Introduction

Top Abstract Introduction Materials and Methods Results Discussion Conclusions References

The endogenous estrogens 17-estradiol (E₂) and estrone (E₁) undergo extensive metabolism in the body (their structures and possible sites for oxidative metabolism are illustrated in Fig. 1). Over the past several decades, a large number of hydroxylated and keto metabolites of E₂ and E₁ have been identified in biological samples from animals or humans (for review, see Zhu and Conney, 1998a). Most of these oxidative estrogen metabolites are believed to be formed by cytochrome P450 (P450) enzymes (Martucci and Fishman, 1993; Zhu and Conney, 1998a). Detailed knowledge of the P450-mediated metabolism of E₂ and E₁ to various oxidative metabolites, in particular those with biological activities, in human liver as well as in extrahepatic target tissues has added significantly to our understanding of the diverse biological actions that are associated with endogenous estrogens. 4-OH-E₂ and 16-OH-E₁, for example, are two unique hydroxyestrogen metabolites that have strong estrogenic hormonal activity and also potential genotoxicity (Swaneck and Fishman, 1988; Zhu and Conney, 1998a; Liehr, 2000). There is growing evidence for an etiological role of these bioactive estrogen metabolites, in particular 4-OH-E₂, in estrogen-induced cancer in animal models and possibly in humans (Fishman et al., 1984; Bradlow et al., 1986; Cavalieri et al., 2000; Liehr, 2000). Recently, some of the other estrogen metabolites (such as 2-methoxyestradiol and 15-OH-E₂) have also been suggested to have unique biological actions that are different from their parent hormone E₂ (Zhu and Conney, 1998a,b).

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Structures of E1 and E2. The R is a keto group for E1 and a -OH group for E2. The possible positions for the NADPH-dependent - or -hydroxylation or keto formation catalyzed by P450 enzymes are indicated by arrowheads. The filled arrowheads indicate the C positions where either - or -hydroxylation or keto formation may take place, and the unfilled arrowheads indicate the C positions where only a hydroxylation (single configuration) may occur. For the C18 position, either hydroxylation or aldehyde formation may occur.

In most animals as well as in humans, liver contains the highest level of total P450-dependent metabolizing enzymes and possibly also the largest number of different P450 isoforms. Recent studies showed that incubations of radiolabeled E₂ with rat or mouse liver microsomes (a crude preparation containing many different P450 isoforms) and NADPH resulted in the formation of at least 15 estrogen metabolites (Suchar et al., 1995, 1996; Zhu et al., 1998). A number of earlier studies have also examined the NADPH-dependent metabolism of E₂ and E₁ by microsomal preparations from human liver (Kerlan et al., 1992; Shou et al., 1997; Huang et al., 1998; Yamazaki et al., 1998). In these studies, however, only a few hydroxylated estrogen metabolites (i.e., products of estrogen 2-, 4-, and 16-hydroxylation) were determined. By using a versatile HPLC separation method coupled with radioactivity detection, we recently characterized the profile of the NADPH-dependent metabolism of E₂ to various hydroxylated or keto metabolites by human liver microsomes (Lee et al., 2001). A large number of hydroxylated or keto metabolites of E₂ were found to be formed by male and female human liver microsomes in vitro, and several uncommon E₂ metabolites were also identified. We describe herein our results on the NADPH-dependent metabolism of E₁ by male and female human liver microsomes. There are several reasons that we also characterized the metabolism of E₁ by human liver microsomes: 1) E₁ is a major endogenous estrogen, and it is more lipophilic than E₂; 2) E₂ and E₁ undergo facile enzymatic interconversion in the body; and 3) it has been widely held for decades that 16-OH-E₂ (estriol, E₃), one of the major hydroxy-E₂ metabolites formed in humans and present in the urine in large quantities during pregnancy, is not the product of direct 16-hydroxylation of E₂ but instead it is formed via 16-hydroxylation of E₁ followed by enzymatic 17-reduction (Fishman, 1983).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion Conclusions References

Chemicals. E₂, E₁, NADPH, dimethyldichlorosilane, and ascorbic acid were purchased from the Sigma Chemical (St. Louis, MO). N,O-bis(trimethylsilyl)trifluoroacetamide containing 1% trimethylchlorosilane was obtained from Pierce Chemical (Rockford, IL). All the organic solvents used in this study were of HPLC grade and obtained from Fisher Scientific (Atlanta, GA).

Human Liver Microsomes and Selectively Expressed Human CYP3A4 and CYP3A5. Liver microsomes from 33 human subjects (21 males and 12 females) were obtained from Human Biologics International (Scottsdale, AZ). According to the supplier, the liver tissues were autopsy samples taken at 18.9 ± 12.4 h (mean ± S.D.) after death. The main causes of death included head trauma, intracerebral bleeding, and/or intracranial hemorrhage. The protein content of each microsomal preparation was already adjusted to 20 mg/ml by the supplier. The catalytic activities for several P450 isoforms in these human liver microsomes were also determined by the supplier based on analyzing the metabolism of selective probe substrates (Table 1). Notably, although the use of these probe substrates provided good estimates of the levels of certain P450 isoforms in a given human liver sample, these probes are not entirely specific for the intended P450 isoform.

View this table: [in this window] [in a new window]   TABLE 1 Information on human liver microsomal preparations used in the study Note: The contents of total P450, b5, and cytochrome c reductase and the activities for various P450 isoforms were determined according the methods described (Pearce et al. (1996).

Assay of the NADPH-Dependent Metabolism of [³H]E₁ by Human Liver Microsomes or Human P450 Isoforms. The assay method for the in vitro metabolism of E₁ was the same as described in our recent study with E₂ as substrate (Lee et al., 2001). Briefly, the reaction mixture consisted of 1 mg/ml of human liver microsomal protein or 140 pmol/ml of human CYP3A4 or CYP3A5, a desired concentration of E₁ (containing 2 µCi of [³H]E₁), 2 mM NADPH, and 5 mM ascorbic acid in a final volume of 0.5 ml of 0.1 M Tris-HCl/0.05 M HEPES buffer, pH 7.4. After a 20-min incubation at 37°C with mild shaking, the reaction was terminated and the mixture was extracted with ethyl acetate. The extracts were dried under nitrogen, and the residues were analyzed for E₁ metabolite composition by HPLC. It should be noted that all glass test tubes used in this study were presilanized with 5% (v/v) dimethyldichlorosilane in toluene to avoid loss of hydroxylated estrogen metabolites due to adsorption to the glass surface (Kushinsky and Anderson, 1974).

HPLC Analysis of [³H]E₁ Metabolites. Analysis of [³H]E₁ metabolites was performed with an HPLC system coupled with in-line UV and radioactivity detections as described in detail in our recent study (Lee et al., 2001). The calculation of the rate of [³H]E₁ metabolite formation was solely based on radioactivity measurements. It should be noted that P450 isoform-mediated formation of hydroxylated and keto metabolites of [³H]E₁ at any of its ³H-labeled positions (i.e., 2, 4, 6, and 7) was known to remove tritium from the substrate [³H]E₁, resulting in the formation of [³H]H₂O. Assuming that each of the numerically labeled positions was evenly labeled, 2- or 4-hydroxylation of [³H]E₁ would result in loss of ~25% of the radioactivity in the products, and hydroxylation at the 6 or 7 position would each cause a 12.5% loss of radioactivity. In this study, the calculated final rates for the formation for 2, 4, 6, or 7-hydoxy or keto E₁ metabolites were adjusted according to the estimated loss of radioactivity in each of these products.

Structural Identification of E₁ Metabolites. The identity of each of the major E₁ metabolites formed by human liver microsomes or selectively expressed CYP3A4 and CYP3A5 was confirmed by comparison of its HPLC retention time, its GC/MS retention time, and its mass fragmentation spectrum with each of the 42 authentic reference compounds (Table 2). For comparison, the mass spectrum for each trimethylsilylated authentic standard was obtained with our GC/MS system operated under the same analytical conditions. The GC/MS analysis of E₁ metabolites as well as their authentic standards was performed as described in detail in our recent study (Lee et al., 2001). For spectrum match-up between the metabolically formed E₁ metabolites and the authentic standards, we used both the built-in library search function of our GC/MS system and the manual comparison of their mass spectra.

View this table: [in this window] [in a new window]   TABLE 2 Structural identification of E1 metabolites formed by human liver microsomes Note: Human liver microsomes were incubated with E1 under conditions as described under Materials and Methods. Each peak was collected from the HPLC, derivatized with N,O-bis(trimethylsilyl)trifluoroacetamide containing 1% trimethylchlorosilane, and then analyzed by GC/MS as described under Materials and Methods. Information for 2-MeO-E2, 2-OH-E3, 6-OH-E2, 6-OH-E2, 6-keto-E2, 6-keto-E3, 7-OH-E2, 7-OH-E1, 7-OH-E2, 11-OH-E1, 11-OH-E2, 11-OH-E1, 11-OH-E2, 11-keto-E1, 12-OH-E1, 12-OH-E2, 12-keto-E2, 14-OH-E1, 14-OH-E2, 15-OH-E1, 15-OH-E2, 15-OH-E3, 15-OH-E1, 15-OH-E2, 16-OH-E2, 16-OH-17-E2, 16-OH-E2, 16-OH-17-E2, 16-keto-E1, and 16-keto-E2 are listed in our recent article (Lee et al., 2001).

Statistical Analysis. Nonpaired or paired two-tailed t tests were performed by using the Microsoft Excel software (Microsoft, Redmond, WA) for statistical evaluation of the differences observed between different groups of data. The Pearson correlation coefficients between the rates of formation of E₁ metabolites and the activities of hepatic P450 isoforms were determined by using the SAS software (SAS Institute Inc., Cary, NC).

	Results

Top Abstract Introduction Materials and Methods Results Discussion Conclusions References

Structural Identification of E₁ Metabolites Formed by Human Liver Microsomes

A representative HPLC profile for the [³H]E₁ metabolite peaks detected after incubation of [³H]E₁ with human liver microsomes and NADPH is shown in Fig. 2. The identity of each of the major E₁ metabolites formed was confirmed through comparison of its HPLC retention time, its GC/MS retention time, and its mass fragmentation spectrum with each of the 42 authentic reference compounds (Table 2). Using this triple match-up method, we confirmed the structural identities of the following E₁ metabolite peaks that were detected by our HPLC system: E₂, 2-OH-E₁, 2-OH-E₂, 4-OH-E₁, 4-OH-E₂, 6-OH-E₁, 6-OH-E₁, 6-keto-E₁, 7-OH-E₁, 16-OH-E₁, and 16-OH-E₁. The GC/MS spectra for the TMS-derivatives of y-OH-E₁, 6-OH-E₁, 16-OH-E₁, and 16-OH-E₁, four representative hydroxy-E₁ metabolites formed by human liver microsomes, are shown in Fig. 3, A to D. 

View larger version (46K): [in this window] [in a new window]   Fig. 2.   Representative HPLC profile for the multiple [3H]E1 metabolites formed by human liver microsomes. The methods for the NADPH-dependent metabolism of E1 by human liver microsomes and for the HPLC separation of the estrogen metabolites were described under Materials and Methods. The designation for y-OH-E1 and y-OH-E2 was based on the evidence discussed in the text. GC/MS data indicated that these metabolites are monohydroxy-E1/E2, but did not match any of the known standards. Peaks M1 and M2 are the unidentified radioactive metabolite peaks formed from [3H]E1. Our GC/MS analysis indicated that they were not the monohydroxylated E2 or E1 metabolites. The shaded area under each peak indicates the time span for the collection of the estrogen metabolite(s) for further structural determination by GC/MS analysis.

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Mass fragmentation spectra (A-D) for the TMS-derivatives of the metabolically formed y-OH-E1, 6-OH-E1, 16-OH-E1, and 16-OH-E1, and the GC quantification of 16-OH-E1 and 16-OH-E1 (E). The methods for the GC/MS analysis of the trimethylsilylated estrogen metabolites were described under Materials and Methods. To quantitatively analyze 16-OH-E1 and 16-OH-E1, the HPLC peak for the two coeluted metabolites was collected and converted to TMS-derivatives. Because 16-OH-E1-TMS and 16-OH-E1-TMS have almost the same mass fragmentation patterns with m/z 286 as the base peak, we selected m/z 286 for the selective detection and quantification of 16-OH-E1-TMS and 16-OH-E1-TMS.

The chemical structures of the two metabolite peaks labeled y-OH-E₁ and y-OH-E₂ (Fig. 2) were not fully identified. Based on their HPLC and GC/MS retention times as well as their mass spectra, these same peaks were also detected when E₂ was used as substrate (Lee et al., 2001), but the ratios of y-OH-E₁ to y-OH-E₂ varied with E₁ or E₂ as substrate. As described in our recent study (Lee et al., 2001), the suggestion that these two metabolite peaks are the y-hydroxylated E₁ and E₂ metabolites was based on the following findings: 1) the GC/MS spectra of their TMS-derivatives suggested that they were monohydroxylated E₁ or E₂ metabolites; 2) the ratios of y-OH-E₁ to y-OH-E₂ formed at different E₂ substrate concentrations were almost the same as the ratios of 2-OH-E₁ to 2-OH-E₂; and 3) the overall similarity between the mass fragmentation patterns of y-OH-E₂ and y-OH-E₁. These same features were also observed in this study for y-OH-E₁ and y-OH-E₂ peaks that were formed when E₁ was the substrate. Because the GC/MS spectra of these two hydroxylated E₁ and E₂ metabolites did not match any of the 42 estrogen standards we analyzed, this would leave only a few possibilities, namely, 1-, 8-, 9-, 12-, or 18-hydroxylated E₁ and E₂ as potential candidate metabolites.

For the radioactive peaks M1 and M2 (Fig. 2), the mass spectra of their TMS-derivatives did not match any of the 42 reference compounds, and no characteristic hydroxy-E₁/E₂ mass fragments (m/z of 430 or 340 for keto-E₂ or hydroxy-E₁ metabolites; 504 or 414 for hydroxy-E₂ metabolites) were found in their mass spectra. It is likely that these radioactive peaks are not hydroxylated or keto estrogen metabolites.

Optimization of Assay Conditions and Effect of Varying Substrate Concentrations on Metabolite Formation

Extraction Efficiency and Reproducibility. An average of 93.7 ± 1.4% (mean ± S.D.) of the total radioactivity was recovered when the incubation mixture was extracted once with 8 ml of ethyl acetate. Statistical analysis of triplicate determinations of several [³H]E₁ metabolites (2-OH-E₁, 2-OH-E₂, 4-OH-E₁, y-OH-E₁, and E₂) by three representative human male liver microsomes (HBI-101, HBI-102, and HBI-105) and three female liver microsomes (HBI-112, HBI-217, and HBI-226) showed highly consistent rates, with average intra-assay variations of <5%.

Effect of Incubation pH. We examined the effect of varying incubation pH (from 4.0 to 10.0) on the NADPH-dependent metabolism of [³H]E₁ by human liver microsomes. The formation of most E₁ metabolites (such as 2-OH-E₁, 4-OH-E₁, and y-OH-E₁) showed the highest velocities at pH 7 to 8, but the conversion of E₁ to E₂ and the formation of some E₂ metabolites such as 2-OH-E₂ were optimal at slightly acidic pH (Fig. 4).

View larger version (32K): [in this window] [in a new window]   Fig. 4.   Effect of the incubation pH on the formation of several [3H]E1 metabolites by a human liver microsomal preparation (HBI-115). The incubation mixtures consisted of human liver microsomes (1 mg/ml of microsomal protein), 20 µM E1 (containing 2 µCi of [3H]E1), 2 mM NADPH, and 5 mM ascorbic acid in a final volume of 0.5 ml of 0.1 M Tris-HCl/0.05 M HEPES buffer with incubation pH as indicated on the x-axis. The incubations were carried out for 20 min at 37°C, and the formation of estrogen metabolites was determined by HPLC analysis as described under Materials and Methods. The shaded area indicates the optimal pH range for metabolic formation of each estrogen metabolite. Each point is the mean of duplicate determinations.

Effect of Different E₁ Concentrations on Metabolite Formation. The effect of different [³H]E₁ concentrations on the rate of E₁ metabolite formation by three representative human liver microsomes (HBI-101, HBI-105, and HBI-107) is shown in Fig. 5. The rates of formation of several E₁ metabolites were maximal at 50 µM substrate concentration, and started to decrease somewhat at substrate concentrations greater than 50 µM. In contrast, the conversion of E₁ to E₂ followed typical Michaelis-Menten's kinetics, with K_m values of 59 to 76 µM.

View larger version (31K): [in this window] [in a new window]   Fig. 5.   Effect of different substrate concentrations on the rate of NADPH-dependent metabolism of [3H]E1 by human liver microsomes. The incubation mixtures consisted of human liver microsomes (1 mg/ml of microsomal protein), a different concentration of E1 (containing 2 µCi of [3H]E1), 2 mM NADPH, and 5 mM ascorbic acid in a final volume of 0.5 ml of 0.1 M Tris-HCl/0.05 M HEPES buffer at pH 7.4. The final substrate concentration was 3.1, 6.3, 12.5, 25, 50, 100, or 200 µM. The incubations were carried out for 20 min at 37°C, and the formation of metabolites was determined by HPLC analysis as described under Materials and Methods. Each point is the mean of duplicate determinations.

NADPH-Dependent Metabolism of [³H]E₁ by 33 Human Liver Microsomes

The results for the NADPH-dependent metabolism of 20 µM [³H]E₁ by human male and female liver microsomes are summarized in Table 3. With both male and female liver microsomes, 2-OH-E₁ was the major hydroxyestrogen metabolite detected, followed by 4-OH-E₁. The average rate of overall 2-hydroxylation (formation of 2-OH-E₁ plus 2-OH-E₂) by all 33 human liver microsomes was 80.0 ± 47.3 pmol/mg of protein/min, and the average rate of overall 4-hydroxylation (formation of 4-OH-E₁ plus 4-OH-E₂) was 19.2 ± 11.9 pmol/mg of protein/min. The average ratio of 4-OH-E₁ formation to 2-OH-E₁ formation (0.228) was almost the same as the ratio of overall 4-hydroxylation to overall 2-hydroxylation (0.243). The average rate of y-OH-E₁ formation was 11.0 ± 9.0 pmol/mg of protein/min. The combined rate of formation of 16-OH-E₁ and 16-OH-E₁ (two coeluted metabolites) was 8.1 ± 6.3 pmol/mg of protein/min, about 12% of the rate of 2-OH-E₁ formation (Table 3). Additional GC/MS analysis of the TMS-derivatives of the collected radioactive HPLC peak (from 23.5 to 26.7 min) corresponding to 16-OH-E₁ and 16-OH-E₁ showed a ratio of ~4:1 for these two E₁ metabolites (Fig. 3E). In addition, formation of large amounts of E₂ from E₁ as substrate was also observed.

Several nonpolar estrogen metabolite peaks (collectively labeled as X in Fig. 2) were consistently detected. Quantitatively, these nonpolar estrogen metabolites constitute a very significant fraction of the total amount of [³H]E₁ substrate metabolized (Table 3). The structures of these nonpolar estrogen metabolites were not identified.

In summary, when [³H]E₁ was incubated with either male or female human liver microsomes and NADPH, the four quantitatively major hydroxyestrogen metabolites were 2-OH-E₁, 4-OH-E₁, y-OH-E₁, and 2-OH-E₂ (in a decreasing order). Several other E₁ metabolites, including 4-OH-E₂, 6-OH-E₁, 16-OH-E₁, and 16-OH-E₁, were also formed in substantial quantities. Small amounts of 6-OH-E₁, 6-keto-E₁, 7-OH-E₁, and y-OH-E₂ were also detected. The overall profiles of E₁ metabolites formed by male or female human liver microsomes were not significantly different.

Correlation of Selective Activities of P450 Isoforms with Rates of Formation of [³H]E₁ Metabolites

To probe which P450 isoform(s) may be responsible for E₁ hydroxylation at specific position(s), we analyzed the correlation coefficients between the rate of formation of each E₁ metabolite and the activities of several selective P450 isoforms. These data are summarized in Table 4 and Fig. 6. The total P450 content in human liver microsomes showed a high degree of correlation (P < 0.001) with the rate of formation of most hydroxyestrogen metabolites, suggesting that P450 enzymes constitute the major catalytic activity in human liver microsomes for the NADPH-dependent metabolism of E₁. Among the P450 isoforms examined, the catalytic activity of CYP3A4 (according to dextromethorphan N-demethylation) or CYP3A4/5 (according to testosterone 6-hydroxylation) showed high degrees of correlation (P < 0.001) with the rate of formation of several E₁ metabolites. In addition, the catalytic activity of CYP2B6 (based on S-mephenytoin N-demethylation) was also highly correlated (P < 0.001) with the rate of formation of several metabolites. Liver microsomes from female human subjects showed somewhat higher degrees of correlation between the rate of formation of most of E₁ metabolites and the corresponding P450 isoform activity compared with male liver microsomes.

View this table: [in this window] [in a new window]   TABLE 4 Correlation coefficients (r) for comparing the formation of various E1 metabolites with the activities of CYP1A2, 2B6, 3A4, or 3A4/5 by measuring the metabolism of probe substrates in 33 human liver microsomes (21 male and 12 female) Note. Microsomal formation of various estrogen metabolites was as described in the legend to Table 3. The CYP1A2, 2B6, 3A4, or 3A4/5 activities in human liver microsomes were determined by the supplier by measuring the activity for caffeine N3-demethylation, S-mephenytoin N-demethylation, dextromethorphan N-demethylation, or testosterone 6-hydroxylation, respectively, according to the published methods (Pearce et al. 1996). Exceptionally high correlation coefficients (r  0.9) are shown in bold. *P < 0.05; **P < 0.01; ***P < 0.001.

View larger version (55K): [in this window] [in a new window]   Fig. 6.   Analysis of the correlation coefficients between the formation of several major hydroxy-E1 metabolites (2-OH-E1, 4-OH-E1, 16-OH-E1 + 16-OH-E1, and y-OH-E1) and the activities for hepatic CYP3A4/5, CYP3A4, and CYP2B6. The rate of formation of 2-OH-E1, 4-OH-E1, 16-OH-E1 + 16-OH-E1, and y-OH-E1 was determined by incubation of human liver microsomes with 20 µM [3H]E1 and 2 mM NADPH as described under Materials and Methods. The Pearson correlation coefficients (r) for liver microsomes from all 21 male subjects (M), from all 12 female subjects (F), and from all male and female subjects combined (T) were determined by using the SAS software (SAS Institute Inc.).

NADPH-Dependent Metabolism of [³H]E₁ by Human CYP3A4 and CYP3A5

Incubations of [³H]E₁ with selectively expressed human CYP3A4 or CYP3A5 in the presence of NADPH resulted in formation of several hydroxylated estrogen metabolites. Figure 7 shows representative HPLC profiles for the radioactive E₁ metabolites formed by human CYP3A4 and CYP3A5 at a 20 µM [³H]E₁ concentration. The overall profiles for the E₁ metabolites formed by CYP3A4 looked similar to those formed by male or female human liver microsomes (compare Fig. 7, top, with Fig. 2). 2-OH-E₁ was the quantitatively major hydroxy-E₁ metabolite formed by CYP3A4, followed by y-OH-E₁ and 4-OH-E₁. Quantitatively, the rate for 2-OH-E₁ formation (mean ± S.D. of triplicate determinations) by CYP3A4 was 282.8 ± 5.9 pmol/nmol of P450/min, and the rate for 4-OH-E₁ formation was 146.9 ± 14.3 pmol/nmol of P450/min, giving a ratio of 4-OH-E₁ formation to 2-OH-E₁ formation of ~0.52.

View larger version (39K): [in this window] [in a new window]   Fig. 7.   Representative HPLC profiles and the calculated rates for the NADPH-dependent [3H]E1 metabolites formed by selectively expressed human CYP3A4 and CYP3A5. The incubation mixtures consisted of 140 pmol/ml of CYP3A4 or CYP3A5, 20 µM E1 (containing 2 µCi of [3H]E1), 2 mM NADPH, and 5 mM ascorbic acid in a final volume of 0.5 ml of 0.1 M Tris-HCl/0.05 M HEPES buffer, pH 7.4. The incubations were carried out for 20 min at 37°C, and the formation of estrogen metabolites was determined by HPLC analysis as described under Materials and Methods. For the descriptions of the peaks labeled y-OH-E1, M1, and M2, refer to the legend to Fig. 2. The rate of formation of each metabolite was calculated by taking into consideration the loss of radioactivity during estrogen metabolite formation and by subtracting the background values obtained with control microsomes (without selectively expressed P450 isoforms).

Selectively expressed human CYP3A5, however, showed a quite different profile for E₁ metabolism (Fig. 7, bottom). 4-OH-E₁ became the quantitatively major hydroxyestrogen metabolite, followed by y-OH-E₁ and 2-OH-E₁. The rate for 2-OH-E₁ formation (mean ± S.D. of triplicate determinations) by CYP3A5 was 37.5 ± 10.4 pmol/nmol of P450/min, whereas the rate for 4-OH-E₁ formation was 72.7 ± 8.1 pmol/nmol of P450/min, almost doubling the rate of 2-OH-E₁ formation.

Several other hydroxy-E₁ metabolites (6-OH-E₁, 16-OH-E₁, and 16-OH-E₁) were also formed in substantial quantities by either CYP3A4 or CYP3A5. In addition, large amounts of nonpolar metabolites (collectively labeled as X in Fig. 7) were formed during incubations of E₁ with CYP3A4 or CYP3A5. However, compared with human liver microsomes, the selectively expressed CYP3A4 or CYP3A5 contained much lower catalytic activity for the formation of E₂ and several hydroxy-E₂ metabolites when E₁ was used as the substrate.

The structural identities for the major hydroxy-E₁ metabolites (2-OH-E₁, 4-OH-E₁, 6-OH-E₁, 16-OH-E₁, 16-OH-E₁, and y-OH-E₁; Fig. 7) that were formed by CYP3A4 and CYP3A5 were also confirmed by GC/MS analysis (data not shown).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion Conclusions References

Liver Microsomal Hydroxylation of E₁ at Various Positions

In a recent study, we characterized the NADPH-dependent metabolism of E₂ to a large number of metabolites by human male and female liver microsomes (Lee et al., 2001). We describe herein our results on the oxidative metabolism of E₁ by the same sets of human liver microsomes. We identified 11 hydroxylated or keto metabolites formed from [³H]E₁. The structural identities of 11 E₁ metabolites were confirmed based on the close match-ups of the HPLC retention times, GC/MS retention times, and mass fragmentation spectra of each formed metabolite with the corresponding authentic reference compound (Table 2). In comparison, 18 oxidative metabolites were identified under the same experimental conditions with E₂ as substrate (Lee et al., 2001). When estimated according to substrate disappearance, the overall rate (292.0 pmol/mg of protein/min) of metabolism of 20 µM E₁ as substrate is slightly slower than the metabolism of 20 µM E₂ as substrate (375.8 pmol/mg of protein/min; Lee et al., 2001). The NADPH-dependent formation of each of the major classes of E₁ metabolites by human liver microsomes is separately discussed below. Our results obtained from this study on the metabolism of E₁ are closely compared with our recent results obtained under the same in vitro metabolic conditions with E₂ as substrate (Lee et al., 2001).

Metabolism at C2 and C4. The 2-hydroxylation of E₂ to a catechol is known to be a major metabolic pathway, whereas its 4-hydroxylation is a quantitatively minor pathway (usually <15% of 2-hydroxylation) in rodent and human livers (Dannan et al., 1986; Kerlan et al., 1992; Suchar et al., 1995; Zhu and Conney, 1998a; Lee et al., 2001). The data of this study showed that 2-hydroxylation of E₁ was also the major hydroxylation pathway with both male and female human liver microsomes. Although the overall rate of metabolism of 20 µM E₁ by these human liver microsomes was slightly slower than that with 20 µM E₂ under the same in vitro metabolism conditions, the average rates of 2- and 4-hydroxylation of E₁ were 80.0 and 19.2 pmol/mg of protein/min, respectively, which were 38 and 76% faster than the rates of 2- and 4-hydroxylation of E₂. These differences were very significant (P < 10⁵, paired two-tailed t test). In addition, the ratio of E₁ 4-hydroxylation to 2-hydroxylation (0.24 ± 0.05) is slightly higher than the ratio with E₂ as substrate (0.20 ± 0.04; P < 0.001, paired two-tailed t test). Therefore, it can be concluded that E₁ is converted to catechol estrogens at a somewhat faster rate than E₂, especially to 4-hydroxyestrogens. The potential importance of 4-hydroxylated estrogen metabolites (such as 4-OH-E₂ and 4-OH-E₁) in hormonal carcinogenesis has received a great deal of attention in the past few years (Liehr, 2000). These catechol estrogens not only serve as intermediates for the formation of reactive chemical species (Cavalieri et al., 2000; Liehr, 2000) but also 4-OH-E₂ may have its own signal transduction pathway that is refractory to E₂ (Das et al., 1997).

Metabolism at C6. It has been known for decades that E₂ could be metabolized to 6-OH-E₂ and 6-OH-E₂ in both animals and humans (Mueller and Rumney, 1957; Breuer et al., 1966). A later study also reported that 6-OH-E₁ and 6-OH-E₁ were the major metabolites formed after incubation of E₁ with porcine uterine endometrial tissues (Maschler et al., 1983). In our recent study when E₂ was the substrate (Lee et al., 2001), formation of both 6- and 6-OH-E₂ by human liver microsomes was observed. The results of our present study showed that 6-OH-E₁ was formed from E₁ in significant quantities by both male and female human liver microsomes, but 6-OH-E₁ was formed only in very small quantities (<2.0 pmol/mg of protein/min).

Metabolism at C16. It has been suggested for many years that 16-OH-E₁ plays an important role in mammary carcinogenesis (Fishman et al., 1984; Bradlow et al., 1986). This estrogen metabolite is not only hormonally active but also chemically reactive and may bind covalently to the estrogen receptor, possibly resulting in sustained hormonal stimulation (Swaneck and Fishman, 1988; Hsu et al., 1991). Because of these biological properties of 16-OH-E₁, several earlier studies have examined the NADPH-dependent 16-hydroxylation of E₁ and E₂ in human subjects (Osborne et al., 1993) or by human hepatic and extrahepatic tissues or cells in vitro (Shou et al., 1997; Huang et al., 1998; Yamazaki et al., 1998). The results of our recent study showed that human liver microsomes only form very minute amounts of 16-OH-E₂ and 16-OH-E₁ when [³H]E₂ was used as substrate (Lee et al., 2001). In the present study, we detected the formation of 16-OH-E₁ and 16-OH-E₁ (two coeluted metabolites) by human liver microsomes at a combined average rate of 8.1 ± 6.3 pmol/mg of protein/min (Table 3). GC/MS analysis of these two 16-hydroxy-E₁ metabolites formed by representative human liver microsomes showed a ratio of ~4:1 between 16-OH-E₁ and 16-OH-E₁ (Fig. 3E). It is estimated that the overall rate of formation of 16-OH-E₁ and 16-OH-E₁ from 20 µM E₁ as substrate was similar to the overall rate of formation of 16-OH-E₂ and 16-OH-E₂ from 20 µM E₂ (6.8 pmol/mg of protein/min; Lee et al., 2001) with the same sets of human liver microsomes under the same reaction conditions. Although 16-hydroxylation is not a major metabolic pathway for E₂ and E₁ in human liver microsomes, our data suggest that E₂ and E₁ are both hydroxylated at the 16- and 16-positions by human liver microsomes. This conclusion is in contrast to the long-held view that 16-hydroxylation only occurs with E₁ as the substrate (Fishman, 1983).

y-Hydroxylation. We found that y-hydroxylation of E₁ is one of the quantitatively major hydroxylation pathways, as also observed recently with E₂ as substrate (Lee et al., 2001). Candidates for the identity of y-OH-E₁ and y-OH-E₂ include 1-, 8-, 9-, 12-, and 18-OH-E₁/E₂ (Lee et al., 2001). More studies are needed to identify the structures of y-OH-E₁ and y-OH-E₂. It will also be of interest to determine whether there are any biological functions that may be associated with these newly identified estrogen metabolites.

Metabolism at C17. It is known that the interconversion between E₂ and E₁ is largely catalyzed by the oxidative and reductive activities of 17-hydroxysteroid dehydrogenases in the presence of suitable cofactors. Our data showed that the optimal pH for the conversion of E₁ to E₂ by human liver microsomes was slightly acidic (pH 6-7; Fig. 4), a condition that is different from the slightly basic pH required for the optimal conversion of E₂ to E₁ by the same microsomes (Lee et al., 2001). At the optimal pH for each reaction, the reductive metabolism of 20 µM E₁ to E₂ had a similar rate as the oxidative metabolism of 20 µM E₂ to E₁; however, at pH 7.4, the conversion of E₁ to E₂ proceeded markedly slower than the conversion of E₂ to E₁. In addition, at pH 7.4, the K_m values for the conversion of E₁ to E₂ (59-76 µM, calculated according to double-reciprocal plot) were much higher than the K_m values for the conversion of E₂ to E₁ (12-20 µM; Lee et al., 2001), suggesting that the reductive activity of 17-hydroxysteroid dehydrogenase(s) has a lower affinity for E₁ than does the oxidative activity for E₂. These results may also suggest that the conversion of E₂ to E₁ probably is preferred under physiological conditions, which is in agreement with several earlier observations showing that E₁ usually has higher tissue and blood levels than E₂ in humans (Judd et al., 1976; Jasonni et al., 1983; Lum et al., 1997).

Formation of Nonpolar Estrogen Metabolites. Large amounts of several unidentified nonpolar metabolites (collectively labeled as X in Fig. 2) were consistently detected after incubation of [³H]E₁ with each of the 33 human liver microsomal preparations. Similar nonpolar metabolites were also observed in our recent study with E₂ as substrate. These nonpolar estrogen metabolites were also formed by selectively expressed human CYP3A4 and CYP3A5 (Fig. 7). However, it should be noted that several other human P450 isoforms (such as CYP1A2, CYP1B1, and CYP2B6) almost completely lacked catalytic activity for the formation of the nonpolar metabolites from E₂ or E₁ as substrate (data not shown). These data suggested that their formation may be selectively mediated by certain P450 isoforms. Additional studies are warranted to gain more knowledge about these nonpolar estrogen metabolites.

Effect of Varying E₁ Concentrations on Metabolite Formation. Our data showed that the rates of formation of most E₁ metabolites reached maximum at 50 µM substrate concentration and then started to decline at higher E₁ concentrations (Fig. 5). In comparison, when E₂ was used as a substrate in our recent study (Lee et al., 2001), the rates of formation of several E₂ metabolites increased continuously with increasing E₂ concentrations up to the highest concentration (200 µM) tested, although the formation of certain metabolites started to plateau at >100 µM E₂. The very different curve patterns observed with E₂ and E₁ as substrate was rather interesting, and they suggest a possible substrate-mediated inhibition of the metabolizing enzymes in the case of E₁. Notably, some of the kinetic features of the substrate-mediated inhibition of human CYP3A4 as well as some other P450 isoforms have recently been reported (Lin et al., 2001).

Interindividual Variations and Effects of Gender on Liver Microsomal Metabolism of E₁

It is known that large interindividual variations exist in the activities of various metabolizing enzymes in human liver (Conney, 1982; Forrester et al., 1992; Shimada et al., 1994; Transon et al., 1996; Lin and Lu, 2001). Such variations are attributable to the effects of both genetic and environmental factors (Conney, 1982; Guengerich and Shimada, 1991; Wrighton and Stevens, 1992). The results of the present study showed that human liver P450-mediated metabolism of E₁ to various hydroxylated or keto metabolites has large interindividual variations (Table 3), and this observation is similar to what was observed in our recent study with E₂ as substrate (Lee et al., 2001).

It is known that some P450 isoforms present in liver or extrahepatic tissues of experimental animals are gender-specific (MacGeoch et al., 1984; Waxman, 1984; Waxman et al., 1985; Bandiera et al., 1986). However, when 30 Caucasians and 30 Japanese liver samples were analyzed (Shimada et al., 1994), no sex-related differences were observed with respect to the contents of several major P450 isoforms and their activities for metabolizing certain xenobiotics. Comparison of 21 male and 12 female human liver microsomes used in this study also showed that CYP1A2, CYP2A6, CYP2B6, CYP2C9, CYP2C19, CYP3A4, CYP3A4/5, and CYP4A11 (several of them are known to have high estrogen-metabolizing activities) did not show gender-related differences in their enzymatic activities. However, CYP2D6 and CYP2E1 (both of which have no detectable activity for estrogen metabolism; Cai et al., 1998) exhibited some gender-related differences (P < 0.05; Table 1). Our further comparison of the rates of formation of various E₁ metabolites by these liver microsomes did not indicate any gender-related differences (Table 3), which has been expected. A similar lack of gender-related differences was also noted in our recent study for the metabolism of E₂ by male and female human liver microsomes (Lee et al., 2001).

Role of Human CYP3A4 and CYP3A5 in Hepatic E₁ Metabolism

Isoforms of the P450 family enzymes are the major catalysts for the NADPH-dependent oxidative metabolism of endogenous and exogenous estrogens to various hydroxylated and keto metabolites in animals and humans. The CYP3A family enzymes are the most abundant P450 isoforms present in human liver (Thummel and Wilkinson, 1998; Guengerich, 1999). An earlier study suggested that CYP3A4 and CYP3A5 in human liver microsomes were responsible for up to 80% of estrogen 2- and 4-hydroxylation (Kerlan et al., 1992). Our data showed that testosterone 6-hydroxylation activity (a selective enzymatic activity marker for CYP3A4/5) was highly correlated with the formation of most hydroxylated E₁ metabolites (P < 0.001). The activity of testosterone 6-hydroxylation showed a somewhat higher degree of correlation with E₁ hydroxylation than did dextromethorphan N-demethylation (a selective marker for CYP3A4 but not for CYP3A5), which suggests that CYP3A5 may also be an important contributor to E₁ metabolism in human liver microsomes.

To further confirm the roles of CYP3A4 and CYP3A5 in hepatic metabolism of E₁, we analyzed the metabolism of [³H]E₁ by selectively expressed human CYP3A4 and CYP3A5. The overall profile of [³H]E₁ metabolites formed by human CYP3A4 was similar to the profiles of metabolites formed by various human liver microsomes. E₁ 2-hydroxylation was the major hydroxylation pathway with CYP3A4. However, the ratio of 4-OH-E₁ formation to 2-OH-E₁ formation by CYP3A4 was ~0.52, which is much higher than the ratio observed with E₂ as substrate (~0.19; Lee et al., 2001). Surprisingly, when selectively expressed human CYP3A5 was the catalyst, E₁ 4-hydroxylation became a quantitatively major metabolic pathway, at rates almost twice as fast as E₁ 2-hydroxylation. The ratio of 4-hydroxylation to 2-hydroxylation for E₁ (1.94) was ~4-fold higher than the ratio for E₂ (0.51; Lee et al., 2001). To our knowledge, this is the first demonstration that human CYP3A5 has a higher catalytic activity for the formation of 4-OH-E₁ than 2-OH-E₁. Because human CYP3A5 is a polymorphic P450 isoform (Wrighton et al., 1989), it will be of considerable interest to determine whether its polymorphism correlates with the amount of 4-hydroxylated estrogens formed in humans and also the risk of estrogen-associated human cancers.

In addition to catechol estrogen metabolites, significant amounts of 6-OH-E₁, 16-OH-E₁, 16-OH-E₁, y-OH-E₁, and a cluster of nonpolar estrogen metabolites were also formed by CYP3A4 and CYP3A5, suggesting an important role for these two P450 isoforms in the hepatic formation these estrogen metabolites. Besides the CYP3A family isoforms, CYP1A2 is another major P450 isoform that has high catalytic activity for NADPH-dependent metabolism of E₁ to catechol estrogens (Shou et al., 1997; Cai et al., 1998; Yamazaki et al., 1998). However, our data showed that the hepatic CYP1A2 activity (using caffeine N3-demethylation as a selective probe) was only weakly correlated with the formation of several E₁ metabolites (Table 4). This may be due to the relatively low levels of CYP1A2 present in most human liver microsomes compared with other abundantly expressed estrogen-metabolizing P450 isoforms.

	Conclusions

Top Abstract Introduction Materials and Methods Results Discussion Conclusions References

Following our recent study of E₂ metabolism by human liver microsomes (Lee et al., 2001), we describe herein our results on the characterization of E₁ metabolism to multiple hydroxylated or keto metabolites by male and female human liver microsomes. The overall profiles for the E₁ metabolites formed by male or female human liver microsomes appeared to be not significantly different from each other. Quantitatively, the major hydroxyestrogen metabolites formed were 2-OH-E₁, 4-OH-E₁, and y-OH-E₁. Several other E₁ metabolites, including 2-OH-E₂, 4-OH-E₂, 6-OH-E₁, 6-keto-E₁, 16-OH-E₁, and 16-OH-E₁ were also formed in substantial quantities. In addition, small amounts of 6-OH-E₁ and 7-OH-E₁ were also detected. Overall, E₁ was metabolized mainly at 2-, 4-, 6-, 16-, and y-positions by NADPH-dependent hepatic enzymes, whereas E₂ was found in our recent study to be metabolized at a greater number of positions (Lee et al., 2001). E₁ is more prone to be metabolized to catechol estrogens (in particular 4-hydroxyestrogens) than E₂ by human liver microsomes. CYP3A family enzymes have a dominant role in hepatic metabolism of E₁ to 2-, 4-, and y-OH-E₁ metabolites. CYP3A5 showed unusually high activity for E₁ 4-hydroxylation, exceeding its activity for 2-hydroxylation by ~100%. This is the first demonstration that human CYP3A5 (a polymorphic P450 isoform; Wrighton et al., 1989) has higher catalytic activity for the formation of 4-OH-E₁ than 2-OH-E₁. It will be of great interest to determine whether its polymorphism correlates with the amount of 4-hydroxylated estrogens formed in humans and also the risk of estrogen-associated human cancers. It will also be of interest to determine the potential physiological or pathophysiological functions that may be associated with the E₁ metabolites identified in our present study. Research in this area may lead to an enhanced understanding of the important diverse biological actions associated with endogenous estrogens.