Inhibition of Carcinogen-Activating Enzymes by 16␣-Fluoro-5-Androsten-17-One

[CANCER RESEARCH 62, 3685–3690, July 1, 2002] Inhibition of Carcinogen-activating Enzymes by 16␣-Fluoro-5-androsten-17-one

Henry P. Ciolino,1 Christopher J. MacDonald, and Grace Chao Yeh Cellular Defense and Carcinogenesis Section, Basic Research Laboratory, Center for Cancer Research, National Cancer Institute at Frederick, NIH, Frederick, Maryland 21702- 1201

ABSTRACT decrease in the incidence of a number of different types of cancer in humans (5, 6). In animal models, DHEA has been shown to inhibit In the present study, we examined the effect of a synthetic analogue of both spontaneous and chemically induced carcinogenesis in rodents the chemopreventive hormone dehydroepiandrosterone, 16␣-fluoro-5- (7–9). Specifically, DHEA inhibits both skin and mammary tumori- androsten-17-one, also known as fluasterone, on the activity and expression of carcinogen-activating enzymes in MCF-7 cells. The increase in genesis caused by the PAH DMBA (10–13). DHEA has been shown cytochrome P450 (CYP) 1A1 and 1B1 activity, as measured by to inhibit DMBA activation in vitro (14) and DMBA-DNA binding in ethoxyresorufin-O-deethylase activity, in cells treated with the carcino- vivo (15–18). The dramatic decline of DHEA levels in humans with gens dimethylbenzanthracene (DMBA) or 2,3,5,7-tetrachlorodibenzo-p- advancing age is, therefore, of substantial concern. However, DHEA dioxin (TCDD), was inhibited by cotreatment with fluasterone. However, supplementation is problematic, because it can be converted to both treatment of the cells with fluasterone after induction with DMBA or testosterone and estrone (19), and it exhibits considerable liver tox- TCDD failed to decrease enzyme activity, indicating that inhibition was icity (20) and hepatocarcinogenicity (21). Thus, DHEA’s clinical not the result of direct enzyme inhibition. Therefore, we examined the usefulness is limited. Recently, a synthetic analogue of DHEA, 16␣- effect of fluasterone on gene expression at the mRNA level. Both DMBA fluoro-5-androsten-17-one (fluasterone), was developed that lacks the and TCDD caused a dramatic increase in the amount of CYP1A1 and liver toxicity and hormone-related side effects of DHEA (22). Fluas- CYP1B1 mRNA, the two major isoforms involved in carcinogen activation in these cells. In cells cotreated with fluasterone, however, there was a terone has been shown to be chemopreventive against DMBA- dose-dependent decrease in CYP1A1 and CYP1B1 mRNA. Fluasterone induced carcinogenesis (23), and, like DHEA, it prevents DMBA also inhibited the basal level of CYP1A1 mRNA but not CYP1B1. Fluas- activation and DMBA-DNA adduct formation (10, 22). However, the terone inhibited the rate of CYP1A1 promoter-controlled transcription, mechanism of this activity is unknown. Therefore, we have tested the indicating that it affects the transcriptional regulation of the gene. Acti- capacity of fluasterone to modulate the effects of DMBA and TCDD nomycin D chase experiments showed that fluasterone also caused an on carcinogen-activating enzyme activity and expression in vitro. increase in the degradation of CYP1A1 mRNA, while leaving CYP1B1 DMBA is a classic model aryl hydrocarbon. TCDD is a widespread mRNA unaffected. These results indicate that fluasterone inhibits the environmental contaminant produced during trash incineration, increase in the expression of CYP1A1 normally caused by exposure to bleaching of paper pulp, synthesis of pesticides, and as a by-product carcinogens by both transcriptional and post-transcriptional mechanisms of combustion during various industrial processes (24, 25). and that CYP1B1 expression is not susceptible to the same post-transcriptional mechanism. MATERIALS AND METHODS

INTRODUCTION Materials. Human breast cancer MCF-7 and human liver carcinoma HepG2 cells were from the American Type Culture Collection (Rockville, Many environmental compounds are carcinogenic only after met- 2 MD). RPMI 1640, glutamine, fetal bovine serum, trypsin/EDTA, and PBS abolic activation. Exposure to carcinogens, such as PAH, causes an were from BioFluids (Rockville, MD). Actinomycin D, ␣-NF, DMBA, EDTA, increase in the expression of the enzymes responsible for this activa- ethoxyresorufin, resorufin, Tris-HCl, and DMSO were from Sigma (St. Louis, tion. These enzymes consist of members of the CYP 1A and 1B MO). [32P]dATP was from DuPont NEN (Boston, MA). TCDD was from the subfamilies. They generate genotoxic epoxide metabolites of the Midwest Research Institute (Kansas City, MO). RT-PCR was performed with parent aryl hydrocarbon, which can bind DNA, forming adducts (1). an Omniscript kit from Qiagen (Valencia, CA). Tris/borate/EDTA gels, run- These adducts, if not repaired, can cause specific mutations leading to ning buffer, and high-density sample buffer were from Novex (San Diego, cellular transformation. Therefore, the activity and expression of CA). Primers for GPDH PCR and the ␤-galactosidase-containing reporter carcinogen-activating enzymes are key components in chemically vector were from Clontech (Palo Alto, CA). TRIzol reagent and Lipo- induced carcinogenesis, and the inhibition of their activity, either by fectAmine were from Life Technologies, Inc. (Gaithersburg, MD). CAT ELISA assay kit was from Boehringer Mannheim (Indianapolis, IN). Fluas- direct enzyme inhibition or through modulation of their expression, is terone was a gift from Dr. Thomas Wang (Phytonutrients Laboratory, USDA, thought to be an important mechanism in the prevention of carcino- Beltsville, MD). genesis (2). Cell Culture. MCF-7 and HepG2 were grown in RPMI 1640 with 2 mM DHEA is the most abundant steroid hormone in humans (3). Al- glutamine and 10% fetal bovine serum and subcultured weekly using 0.25% though its physiological function remains unclear, it is associated with trypsin/0.05% EDTA. All experiments were carried out on confluent cultures a number of beneficial health effects in humans (4). A considerable in growth medium, unless otherwise noted. body of evidence indicates that DHEA is also associated with a Assay of EROD Activity. Confluent MCF-7 or HepG2 cells in 24-well plates were treated with 1 ml of growth medium containing 1 ␮M DMBA or 1nM TCDD in the presence of DMSO (vehicle control) or fluasterone for 24 h. Received 1/11/02; accepted 5/2/02. The costs of publication of this article were defrayed in part by the payment of page The final DMSO concentration in both control and treated cultures for this and charges. This article must therefore be hereby marked advertisement in accordance with all other experiments was 0.1%. The medium was removed, and the wells were 18 U.S.C. Section 1734 solely to indicate this fact. washed two times with fresh growth medium. EROD activity was determined 1 To whom requests for reprints should be addressed, at Basic Research Laboratory, in intact cells using 1 ␮M ethoxyresorufin in growth medium as a substrate in Building 560/Room 12-05, National Cancer Institute at Frederick, NIH, Frederick, MD 21702-1201. Phone: (301) 846-5160; Fax: (301) 846-6709; E-mail: [email protected]. the presence of 1.5 mM salicylamide to inhibit conjugating enzymes. The assay 2 The abbreviations used are: PAH, polycyclic aromatic hydrocarbon; CAT, chloram- was carried out at 37°C. The fluorescence of resorufin generated from the phenicol acetyltransferase; CYP, cytochrome P450; DHEA, dehydroepiandrosterone; conversion of ethoxyresorufin by CYP1A1/CYP1B1 was measured every 10 DMBA, dimethylbenzanthracene; EROD, ethoxyresorufin-O-deethylase; G6PDH, min for 60 min with a CytoFluor II multiwell fluorescence plate reader glucose-6-phosphate dehydrogenase; GPDH, glyceraldehyde-3-phosphate dehydrogenase; ␣-NF, ␣-naphthoflavone; RT-PCR, reverse transcription-PCR; TCDD, 2,3,5,7-tetra- (PerSeptive Biosystems, Framingham, MA), with an excitation wavelength of chlorodibenzo-p-dioxin; XRE, xenobiotic-responsive element. 530 nm and emission at 590 nm. 3685

Statistical Analysis. Statistical analyses were performed using StatView Statistical Analysis software (SAS Institute). Differences between group mean values were determined by a one-factor ANOVA, followed by Fisher’s protected least-significant difference post hoc analysis for pairwise comparison of means.

RESULTS Fluasterone Inhibits CYP1A1/1B1 Enzyme Activity in Intact Cells. Carcinogen-activating enzyme activity was measured in intact MCF-7 cells by the EROD assay, which is specific for CYP1A1/1B1. The treatment of MCF-7 cells with 1 ␮M DMBA for 24 h resulted in an increase in activity from undetectable levels to 1.38 Ϯ 0.13 pmol/min/105 cells. In cells coincubated with DMBA and fluasterone,

there was a dose-dependent decrease in EROD activity, with a IC50 of ϳ0.5 ␮M (Fig. 1A). In cells treated with 1 nM TCDD, the most potent inducer of these enzymes, there was an increase in EROD activity to 16.72 Ϯ 0.43 pmol/min/105 cells. This was also decreased in a

dose-dependent fashion by coincubation with fluasterone, with a IC50 of ϳ2.5 ␮M (Fig. 1B). Fluasterone and DHEA also inhibited TCDD- induced EROD activity in human liver HepG2 cells (Fig. 1C). MCF-7 cells were preincubated with DMBA or TCDD and postin- cubated for 3 h with fluasterone. As shown in Table 1, ␣-NF, a known inhibitor of EROD activity, caused a significant decrease in activity, but fluasterone had no effect. In microsomes isolated from cells that had been incubated with TCDD to induce activity, there was likewise no effect of fluasterone on EROD activity, whereas ␣-NF completely abolished activity. Fluasterone Inhibits CYP1A1 and CYP1B1 mRNA Levels. MCF-7 cells were treated with DMBA or TCDD with or without fluasterone for 6 h, and the amount of CYP1A1 mRNA was determined by RT-PCR. As shown in Fig. 2A, there was a 4-fold increase in CYP1A1 mRNA in DMBA-treated cells compared with DMSO- treated cells. In cells coincubated with fluasterone, this increase was Fig. 1. Effect of fluasterone on cellular EROD activity induced by DMBA (A), TCDD significantly inhibited at all doses tested. In cells treated with TCDD, ␮ (B), or TCDD in HepG2 cells (C). MCF-7 or HepG2 cells were incubated with 1 M there was a 9-fold increase in CYP1A1 mRNA. This increase was also DMBA or 1 nM TCDD for 24 h in the presence of the indicated concentrations of fluasterone, and cellular EROD activity was determined. n ϭ 4 Ϯ SE. There was a inhibited by coincubation with fluasterone in a concentration-depen- significant decrease of EROD activity in the presence of Ն0.25 ␮M fluasterone in A and dent manner (Fig. 2B). Ն1 ␮M in B (P Ͻ 0.05). The other major carcinogen-activating enzyme expressed in MCF-7 cells is CYP1B1, which is also induced by carcinogen exposure. DMBA caused a 2.4-fold increase in CYP1B1 mRNA (Fig. 3A), To determine the direct effect of fluasterone on EROD activity, MCF-7 cells whereas TCDD increased CYP1B1 mRNA by 4.5-fold (Fig. 3B). ␮ were incubated with 1 M DMBA or 1 nM TCDD for 24 h to induce enzyme Coincubation with fluasterone also inhibited this induction, albeit less expression. The cells were then washed extensively and incubated with DMSO dramatically than for CYP1A1. (vehicle control), 10 ␮M fluasterone, or 5 ␮M ␣-NF as a positive control for 3 h, and EROD activity was determined. In addition, the effect of fluasterone Unlike enzyme activity, which is not detectable in uninduced cells, on EROD activity in microsomes isolated from TCDD-treated MCF-7 cells was measured as described (26). RT-PCR. MCF-7 cells were grown in six-well plates and treated with Table 1 Lack of a direct inhibitory effect of fluasterone on EROD activity DMBA (1 ␮M) or TCDD (1 nM) in the presence of DMSO (control) or A. Cellular activitya fluasterone for 6 h. Isolation of total RNA; cDNA synthesis; semiquantitative RT-PCR for CYP1A1, CYP1B1, and GPDH mRNA; and analysis of results Inducer Competitor Activity (pmol/min/105 cells) were performed as described previously (27). Primer sequences for CYP1A1 DMBA None (DMSO) 0.79 Ϯ 0.08 and CYP1B1 were described in Dohr et al. (28). cDNA was synthesized from ϩ␣-NF (1 ␮M) 0.27 Ϯ 0.04 ϩ ␮ Ϯ 2 ␮g of total RNA using a Omniscript RT-PCR kit as instructed. A cycle Fluasterone (10 M) 0.70 0.04 TCDD None (DMSO) 14.77 Ϯ 0.60 number that fell within the linear range of response for CYP1A1 (27 cycles for ϩ␣-NF 6.38 Ϯ 0.33 the determination of CYP1A1 mRNA stability and basal mRNA levels, 24 ϩFluasterone 16.71 Ϯ 0.39 cycles otherwise), CYP1B1 (24 cycles for mRNA stability and basal mRNA B. Microsomal activity levels, 22 cycles otherwise), and GPDH (17 cycles) was used. Transient Transfections. XRE-controlled CAT transcription was deter- Competitor Activity (fmol/min/10 ␮g) mined as described previously (29). None (DMSO) 290 Ϯ 20 Determination of mRNA Stability. MCF-7 cells were pretreated with 1 ␣-NF (1 ␮M) Not detectable Fluasterone (10 ␮M) 230 Ϯ 20 ␮M DMBA for 24 h to induce CYP1A1 and CYP1B1 expression. The cells a were then washed extensively and incubated with media containing 5 ␮g/ml MCF-7 cells were incubated with DMBA (1 ␮M) or TCDD (1 nM) for 24 h before the addition of ␣-NF or fluasterone, and the EROD activity in intact cells was determined. the transcription inhibitor actinomycin D in the presence of DMSO (control) or b MCF-7 cells were treated with 1 nM TCDD, and microsomes were isolated. Micro- fluasterone. After 4 h, the total RNA was isolated, and RT-PCR for CYP1A1 somes (10 ␮g) were incubated with ␣-NF or fluasterone, and the EROD activity was and CYP1B1 was performed as described above. determined. For both A. and B., n ϭ 4 Ϯ SE. 3686

Fluasterone Decreases the Stability of CYP1A1 but not CYP1B1 mRNA. The effect of fluasterone on the stability of CYP1A1 and CYP1B1 mRNA was examined. Expression was induced with DMBA, after which, additional transcription was inhibited by the addition of the RNA synthesis inhibitor actinomycin D. Cells were treated with DMSO or fluasterone for 4 h, and the amount of mRNA was determined by RT-PCR. There was a concentration- dependent decrease in the amount of CYP1A1 mRNA in cells incubated with fluasterone but no change in the amount of CYP1B1 mRNA (Fig. 6).

DISCUSSION The chemopreventive activity of the steroid hormone DHEA is believed to result from its inhibitory action on G6PDH, the rate-

Fig. 2. Effect of fluasterone on CYP1A1 mRNA induced by DMBA (A) or TCDD (B). MCF-7 cells were treated for 6 h with DMSO (Control), 1 ␮M DMBA, or 1 nM TCDD in the presence of DMSO or fluasterone at the indicated concentrations. Total RNA was isolated, cDNA was synthesized, and the amount of mRNA was determined by PCR. The results were visualized and quantified by phosphoimaging. For graph, the levels of CYP1A1 were normalized to GPDH level. n ϭ 3 Ϯ SE. There was a significant decrease in CYP1A1 mRNA in cells treated with fluasterone (P Ͻ 0.05). there is a small but measurable expression of CYP1A1 mRNA in untreated cells. This was measured by RT-PCR by slightly increasing the cycle number used. As seen in Fig. 4, there was a significant inhibition of the basal level of CYP1A1 mRNA in MCF-7 cells incubated with fluasterone. Fluasterone did not affect the basal expression of CYP1B1 (Fig. 4). Fluasterone Affects the Rate of XRE-controlled Transcription. Treatment of MCF-7 cells with TCDD or DMBA caused a 4.4- or 2-fold increase, respectively, in the rate of transcription of a CAT reporter vector controlled by the XRE (Fig. 5A). Cotreatment with fluasterone decreased the rate of transcription in a dose-dependent Fig. 3. Effect of fluasterone on CYP1B1 mRNA induced by DMBA (A) or TCDD (B). Cells were treated and analyzed for CYP1B1 mRNA as described in the legend for Fig. manner. Treatment of transfected cells with fluasterone also inhibited 2. For graph, the level of CYP1B1 was normalized to GPDH level. n ϭ 3 Ϯ SE. There the basal rate of transcription (Fig. 5B). was no significant decrease in CYP1B1 in cells treated with fluasterone. 3687

induced EROD activity (Fig. 1C), although both compounds have higher

IC50s in HepG2 cells than in MCF-7 cells. This may be attributable to the presence of sulfating enzymes in cells of liver origin, such as HepG2 cells, because the sulfated form of DHEA does not inhibit EROD activity (33). Thus, the relative effectiveness of both DHEA and fluasterone may be cell type dependent. When fluasterone was added to MCF-7 cells that had been pretreated with either DMBA or TCDD, it did not inhibit EROD activity (Table 1). Nor could it inhibit the EROD activity present in microsomes isolated from treated cells. Table 1 indicates that fluasterone, like DHEA, does not directly inhibit CYP enzyme activity, as do other compounds, such as ␣-NF. Therefore, we investigated the effect of fluasterone on CYP expression. DMBA or TCDD treatment of cells causes an increase in the levels of CYP1A1 mRNA, as determined by RT-PCR (Fig. 2). Cotreatment with fluasterone inhibited this increase. As for EROD activity, fluasterone was more effective in inhibiting DMBA-induced CYP1A1 mRNA than in inhibiting TCDD-induced mRNA. This may be because TCDD, being a more potent inducer of CYP1A1 expression, causes much higher amounts of CYP1A1. We also measured the mRNA of the other major carcinogen-activating enzyme in MCF-7 cells, CYP1B1. The level of induction of CYP1B1 mRNA by DMBA or TCDD (Fig. 3) was not as profound as CYP1A1. Cotreatment with Fig. 4. Inhibition of basal CYP1A1 mRNA levels by fluasterone. MCF-7 cells were treated with the indicated concentrations of fluasterone for 24 h. CYP1A1, CYP1B1, and fluasterone also inhibited the increase in CYP1B1 mRNA, although GPDH mRNA were determined by RT-PCR. n ϭ 3 Ϯ SE. There was a significant decrease in basal CYP1A1 mRNA in cells treated with all concentrations of fluasterone compared with controls (P Ͻ 0.05) and no significant difference in CYP1B1 mRNA.

limiting enzyme in the pentose-phosphate pathway, which generates NADPH. The resulting depletion of NADPH, which is a cofactor for CYP enzyme activity, would reduce CYP activity and thereby decrease carcinogen activation. However, inhibition of G6PDH activity and NAPDH depletion in vivo by DHEA has not been observed (30–32), and, furthermore, the concentrations of DHEA needed to ␮ inhibit G6PDH in vitro are very high (IC50 of 18.7 M; Ref. 10). Thus, the mechanism of DHEA’s chemopreventive activity toward DMBA is uncertain. Recently, we demonstrated that DHEA inhibits carcinogen-induced CYP1A1 activity and expression by a post-transcriptional mechanism, i.e., by destabilizing CYP1A1 mRNA (33). The utility of DHEA treatment, however, is severely limited by its side effects. An analogue of DHEA, fluasterone, was therefore developed, which does not cause these side effects in animal models. Fluasterone has completed Phase I trials and has been shown to be well tolerated, and it is currently in several Phase II trials (34). Thus, it is important to understand the basic biochemical mechanism(s), whereby it exerts its effects. In the present study, we have examined the effects of fluasterone on CYP enzyme activity and expression in vitro. MCF-7 cells were chosen as a model system because they have been used extensively to study CYP expression (28, 35, 36) and because our previous study on DHEA (33) was carried out in this cell line under the same conditions, allowing direct comparisons of the relative efficacy of fluasterone with DHEA. Treatment of MCF-7 cells with DMBA or TCDD results in a profound increase in CYP enzyme activity, as measured by EROD assay. Both CYP1A1 and CYP1B1 contribute to EROD activity, although the specific activity of CYP1A1 is ϳ10-fold higher than CYP1B1 (37). In cells cotreated with fluasterone, there was a concentration-dependent decrease

in EROD activity (Fig. 1A). The IC50 toward DMBA-induced EROD activity was 0.5 ␮M, which is ϳ5-fold higher than for DHEA (33). The Fig. 5. Fluasterone inhibits DMBA- (A) or TCDD-induced and basal CAT transcription IC for the inhibition of TCDD-induced EROD activity was ϳ2.5 ␮M 50 (B) mediated by the XRE. MCF-7 cells were transfected with a CAT reporter vector (Fig. 1B), compared with 1 ␮M for DHEA. Thus, the parent molecule controlled by the XRE and with a vector containing ␤-Gal. Transfected cells were treated DHEA is a more effective inhibitor of carcinogen-induced EROD activ- with 1 nM TCDD (solid bars)or1␮M DMBA (hatched bars)for6hinthepresence of the indicated concentrations of fluasterone. CAT transcription was normalized to ␤- ity than fluasterone. However, in the HepG2 human hepatic cancer cell galactosidase transcription. n ϭ 4 Ϯ SE. There was a significant difference in transcrip- line, fluasterone is more effective than DHEA in inhibiting TCDD- tion in cells treated with 5 or 10 ␮M fluasterone (P Ͻ 0.05). 3688

effect on CYP1B1 mRNA stability. These results, along with Fig. 4, suggest that there is a fundamental difference in the regulation of CYP1A1 and CYP1B1 mRNA at the post-transcriptional level, although the nature of the mechanism involved is unknown. The com- bination of transcriptional and post-transcriptional inhibition may be reason that fluasterone is more effective in inhibiting CYP1A1 than CYP1B1 expression. These data demonstrate that fluasterone inhibits carcinogen-activating enzyme activity and expression in vitro by both transcriptional and post-transcriptional mechanisms. These results suggest that the chemopreventive properties of fluasterone toward PAH-induced carcinogenesis may be attributable, in part, to its effect on the expression of CYP enzymes. These results are particularly important because fluasterone is now in Phase II clinical trials to measure its effectiveness in vivo. This study does not exclude other chemopreventive mechanisms of action, e.g., fluasterone, in addition to inhibiting the initiation of DMBA-induced mammary and skin tumorigenesis, has also been shown to inhibit 12-O-tetradecanoylphorbol-13-acetate-promoted skin papilloma formation when administered after initiation by DMBA (10), suggesting that it also affects biochemical mechanisms involved in the promotion phase.

Fig. 6. Effect of fluasterone on the stability of CYP1A1 and CYP1B1 mRNA. MCF-7 REFERENCES cells were incubated with 1 ␮M DMBA for 24 h to induce CYP1A1 and CYP1B1 expression, then washed three times in growth medium. The cells were then incubated for 1. Peltonen, K., and Dipple, A. Polycyclic aromatic hydrocarbons: chemistry of DNA ␮ 4 h in growth medium without DMBA in the presence of 5 g/ml actinomycin D and the adduct formation. J. Occup. Environ. Med., 37: 52–58, 1995. indicated concentrations of fluasterone. CYP1A1, CYP1B1, and GPDH mRNA were 2. Wattenberg, L. An overview of chemoprevention: current status and future prospects. ϭ Ϯ determined by RT-PCR. n 3 SE. For graph, the levels of CYP1A1 and CYP1B1 were Proc. Soc. Exp. Biol. Med., 16: 133–141, 1997. normalized to GPDH level. There was a significant decrease in CYP1A1 mRNA in cells 3. Ebeling, P., and Koivisto, V. A. Physiological importance of dehydroepiandrosterone. Ͻ treated with fluasterone (P 0.05) but no change in CYP1B1 mRNA. Lancet, 343: 1479–1481, 1994. 4. Svec, F., and Porter, J. R. The actions of exogenous dehydroepiandrosterone in experimental animals and humans. Proc. Soc. Exp. Biol. Med., 218: 174–191, 1998. 5. Regelson, W., and Kalimi, M. Dehydroepiandrosterone (DHEA)–the multifunctional not as dramatically as for CYP1A1. These results demonstrate that the steroid. II. Effects on the CNS, cell proliferation, metabolic and vascular, clinical and inhibition of carcinogen-induced CYP enzyme activity is attributable other effects. Mechanism of action? Ann. N. Y. Acad. Sci., 719: 564–575, 1994. 6. Rose, D. P., Stauber, P., Thiel, A., Crowley, J. J., and Milbrath, J. R. Plasma to the inhibition of CYP expression. The level of CYP1A1 and dehydroepiandrosterone sulfate, androstenedione and cortisol, and urinary free cor- CYP1B1 in MCF-7 cells is low but measurable in the absence of tisol excretion in breast cancer. Eur. J. Cancer, 13: 43–47, 1977. carcinogen treatment. Fluasterone, like DHEA, caused a decrease in 7. Hursting, S. D., Perkins, S. N., Haines, D. C., Ward, J. M., and Phang, J. M. Chemoprevention of spontaneous tumorigenesis in p53-knockout mice. Cancer Res., the amount of basal CYP1A1. Basal CYP1B1 levels, on the other 55: 3949–3953, 1995. hand, were unaffected (Fig. 4). 8. Nyce, J. W., Magee, P. N., Hard, G. C., and Schwartz, A. G. Inhibition of 1, 2-dimethylhydrazine-induced colon tumorigenesis in Balb/c mice by dehydroepi- CYP1A1 and CYP1B1 expression are controlled at the transcrip- androsterone. Carcinogenesis (Lond.), 5: 57–62, 1984. tional level by the aryl hydrocarbon receptor, a cytosolic protein that 9. McCormick, D. L., Rao, K. V., Johnson, W. D., Bowman-Gram, T. A., Steele V. E., binds carcinogens and activates CYP transcription by binding to the Lubet, R. A., and Kellof, G. J. Exceptional chemopreventive activity of low-dose dehydroepiandrosterone in the rat mammary gland. Cancer Res., 56: 1724–1726, XRE contained in the promoter region of the CYP1A1 and CYP1B1 1996. genes. We demonstrated previously that DHEA does not affect XRE- 10. Schwartz, A. G., Fairman, D. K., Polansky, M., Lewbart, M. L., and Pashko, L. L. mediated transcription, whether it was induced by TCDD or DMBA Inhibition of 7, 12-dimethylbenz[a]anthracene-initiated and 12-O-tetradecanoyl-phor- bol-13-acetate-promoted skin papilloma formation in mice by dehydro-epiandros- (33). We tested the effect of fluasterone on XRE-mediated transcrip- terone and two synthetic analogs. Carcinogenesis (Lond.), 10: 1809–1813, 1989. tion and found that it, in contrast to DHEA, caused a modest inhibition 11. Pashko, L. L., Hard, G. C., Rovito, R. J., Williams, J. R., Sobel, E. L., and Schwartz, A. G. Inhibition of 7, 12-dimethylbenz(a)anthracene-induced skin papillomas and of both TCDD- or DMBA-induced transcription (Fig. 5A) and inhib- carcinomas by dehydroepiandrosterone and 3-␤-methylandrost-5-en-17-one in mice. ited the basal rate of transcription (Fig. 5B). Structurally, fluasterone Cancer Res., 45: 164–166, 1985. differs from DHEA only in the presence of a single fluorine atom, 12. Li, S, Yan, X., Belanger, A., and Labrie, F. Prevention by dehydroepiandrosterone of the development of mammary carcinoma induced by 7, 12-dimethylbenz(a)anthra- which prevents the metabolism of the compound to other steroids. cene (DMBA) in the rat. Breast Cancer Res. Treat., 29: 203–217, 1994. This difference may account for the inhibitory effect on aryl hydro- 13. Luo, S., Labrie, C., Belanger, A., and Labrie, F. Combined effects of dehydroepi- carbon receptor-mediated transcription. Another possibility is that androsterone and EM-800 on bone mass, serum lipids, and the development of dimethylbenz(a)anthracene-induced mammary carcinoma in the rat. Endocrinology, DHEA itself would inhibit transcription but that it is quickly metab- 138: 3387–3394, 1997. olized by MCF-7 cells. But although fluasterone at 1 ␮M caused a 14. Schwartz, A. G., and Perantoni, A. Protective effect of dehydroepiandrosterone against aflatoxin B1- and 7, 12-dimethylbenz(a)anthracene-induced cytotoxicity and significant decrease in DMBA- or TCDD-induced CYP1A1 mRNA, it transformation in cultured cells. Cancer Res., 5: 2482–2487, 1975. caused only a slight decrease in transcription. Thus, the inhibition of 15. Kim, S. H., Han, H. M., Kang, S. Y., Jung, K. K., Kim, T. G., Oh, H. Y., Lee, Y. K, transcription, although it may contribute to the overall inhibition of and Rheu, H. M. Modulation of chemical carcinogen-induced unscheduled DNA synthesis by dehydroepiandrosterone (DHEA) in the primary rat hepatocytes. Arch. CYP expression by fluasterone, is not the only mechanism. Therefore, Pharm. Res., 22: 474–478, 1999. we also examined the effect of fluasterone on the stability of CYP 16. Ikegwuonu, F. I., and Jefcoate, C. R. Evidence for the involvement of the fatty acid mRNA. As shown in Fig. 6, fluasterone caused a concentration- and peroxisomal beta-oxidation pathways in the inhibition by dehydroepiandrosterone (DHEA) and induction by 2, 3, 7, 8-tetrachlorodibenzo-p-dioxin (TCDD) and dependent decrease in CYP1A1 mRNA in the presence of the RNA benz(a)anthracene (BA) of cytochrome P4501B1 (CYP1B1) in mouse embryo fibro- polymerase II inhibitor actinomycin D after induction by DMBA. This blasts (C3H10T1/2 cells). Mol. Cell. Biochem., 198: 89–100, 1999. 17. Prasanna, H. R., Hart, R. W., and Magee, P. N. Differential effects of dehydro- agrees with our previous study (33), which demonstrated that DHEA epiandrosterone and clofibrate on the binding of 7, 12-dimethylbenz(a)anthracene to decreased CYP1A1 mRNA stability. Interestingly, fluasterone had no hepatic DNA in vivo–a preliminary study. Drug Chem. Toxicol., 12: 327–335, 1989. 3689

18. Pashko, L. L., and Schwartz, A. G. Effect of food restriction, dehydroepiandrosterone, 29. Ciolino, H. P., Daschner, P. J., and Yeh, G. C. Dietary flavonols quercetin and or obesity on the binding of [3H]-7, 12-dimethylbenz(a)anthracene to mouse skin kaempferol are ligands of the aryl hydrocarbon receptor that affect CYP1A1 tran- DNA. J. Gerontol., 38: 8–12, 1983. scription differentially. Biochem. J., 340: 715–722, 1999. 19. Longcope, C. Dehydroepiandrosterone metabolism. J. Endocrinol., 150: 125–127, 1996. 30. Perkins, S. N., Hursting, S. D., Haines, D. C., James, S. J., Miller, B. J., and Phang, 20. Bellei, M., Battelli, D., Fornieri, C., Mori, G., Muscatello, U., Lardy, H., and J. M. Chemoprevention of spontaneous tumorigenesis in nullizygous p53-deficient Bobyleva, V. Changes in liver structure and function after short-term and long-term mice by dehydroepiandrosterone and its analog 16-␣-fluoro-5-androsten-17-one. Car- treatment of rats with dehydroepiandrosterone. J. Nutr., 122: 967–976, 1992. cinogenesis (Lond.), 18: 989–994, 1997. 21. Rao, M. S., Subbarao, V., Yeldandi, A. V., and Reddy, J. K. Hepatocarcinogenicity 31. Casazza, J. P., Schaffer, W. T., and Veech, R. L. The effect of dehydroepiandros- of dehydroepiandrosterone in the rat. Cancer Res., 52: 2977–2979, 1992. terone on liver metabolites. J. Nutr., 116: 304–310, 1986. 22. Schwartz, A. G., Lewbart, M. L., and Pashko, L. L. Novel dehydroepiandrosterone 32. Cleary, M. P. Effect of dehydroepiandrosterone treatment on liver metabolism in rats. analogues with enhanced biological activity and reduced side effects in mice and rats. Int. J. Biochem., 22: 205–210, 1990. Cancer Res., 48: 4817–4822, 1988. 33. Ciolino, H. P., and Yeh, G. C. The steroid hormone dehydroepiandrosterone inhibits 23. Schwartz, A. G., and Pashko, L. L Cancer prevention with dehydroepiandrosterone CYP1A1 expression in vitro by a post-transcriptional mechanism. J. Biol. Chem., and non-androgenic structural analogs. J. Cell. Biochem., 22: 210–217, 1995. 274: 35186–35190, 1999. 24. Dyke, P. H., Foan, C., Wenborn, M., and Coleman, P. J. A review of dioxin releases 34. Schwartz, A. G., and Pashko, L. L. Potential therapeutic use of dehydroepiandros- to land and water in the UK. Sci. Total Environ., 207: 119–131, 1997. 25. Quass, U., Fermann, M. W., and Broker, G. Steps towards a European dioxin terone and structural analogs. Diabetes Technol. Ther., 3: 221–224, 2001. emission inventory. Chemosphere, 40: 1125–1129, 2000. 35. Wang, X., Thomsen, J. S., Santostefano, M., Rosengren R., Safe, S., and Perdew, 26. Ciolino, H. P., and Yeh, G. C. The flavonoid galangin is an inhibitor of CYP1A1 G. H. Comparative properties of the nuclear aryl hydrocarbon (Ah) receptor complex activity and an agonist/antagonist of the aryl hydrocarbon receptor. Br. J. Cancer, 79: from several human cell lines. Eur. J. Pharmacol., 293: 191–205, 1995. 1340–1346, 1999. 36. Moore, M., Wang, X., Lu, Y. F., Wormke, M., Craig, A., Gerlach, J. H, Burghardt, 27. MacDonald, C. J., Ciolino, H. P., and Yeh, G. C. Dibenzoylmethane modulates aryl R., Barhoumi, R., and Safe, S. Benzo[a]pyrene-resistant MCF-7 human breast cancer hydrocarbon receptor function and expression of cytochromes P50 1A1, 1A2, and cells. A unique aryl hydrocarbon-nonresponsive clone. J. Biol. Chem., 269: 11751– 1B1. Cancer Res., 61: 3919–3924, 2001. 11759, 1994. 28. Dohr, O., Vogel, C., and Abel, J. Different response of 2, 3, 7, 8-tetrachlorodibenzo- 37. Doostdar, H., Burke, M. D., and Mayer, R. T. Bioflavonoids: selective substrates p-dioxin (TCDD)-sensitive genes in human breast cancer MCF-7 and MDA-MB 231 and inhibitors for cytochrome P450 CYP1A and CYP1B1. Toxicology, 144: cells. Arch. Biochem. Biophys., 321: 405–412, 1995. 31–38, 2000.

3690

Henry P. Ciolino, Christopher J. MacDonald and Grace Chao Yeh

Cancer Res 2002;62:3685-3690.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/62/13/3685

Cited articles This article cites 33 articles, 8 of which you can access for free at: http://cancerres.aacrjournals.org/content/62/13/3685.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/62/13/3685.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/62/13/3685. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.