In Vitro Model of Mammary Estrogen Metabolism: Structural and

Aug 18, 2004 - In the case of 2-OHE2, methylation occurred at either the 2-OH or the ...... Grammatical Immune System Evolution for Reverse Engineerin...
0 downloads 0 Views 106KB Size
Download PDF
1258

Chem. Res. Toxicol. 2004, 17, 1258-1264

In Vitro Model of Mammary Estrogen Metabolism: Structural and Kinetic Differences between Catechol Estrogens 2- and 4-Hydroxyestradiol Sheila Dawling,† David L. Hachey,‡,§ Nady Roodi,† and Fritz F. Parl*,†,§ Departments of Pathology and Pharmacology and Vanderbilt Ingram Cancer Center, Vanderbilt University, Nashville, Tennessee 37232 Received May 14, 2004

Estrogens and their oxidative metabolites, the catechol estrogens, have been implicated in the development of breast cancer; yet, relatively little is known about estrogen metabolism in the breast. To determine how the parent hormone, 17β-estradiol (E2), is metabolized, we used recombinant, purified phase I enzymes, cytochrome P450 (CYP) 1A1 and 1B1, with the phase II enzymes catechol-O-methyltransferase (COMT) and glutathione S-transferase P1 (GSTP1), all of which are expressed in breast tissue. We employed both gas and liquid chromatography with mass spectrometry to measure E2, the catechol estrogens 2-hydroxyestradiol (2-OHE2) and 4-hydroxyestradiol (4-OHE2), as well as methoxyestrogens and estrogen-GSH conjugates. The oxidation of E2 to 2-OHE2 and 4-OHE2 was exclusively regulated by CYP1A1 and 1B1, regardless of the presence or concentration of COMT and GSTP1. COMT generated two products, 2-methoxyestradiol and 2-hydroxy-3-methoxyestradiol, from 2-OHE2 but only one product, 4-methoxyestradiol, from 4-OHE2. Similarly, GSTP1 yielded two conjugates, 2-OHE21-SG and 2-OHE2-4-SG, from the corresponding quinone 2-hydroxyestradiol-quinone and one conjugate, 4-OHE2-2-SG, from 4-hydroxyestradiol-quinone. Using the experimental data, we developed a multicompartment kinetic model for the oxidative metabolism of the parent hormone E2, which revealed significant differences in rate constants for its C-2 and C-4 metabolites. The results demonstrated a tightly regulated interaction of phase I and phase II enzymes, in which the latter decreased the concentration of catechol estrogens and estrogen quinones, thereby reducing the potential of these oxidative estrogen metabolites to induce DNA damage.

Introduction The two major estrogens, E21 and E1, are ligands for the estrogen receptor and substrates for oxidizing phase I enzymes, CYP1A1 and CYP1B1. In their dual role of ligand and substrate, estrogens have been implicated in the development of breast cancer by simultaneously stimulating cell proliferation and gene expression via the estrogen receptor and by causing DNA damage via their oxidation products, the 2-OH and 4-OH catechol estrogens (1-3). The latter are produced in a series of linked oxidation reactions that have been proposed by several investigators to form the oxidative estrogen metabolism pathway (1, 2). The pathway starts with E2 and E1, which are oxidized to the 2-OH and 4-OH catechol estrogens by CYP1A1 and CYP1B1 (4, 5). These enzymes are postulated to further oxidize the catechol estrogens to unstable semiquinones and quinones. The estrogen quino* To whom correspondence should be addressed. Tel: 615-343-9117. Fax: 615-343-9563. E-mail: [email protected]. † Department of Pathology. ‡ Department of Pharmacology. § Vanderbilt Ingram Cancer Center. 1 Abbreviations: CYP1A1, cytochrome P450 1A1; CYP1B1, cytochrome P450 1B1; COMT, catechol-O-methyltransferase; GSTP1, glutathione S-transferase P1; E2, 17β-estradiol; E1, estrone; 2-OHE2, 2-hydroxyestradiol; 4-OHE2, 4-hydroxyestradiol; E2-2,3-Q, 2-hydroxyestradiol-quinone; E2-3,4-Q, 4-hydroxyestradiol-quinone; 2-MeOE2, 2-methoxyestradiol; 2-OH-3-MeOE2, 2-hydroxy-3-methoxyestradiol; 4-MeOE2, 4-methoxyestradiol; SAM, S-adenosyl methionine; GSH, glutathione; -SG, glutathione moiety.

nes then form Michael addition products with deoxynucleosides (6-8). The catechol estrogens and their estrogen quinones/semiquinones also undergo redox cycling, which results in the production of reactive oxygen species capable of causing oxidative DNA damage (9-12). Thus, P450-mediated estrogen metabolism is expected to lead to the formation of both oxidative and estrogen DNA adducts, all of which have been shown to possess mutagenic potential (13-15). Although other phase I enzymes, such as CYP1A2 and CYP3A4, are involved in hepatic and extrahepatic estrogen oxidation, CYP1A1 and CYP1B1 display the highest levels of expression in breast tissue (16-18). In turn, CYP1B1 exceeds CYP1A1 in its catalytic efficiency as E2 hydroxylase and differs from CYP1A1 in its principal site of catalysis (4). CYP1B1 has its primary hydroxylation activity at the C-4 position of E2, whereas CYP1A1 has its primary activity at the C-2 position. The 4-hydroxylation activity of CYP1B1 has received particular attention because of experimental evidence that 4-OH catechol estrogens are more carcinogenic than the 2-OH isomers. Treatment with 4-OHE2, but not 2-OHE2, induced renal cancer in Syrian hamster (9, 19). The administration of E2, 2-OHE2, and 4-OHE2 induced endometrial carcinomas in 7, 12, and 66%, respectively, of treated CD-1 mice (20). Finally, the examination of microsomal E2 hydroxylation activity in human breast cancer showed significantly higher 4-OHE2/2-OHE2 ratios in tumor tissue than in

10.1021/tx0498657 CCC: $27.50 © 2004 American Chemical Society Published on Web 08/18/2004

Mammary Metabolism of Catechol Estrogens

adjacent normal breast tissue (21), while the latter tissue samples contained 4-fold higher levels of 4-OHE2 than normal tissue from benign breast biopsies (22). All of these findings support a causative role of 4-OHE2 in carcinogenesis but leave open the question why 2-OHE2 and 4-OHE2 differ in carcinogenic activity, given that both are derived from E2 and share the catechol structure. Recent evidence presented by Cavalieri and coworkers indicates that the corresponding quinones, E2-2,3-Q and E2-3,4-Q, form different types of DNA adducts, the latter generating predominantly depurinating adducts that are detectable in human breast tissue (23-25). In addition to phase I enzymes, breast epithelial cells also express phase II enzymes, such as COMT and GSTP1 (26, 27). Both COMT and GSTP1 are also thought to be involved in mammary estrogen metabolism (28, 29). Specifically, COMT catalyzes the methylation of catechol estrogens to methoxyestrogens, which would lower the level of catechol estrogens available for conversion to estrogen quinones (30). We recently showed that methoxyestrogens also exert feedback inhibition on CYP1A1and CYP1B1-mediated oxidative estrogen metabolism, further reducing the potential for estrogen-induced DNA damage (31). In turn, the estrogen quinones undergo conjugation with GSH via the catalytic action of GSTP1 (32, 33). The formation of GSH-estrogen conjugates would reduce the level of estrogen quinones and similarly decrease the potential for DNA damage. Therefore, it is postulated that the phase II enzymes reduce the genotoxicity of the oxidative estrogen metabolism pathway by alternate reactions with catechol estrogens and estrogen quinones produced by CYP1A1 and CYP1B1 (28, 29). However, it is uncertain how phase I and phase II enzymes interact in the pathway. To date, studies of estrogen metabolism have analyzed single enzymes with simple substrate-product kinetics. In this study, we examined the interaction of CYP1A1 and CYP1B1 with COMT and GSTP1 in estrogen metabolism. Because of the complexity of the metabolic pathway, we used recombinant purified enzymes and employed both gas and liquid chromatography with mass spectrometry (GC/MS and LC/MS) to measure the multiple metabolites. Starting with the parent hormone E2 as the substrate, we could define structural and kinetic differences in the metabolic pathways of 2-OHE2 and 4-OHE2. Using the experimental data, we developed a multicompartment kinetic model for the oxidative pathway. The model allowed the prediction of relative metabolite levels that matched actual measurements reported in the literature. The model further demonstrated a tightly regulated interaction of phase I and phase II enzymes, in which the latter decreased the concentration of catechol estrogens and estrogen quinones, thereby reducing the potential of these oxidative estrogen metabolites to induce DNA damage.

Materials and Methods Chemicals. E2, catechol estrogens (2-OHE2 and 4-OHE2), and methoxyestrogens (2-MeOE2, 2-OH-3-MeOE2, and 4-MeOE2) were obtained from Steraloids (Newport, RI). Deuterated E2 was obtained from CDN Isotopes (Pointe-Claire, Quebec). EstrogenGSH conjugates 2-OHE2-4-SG, 2-OHE2-1-SG, 4-OHE2-2-SG, and 4-OHE2-2-SG-d4 were synthesized as previously described (33). Expression and Purification of Recombinant Enzymes. Purified recombinant CYP1A1, CYP1B1, NADPH-P450 reduc-

Chem. Res. Toxicol., Vol. 17, No. 9, 2004 1259 tase, COMT, and GSTP1 were prepared as previously described (5, 30, 31, 33). Assay of Enzyme Activities. Recombinant CYP1A1 (85 pmol), CYP1B1 (165 pmol), NADPH-P450 reductase (500 pmol), COMT (125 pmol), and GSTP1 (500 pmol) were mixed with 80 µg of L-R-dilauroyl-sn-glycero-3-phosphocholine in 0.5 mL of 100 mM potassium phosphate buffer (pH 7.4) containing 10 µM E2, 5 mM glucose 6-phosphate, 5 mM MgCl2, and 1 mM ascorbate. Reactions were initiated by adding 100 µM SAM, 100 µM GSH, glucose-6-phosphate dehydrogenase (0.5 u/mL), and NADP+ to a final concentration of 0.5 mM. Reactions proceeded up to 30 min with gentle shaking at 37 °C. The reactions were terminated at 0, 2, 5, 10, 20, and 30 min by transferrring 400 µL aliquots to 2 mL of chilled CH2Cl2 for E2, catechol estrogen, and methoxyestrogen analysis and 50 µL aliquots to 250 µL of chilled acetone for estrogen-GSH conjugate analysis. GC/MS Analysis. E2, 2-OHE2, 4-OHE2, 2-MeOE2, 2-OH-3MeOE2, and 4-MeOE2 were determined as previously described (5, 30). LC/MS Analysis. 2-OHE2-1-SG, 2-OHE2-4-SG, and 4-OHE22-SG were determined as previously described (33). Kinetic Analysis. Rate constants for the individual oxidation and conjugation steps in the overall pathway of E2 metabolism were determined using the SAAM II program (SAAM Institute, Inc., Seattle, WA). The system was modeled as a nonsteady state compartmental model using Akaike and Bayesian information criteria to evaluate the best model structure and the quality of the fitted parameters (34). During the numerical fitting process, the data for the various metabolites were weighted as the reciprocal of their molar concentrations. Parameters derived from the model were the individual rate constants (min-1), and the corresponding variance was expressed as a percentage of the parameter value.

Results To ascertain the effect of COMT on the CYP-mediated hydroxylation of E2 to catechol estrogens, we varied the molar ratio of COMT:CYP from 0.25 to 2.0; that is, we varied the amount of COMT from 60 to 500 pmol per reaction while holding the concentration of CYP1A1 and CYP1B1 constant at 250 pmol. The concentration of E2 remaining after a 10 min incubation of 10 µM E2 was not altered by increasing amounts of COMT (Figure 1A). In contrast, the concentration of 4-OHE2 showed an 8-fold decline associated with a proportional increase in 4-MeOE2. The concentration of 2-OHE2 decreased 2-fold without a noticeable change in the concentration of 2-MeOE2 and 2-OH-3-MeOE2 (Figure 1B,C). We performed a similar experiment with GSTP1, in which the concentration of GSTP1 was varied from 125 to 1000 pmol per reaction. Again, the concentration of E2 remaining after a 10 min incubation was not affected by increasing amounts of GSTP1 (result not shown), while the concentrations of catechol estrogens and estrogenGSH conjugates were affected as recently described (33). A 30 min time-course experiment showed nearly complete disappearance of 10 µM E2 by 10 min in the presence of CYP1A1 and CYP1B1 without a discernible effect of added COMT, GSTP1, or both phase II enzymes (Figure 2). This was in contrast to the catechol estrogen concentration, which was affected by the addition of COMT and GSTP1 (Figure 3). In the presence of CYP1A1 and CYP1B1, both 2-OHE2 and 4-OHE2 showed a steep increase, peaking at about 4 µM around 5 min, and then slowly declined to 3 µM at 30 min. The addition of COMT, GSTP1, or both caused a smaller increase in catechol estrogen concentration and a more pronounced decline. However, there were noticeable differences both in the

1260

Chem. Res. Toxicol., Vol. 17, No. 9, 2004

Dawling et al.

Figure 3. Metabolism of E2 to 2-OHE2 (A) and 4-OHE2 (B) as a function of time in the presence of CYP (250 pmol) alone (9), CYP plus COMT (125 pmol) (2), CYP plus GSTP1 (500 pmol) (1), and the combination of CYP, COMT, and GSTP1 ([). Each reaction contained 10 µM E2, 100 µM SAM, 100 µM GSH, and the indicated purified recombinant enzymes. Reactions proceeded for 0, 2, 5, 10, 20, and 30 min at 37 °C and were analyzed by GC/MS. Data are represented as means of two replicate assays.

Figure 1. Effect of increasing concentrations of COMT on CYPmediated E2 metabolism. Each reaction contained 10 µM E2, 100 µM SAM, 250 pmol of CYP (85 pmol of CYP1A1 and 165 pmol of CYP1B1), and 60, 125, 250, and 500 pmol of COMT. Reactions proceeded for 10 min at 37 °C and were analyzed for concentrations of E2 (A), catechol estrogens (B), and methoxyestrogens (C) by GC/MS as described under the Materials and Methods. Data are represented as means of two replicate assays.

Figure 2. Metabolism of E2 as a function of time in the presence of CYP (250 pmol) alone (9), CYP plus COMT (125 pmol) (2), CYP plus GSTP1 (500 pmol) (1), and the combination of CYP, COMT, and GSTP1 ([). Each reaction contained 10 µM E2, 100 µM SAM, 100 µM GSH, and the indicated purified recombinant enzymes. Reactions proceeded for 0, 2, 5, 10, 20, and 30 min at 37 °C and were analyzed by GC/MS. Data are represented as means of two replicate assays.

effect of COMT and GSTP1 and in the response of 2-OHE2 and 4-OHE2. COMT was more effective than GSTP1 in lowering the concentration of 2-OHE2 and 4-OHE2. In the case of 4-OHE2, the effect of COMT and

GSTP1 was additive, leading to its complete disappearance at 30 min when both enzymes were added (Figure 3B). In contrast, 2-OHE2 remained at a level of 0.5 µM in the presence of either COMT or GSTP1 or both (Figure 3A). The addition of GSTP1 lowered the production of methoxyestrogens by COMT (Figure 4) just as the addition of COMT decreased the formation of estrogenGSH conjugates (Figure 5). However, there were noticeable differences between the methoxy and GSH conjugates. The levels of 2-MeOE2 and 2-OH-3-MeOE2 showed a steep rise followed by a rapid decline (Figure 4A,B), while 4-MeOE2 displayed a larger increase and a much slower decline (Figure 4C). Among the three GSH conjugates, 4-OHE2-2-SG was produced in largest amounts and its production was least affected by the addition of COMT (Figure 5A-C). The oxidative pathway of E2 metabolism yielded eight metabolites measured by GC/MS and LC/MS (Figure 6). The complexity of the system precludes kinetic analysis by a simple precursor-product model; thus, we developed a more complex multicompartment model to handle the data. Two experimental findings required special consideration. First, the transient semiquinones and quinones were not quantifiable due to their short half-lives (32). Therefore, we treated the path leading from 2-OHE2 and 4-OHE2 to GSH conjugates in the model as an irreversible catabolic pathway. Second, unpublished observations from our lab indicate that the total mass balance of E2 in these and similar experiments was never fully accounted for by the metabolites identified in Figure 6. Most of the unaccounted estrogen metabolites appear to be derived from the reactive quinone intermediates, which are known to form adducts with proteins (and DNA) (35, 36). In the model, we included elements in the

Mammary Metabolism of Catechol Estrogens

Figure 4. Production of methoxyestrogens 2-MeOE2 (A), 2-OH3-MeOE2 (B), and 4-MeOE2 (C) as a function of time in the presence of CYP plus COMT (9) and the combination of CYP, COMT, and GSTP1 (2). Each reaction contained 10 µM E2, 100 µM SAM, 100 µM GSH, and the indicated purified recombinant enzymes. Reactions proceeded for 0, 2, 5, 10, 20, and 30 min at 37 °C and were analyzed by GC/MS. Data are represented as means of two replicate assays.

data set for these unaccounted metabolites in the respective 2-OH and 4-OH hydroxylation pathways. The rates of the CYP-mediated 2-OH and 4-OH hydroxylation reactions were similar (k ) 0.13 ( 27 and 0.11 ( 9% min-1, respectively) but lower than the COMT-mediated conversion of the catechol estrogens to the methoxyestrogens 2-MeOE2, 2-OH-3-MeOE2, and 4-MeOE2 (k ) 0.45 ( 87, 0.15 ( 82, and 0.18 ( 8% min-1, respectively). However, the CYPs also catalyzed the O-demethylation of methoxyestrogens, primarily 2-MeOE2 and 2-OH-3MeOE2 (k ) 0.52 ( 79 and 0.33 ( 70% min-1, respectively). The GSTP1-mediated formation of estrogen-GSH conjugates from the quinone intermediates had the lowest rate constants with GSTP1 being slightly more reactive toward E2-3,4-Q (k ) 0.035 ( 23% min-1) than E2-2,3-Q (0.023 ( 41 and 0.002 ( 40% min-1). As mentioned, the transient semiquinones and quinones were not quantifiable; therefore, rate constants not calculated. We performed a recovery analysis of input E2 by adding the measured levels of E2 and metabolites at each time point and for each of the enzyme combinations. There was a progressive loss from the system, which increased with both reaction time and number of enzymes added. At 30 min, the recovery of input E2 was 54% in the presence of CYPs and 24% in the presence of CYPs,

Chem. Res. Toxicol., Vol. 17, No. 9, 2004 1261

Figure 5. Production of estrogen-GSH conjugates 2-OHE24-SG (A), 2-OHE2-1-SG (B), and 4-OHE2-2-SG (C) as a function of time in the presence of CYP plus GSTP1 (9) and the combination of CYP, COMT, and GSTP1 (2). Each reaction contained 10 µM E2, 100 µM SAM, 100 µM GSH, and the indicated purified recombinant enzymes. Reactions proceeded for 0, 2, 5, 10, 20, and 30 min at 37 °C and were analyzed by LC/MS as described under the Materials and Methods. Data are represented as means of two replicate assays.

COMT, and GSTP1. This irreversible loss, presumably by binding of the reactive quinones to protein, was significantly higher for the 2-OHE2 than the 4-OHE2 pathway, i.e., k ) 0.54 ( 44 and k ) 0.016 ( 33% min-1, respectively, in the presence of all four enzymes (Figure 6). The 4-hydroxylation-derived 4-MeOE2 and 4-OHE2SG constituted the bulk of the quantifiable, nonproteinbound metabolites after 30 min of reaction.

Discussion The steroids E2 and E1 are characterized by an aromatic A ring with a hydroxyl group at C-3. The enzymes involved in mammary estrogen metabolism act principally on the A ring, starting with CYP1A1 and CYP1B1, which hydroxylate the vicinal carbons, C-2 and C-4, to form catechol estrogens. The latter are substrates for COMT, which catalyzes the transfer of a methyl group from the methyl donor SAM to one hydroxyl moiety of the catechol ring. In the case of 2-OHE2, methylation occurred at either the 2-OH or the 3-OH group, resulting

1262

Chem. Res. Toxicol., Vol. 17, No. 9, 2004

Figure 6. Estrogen metabolism pathway is regulated by oxidizing phase I and conjugating phase II enzymes. CYP1A1 and CYP1B1 catalyze the oxidation of E2 to catechol estrogens 2-OHE2 and 4-OHE2. The catechol estrogens are either methylated by COMT to methoxyestrogens (2-MeOE2, 2-OH-3-MeOE2, and 4-MeOE2) or further oxidized by CYPs to semiquinones (E22,3-SQ and E2-3,4-SQ) and quinones (E2-2,3-Q and E2-3,4-Q). The estrogen quinones are either conjugated by GSTP1 to GSH conjugates (2-OHE2-1-SG, 2-OHE2-4-SG, and 4-OHE2-2-SG), or they form quinone-DNA adducts or oxidative DNA adducts via quinone-semiquinone redox cycling. Kinetic analysis was performed to calculate individual rate constants, which are expressed as k ) min-1. The respective coefficients of variation are shown as percentages. Semiquinones and quinones were not quantitated; therefore, k values could not be calculated. There is an irreversible loss from the system as indicated by the dashed arrows.

in the formation of 2-MeOE2 and 2-OH-3-MeOE2. In contrast, methylation of 4-OHE2 occurred only at the 4-OH group, yielding 4-MeOE2 (30). The main structural difference between 2-OH and 4-OH catechol estrogens is the proximity of the 4-OH group to the B ring of the steroid. Clearly, the 2-OH and 3-OH groups in 2-OHE2 appear to be similar in reactivity, whereas the 3-OH and 4-OH groups in 4-OHE2 differ in reactivity to the point that in the latter only the 4-OH group becomes methylated. The difference in regiospecific reactivity is not limited to the catechol estrogens but extends to the corresponding quinones, which are substrates for GSTP1mediated GSH conjugation (33). The GSH attachment occurs through its cysteine moiety, with the cysteine sulfur binding to an A ring carbon vicinal to the catechol carbons, i.e., either C-1 or C-4 in 2-OHE2, yielding 2-OHE2-1-SG and 2-OHE2-4-SG, whereas conjugation at C-2 in 4-OHE2 resulted in 4-OHE2-2-SG (33, 37). Thus, the phase I enzymes metabolize E2 to two products, 2-OHE2 and 4-OHE2, and further to E2-2,3-Q and E2-3,4Q, while the phase II enzymes produce two conjugates each from 2-OHE2 and E2-2,3-Q but only one each from 4-OHE2 and E2-3,4-Q (Figure 6). The molar ratios of CYP1A1, CYP1B1, COMT, and GSTP1 in breast tissue are unknown but likely to vary since CYPs and GSTs are inducible by numerous agents (38, 39). Nevertheless, we can assume that the concentration of CYP1B1 is equal to or greater than that of CYP1A1 based on mRNA expression levels, the ratio of 4-OHE2/2-OHE2 in breast tissue, and the observation that 2-OHE2 is produced by both CYP isoforms, whereas 4-OHE2 is produced only by CYP1B1 (4, 5, 16, 17, 21, 40). Therefore, we held the concentrations of CYP1B1 and CYP1A1 constant at 165 and 85 pmol, respectively, for a total CYP concentration of 250 pmol and varied the molar

Dawling et al.

ratios of the phase II enzymes, e.g., 60-500 pmol COMT. As shown in Figures 1A and 2, the hydroxylation of E2 was exclusively regulated by the phase I enzymes CYP1A1 and CYP1B1, regardless of the presence or concentration of phase II enzymes. This is in contrast to the catechol estrogen concentration, which was influenced by the presence of COMT and GSTP1 (Figure 3). Catechol estrogens are substrates for both P450 enzymes and COMT, leading to the formation of estrogen quinones and methoxyestrogens, respectively. Thus, one would expect a decrease in catechol estrogen concentration upon addition of COMT. However, GSTP1 addition also lowered the concentration of catechol estrogens, although the latter are not GSTP1 substrates (33). This decrease in the presence of GSTP1 is best explained by enhanced conversion of catechol estrogens to estrogen quinones, which are GSTP1 substrates and become depleted by GSTP1-mediated formation of estrogen-GSH conjugates. Methoxyestrogens are products of the COMT-mediated methylation of catechol estrogens as well as substrates for the CYP1A1- and CYP1B1-mediated demethylation to catechol estrogens (30, 31). The sum of both reactions in the combined presence of CYP1A1, CYP1B1, and COMT yielded a different kinetic profile for 4-MeOE2, which showed a severalfold higher increase as compared to 2-MeOE2 and 2-OH-3-MeOE2 (Figures 1C and 4). Moreover, the increase in 4-MeOE2 was sustained during the 30 min reaction, whereas 2-MeOE2 and 2-OH-3MeOE2 rapidly declined to one-tenth of their respective peak levels. Thus, the difference in regiospecific reactivity noted for 4-OHE2 and E2-3,4-Q as compared to 2-OHE2 and E2-2,3-Q appears to extend to a different reactivity of 4-MeOE2 as compared to the other two methoxyestrogens. Catechol estrogens and estrogen quinones occupy pivotal positions in the oxidative estrogen metabolism pathway. Using GC/MS, we could follow the production and disappearance of catechol estrogens. Ideally, the measurement of estrogen quinones would help quantify their role, but they are highly reactive with short halflives due to the strained 1,2-diketone functionality inherent in o-quinones (41). Thus, estrogen quinones are too labile to be reliably quantified, and we used production of the stable GSH-estrogen conjugates as surrogate markers (33). Again, we noticed a difference in the formation of conjugates. Just as COMT in the presence of CYPs yielded more 4-MeOE2 than 2-MeOE2 and 2-OH3-MeOE2, GSTP1 produced more 4-OHE2-2-SG than 2-OHE2-4-SG and 2-OHE2-1-SG. Despite the complexity of oxidative E2 metabolism, the pathway was amenable to kinetic analysis using a multicompartment model (Figure 6). The CYPs appeared to be less active than COMT based on a comparison of the respective rate constants for the hydroxylation and O-methylation reactions. However, the CYPs also catalyze the O-demethylation of the methoxyestrogens 2-MeOE2 and 2-OH-3-MeOE2 with rather high rate constants. Moreover, the CYPs catalyze the oxidation of catechol estrogens to estrogen semiquinones and quinones. Although the respective rate constants could not be calculated due to our inability to measure the latter compounds, the CYPs are overall the most active enzymes, initiating as well as driving the pathway. As mentioned, the pathway displays asymmetry with respect to the number of metabolites derived from 2-OHE2 and 4-OHE2, which appears to be due to structural differences

Mammary Metabolism of Catechol Estrogens

between the two catechols. The kinetic analysis further documents the asymmetry of the pathway. For example, the 4-hydroxylation-derived 4-MeOE2 and 4-OHE2-SG constituted the bulk of the measured metabolites after 30 min of reaction. Irreversible loss from the system, most likely due to binding of the reactive quinones to protein, was much higher for the 2-OHE2 (k ) 0.54 ( 44% min-1) than the 4-OHE2 (k ) 0.016 ( 33% min-1) pathway. This is consistent with the shorter half-life of 42 s for E1-2,3-Q as compared to 12 min for E1-3,4-Q, determined at pH 7.4 and 37 °C (32). Experimental evidence in several animal models indicates that 4-OH catechol estrogens are more carcinogenic than their 2-OH isomers (9, 19, 20). The determination of estrogen and estrogen metabolites in normal and malignant human breast tissue revealed that the concentration of 4-OHE2 in tumor tissue was more than twice as high as that of any other estrogen (42). The examination of microsomal E2 hydroxylation activity in human breast cancer showed significantly higher 4-OHE2/2-OHE2 ratios in tumor tissue than in adjacent normal breast tissue (21), while the latter tissue samples contained 4-fold higher levels of 4-OHE2 than normal tissue from benign breast biopsies (22). These data support a mechanistic role of 4-OHE2 in tumor development. One can speculate that the high carcinogenic activity of 4-OHE2 may be due to the greater stability of its quinone E2-3,4-Q, which might allow more time for diffusion of the reactive intermediate away from the site of formation to interact with cellular macromolecules such as DNA. In contrast, the low carcinogenicity of 2-OHE2 may be related to the decreased stability of its quinone E2-2,3-Q, which would limit diffusion and restrict DNA contact (41). The model of mammary estrogen metabolism presented here has several limitations. There are other GST isozymes that are expressed in breast tissue, namely, GSTM1 and GSTA1 (26, 43), which were not included in our analysis. However, neither GSTM1 nor GSTA1 are consistently expressed in breast tissue and GSTP1 is the predominant GST isoform catalyzing the GSH conjugation of mammary estrogen quinones (26, 43). Besides COMT, there are two other classes of phase II enzymes capable of conjugating catechol estrogens, namely, the sulfotransferases and UDP-glucuronosyltransferases. However, it is thought that catechol estrogens are conjugated predominantly to methyl conjugates and to a lesser extent sulfate and glucuronide conjugates (28, 29). Therefore, on the basis of our current knowledge, COMT and GSTP1 are the most important phase II enzymes involved in mammary estrogen metabolism. The same can be said for CYP1A1 and CYP1B1 as phase I enzymes. Although other P450 enzymes, such as CYP1A2 and CYP3A4, are involved in hepatic and extrahepatic estrogen oxidation, CYP1A1 and CYP1B1 display the highest levels of expression in breast tissue (16). The model cannot duplicate the subcellular localization of the enzymes, which is likely to affect the disposition of E2 and its metabolites. CYPs and GSTs are microsomal enzymes, whereas COMT is localized predominantly in the cytosol with a small percentage in the microsomal fraction (38, 39, 44). The hydrophobicity of E2 permits its preferential partitioning into lipid membranes such as the endoplasmic reticulum, which facilitates access to CYP1A1 and CYP1B1, resulting in oxidation to catechol estrogens. The resultant catechols are also lipophilic in nature and may be retained in the hydrophobic environment of the

Chem. Res. Toxicol., Vol. 17, No. 9, 2004 1263

endoplasmic reticulum, where they undergo oxidation to quinones and subsequent conjugation by GSTP1 to hydrophilic GSH conjugates. The fraction of catechol estrogens that escapes the endoplasmic reticulum into the cytosol may be conjugated by COMT, although the microsomal fraction of the latter enzyme may also participate in the formation of methoxyestrogens. Despite the limitations, we believe that our model provides mechanistic insight into mammary estrogen metabolism as indicated by comparison with actual measurements of estrogens in breast tissue. Two groups have measured the parent hormones, catechols, and methoxy and GSH conjugates in breast tissue from reduction mammoplasties, benign breast biopsies, and breast cancers (22, 42). Despite differences in analytical methodology, types of tissues analyzed, and up to 100-fold variation in absolute values measured, both groups showed mostly higher tissue concentrations of C-4 than C-2 metabolites as predicted by the model. Parent hormones accounted for only a small percentage of the measured estrogens, consistent with our data, which showed rapid disappearance of E2 as result of CYP-mediated metabolism. In summary, we have examined mammary estrogen metabolism using recombinant purified phase I and phase II enzymes known to be expressed in breast tissue. We demonstrate a tightly regulated pathway of E2 oxidation to catechol estrogens and estrogen quinones that is catalyzed by CYP1A1 and CYP1B1. The phase II enzymes COMT and GSTP1 have no effect on the oxidation of E2 but act on the catechol estrogens and estrogen quinones, respectively, each by forming three conjugates. The experiments uncovered striking differences in the regiospecific reactivity of C-2 and C-4 metabolites, which provide a possible explanation for the greater carcinogenic activity of 4-OHE2 as compared to 2-OHE2. We developed a multicompartment kinetic model that defines the interaction of phase I and II enzymes in the pathway, in which the latter decrease the concentration of catechol estrogens and estrogen quinones, thereby reducing the potential of these oxidative estrogen metabolites to induce DNA damage.

Acknowledgment. This work was supported in part by NIH 1R01 CA83752, 5P30 CA68485, and 5P30 ES00267.

References (1) Yager, J. D., and Liehr, J. G. (1996) Molecular mechanisms of estrogen carcinogenesis. Annu. Rev. Pharmacol. Toxicol. 36, 203232. (2) Cavalieri, E. L., Stack, D. E., Devanesan, P. D., Todorvic, R., Dwivedy, I., Higginbotham, S., Johansson, S. L., Patil, K. D., Gross, M. L., Gooden, J. K., Ramanathan, R., and Cerny, R. L. (1997) Molecular origin of cancer: Catechol estrogen-3,4-quinones as endogenous tumor initiators. Proc. Natl. Acad. Sci. U.S.A. 94, 10937-10942. (3) Parl, F. F. (2000) Estrogens, Estrogen Receptor and Breast Cancer, IOS Press, Amsterdam. (4) Hayes, C. L., Spink, D. C., Spink, B. C., Cao, J. Q., Walker, N. J., and Sutter, T. R. (1996) 17β-Estradiol hydroxylation catalyzed by human cytochrome P450 1B1. Proc. Natl. Acad. Sci. U.S.A. 93, 9776-9781. (5) Hanna, I. H., Dawling, S., Roodi, N., Guengerich, F. P., and Parl, F. F. (2000) Cytochrome P450 1B1 (CYP1B1) pharmacogenetics: Association of polymorphisms with functional differences in estrogen hydroxylation activity. Cancer Res. 60, 3440-3444. (6) Stack, D. E., Byun, J., Gross, M. L., Rogan, E. G., and Cavalieri, E. L. (1996) Molecular characteristics of catechol estrogen quinones in reactions with deoxyribonucleosides. Chem. Res. Toxicol. 9, 851-859.

1264

Chem. Res. Toxicol., Vol. 17, No. 9, 2004

(7) Akanni, A., and Abul-Hajj, Y. J. (1997) Estrogen-nucleic acid adducts: Reaction of 3,4-estrone o-quinone radical anion with deoxyribonucleosides. Chem. Res. Toxicol. 10, 760-766. (8) Bolton, J. L., Pisha, E., Zhang, F., and Qiu, S. (1998) Role of quinoids in estrogen carcinogenesis. Chem. Res. Toxicol. 11, 11131127. (9) Liehr, J. G., Fang, W. F., Sirbasku, D. A., and Ari-Ulubelen, A. (1986) Carcinogenicity of catechol estrogens in Syrian hamsters. J. Steroid Biochem. 24, 353-356. (10) Liehr, J. G., and Roy, D. (1990) Free radical generation by redox cycling of estrogens. Free Radical Biol. Med. 8, 415-423. (11) Nutter, L. M., Wu, Y. Y., Ngo, E. O., Sierra, E. E., Gutierrez, P. L., and Abul-Hajj, Y. J. (1994) An o-quinone form of estrogen produces free radicals in human breast cancer cells: correlation with DNA damage. Chem. Res. Toxicol. 7, 23-28. (12) Han, X., and Liehr, J. G. (1995) Microsome-mediated 8-hydroxylation of guanine bases of DNA by steroid estrogens: correlation of DNA damage by free radicals with metabolic activation to quinones. Carcinogenesis 16, 2571-2574. (13) Shibutani, S., Takeshita, M., and Grollman, A. (1991) Insertion of specific bases during DNA synthesis past the oxidationdamaged base 8-oxodG. Nature 349, 431-434. (14) Terashima, I., Suzuki, N., and Shibutani, S. (2001) Mutagenic properties of estrogen-quinone derived DNA adducts in Simian kidney cells. Biochemistry 40, 166-172. (15) Chakravarti, D., Mailander, P. C., Li, K. M., Higginbotham, S., Zhang, H. L., Gross, M. L., Meza, J. L., Cavalieri, E. L., and Rogan, E. G. (2001) Evidence that a burst of DNA depurination in SENCAR mouse skin induces error-prone repair and forms mutations in the H-ras gene. Oncogene 20, 7945-7953. (16) Huang, Z., Fasco, M. J., Figge, H. L., Keyomarsi, K., and Kaminsky, L. S. (1996) Expression of cytochromes P450 in human breast tissue and tumors. Drub Metab. Dispos. 24, 899-905. (17) Spink, D. C., Spink, B. C., Cao, J. Q., DePasquale, J. A., Pentecost, B. T., Fasco, M. J., Li, Y., and Sutter, T. R. (1998) Differential expression of CYP1A1 and CYP1B1 in human breast epithelial cells and breast tumor cells. Carcinogenesis 19, 291-298. (18) McFayden, M. C. E., Breeman, S., Payne, S., Stirk, C., Miller, I. D., Melvin, W. T., and Murray, G. I. (1999) Immunohistochemical localization of cytochrome P450 CYP1B1 in breast cancer with monoclonal antibodies specific for CYP1B1. J. Histochem. Cytochem. 47, 1457-1464. (19) Li, J. J., and Li, S. A. (1987) Estrogen carcinogenesis in Syrian hamster tissues: Role of metabolism. Fed. Proc. 46, 1858-1863. (20) Newbold, R. R., and Liehr, J. G. (2000) Induction of uterine adenocarcinoma in CD-1 mice by catechol estrogens. Cancer Res. 60, 235-237. (21) Liehr, J. G., and Ricci, M. J. (1996) 4-Hydroxylation of estrogens as marker of human mammary tumors. Proc. Natl. Acad. Sci. U.S.A. 93, 3294-3296. (22) Rogan, E. G., Badawi, A. F., Devanesan, P. D., Meza, J., Edney, J. A., West, W. W., Higginbotham, S. M., and Cavalieri, E. L. (2003) Relative imbalances in estrogen metabolism and conjugation in breast tissue of women with carcinoma: potential biomarkers of susceptibility to cancer. Carcinogenesis 24, 697-702. (23) Cavalieri, E. L., Li, K. M., Balu, N., Saeed, M., Devanesan, P., Higginbotham, S., Zhao, J., Gross, M. L., and Rogan, E. G. (2002) Catechol ortho-quinones: the electrophilic compounds that form depurinating DNA adducts and could initiate cancer and other diseases. Carcinogenesis 23, 1071-1077. (24) Markushin, Y., Zhong, W., Cavalieri, E. L., Rogan, E. G., Small, G. J., Yeung, E. S., and Jankowiak, R. (2003) Spectral characterization of catechol estrogen quinone (CEQ)-derived DNA adducts and their identification in human breast tissue extract. Chem. Res. Toxicol. 16, 1107-1117. (25) Yue, W., Santen, R. J., Wang, J. P., Li, Y., Verderame, M. F., Bocchinfuso, W. P., Korach, K. S., Devanesan, P., Todorovic, R., Rogan, E. G., and Cavalieri, E. L. (2003) Genotoxic metabolites of estradiol in breast: Potential mechanism of estradiol induced carcinogenesis. J. Steroid Biochem. Mol. Biol. 86, 477-486. (26) Shea, T. C., Claflin, G., Comstock, K. E., Sanderson, B. J., Burstein, N. A., Keenan, E. J., Mannervik, B., and Henner, W.

Dawling et al.

(27)

(28) (29) (30)

(31) (32)

(33)

(34) (35) (36)

(37)

(38)

(39) (40)

(41)

(42)

(43)

(44)

D. (1990) Glutathione transferase activity and isoenzyme composition in primary human beast cancers. Cancer Res. 50, 68486853. Weisz, J., Fritz-Wolz, G., Gestl, S., Clawson, G. A., Creveling, C. R., Liehr, J. G., and Dabbs, D. (2000) Nuclear localization of catechol-o-methyltransferase in neoplastic and nonneoplastic mammary epithelial cells. Am. J. Pathol. 156, 1841-1848. Zhu, B. T., and Conney, A. H. (1998) Functional role of estrogen metabolism in target cells: Review and perspectives. Carcinogenesis 19, 1-27. Raftogianis, R., Creveling, C., Weinshilboum, R., and Weisz, J. (2000) Estrogen metabolism by conjugation. J. Natl. Cancer Inst. Monogr. 27, 113-124. Dawling, S., Roodi, N., Mernaugh, R. L., Wang, X. Y., and Parl, F. F. (2001) Catechol-O-methyltransferase (COMT)-mediated metabolism of catechol estrogens: comparison of wild-type and variant COMT isoforms. Cancer Res. 61, 6716-6722. Dawling, S., Roodi, N., and Parl, F. F. (2003) Methoxyestrogens exert feedback inhibition on cytochrome P450 1A1 and 1B1. Cancer Res. 63, 3127-3132. Iverson, S. L., Shen, L., Anlar, N., and Bolton, J. L. (1996) Bioactivation of estrone and its catechol metabolites to quinoidglutathione conjugates in rat liver microsomes. Chem. Res. Toxicol. 9, 492-499. Hachey, D. L., Dawling, S., Roodi, N., and Parl, F. F. (2003) Sequential action of phase I and II enzymes cytochrome P450 1B1 and glutathione S-transferase P1 in mammary estrogen metabolism. Cancer Res. 63, 8492-8499. Li, W., and Nyholt, D. R. (2001) Marker selection by Akaike information criterion and Bayesian information criterion. Genet. Epidemiol. 21 (Suppl.), S272-S277. Tabakovic, K., and Abul-Hajj, Y. J. (1994) Reaction of lysine with estrone 3,4-o-quinone. Chem. Res. Toxicol. 7, 696-701. Cao, K., Stack, D. E., Ramanathan, R., Gross, M. L., Rogan, E. G., and Cavalieri, E. L. (1998) Synthesis and structure elucidation of estrogen quinones conjugated with cysteine, N-acetylcysteine, and glutathione. Chem. Res. Toxicol. 11, 909-916. Ramanathan, R., Cao, K., Cavalieri, E., and Gross, M. L. (1998) Mass spectrometric methods for distinguishing structural isomers of glutathione conjugates of estrone and estradiol. J. Am. Soc. Mass. Spectrom. 9, 612-619. Hayes, J. D., and Pulford, D. J. (1995) The glutathione Stransferase supergene family: Regulation of GST and the contribution of the isoenzymes to cancer chemoprotection and drug resistance. Crit. Rev. Biochem. Mol. Biol. 30, 445-600. Guengerich, F. P. (2001) Common and uncommon cytochrome P450 reactions related to metabolism and chemical toxicity. Chem. Res. Toxicol. 14, 611-650. Shimada, T., Hayes, C. L., Yamazaki, H., Amin, S., Hecht, S. S., Guengerich, F. P., and Sutter, T. R. (1996) Activation of chemically diverse procarcinogens by human cytochrome P-450 1B1. Cancer Res. 56, 2979-2984. Tabakovic, K., Gleason, W. B., Ojala, W. H., and Abul-Hajj, Y. J. (1996) Oxidative transformation of 2-hydroxyestrone. Stability and reactivity of 2,3-estrone quinone and its relationship to estrogen carcinogenicity. Chem. Res. Toxicol. 9, 860-865. Castagnetta, L. A. M., Granata, O. M., Traina, A., Ravazzolo, B., Amoroso, M., Miele, M., Bellavia, V., Agostara, B., and Carruba, G. (2002) Tissue content of hydroxyestrogens in relation to survival of breast cancer patients. Clin. Cancer Res. 8, 31463155. Cairns, J., Wright, C., Cattan, A. R., Hall, A. G., Cantwell, B. J., Harris, A. L., and Horne, C. H. (1992) Immunohistochemical demonstration of glutathione S-transferases in primary human breast carcinomas. J. Pathol. 166, 19-25. Ulmanen, I., Peranen, J., Tenhunen, J., Tilgmann, C., Karhunen, T., Panula, P., Bernasconi, L., Aubry, J. P., and Lundstrom, K. (1997) Expression and intracellular localization of catechol Omethyltransferase in transfected mammalian cells. Eur. J. Biochem. 243, 452-459.

TX0498657