as TCDD have been reported, for instance, for MCF-7 human breast cancer cells (Spink et al., 1992; Arcaro et al., 1999). In our hepatocyte culture system, analyses of medium E2 concentrations from control and induced cultures did not reveal significant differences of hormone levels, thereby indicating that the elevated CYP 1A levels were not associated with enhanced E2 biotransformation. Anderson et al. (1996a) observed that 1 h after rainbow trout liver cells were exposed to 1 mM E2, approximately 99% of the steroid was converted to water soluble compounds, regardless whether the cells were CYP 1A-induced or not. The other possible mechanisms to explain the AhR-mediated antiestrogenicity implicate interactions of the activated AhR or of AhR-dependent modulatory factors with estrogen-dependent genes or DNA elements, including down-regulation of ER expression by activated AhR, or impaired binding of activated ER to estrogen-responsive elements on the DNA. Several lines of evidence indicate that AhR ligands can lead to a decrease in the levels of nuclear ER (Zacharewski et al., 1991; DeVito et al., 1992; Wang et al., 1993; Tian et al., 1998), possibly by inducing the interaction of the activated AhR with XREs present upstream of the ER gene (White and Gasiewicz, 1993). In rainbow trout, a reduction of the E2binding sites was observed when the fish were treated simultaneously with E2 and bNF (Anderson et al., 1996b), an effect that could have been caused by bNF-induced downregulation of hepatic ER. AhR-mediated antiestrogenic response due to the interaction with E2-responsive sequences (ERE) in the DNA by targeted interaction with an overlapping XRE was reported by several authors from studies with mammalian systems (Krishnan et al., 1995; Kharat and Saatcioglus, 1996). In the present study, 8MP was used to explore if

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

the antiestrogenic action of bNF requires DNA binding. 8MP has been described in Hepa 1c1c7 cells (Jeong et al., 1995) to be an inhibitor of the CYP 1A induction provoked by TCDD, by interfering with the AhR binding to the XRE of the DNA. In rainbow trout hepatocytes, 8MP inhibited completely the bNF induced EROD activity, indicating the effectivity of this substance to inhibit the AhR-mediated CYP 1A induction. This effect could be explained by a mechanism similar to that described for Hepa 1c1c7 cells. At the same time, 8MP eliminated the bNF-induced inhibition of the VTG production. This finding may indicate that bNF exerts its antiestrogenic effect

89

through the interaction of the activated AhR with a XRE present in the regulatory regions of the VTG gene. The question arises on the in vivo physiological relevance of the herein reported in vitro observations of PAH antiestrogenicity. In oviparous vertebrates, hepatic vitellogenesis is a key process for the growth of the oocytes. VTG is synthesized in the liver under stimulation of E2, transported via the vascular system to the ovary and incorporated through specific receptors in developing oocytes (Mommsen and Walsh, 1988). The VTG in the egg is the major source of nutrition for the developing embryo. Any perturbation of this system by

Fig. 9. Effect of 8MP on EROD induction and the antiestrogenic activity of bNF. bNF led to an increase of EROD (A); and a reduction of the E2-dependent vitellogenin production (B). At the concentrations of 10 and 20 mM, 8MP inhibited the EROD induction by bNF (C and E) and the antivitellogenic effect of bNF (D and F). Asterisks indicate significant differences with respect to the E2-control (PB0.05).

90

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

a toxicant could have detrimental influences on oogenesis, fecundity, embryonic development, egg hatching, and larval survival. Changes in egg production and larval survival could in turn affect recruitment and thereby constitute a population level effect (Cairns et al., 1984). In several field studies, relations between exposure of fish to AhR ligands and reproductive disturbances, particularly reduced plasma steroid levels and impaired gonad development, have been observed (Kime, 1995; Johnson et al., 1998; Munkittrick et al., 1998). Laboratory studies showed that Aroclor 1254 or benzo(a)pyrene are able to reduce plasma vitellogenin levels in fish in vivo (Chen et al., 1985; Thomas, 1989). However, a number of observations indicate that the relation between AhR binding of xenobiotics and antiestrogenic activity cannot expected to be a simple, deterministic relationship. For instance, Donohoe et al. (1999) found that 3,3%,4,4%,5,5%-hexachlorobiphenyl, although inducing CYP 1A, failed to evoke antiestrogenic responses in rainbow trout in vivo. Interestingly, Anderson et al. (1996b) observed that the antiestrogenic potency of bNF in rainbow trout in vivo was influenced by the relative concentrations of bNF and E2. In their in vitro studies, Anderson et al. (1996a) observed that PCBs with low CYP 1A induction potency did not act as antiestrogens but even enhanced the cellular response to E2 treatment. These authors speculated that the antiestrogenic effect may depend on the dose, with low, hardly CYP 1A-inducing doses exerting a pro-estrogenic activity, and high, strongly CYP 1A-inducing doses leading to antiestrogenic effects. Contrary to Anderson et al. (1996a,b), who tested one or two concentrations of the AhR ligands, we used in the present study, complete concentration – response curves, but we did not notice any pro-estrogenic activity at low PAH concentrations. Remarkably, in our in vitro experiments the antiestrogenicity of 3MC and bNF was expressed only at the higher concentrations of the test compounds, when EROD levels approached the satiation niveau, i.e. antiestrogenic responses became evident at much higher concentrations of AhR ligands than required for CYP 1A induction. Identical findings have been reported by Smeets (1999) from his

study on the antiestrogenicity of PCBs in carp hepatocytes in vitro. The distance between the CYP 1A and the antiestrogenic threshold leads to the question if the antiestrogenic potential of AhR ligands such as PCBs and PAHs will be of toxicological relevance in the in vivo situation.

Acknowledgements This work was financially supported by European Commission, Contract No. ENV4-CT960223, and Contract No. ENV4-CT-965042 to Jose´ Marı´a Navas.

References Anderson, M.J., Miller, M.R., Hinton, D.E., 1996a. In vitro modulation of 17b-estradiol-induced vitellogenin synthesis: effects of cytochrome P450 1A1 inducing compounds on rainbow trout (Oncorhynchus mykiss) liver cells. Aquat. Toxicol. 34, 327 – 350. Anderson, M.J., Olsen, H., Matsumura, F., Hinton, D.E., 1996b. In vivo modulation of 17b-estradiol-induced vitellogenin synthesis and estrogen receptor in rainbow trout (Oncorhynchus mykiss) liver cells by b-naphthoflavone. Toxicol. Appl. Pharmacol. 137, 210 – 218. Arcaro, K.F., O’Keefe, P.W., Yang, Y., Clayton, W., Gierthy, J.F., 1999. Antiestrogenicity of environmental polycyclic aromatic hydrocarbons in human breast cancer cells. Toxicology 133, 115 – 127. Cairns, V.W., Hodson, P.V., Nriagu, J.O., 1984. Contaminant effects on fisheries. In: Nriagu, J.O. (Ed.), Advances in Environmental Science and Technology. Wiley, New York, p. 16. Chaloupka, K., Krishnan, V., Safe, S., 1992. Polynuclear aromatic hydrocarbon carcinogens as antiestrogens in MCF-7 human breast cancer cells: role of the Ah receptor. Carcinogenesis 13, 2233 – 2239. Chen, T.T, Reid, P.C., van Beneden, R., Sonstegard, R.A., 1985. Effect of Aroclor 1254 and Mirex on estradiol-induced vitellogenin production in juvenile rainbow trout (Salmo gairdneri ). Can. J. Fish Aquat. Sci. 43, 169 – 173. Colborn, T., Clement, C., 1992. Chemically Induced Alterations in Sexual and Functional Development: The Wildlife/Human Connection. Princeton Scientific Publishing, Princeton, NJ. DeVito, M.J., Thomas, T., Martin, E., Umbreit, H., Gallo, M.A., 1992. Antiestrogenic action of 2,3,7,8-tetrachlorodibenzo-p-dioxin: tissue-specific regulation of estrogen receptor in CD1 mice. Toxicol. Appl. Pharmacol. 113, 284 – 292.

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92 Donohoe, R.M., Wang-Buhler, J.L., Buhler, D.R., Curtis, L.R., 1999. Effects of 3,3%,4,4%,5,5%-hexachlorobiphenyl on cytochrome P450 1A and estrogen-induced vitellogenesis in rainbow trout (Oncorhynchus mykiss). Environ. Toxicol. Chem. 18, 1046 – 1052. Ebright, R.H., Wong, J.R., Chen, L.B., 1986. Binding of 2-hydroxy-benzo(a)pyrene to estrogen receptors in rat cytosol. Cancer Res. 46, 2349–2351. Facemire, C.F., Gross, T.S., Guillette, L.J. Jr, 1995. Reproductive impairment in the Florida panther: nature or nurture? Environ. Health Perspect. 103, 79–86. Gasiewicz, T.A., Rucci, G., 1991. a-Naphthoflavone acts as an antagonist of 2,3,7,8-tetrachlorodibenzo-p-dioxin by forming an inactive complex with the Ah receptor. Mol. Pharmacol. 40, 607 – 612. Gierthy, J.F., Bennet, J.A., Bradley, L.M., Cutler, D.S., 1993. Correlation of in vitro and in vivo growth suppression of MCF-7 human breast cancer by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Cancer Res. 53, 3149–3153. Guillette, L.J. Jr, Gross, T.S., Masson, G.R., Matter, J.M., Percival, H.J., Woodward, A.R., 1994. Developmental abnormalities of the gonad and abnormal sex hormone concentrations in juvenile alligators from contaminated and control lakes in Florida. Environ. Health Perspect. 102, 680 – 688. Jeong, H.G., Yun, C.-H., Jeon, Y.J., Lee, S.S., Yang, K.-H., 1995. Suppression of cytochrome P450 (Cyp1a-1 ) induction in mouse hepatoma hepa-1c1c7 cells by methoxysoralen. Biochem. Biophys. Res. Commun. 208, 1124–1130. Johnson, L.L., Misitano, D., Sol, S.Y., Nelson, G.M., French, B., Ylitalo, G.M., Hom, T., 1998. Contaminant effects on ovarian development in rock sole from Puget Sound, Washington. Trans. Am. Fish Soc. 127, 375–392. Kavlock, R.J., Daston, G.P., DeRosa, C., Fenner-Crisp, P., Gray, L.E., Kaatari, S., Lucier, G., Luster, M., Mac, M.J., Mazka, C., Miller, R., Moore, J., Rolland, R., Scott, G., Sheehan, D., Sinks, T., Tilson, H.A., 1996. Research needs for the risk assessment of endocrine disruptors: a report of the US-EPA sponsored workshop. Environ. Health Perspect. 102 (Suppl. 4), 715–740. Kennedy, S.W., Jones, S.P., Bastien, L.J., 1995. Efficient analysis of cytochrome P450 1A catalytic activity, porphyrins, and total proteins in chicken embryo hepatocyte cultures with a fluorescence plate reader. Anal. Biochem. 226, 362– 370. Kharat, I., Saatcioglus, F., 1996. Antiestrogenic effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin are mediated by direct transcriptional interference with the liganded estrogen receptor. J. Biol. Chem. 271, 10533–10537. Kime, D.E., 1995. The effects of pollution on reproduction in fish. Rev. Fish Biol. Fish. 5, 52–96. Koley, A.P., Buters, J.T.M., Robinson, R.C., Markowitz, A., Friedman, F.D., 1997. Differential mechanisms of cytochrome P450 inhibition and activation by a-naphthoflavone. J. Biol. Chem. 272, 3149–3152. Krishnan, V., Safe, S., 1993. Polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs) and dibenzofurans

91

(PCDFs) as antiestrogens in MCF-7 human breast cancer cells: quantitative structure – activity relationships. Toxicol. Appl. Pharmacol. 120, 55 – 61. Krishnan, V., Porter, W., Santostefano, M., Wang, X., Safe, S., 1995. Molecular mechanism of inhibition of estrogeninduced cathepsin D gene expression by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in MCF-7 cells. Mol. Cell. Biol. 15, 6710 – 6719. MacGregor, J.I., Jordan, C.V., 1998. Basic guide to the mechanisms of antiestrogenic action. Physiol. Rev. 50, 152 – 196. Merchant, M., Krishnan, V., Safe, S., 1993. Mechanism of action of a-naphthoflavone as an Ah receptor antagonist in MCF-7 human breast cancer cells. Toxicol. Appl. Pharmacol. 120, 179 – 185. Mommsen, T.P., Walsh, P.J., 1988. Vitellogenesis and oocyte assembly. In: Hoar, W.S., Randall, D.J. (Eds.), Fish Physiology, XIA. Academic Press, New York, pp. 347 – 406. Munkittrick, K.R., McMaster, M.E., McCarthy, L.H., Servos, M.R., van der Kraak, G.J., 1998. An overview of recent studies on the potential of pulp-mill effluents to alter reproductive parameters in fish. J. Toxicol. Environ. Health B1, 347 – 371. Navas, J.M., Segner, H., 1998. Antiestrogenic activity of anthropogenic and natural chemicals. Environ. Sci. Pollut. Res. 5, 75 – 82. Nicolas, J.M., 1999. Vitellogenesis in fish and the effects of polycyclic aromatic hydrocarbon contaminants. Aquat. Toxicol. 45, 77 – 90. Safe, S., 1994. Dietary and environmental estrogens and antiestrogens and their possible role in human disease. Environ. Sci. Pollut. Res. 1, 29 – 33. Safe, S., Astroff, B., Harris, M., Zacahrewski, T., Dickerson, R., Romkes, M., Biegel, L., 1991. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) and related compounds as antioestrogens: characterization and mechanism of action. Pharmacol. Toxicol. 69, 400 – 409. Scholz, S., Segner, H., 1998. Induction of CYP 1A in primary cultures of rainbow trout (Oncorhynchus mykiss) liver cells: concentration – response relationships of four model substances. Ecotoxciol. Environ. Saf. 43, 252 – 260. Scholz, S., Behn, I., Honeck, H., Hauck, C., Braunbeck, T., Segner, H., 1997. Development of a monoclonal antibody for ELISA of CYP 1A in primary cultures of rainbow trout (Oncorhynchus mykiss) hepatocytes. Biomarkers 2, 287 – 294. Schrag, B., Ensenbach, U., Navas, J.M., Segner, H., 1998. Evaluation of xenoestrogenic effects in fish on different organization levels. In: Del Mazo, J. (Ed.), Reproductive Toxicology. Plenum Press, New York, pp. 207 – 214. Sharpe, R.M., Skakkebaek, N.E., 1993. Are oestrogens involved in falling sperm counts and disorders of the male reproductive tract? Lancet 341, 1392 – 1395. Smeets, J.M.W., 1999. In vitro assays for effects of contaminants on fish. Dissertation, University of Utrecht, The Netherlands. Spink, D.C., Eugster, H.P., Lincoln, D.W., Schuetz, J.D., Schuetz, E.G., Johnson, J.A., Kaminsky, L.S., Gierthy,

92

J.M. Na6as, H. Segner / Aquatic Toxicology 51 (2000) 79–92

J.F., 1992. 17b-Estradiol hydroxylation catalyzed by human cytochrome P450 1A1: a comparison of the activities induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin in MCF-7 cells with those from heterologous expression of the cDNA. Arch. Biochem. Biophys. 293, 342–348. Sumpter, J.P., Jobling, S., Tyler, C.R., 1996. Oestrogenic substances in the aquatic environment and their potential impact on animals, particularly fish. In: Taylor, E.W. (Ed.), Toxicology of Aquatic Pollution. Cambridge University Press, Cambridge, pp. 205–225. Thomas, P., 1989. Effects of Aroclor 1254 and cadmium on reproductive endocrine function and ovarian growth in Atlantic croaker. Mar. Environ. Res. 28, 499–503. Thomas, P., Smith, J., 1993. Binding of xenobiotics to the estrogen receptor of spotted seatrout: a screening assay for potential estrogenic effects. Mar. Environ. Res. 35, 147– 151. Tian, Y., Ke, S., Thomas, T., Meeker, R.J., Gallo, M.A., 1998. Transcriptional suppression of estrogen receptor gene expression by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). J. Steroid Biochem. Mol. Biol. 67, 17–24. Tsai, M.J., O’Malley, B.W., 1994. Molecular mechanisms of action of steroid/thyroid receptor superfamily members. Annu. Rev. Biochem. 63, 451–486.

.

Wang, X., Porter, W., Krishnan, V., Narasimhan, T.R., Safe, S., 1993. Mechanism of 2,3,7,8-tetrachlorodibenzo-pdioxin (TCDD)-mediated decrease of nuclear estrogen receptor in MCF-7 human breast cancer cells. Mol. Cell. Endocrinol. 96, 159 – 166. Wannemacher, R., Rebstock, E., Kulzer, E., Schrenk, D., Bock, W.K., 1992. Effects of 2,3,7,8-tetrachlorodibenzo-pdioxin on reproduction and oogenesis in zebrafish (Brachydanio rerio). Chemosphere 24, 1361 – 1368. White, T.E., Gasiewicz, T.A., 1993. The human estrogen receptor structural gene contains a DNA sequence that binds activated mouse and human Ah receptors: a possible mechanisms of estrogen receptor regulation by 2,3,7,8-tetrachlorodibenzo-p-dioxin. Biochem. Biophys. Res. Commun. 193, 956 – 962. Zacharewski, T., Harris, M., Safe, S., 1991. Evidence for a possible mechanism of action of 2,3,7,8-tetrachlorodibenzo-p-dioxin-mediated decrease of nuclear estrogen receptor levels in wild-type and mutant Hepa 1c1c7 cells. Biochem. Pharmacol. 41, 1931 – 1939. Zacharewski, T.R., Bondi, K.L., McDonell, P., Wu, Z.F., 1994. Antiestrogenic effect of 2,3,7,8-tetrachlorodibenzo-pdioxin on 17b-estradiol-induced pS2 expression. Cancer Res. 54, 2707 – 2713.