Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Among the environmental contaminants of most concern due to their toxicity, including carcinogenicity, are heavy metals, such as arsenic, lead, and mercury, and polycyclic aromatic hydrocarbons (PAHs¹) typified by benzo[a]pyrene (BAP). PAHs and heavy metals are often cocontaminants in the environment, yet there have been relatively few studies of the combined toxic and particularly the carcinogenic effects elicited by PAHs and heavy metals. The carcinogenicity of BAP and other PAHs is a consequence of their metabolic activation catalyzed by cytochromes P450 (P450) and epoxide hydrolase (Wood et al., 1976; Kapitulnik et al., 1978). Covalent adducts formed by the reaction of PAH diol epoxide metabolites with guanine in mutational hotspots of critical genes such as that of the p53 tumor suppressor (Denissenko et al., 1996), if not efficiently repaired, may initiate tumorigenesis.

Several members of the P450 superfamily in conjunction with epoxide hydrolase have been shown to catalyze the metabolism of BAP to carcinogenic intermediates. In extrahepatic tissues, CYP1A1 and CYP1B1 are thought to be the most important enzymes in catalyzing the formation of mutagenic intermediates from BAP and a number of other PAHs, including several that are potent mammary gland carcinogens in rodents. CYP1B1 appears to be more active than CYP1A1 in the conversion of a number of PAHs to genotoxic intermediates (Shimada et al., 1996). In the presence of epoxide hydrolase, both CYP1A1 and CYP1B1 catalyze the conversion of BAP to its 7,8-dihydrodiol, and both enzymes can in turn metabolically activate this BAP metabolite to a mutagenic form (Shimada et al., 1996, 1999; Kim et al., 1998). 7,12-Dimethylbenz[a]anthracene-induced carcinogenesis, an extensively studied model of PAH-induced carcinogenesis, recently was found to be dependent on CYP1B1: the CYP1B1 knockout mouse is refractory to 7,12-dimethylbenz[a]anthracene-induced carcinogenesis (Buters et al., 1999).

PAHs are not only substrates of human CYP1A1 and CYP1B1 but also are inducers of these enzymes through binding to and activation of the aromatic hydrocarbon receptor (AhR). The regulation of expression of both CYP1A1 and CYP1B1 is complex because gene transcription not only involves the AhR but also a number of transcription factors, and is potentially influenced by the actions of transcriptional coactivators and corepressors. For reasons that are not currently known, in some human Ah-responsive cells, CYP1A1 is inducible but not CYP1B1, in other cells CYP1B1 is inducible but not CYP1A1, and in yet others both enzymes are inducible (Christou et al., 1994; Spink et al., 1994, 1998b; Döhr et al., 1995; Kress and Greenlee, 1997). There are several potential points, including transcription, translation, and incorporation of the heme prosthetic group, at which other chemicals, including environmental cocontaminants, could potentially modulate expression and catalytic activity of CYP1A1, CYP1B1, and other P450s. In several recent studies, an effect of arsenite on expression of cytochromes P450 in hepatocyte cultures was reported (Jacobs et al., 1999; Vakharia et al., 2001), although the mechanism responsible for this effect has not been determined.

Human CYP1A1 and CYP1B1 are also 17-estradiol (E₂) hydroxylases. CYP1A1 is primarily an E₂ 2-hydroxylase (Spink et al., 1992), whereas CYP1B1 is primarily an E₂ 4-hydroxylase with a lesser activity at C-2 (Spink et al., 1994; Hayes et al., 1996). The resultant catechol estrogens, 2- and 4-hydroxyestradiol (2- and 4-OHE₂), are converted to 2- and 4-methoxyestradiol (2- and 4-MeOE₂) by the action of constitutively expressed catechol O-methyltransferase. This metabolism of E₂ can be used as a probe of the AhR-regulated expression of CYP1A1 and CYP1B1 in human breast epithelial and breast tumor cells (Spink et al., 1998b). In this study, we used the T-47D line of estrogen receptor--positive breast tumor cells, which rapidly metabolize BAP (Merrick et al., 1985) and express both CYP1A1 (Vickers et al., 1989) and CYP1B1 (Spink et al., 1998b) in response to exposure to Ah-receptor activation, to investigate the effects of BAP alone or in combination with NaAsO₂ on the expression of CYP1A1 and CYP1B1.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture and Treatments. T-47D cells were obtained from the American Type Culture Collection (Rockville, MD) and were maintained in a 37°C incubator with a humidified atmosphere containing 5% CO₂. For experimental protocols, the medium was DC₅ (Gierthy et al., 1996) consisting of Dulbecco's modified Eagle's medium, without phenol red, and with 5% bovine calf serum (Cosmic calf serum; Hyclone Laboratories, Logan, UT), 10 mM nonessential amino acids, 2 mM L-glutamine, and 10 µg/l insulin. BAP was added to the cultures alone or in combination with varying concentrations of NaAsO₂. The vehicle for BAP was dimethyl sulfoxide (DMSO), which was used at a highest concentration of 0.1% (v/v).

RNA Isolation and RT-PCR. Total RNA was isolated by phenol-guanidinium thiocyanate extraction (Chomczynski and Sacchi, 1987), and portions (2.5-5 µg) were primed with oligo-dT, reverse-transcribed using Superscript II reverse transcriptase (Invitrogen, Carlsbad, CA) and treated with RNase H (Invitrogen) according to the manufacturer's protocol. Primers for the amplification of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and CYP1A1 cDNAs were as previously reported (Fasco et al., 1995). The primers for the amplification of CYP1B1 cDNA were those used previously (Spink et al., 1998a). Competitor DNAs were synthesized by the method of Förster (1994). Each amplified sequence spanned an exon-exon junction. PCR of the CYP1A1, CYP1B1, and GAPDH cDNAs was carried out as described (Spink et al., 1998a) with core reagents from PerkinElmer (Foster City, CA), but with the addition of TaqStart antibody (CLONTECH, Palo Alto, CA) and Taq extender (Stratagene, La Jolla, CA), and were amplified with the same PCR-plus-competitor mix used for the calibration curve at dilutions that fell within the linear range of the competitor curve. PCR products were resolved on a 2% 3:1 NuSieve (FMC Bioproducts, Rockland, ME) agarose gel, followed by staining with 0.75 µg/ml ethidium bromide for photography and quantitation by densitometric scanning of the photographic negative. For the determination of heme oxygenase 1 (HO-1) mRNA levels, a real-time PCR technique was used (for review, see Bustin, 2000). A 271-base pair fragment of the HO-1 cDNA was amplified by using the LightCycler System (Roche Molecular Biochemicals, Indianapolis, IN) with the primers of Premkumar et al. (1995). Quantitation was achieved by monitoring the fluorescence of intercalated SYBR Green I at each cycle. All mRNA levels were analyzed in triplicate, each representing three separate reverse transcriptions of RNA isolated from individual wells of control or treated T-47D cells.

Assay of Cellular Estrogen Metabolism. For the analysis of E₂ metabolism, confluent cultures of T-47D cells in six-well plates were exposed to the indicated concentrations of BAP, NaAsO₂, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), or the solvent vehicle 0.1% v/v DMSO in 2 ml of medium per well. After the specified time period for enzyme induction, media were replaced with media containing 1 or 5 µM E₂. At the indicated times, these media were recovered and treated with b-glucuronidase/sulfatase for hydrolysis of the metabolite conjugates as previously described (Spink et al., 1994). These enzyme-treated samples were subjected to solid-phase extraction and preparation of the metabolite-trimethylsilyl derivatives. The metabolite derivatives were analyzed by gas chromatography/mass spectrometry with stable-isotope dilution and selected-ion monitoring as described (Spink et al., 1994, 1998b). Rates of metabolite formation were normalized to cellular protein content as determined by the bicinchoninic acid assay method (Smith et al., 1985) by using a commercial reagent (Pierce Chemical, Rockford, IL).

Determination of Microsomal E₂ Hydroxylase Activity. To evaluate the inhibitory effects of BAP and NaAsO₂ on human CYP1A1 and CYP1B1, the E₂ hydroxylase activities of the two enzymes were determined essentially as described (Spink et al., 1992). Human recombinant CYP1A1 and CYP1B1, both coexpressed with human NADPH/cytochrome P450 oxidoreductase in Sf9 cells (Supersomes), were obtained from GENTEST (Woburn, MA). Microsomal incubations (250 µl) contained 5 mM MgCl₂, 10 µM of E₂ as substrate, 10 pmol of cDNA-expressed CYP1A1 or CYP1B1, and 2 mM ascorbic acid, and were buffered at pH 7.4 with 100 mM sodium phosphate. The reaction mixtures were preincubated at 37°C for 5 min in the presence of added BAP or NaAsO₂, and enzymatic reactions were initiated with NADPH at a final concentration of 1.4 mM. After 10 min, the reactions were terminated by the addition of two volumes of ice-cold 30 mM ascorbic acid, addition of internal standards, and immediate extraction with ethyl acetate. The extracts were dried over anhydrous sodium sulfate, evaporated to dryness under N₂, and trimethylsilyl derivatives were prepared. The derivatized samples were analyzed by gas chromatography/mass spectrometry with selected-ion monitoring.

Western Immunoblot Analysis of NADPH-P450 Reductase (CPR) and CYP1A1. Cultures in six-well plates were exposed to BAP and NaAsO₂ as indicated, after which the cell monolayers were washed with phosphate-buffered saline and solubilized by addition of gel sample buffer (250 µl/well), which did not contain reducing agent or dye. Equal amounts of protein from each well (2 µg for CPR, 4 µg for CYP1A1) were run on 10% Bis-Tris polyacrylamide gels (NuPAGE; Invitrogen) together with a series of standards consisting of either purified, recombinant human CPR (Panvera, Madison, WI) or human CYP1A1 (Supersomes from GENTEST). The proteins were transferred to Immobilon membranes (Millipore, Bedford, MA) that were then blocked with 5% nonfat dry milk in Tris-buffered saline containing 0.05% Tween 20 (Bio-Rad, Hercules, CA). The primary antibodies were goat anti-rat CPR (GENTEST) and goat anti-rat CYP1A1 (Daiichi, Tokyo, Japan), which were used at dilutions of 1:1000 and 1:2000, respectively, in Tris-buffered saline containing 0.05% Tween 20 and 0.5% nonfat dry milk. The secondary antibody was horseradish peroxidase conjugated-donkey anti-goat IgG (Cruz Marker compatible; Santa Cruz Biotechnology, Santa Cruz, CA) used at 1:7500 dilution for analysis of CPR or 1:20,000 for CYP1A1. Immunoreactive proteins were detected by using the SuperSignal West Pico Chemiluminescent Substrate (Pierce Chemical), and exposed films were subjected to scanning densitometry on a Pharmacia LKB ImageMASTER densitometer (Amersham Biosciences, Inc., Piscataway, NJ). The resulting band intensities for the detected proteins fell within linear standard curves prepared over the 1.3- to 20-fmol range for CPR and over the 0.1- to 2.0-fmol range for CYP1A1, and were thus converted to molar amounts of CPR and CYP1A1.

Determination of BAP Uptake. The extent of uptake of BAP into T-47D cells was evaluated by determining the level of BAP remaining in the medium and the amount that could be recovered from the cells at various times after exposure. BAP was extracted from the media samples by use of Sep-Pak C₁₈ cartridges (Waters, Milford, MA), and BAP was recovered from the trypsinized cells by sonication of the cell pellets in the presence of acetonitrile. The BAP levels in these extracts were determined by reversed phase, high-performance liquid chromatography (Vakharia et al., 2001).

Determination of Cell Viability. The effects on cell viability of exposure to 3 µM BAP alone or in combination with varying concentrations of NaAsO₂ for 12 h were assessed in 96-well plates, as described by Borenfreund et al. (1988), by using 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT).

Statistical Analyses. Statistical differences among treatment groups were evaluated by analysis of variance and use of the Student-Newman-Keuls test for multiple comparisons. These evaluations were performed using the SigmaStat software package, version 2.0 (Jandel Scientific, San Rafael, CA). For determination of EC₅₀ values for mRNA induction, concentration-response data were fit to the four-parameter sigmoidal function,
by using the SigmaPlot (Jandel Scientific) program.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

The time course of BAP uptake, CYP1A1 and CYP1B1 mRNA levels, and the rates of the 2- and 4-hydroxylation pathways of E₂ metabolism in T-47D cells were determined after a single exposure to medium containing 3 µM BAP. There was a rapid disappearance of BAP from the medium of T-47D cultures because only 30% of the initial BAP could be recovered from the medium at 3 h (Fig. 1A). BAP was rapidly taken up in these cells and was apparently rapidly metabolized because at 3 h 20% of the initial BAP was present in the cells, but the level declined rapidly thereafter (Fig. 1B). With extended incubation the cultures became depleted of BAP because by 12 h only 5% of the original BAP could be recovered from the medium, and BAP could no longer be detected in extracts of the cells.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Time course of BAP uptake, CYP1A1 and CYP1B1 mRNA levels, and the rates of the 2- and 4-hydroxylation pathways of E2 metabolism in T-47D cells after BAP exposure. Cultures of T-47D cells in six-well plates were exposed to a single application of medium containing 3 µM BAP (2 ml/well). At the indicated times, the levels of BAP in the medium (A) and in the cells (B) were determined. ND indicates that BAP was not detectable (<0.0001 nmol). RNA was isolated at the indicated times, and CYP1A1 (C) and CYP1B1 (D) mRNA levels were determined by competitive RT-PCR. In E and F, media were replaced at the indicated times with media containing 1 µM E2 with 0.1% DMSO as the vehicle. After 6-h assay periods indicated by the horizontal bars, media were recovered, conjugates were hydrolyzed, and rates of 2-MeOE2 (E) and 4-MeOE2 (F) formation were determined relative to cellular protein content as described under Materials and Methods. The curves bordering the shaded areas in E and F represent approximations of the theoretical limits for the time course of induction and decline of the metabolic rates as defined by these assay periods. In each panel, data are the means of triplicate determinations, each replicate representing a well of cells. Standard errors are shown when they are larger than the symbols.

The time courses of the effects of BAP exposure on the levels of CYP1A1 and CYP1B1 mRNAs in T-47D cells were determined by using competitive RT-PCR. After exposure to 3 µM BAP, the levels of both the CYP1A1 (Fig. 1C) and CYP1B1 (Fig. 1D) mRNAs increased for the first 9 h, but declined steadily thereafter, returning to near-basal levels by 48 h. The rates of 2-MeOE₂ (Fig. 1E) and 4-MeOE₂ (Fig. 1F) synthetic activities, reflecting the activities of CYP1A1 and CYP1B1, respectively, also showed transient elevations after exposure to 3 µM BAP. Although the data for 2- and 4-MeOE₂ formation are very precise, they should be seen as representing mean velocities, averaged over 6-h periods. Uncertainty of the exact nature of the time course of the induction of the activities of CYP1A1 and CYP1B1 arises from probable changes in metabolic velocities during these 6-h assay periods. Therefore, theoretical limits for the rates of change of the metabolic velocities at the beginning and the end of each assay period were used to approximate ranges of possible curves that describe the time course of induction and decline of E₂ metabolism. These ranges are depicted by the shaded areas within the curves (Fig. 1, E and F). From these curves it is estimated that the rates of 2- and 4-MeOE₂ formation reached maxima by 9 to 16 h, and by 48 h, the rates of 2- and 4-MeOE₂ synthetic activities had returned to near-basal levels.

The concentration-response effects of a 12-h exposure to BAP on the 2- and 4-hydroxylation pathways of E₂ metabolism in T-47D cells were evaluated (Fig. 2). The induction curve is bell shaped, with the highest metabolic rates observed from exposure to 3 µM BAP, and lower metabolic rates were observed with BAP concentrations lower and higher than 3 µM. Because this type of concentration-response curve often reflects enzyme inhibition by the higher levels of the inducing agent, the effects of BAP on E₂ metabolism induced by TCDD were investigated. Addition of BAP to TCDD-treated T-47D cultures during the E₂ metabolism assay period markedly inhibited metabolite formation when the BAP was present at 3 and 10 µM (Fig. 3). At 10 µM BAP, the rates of TCDD-induced 2- and 4-MeOE₂ formation were both inhibited by more than 90%. In contrast to the concentration-response curves for formation of 2- and 4-MeOE₂, concentration-response curves for CYP1A1 and CYP1B1 mRNA levels are sigmoidal, with no apparent inhibitory effects at the higher BAP levels (Fig. 4). From these data, EC₅₀ values of 1.75 ± 0.07 and 1.88 ± 0.16 µM BAP were determined for the induction of the CYP1A1 and CYP1B1 mRNAs, respectively.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Effect of varying BAP concentration on the 2- and 4-hydroxylation pathways of E2 metabolism in T-47D cells. Confluent cultures of T-47D cells were exposed for 12 h to medium containing the indicated concentrations of BAP, after which media were replaced with media containing 1 µM E2. After a further 6 h, media were recovered, conjugates were hydrolyzed, and levels of 2-MeOE2 () and 4-MeOE2 () were determined. Rates of metabolite formation are expressed relative to cellular protein content. Data are the means ± standard error of triplicate determinations.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Effect of varying BAP concentration on TCDD-induced estrogen metabolism in T-47D cells. Cultures were exposed to 10 nM TCDD or 0.1% (v/v) DMSO for 72 h. Media were then replaced with media containing 1 µM E2 and the indicated concentration of BAP. After a further 6 h, media were recovered, conjugates were hydrolyzed, and levels of 2-MeOE2 () and 4-MeOE2 () were determined. Rates of metabolite formation are expressed relative to cellular protein content. Data are the means ± standard error of triplicate determinations.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Effect of varying BAP concentrations on the levels of CYP1A1 and CYP1B1 mRNAs in T-47D cells. Confluent cultures of T-47D cells were exposed for 12 h to medium containing the indicated concentrations of BAP, after which RNA was extracted and reverse transcribed, and CYP1A1, CYP1B1, and GAPDH sequences were amplified by competitive PCR. Data are presented as the ratios of CYP1A1 (A) or CYP1B1 (B) mRNA levels to those of GAPDH mRNA and are expressed relative to the DMSO control. Data are the means ± standard error of triplicate determinations.

To examine the effects of NaAsO₂ on the induction of E₂ metabolism in T-47D cells, the compound was added to cultures at various concentrations in combination with 3 µM BAP. NaAsO₂ at 1, 3, or 10 µM added in the presence of 3 µM BAP did not affect cell viability as determined by the MTT assay (Fig. 5A). These levels of NaAsO₂ in combination with BAP did not significantly affect the levels of induced CYP1A1 (Fig. 5B) or CYP1B1 (Fig. 5C) mRNAs. Exposure of T-47D cells to NaAsO₂ in combination with 3 µM BAP had marked effects on the formation of 2- and 4-MeOE₂, reflecting decreases in the activities of CYP1A1 and CYP1B1, respectively. Exposure to 10 µM NaAsO₂ in combination with 3 µM BAP had very similar effects on the CYP1A1- and CYP1B1-catalyzed activities because the rates of 2-MeOE₂ (Fig. 5D) and 4-MeOE₂ (Fig. 5E) production were both significantly reduced at 1 µM NaAsO₂ and here reduced by 87 and 91%, respectively, at 10 µM NaAsO₂ compared with that elicited by 3 µM BAP alone.

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Effects of BAP plus varying concentrations of NaAsO2 on cell viability, the induction of CYP1A1 and CYP1B1 mRNAs, and estrogen metabolism in T-47D cells. Confluent cultures of T-47D cells were exposed to medium containing the DMSO vehicle or 3 µmM BAP plus the indicated concentrations of NaAsO2 for 12 h. A, cell viability as assessed by the MTT assay. B, levels of CYP1A1 mRNA expressed relative to those of GAPDH. C, levels of CYP1B1 mRNA expressed relative to those of GAPDH. D, rates of 2-MeOE2 formation, determined in a 6-h assay, expressed relative to cellular protein content. E, rates of 4-MeOE2 formation determined in a 6-h assay, expressed relative to cellular protein content. Data are the means ± standard errors of six (A) or three (B-E) determinations. Significant differences from control (*, p < 0.05; **, p < 0.01; ***, p < 0.001) and from the 3 µM BAP treatment group (, p < 0.05; , p < 0.01; , p < 0.001) are indicated.

Direct effects of BAP and NaAsO₂ on CYP1A1 and CYP1B1 activities were investigated in microsomal reactions with cDNA-expressed enzymes. Consistent with the effects of added BAP on the activities of CYP1A1 and CYP1B1 induced by TCDD in T-47D cells (Fig. 3), BAP potently inhibited both CYP1A1 and CYP1B1. Addition of 10 µM BAP to microsomal incubations caused 96% inhibition of CYP1A1-catalyzed E₂ 2-hydroxylase activity (Fig. 6A) and 99% inhibition of CYP1B1-catalyzed E₂ 4-hydroxylase activity (Fig. 6B). NaAsO₂ also inhibited the activities of cDNA-expressed CYP1A1 and CYP1B1. Addition of 10 µM NaAsO₂ caused 21 and 43% inhibition of the CYP1A1 and CYP1B1 E₂ hydroxylase activities, respectively.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Effects of BAP and NaAsO2 on E2 hydroxylation catalyzed by human CYP1A1 and CYP1B1. Microsomal incubations containing 10 pmol of cDNA-expressed CYP1A1 (A) or CYP1B1 (B) were preincubated without additions or with 10 µM BAP or 10 µM NaAsO2 for 5 min before a 10-min assay of E2 hydroxylase activity as described under Materials and Methods. Data are the means ± standard error of triplicate determinations and are expressed relative to control. Significant differences from control are indicated (*, p < 0.05; ***, p < 0.001).

Because NaAsO₂ appears to both inhibit CYP1A1 and CYP1B1 activities and to diminish the induction of the two enzymes, an experiment was designed to evaluate the relative importance of these two effects of NaAsO₂ in T-47D cells. TCDD was used as the inducer of CYP1A1 and CYP1B1 to avoid the confounding effects of enzyme inhibition by BAP. As previously reported (Spink et al., 1998b), TCDD caused a marked elevation in the rates of formation of 2-MeOE₂, 4-MeOE₂, 6-OHE₂, and 15-OHE₂ as a consequence of the induction of CYP1A1 and CYP1B1 (Fig. 7). Addition of 10 µM NaAsO₂ together with 10 nM TCDD during the 12-h period for enzyme induction resulted in significant reductions in the rates of E₂ metabolism determined during the assay phase. Not only were rates of 2-MeOE₂ and 4-MeOE₂ reduced, but also rates of formation of 6-OHE₂, and 15-OHE₂, two additional products of CYP1A1-catalyzed E₂ metabolism (Spink et al., 1992), were reduced relative to rates of cultures that received TCDD only. However, if 10 µM NaAsO₂ was added in the E₂ metabolism phase of the experiment rather than the induction phase, no significant changes in the rates of E₂ metabolism were observed, suggesting that the effects of arsenite on E₂ metabolism in T-47D cultures were primarily on the levels of TCDD-induced CYP1A1 and CYP1B1 and not as a result of inhibiting their activities.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Effects of NaAsO2 on TCDD-induced metabolism of E2 in T-47D cells. Confluent cultures of T-47D cells were exposed to 10 nM TCDD, 10 nM TCDD plus 10 µM NaAsO2, or received no additions (0.1% DMSO only) as indicated. After a 12-h period for enzyme induction, media were replaced with media containing 5 µM E2 or 5 µM E2 plus 10 µM NaAsO2. After incubation for an additional 3 h, these media were recovered and E2 metabolites were determined as described under Materials and Methods. TCDD and/or NaAsO2 were added in the 12-h induction phase (I) or in the 3-h E2 metabolism phase (A) of the experiment as indicated. The rates of formation of 4-MeOE2 (), 2-MeOE2 (), 6-OHE2 (), and 15-OHE2 () were as indicated. Significant differences from control (*, p < 0.05; ***, p < 0.001) and from the10 nM TCDD (I) treatment group (, p < 0.001) are indicated.

Reductions in the levels of the CYP1A1 and CYP1B1 proteins after cotreatment with arsenite would explain the reduction in the BAP- and TCDD-induced E₂ metabolism in T-47D cells. Additionally, reduction in the levels of NADPH-cytochrome P450 reductase could also result in diminished rates of BAP-induced E₂ metabolism. To investigate the effects of cotreatments with BAP and arsenite on the levels of CPR and CYP1A1, Western immunoblot analyses of the two proteins in cell lysates were performed. Exposure to 3 µM BAP did not significantly affect the level of CPR in T-47D cells, nor did combined treatments of 1, 3, or 10 µM NaAsO₂ together with 3 µM BAP (Fig. 8A), but these treatments had marked effects on the levels of CYP1A1 protein (Fig. 8B). In untreated T-47D cells, CYP1A1 protein was not detectable. After exposure to 3 µM BAP, CYP1A1 was highly induced, and the combined treatments of 3 or 10 µM NaAsO₂ together with 3 µM BAP resulted in significantly reduced levels of immunoreactive CYP1A1 compared with exposure to 3 µM BAP alone. Western immunoblot analyses of CYP1B1 expression in T-47D cells were performed in a similar manner to those for CYP1A1 and CPR. Although these analyses were suggestive of a similar effect of arsenite in reducing the BAP-induced level of CYP1B1, the low levels of CYP1B1 expressed in T-47D cells did not allow quantitative measurements of CYP1B1 expression with the available antibody (data not shown). Thus, arsenite was shown to diminish the activities of both CYP1A1 and CYP1B1 in cellular E₂ metabolism studies (Figs. 5, D and E, and 7); however, the effect of arsenite on reducing immunoreactive P450 protein was only clearly demonstrated for CYP1A1 (Fig. 8B).

View larger version (27K): [in this window] [in a new window]   Fig. 8.   Effects of BAP and NaAsO2 on the levels of CPR and CYP1A1 in T-47D cells. Confluent cultures of T-47D cells were exposed to medium containing the indicated concentrations of BAP and NaAsO2 for 12 h. Cellular proteins were then solubilized and subjected to electrophoresis and Western immunoblotting for CPR (A) and CYP1A1 (B) with chemiluminescence detection and quantitation by scanning densitometry. A chemiluminescence image of a representative CPR Western immunoblot with 20, 13, 9.8, 6.5, 2.6, 1.3, and 0 fmol of purified human recombinant human CPR analyzed in lanes 1 to 7, respectively, and samples from T-47D cultures exposed to the DMSO vehicle only in lane 8, 3 µM BAP in lane 9, or 3 µM BAP plus 1, 3, or 10 µM NaAsO2 in lanes 10 to 12, respectively, is shown (A, inset). The bar graph (A) shows mean levels (± standard error) of CPR in T-47D cell extracts expressed relative to total cellular protein content of three separate determinations. A chemiluminescence image of a representative CYP1A1 Western immunoblot with 2, 1, 0.75, 0.50, 0.25, 0.10, and 0 fmol of cDNA-expressed human CYP1A1 analyzed in lanes 1 to 7, respectively, and samples from T-47D cultures exposed to the DMSO vehicle only in lane 8, 3 µM BAP in lane 9, or 3 µM BAP plus 1, 3, or 10 µM NaAsO2 in lanes 10 to 12, respectively, is shown (B, inset). The bar graph (B) shows mean levels (± standard error) of CYP1A1 in T-47D cell extracts expressed relative to total cellular protein content of three separate determinations. ND indicates that CYP1A1 was not detected [<0.005 pmol(mg protein)1]. Significant differences from treatment with 3 µM BAP alone (, p < 0.05) are indicated.

Because the induction of HO-1 has been observed in response to arsenite exposure in vivo (Falkner et al., 1993) and in vitro (Jacobs et al., 1999), concomitant with the reduction of P450 activities, we investigated whether HO-1 mRNA was induced under the conditions of the experiment in Fig. 5, in which rates of CYP1A1- and CYP1B1-catalyzed E₂ metabolism were diminished by added NaAsO₂. Real-time PCR was used to quantitate the levels of HO-1 mRNA in T-47D cultures exposed to 3 µM BAP or to 1, 3, or 10 µM NaAsO₂ together with 3 µM BAP. Exposure to 3 µM BAP alone caused a 3.3-fold elevation in HO-1 mRNA (Fig. 9). Coexposure to 1, 3, or 10 µM NaAsO₂ together with 3 µM BAP caused 8.1-, 66-, and 155-fold elevations in the level of HO-1 mRNA in T-47D cells.

View larger version (14K): [in this window] [in a new window]   Fig. 9.   Effects of BAP and NaAsO2 on HO-1 mRNA levels in T-47D cells. Confluent cultures of T-47D cells were exposed to medium containing the DMSO vehicle or 3 µM BAP plus the indicated concentrations of NaAsO2 for 12 h. RNA was then isolated, reverse transcribed, and aliquots of the cDNAs were subjected to real-time PCR with primers specific for human HO-1. Levels of HO-1 mRNA are expressed relative to total cellular RNA content. Data are the means ± standard error of three determinations. Significant differences from control (*, p < 0.05) are indicated.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The ability of T-47D cells to rapidly metabolize BAP was initially reported by Merrick et al. (1985), who found that 95% of the BAP initially present at 4 µM was metabolized within 24 h. Dihydrodiols, tetraols, quinones, and sulfated compounds were detected as BAP metabolites, as were DNA adducts (Pruess-Schwartz et al., 1986). Our results showed a similar rapid uptake and metabolism of BAP in T-47D cells concomitant with induction of CYP1A1 and CYP1B1, which is most likely responsible for the BAP metabolism. Comparison of the time courses of the BAP levels and the levels of CYP1A1- and CYP1B1-catalyzed activities indicates that the activities of the two enzymes are dependent on the presence of unmetabolized BAP. Likewise, the P450 mRNA levels were dependent on the presence of BAP, suggesting a short (i.e., 1-3 h) half-life for the CYP1A1 and CYP1B1 mRNAs after metabolic removal of the BAP inducer. The rapid turnover of the mRNAs and activities in T-47D cells after exposure to BAP suggests that these cells may provide a system to investigate P450 mRNA and protein catabolism. In recent studies with HepG2 hepatoma cells, the CYP1A1 mRNA had a similar rapid turnover, with a reported half-life of 2.4 h (Lekas et al., 2000).

The kinetics and concentration-response relationships of BAP-mediated induction of CYP1A1 and CYP1B1 are complex because PAHs are inducers of, substrates for, and inhibitors of CYP1A1 and CYP1B1 (Shimada et al., 1998; Willett et al., 1998; Arcaro et al., 1999). The enzyme inhibitory effects result in bell-shaped concentration-response curves for apparent enzyme induction in vitro (Willett et al., 1998; Arcaro et al., 1999). The shapes of concentration-response curves vary markedly with time of incubation (Arcaro et al., 1999). This has been attributed to the removal of enzyme inducers and enzyme inhibitors through metabolism. These effects make estimation of toxic equivalency factors for PAHs by measurement of induced enzyme activities difficult. Determination of the effects of BAP and other PAHs on CYP1A1 and CYP1B1 levels by measurement of enzyme activities is only of value at low concentrations of BAP, where its inhibitory effects are minimal.

Coexposure of T-47D cells to NaAsO₂ and BAP resulted in a marked diminution of the activities of CYP1A1 and CYP1B1 relative to those observed with BAP alone. These effects were observed at arsenite concentrations below those that affected cell viability. Although both CYP1A1- and CYP1B1-catalyzed E₂ hydroxylase activities in microsomal incubations were inhibited by arsenite, the magnitude of inhibition was not sufficient to account for the reduction of BAP-induced CYP1A1- and CYP1B1-catalyzed activities in T-47D cells by arsenite. This was most evident for CYP1A1 because addition of 10 µM NaAsO₂ inhibited CYP1A1-catalyzed E₂ 2-hydroxylation in microsomal assays by only 21% (Fig. 6A), whereas addition of 10 µM NaAsO₂ to the culture medium caused an 86% reduction in the rate of the BAP-induced E₂ 2-hydroxylation pathway (Fig. 5D). Further evidence that the effect of arsenite on CYP1A1 and CYP1B1 activities was on the levels of active enzyme rather than a direct inhibitory effect or modulation of the inhibitory effect of BAP was obtained in the experiments with TCDD. In these experiments, arsenite was only effective in the induction phase of the experiment and not during the E₂ metabolism assay phase (Fig. 7). The reduction in the levels of immunoreactive CYP1A1 observed on Western blots correlated well with the reduction in 2-MeOE₂ formation in response to coexposure to BAP and NaAsO₂. Because the levels of CPR were not affected by these coexposures (Fig. 8A), the effects of NaAsO₂ on P450s do not appear to be reflective of a general effect on cellular proteins.

Arsenite has been previously shown to reduce the levels of hepatic P450s of the CYP1A (Falkner et al., 1993; Jacobs et al., 1999; Vakharia et al., 2001), CYP2B (Jacobs et al., 1999), and CYP3A families (Jacobs et al., 1999). The results reported here indicate that the effect of arsenite on the expression of cytochromes P450 is retained in a transformed, immortalized cell line derived from a human breast carcinoma. The fact that arsenite diminishes expression of P450s of several gene families suggests a general mechanism. One mechanism that has been offered to explain the effect is the induction of heme oxygenase by arsenite (Sardana et al., 1981; Falkner et al., 1993), which would reduce the level of active P450. Several heavy metal species are known to induce HO-1; arsenite is particularly potent in this regard. Results reported here show a marked induction of HO-1 mRNA in T-47D human breast cells coexposed to arsenite and BAP. The induction of HO-1 by arsenite and the subsequent reduction in the availability of heme as the prosthetic group for P450s may explain the reduced P450 activities in T-47D cells, hepatocyte cultures, and in vivo. However, recent studies by Jacobs et al. (1999) suggest that induction of HO-1 may not completely explain the effect of arsenite on P450 expression. HO-1 is also induced by heme in hepatocyte cultures, but treatment with heme did not result in reduced P450 levels, whereas cotreatment of hepatocyte cultures with arsenite and heme diminished levels of CYP2B1/2 (Jacobs et al., 1999). The authors concluded that a mechanism in addition to the induction of HO-1 must be involved. Further studies are underway to resolve the role of HO-1 in the effects of arsenite reported here.

The observation that the induction of CYP1A1- and CYP1B1-catalyzed activities by BAP are markedly diminished by cotreatment with arsenite, but the levels of the mRNAs encoding CYP1A1 and CYP1B1 are not significantly affected indicates that the primary effects of arsenite are not at the level of transcription of the two enzymes or mediated by effects on mRNA stability or catabolism. However, these results do not rule out possible effects of arsenite on translation. Effects of arsenite at the level of mRNA translation have been reported, although they are thought to result in increased rather than decreased rates of translation. Arsenite and other stress inducers increase the level of phosphorylation of initiation factor eIF4E, a protein complex that binds the 7-methylguanosine cap structure of mRNAs. Phosphorylation of eIF4E increases its affinity for the mRNA cap structure, which increases the rate of activation of translation (Wang et al., 1998). We are not aware of any reports indicating that phosphorylation of eIF4E decreases the rates of translation of specific mRNAs.

Another point at which arsenite might affect the rates of CYP1A1- and CYP1B1-catalyzed activities is the rate of protein turnover. Our results show rapid decline in the rates of the BAP-induced E₂ 2- and 4-hydroxylation pathways after metabolic removal of the BAP, suggesting rapid degradation of CYP1A1 and CYP1B1. Although there are no data on P450 turnover in T-47D cells, rates of P450 turnover in rat hepatocytes were found to differ among the P450s. Proteasome-mediated degradation was demonstrated for CYP1A2, CYP2E1, CYP3A, and CYP4A (Roberts, 1997), but appears to require prior labilization of the P450s because it can be stimulated in cell-free experiments by processes such as incubation of the P450 with a suicide substrate or denaturation by freezing and thawing. Arsenite is known to react with cysteine residues of some proteins, resulting in loss of function (Kapahi et al., 2000). Whether proteasome-mediated degradation is involved in the turnover of human CYP1A1 and CYP1B1 in extrahepatic cells and whether arsenite might affect this process by influencing factors such as protein conformation and heme incorporation are unknown and will be investigated in future studies.

Although the underlying mechanism is not fully resolved, our results suggest that coexposure to arsenite in combination with PAHs may lessen the rate of PAH metabolism by reducing the induced levels and thus activities of both CYP1A1 and CYP1B1. This reduction in P450 activity may be a consequence of cross talk among stress-response signaling pathways that are activated to limit cellular damage caused by exposure to the toxicants. The diminution of CYP1A1 and CYP1B1 activities by arsenite in cells of extrahepatic origin may actually lower the rate of conversion of PAH to DNA-binding and mutagenic species in a coexposure situation. Due to the rapid metabolism of BAP, it is difficult to compare levels of BAP that induce CYP1A1 and CYP1B1 in cultured cells in vitro with those that might occur in humans. The level of arsenite that elicits significant effects in T-47D cells (1 µM, or 134 µg/l NaAsO₂) appears to be within the range that could occur in humans. Blood levels of 8 to 15 µg/l (Concha et al., 1998) and urinary levels of 133 to1893 µg/l (Hopenhayn-Rich et al., 1996) of total arsenic have been reported in exposed individuals. Cellular levels of arsenic may exceed those in circulation. This PAH-arsenite interaction is likely to be one of many such interactions resulting from exposure to complex environmental mixtures that may affect carcinogenesis.