Introduction

Top Abstract Introduction Materials and methods Results Discussion References

Mifepristone (RU486), (11, 17)-11-[4-(dimethylamino)-phenyl]-17-hydroxy-17-(1-propynyl)-estra-4,9-dien-3-one (Fig. 1), has been used clinically in Europe and other countries as an antiprogestin agent for medical abortion in the first trimester of pregnancy (Spitz and Bardin, 1993). Several other potential applications have also been investigated. They include the treatment of breast cancer, prostate cancer, meningioma, and uterine leiomyoma; the induction of labor after fetal death or at the end of the third trimester; and the treatment of Cushing's syndrome (Spitz and Bardin, 1993). Mifepristone has been shown to be metabolized to several metabolites by cytochrome P-450 (CYP) (Chasserot-Golaz et al., 1990). Recently, CYP-3A4 was identified as the principal enzyme catalyzing the N-demethylations of the 11-dimethylaminophenyl group and the hydroxylation of the 17-propynyl group, the three major metabolic pathways in human liver microsomes (Jang et al., 1996). The hydroxylation of the 17-propynyl group catalyzed by CYP-3A4 suggests that the carbon-carbon triple bond may be positioned at the active site of the enzyme. The oxidation of carbon-carbon triple bonds by CYPs is believed to form an intermediate that can irreversibly modify critical active site moieties and thus inactivate the enzyme in a process characterized as mechanism-based inactivation (Ortiz de Montellano and Correia, 1995).

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Structure of mifepristone (RU486). *Site of tritium labeling.

Mechanism-based inactivators of CYPs have drawn a great amount of attention because, first, derivatives of the inactivators covalently label moieties in the active site of the enzyme. Determination of the structure of the heme adduct has provided information that is very valuable for understanding the topology of the active site and the mechanism of action of the enzyme as well as for drug design (Ortiz de Montellano and Correia, 1995). Structural identification of modified active site peptides has been facilitated in recent years by the emergence of matrix-assisted laser desorption ionization and electrospray mass spectrometry and some other new techniques (Roberts et al., 1994 and He et al., 1996a). The use of mechanism-based inactivators is rapidly becoming an important approach for characterizing the peptides or amino acid residues involved in the active site of CYPs. Second, the inhibition of the catalytic activity of various enzymes is more specific than that seen with other inhibitors. Such features of mechanism-based inactivators have greatly facilitated the design of drugs such as lanosterol 14-demethylase inhibitors developed as fungicide agents (Frye and Robinson, 1990; Ortiz de Montellano and Correia, 1995) and aromatase inhibitors for the treatment of breast cancer (Marcotte and Robinson, 1982; Ortiz de Montellano and Correia, 1995). Large numbers of acetylenes, particularly those synthetic steroids such as gestodene, norethisterone, ethinyl estradiol, norgestrel, and danazol have been demonstrated to cause mechanism-based inactivation of CYPs (Guengerich, 1990; Ortiz de Montellano and Correia, 1995). However, most of the alkynes that inactivate CYPs are terminal acetylenes. Recently, studies have shown that internal acetylenes such as several different methyl-substituted aryl acetylenes (propynylaryl acetylenes) and 10-dodecynoic acid also cause mechanism-based inactivation of CYPs (Foroozesh et al., 1997; Helvig et al., 1997). Our present study demonstrates that mifepristone, an internal acetylene that has a methyl group that substitutes for the hydrogen on the external carbon of the triple bond, is a potent and selective mechanism-based inactivator of CYP-3A4 via irreversible modification of the apoprotein.

	Materials and Methods

Top Abstract Introduction Materials and methods Results Discussion References

Chemicals. NADPH, L--dilauroyl- and L--dioleyl-sn-glycero-3-phosphocholines, phosphatidyl serine, catalase, glutathione (GSH), -aminolevulinic acid hydrochloride, testosterone, 6- and 11- hydroxytestosterone, p-nitrophenol, -naphthoflavone (-NF), phenobarbital (PB), pyridine, and dexamethasone (DEX) were purchased from Sigma Chemical Co. (St. Louis, MO). Isopropyl -D-thiogalactoside was purchased from Calbiochem Corp. (La Jolla, CA). 7-Ethoxy-4-trifluoromethylcoumarin (EFC), resorufin, methoxyresorufin, and benzyloxyresorufin were purchased from Molecular Probes, Inc. (Eugene, OR). Mifepristone was a gift from Dr. Paul R. Housley at the University of South Carolina Medical School and [³H]mifepristone, originally obtained from ROUSSEL-UCLAF (France), was a gift from Dr. William B. Pratt at the University of Michigan. The specific activity of the [³H]mifepristone is 38.4 Ci/mmol.

Expression of CYP-3A4 and Purification of Expressed Enzyme. A 5' end-modified CYP-3A4 cDNA constructed as described elsewhere (Gillam et al., 1993) and inserted into the pCW vector was obtained from Dr. R.W. Estabrook (University of Texas Southwestern Medical Center, Dallas, TX). The CYP-3A4-containing vector was transformed into MV1304 cells. Growth of the transformed Escherichia coli was carried out in modified Terrific Broth and the expression of CYP-3A4 was induced by the addition of 1 mM isopropyl -D-thiogalactoside. -Aminolevulinic acid (0.5 mM) was added to increase heme synthesis. The membrane fraction was prepared from the bacterial cells by sonication after treatment with lysozyme and subsequently isolated from the bacterial cell homogenate by differential centrifugation. CYP-3A4 was purified to homogeneity using the method described by Gillam et al. (1993).

Coexpression of CYP-2D6 and NADPH-CYP Reductase in E. coli and Determination of CYP-2D6 Activity. The pCW vector containing CYP-2D6 cDNA and the pACYC vector containing human NADPH-CYP reductase were obtained from Dr. Thomas Friedberg (University of Dundee, UK). The two vectors were transformed into JM109 cells. Growth of the transformed E. coli and the preparation of the bacterial membrane were carried out using the same conditions as previously described for CYP-3A4. To measure the enzyme activity, the bacterial membrane containing 50 pmol CYP-2D6 was incubated with EFC at various concentrations in 0.5 ml of 0.1 M potassium phosphate buffer (pH 7.4) containing 10 mM MgCl₂ and 0.1 mM EDTA at 37°C for the time period indicated. The reaction was started by addition of NADPH at a final concentration of 1 mM and stopped by addition of 0.3 ml of cold acetonitrile. 7-Hydroxy-4-trifluoromethylcoumarin was determined by fluorimetric detection using an excitation wavelength of 410 nm and an emission wavelength of 510 nm on a SLM-AMINCO spectrofluorometer (SLM-AMINCO, Urbana, IL). The EFC O-deethylation activity was linear with time during the 30- min incubation. The values of K_m and V_max were determined to be 142 µM and 14 nmol of 7-hydroxy-4-trifluoromethylcoumarin formed/nmol/min.

Isolation of CYP-2B1, NADPH-CYP Reductase, and Cytochrome b₅ and Preparation of Liver Microsomes. Male Fischer 344 rats (161-190 g; Harlan Sprague-Dawley, Indianapolis, IN) were treated i.p. with -NF (80 mg/kg), PB (80 mg/kg), pyridine (100 mg/kg), or DEX (100 mg/kg) for 3 to 4 days, respectively. Liver microsomes were prepared by differential centrifugation. CYP-2B1, NADPH-CYP reductase, and cytochrome b₅ were purified from liver microsomes of phenobarbital-treated rats by the methods described previously (Strobel and Dignam, 1978; Waxman and Walsh, 1982)

Mifepristone-Mediated Inactivation of CYP-3A4 in a Reconstituted System. CYP-3A4 (0.5 nmol) was reconstituted with 20 µg of a mixture (1:1:1) of L--dilauroyl- and L--dioleyl-sn-glycero-3-phosphocholines and phosphatidyl serine, 200 µg of cholic acid, 1 nmol of NADPH-CYP reductase, 0.5 nmol of cytochrome b₅, 500 U of catalase, 2 µmol of GSH, 30 mM MgCl₂, 0.5 mM EDTA, and 20% glycerol in a final volume of 1 ml of 50 mM HEPES buffer (pH 7.5). Reactions with various concentrations of mifepristone were initiated by the addition of 1 mM NADPH and stopped by cooling on ice. The incubations were performed at 37°C for the time periods indicated. At the end of the incubation, 0.2 ml of the incubation mixture was diluted into 0.8 ml of 50 mM HEPES buffer (pH 7.5) containing 20% glycerol and 0.5 mM EDTA. The spectra were recorded between 330 to 700 nm against the diluting buffer as the reference on a DW2-OLIS spectrophotometer in the split beam mode. An aliquot of 0.25 ml was used for the determination of the P-450 content by the method of Omura and Sato (1964). Additional aliquots were taken for determination of testosterone 6-hydroxylation activity and high-performance liquid chromatography (HPLC) analysis as described below.

Determination of Testosterone 6-Hydroxylation Activity. An aliquot (0.05 ml) of the incubation mixture was diluted into 0.95 ml of 50 mM HEPES buffer (pH 7.5) containing 200 µM testosterone, 500 U of catalase, 2 µmol of GSH, 30 mM MgCl₂, 0.5 mM EDTA, and 20% glycerol in a final volume of 1 ml, and incubated for 10 min at 37°C. The reaction was started by the addition of 1 mM NADPH and was stopped by addition of an equal volume of ethyl acetate. 6-Hydroxytestosterone was determined by HPLC on a C₁₈ column (Microsorb-MV, 5 µm, 4.6 × 150 mm; Rainin, Woburn, MA) eluted isocratically with a mobile phase of 65% methanol at flow rate of 1 ml/min and the eluate was monitored by UV detection at 254 nm.

HPLC Analysis of Mifepristone-Inactivated CYP-3A4. After a 15-min incubation of CYP-3A4 with 1 µM [³H]mifepristone in the reconstituted system as described above, 200 µL of the reaction mixture was directly analyzed on a Poros column (R/H, 4.6 × 100 mm, S/N 195; PerSeptive Biosystems, Framingham, MA) as described previously (Roberts et al., 1993). The eluate was monitored by visible-UV detection at 214 and 405 nm simultaneously.

SDS-PAGE Analysis of Mifepristone-Inactivated CYP-3A4 and Autoradiography. The incubation mixtures containing [³H]mifepristone-inactivated CYP-3A4 or the [³H]mifepristone-NADPH control were directly analyzed by SDS-PAGE on a 7.5% gel. After staining with Coomassie blue and destaining, the gel was treated with autofluor (National Diagnostics, Atlanta, GA) for 30 min. The dried gel was exposed to Kodak Scientific Imaging film (Eastman Kodak Co., Rochester, NY) at 80°C for a week before developing.

Covalent Binding of Mifepristone to apoCYP. CYP-3A4 was inactivated by incubating with 50 µM mifepristone and 0.5 µM [³H]mifepristone in the reconstituted system at 37°C for 15 min. Aliquots of the incubation mixture containing 50 to 125 pmol of purified CYP-3A4 were taken and added to 10 mg of bovine serum albumin as carrier protein after the reaction was stopped by cooling in ice. The covalent binding of mifepristone to the protein was determined according to the method described elsewhere (Chan et al., 1993). The protein was then immediately precipitated by adding a 10-fold volume of 5% sulfuric acid in methanol. The pellet was washed with the same solvent until no [³H]mifepristone was detectable in the supernatant and then it was finally washed with 80% methanol. The pellet was then dissolved in 100 µL of 1 N NaOH at 60°C. An aliquot was neutralized with an equal amount of HCl before scintillation counting.

Detection of Heme Adduct Formation. The method described by He et al. (1996b) was used to check for the possible formation of heme adducts. The incubation mixture was extracted with butanone containing 10% trifluoroacetic acid (TFA). The extracts were dried using a speed vacuum and then analyzed by HPLC on a C₄ column (VYDAC 214TP54, 4.6 × 250 mm; VYDAC, Hesperia, CA) eluted with solvent A (0.1% TFA in water) and solvent B (95% acetonitrile and 0.1% TFA) using a linear gradient from 30 to 70% B within 30 min. The eluate was monitored by UV-visible detection at 400 and 415 nm. The spectrum of the inactivated CYP-3A4 was also recorded against the sample of NADPH control or the sample with NADPH added but kept in ice.

Effect of Mifepristone on Activities of Several Other CYPs. The activities of CYPs 1A, 2B, 2E1, and 3A2 were determined using methoxyresorufin O-demethylation, benzyloxyresorufin O-dealkylation, p-nitrophenol hydroxylation, and testosterone 6-hydroxylation with microsomes from -NF-, PB-, pyridine-, and DEX-induced rat livers, respectively (Reinke and Moyer, 1985; Sonderfan et al., 1987; Nerurkar et al., 1993). The activity of CYP-2D6 was determined using EFC O-deethylation by the expressed bacterial membrane containing CYP-2D6 and NADPH-CYP reductase using the method described above. Rat liver microsomes or the bacterial membrane containing CYP-2D6 and NADPH-CYP reductase were incubated with 1, 10, and 100 µM mifepristone in the presence of methoxyresorufin (5 µM), benzyloxyresorufin (5 µM), p-nitrophenol (100 µM), testosterone (200 µM), and EFC (200 µM) for 15 min at 37°C, respectively. The reactions were initiated by the addition of 1 mM NADPH and stopped by cooling on ice. The effect of mifepristone on the activities of CYP-2B1 and CYP-3A4 were carried out using the same conditions used for the purified CYP-2B1 reconstituted system (He et al., 1996a) and for the purified CYP-3A4 reconstituted system as described above, respectively.

	Results

Top Abstract Introduction Materials and methods Results Discussion References

NADPH-, Time-, and Concentration-Dependent Inactivation of CYP-3A4 by Mifepristone. The testosterone 6-hydroxylation activity of CYP-3A4 decreased by 81% after incubation of the enzyme with 50 µM mifepristone for 15 min in the presence of NADPH in the reconstituted system containing purified CYP-3A4, CYP-NADPH reductase, cytochrome b₅, and lipids (Table 1). As shown in Fig. 2, the inactivation exhibited pseudo-first-order kinetics when several different concentrations of mifepristone were used. Linear regression analysis of the data in Fig. 2 was used to determine the initial rate constants for inactivation (k_obs). Double-reciprocal plots of the values for k_obs and the mifepristone concentrations gave a maximal rate constant (k_inactivation) for inactivation of 0.089 min¹ and a concentration of inactivator required for half-maximal inactivation (K_I) of 4.7 µM (Walsh, 1984).

View this table: [in this window] [in a new window]   TABLE 1 Mifepristone-mediated inactivation of CYP-3A4 in a reconstituted system CYP-3A4 (0.5 nmol/ml) was incubated with 50 µM mifepristone in a reconstituted system at 37°C for 15 min as described in Methods and Materials. Aliquots of 50 µl of incubation mixture were diluted into 0.95 ml of 50 mM HEPES buffer (pH 7.5) for determinations of testosterone 6-hydroxylation activity.

View larger version (22K): [in this window] [in a new window]   Fig. 2.   Time- and concentration-dependent inhibition of testosterone 6-hydroxylation activity of CYP-3A4 by mifepristone in a reconstituted system. CYP-3A4 was first incubated with 0 µM (), 5 µM (), 10 µM (), 20 µM (), and 30 µM (×) mifepristone in the reconstituted system for 0, 2, 5, and 10 min at 37°C, respectively. The incubation mixture was then diluted 20-fold with 50 mM HEPES buffer (pH 7.5) containing 200 µM testosterone and the appropriate components and incubated for 10 min at 37°C. 6-Hydroxytestosterone was determined by HPLC. The experimental details are described in Materials and Methods.

Changes in Reduced-CO Difference Spectrum and Absolute Spectrum of Mifepristone-Inactivated CYP-3A4. As shown in Fig. 3A, the reduced-CO difference spectrum of CYP-3A4 decreased by 76% following incubation with mifepristone in the presence of NADPH in the reconstituted system, whereas the catalytic activity of the enzyme decreased by 81%. Although there was a small peak that appeared at 425 nm in the spectrum of the inactivated enzyme, it was much less than expected for the conversion of P-450 to a P-420 form. However, the Soret peak of the inactivated CYP-3A4 was found to be slightly higher than that of the +mifepristone/NADPH control or the mifepristone/+NADPH controls (Fig. 3B), indicating that the loss of the reduced-CO difference spectrum was not due to heme destruction.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Reduced-carbon monoxide difference spectra (A) and UV-visible spectra (B) of reconstituted CYP-3A4 reaction mixture incubated with mifepristone in the absence of NADPH (a), without mifepristone in the presence of NADPH (b), and with mifepristone in the presence of NADPH (c), respectively. CYP-3A4 (0.5 nmol/ml) was incubated with 50 µM mifepristone in the reconstituted system at 37°C for 15 min as described in Methods and Materials. Aliquots of 0.25 or 0.2 ml of the incubation mixtures were diluted into 1.75 or 0.8 ml of 50 mM HEPES buffer (pH 7.5) containing 20% glycerol and 0.5 mM EDTA for recording the reduced carbon monoxide P-450 difference and UV-visible spectra, respectively.

Attempts to Detect Heme Adduct Formation. No new heme-derived peaks were detected at 400 nm or 415 nm in HPLC chromatograms of the TFA-butanone extracts of mifepristone-inactivated CYP-3A4 when compared with the NADPH control (data not shown). In addition, the peak containing the heme dissociated from the mifepristone-inactivated CYP-3A4 was almost identical with that of the NADPH control when the incubation mixtures were analyzed by HPLC on a Poros column (Fig. 4A). There was also no indication of heme adduct formation as detected by absorption at 445 nm when the sample of the +mifepristone/+NADPH was scanned against the sample of the +mifepristone/NADPH as the reference (data not shown).

View larger version (27K): [in this window] [in a new window]   Fig. 4.   HPLC profiles of the CYP-3A4 reconstituted system after incubation with 50 µM mifepristone in the absence (dotted line) or presence (solid line) of NADPH. The eluate was monitored at 405 nm (A), 214 nm (B), and by radioactivity (C). Peaks a, b, c, and d in B represent short NADPH-CYP reductase, NADPH-CYP reductase, cytochrome b5, and CYP-3A4, respectively. The experimental details were as described in Materials and Methods.

HPLC and SDS-PAGE Analysis of Mifepristone-Inactivated CYP-3A4. ApoCYP-3A4 was shown to be the only major protein with counts from ³H-mifepristone when a sample incubated in the presence of NADPH was analyzed by HPLC on a Poros column (Fig. 4C). The large radioactive peak eluting just before 7 min was mifepristone. [³H]Mifepristone was also shown to slightly associate with cytochrome b₅; this may reflect the ability of some of the reactive intermediate to escape from the enzyme active site. Some of the apoCYP-3A4 of the sample incubated with NADPH and mifepristone did not come off of the column, whereas reductase, cytochrome b₅, and heme were fully recovered when compared with the amounts of these proteins that eluted in the NADPH control (Fig. 4B). SDS-PAGE analysis also showed that only apoCYP-3A4 was labeled with [³H]mifepristone (Fig. 5).

View larger version (13K): [in this window] [in a new window]   Fig. 5.   SDS-PAGE analysis of the CYP-3A4 reconstituted reaction mixture after incubation with [3H]mifepristone in the presence or absence of NADPH. A, Coomassie blue-stained gel; B, autoradiogaphy of the gel. The experimental details were as described in Materials and Methods.

Stoichiometry of Binding of Mifepristone to apoCYP-3A4. The stoichiometry for the binding of the mifepristone was determined to be 1.02 ± 0.15 (n = 3), approximately 1 mol of mifepristone bound per mole of inactivated CYP-3A4. Because SDS-PAGE and HPLC analyses of the inactivated CYP-3A4 showed that [³H]mifepristone was covalently bound to the CYP protein and there were no signs of heme adduct formation or heme adducts linked to apoCYP, the stoichiometry measured for covalent binding was to the apoCYP-3A4.

Selectivity of Inhibition of P-450 Activities by Mifepristone. Approximately 22% of the methoxyresorufin O-demethylation activity (CYP-1A) in -NF-induced rat liver microsomes was inhibited by 100 µM mifepristone (Fig. 6). Approximately 46% and 63% of the benzyloxyresorufin O-dealkylation activities of PB-induced rat liver microsomes (CYP-2B) and purified CYP-2B1 were also inhibited by higher concentrations of mifepristone (100 µM), respectively. Approximately 78% of the activity of CYP-2D6 was inhibited by 100 µM mifepristone. The activities for the P-450 1A, 2B, and 2D6 enzymes were shown to be restored by more than 90% by extensively diluting the reaction mixture with appropriate buffer and increasing the substrate concentration in the secondary reactions. Mifepristone did not inhibit p-nitrophenol hydroxylation activity in pyridine-induced rat liver microsomes. The testosterone 6-hydroxylation activities of DEX-induced rat liver microsomes (CYP-3A2) and purified CYP-3A4 were inhibited by 96% and 92%, respectively, by 100 µM mifepristone. The loss of the CYP-3A2 and -3A4 activities was irreversible.

View larger version (37K): [in this window] [in a new window]   Fig. 6.   Inhibition of the activities of CYPs 1A, 2B, 2D6, 2E1, and 3A by mifepristone. -NF-, PB-, pyridine- and DEX-treated rat liver microsomes, E. coli membranes containing CYP-2D6 and NADPH-CYP reductase, and purified CYP-2B1 or CYP-3A4 reconstituted with NADPH-CYP reductase and/or cytochrome b5 were incubated with mifepristone (1, 10, and 100 µM) in the presence of the appropriate substrates of the CYPs at 37°C for the appropriate time period, respectively. The activities of the CYPs were determined using the methods described in Materials and Methods. The experiments were carried out in duplicate. *Inhibition of activities of CYPs 1A, 2B, 2B1, and 2D6 by mifepristone was reversible. The activities of the enzymes could be restored more than 90% by extensive dilution of the primary reaction mixtures and by increasing the concentrations of the substrates in the secondary reactions. **Inhibition of activities of CYPs 3A2 and 3A4 was irreversible.

	Discussion

Top Abstract Introduction Materials and methods Results Discussion References

The results presented in this article demonstrate that mifepristone is a mechanism-based inactivator of CYP-3A4. The loss of the catalytic activity of the enzyme is time- and concentration-dependent as well as requiring metabolism of mifepristone. The values of K_I and k_inactivation are 4.7 µM and 0.089 min¹, respectively, indicating that mifepristone is one of the most potent mechanism-based inactivators of CYP-3A4 investigated so far. Comparable K_I and k_inactivation values for the four other mechanism-based inactivators of CYP-3A4 that have been investigated in detail are: 46 µM and 0.4 min¹, respectively, for gestodene (Guengerich, 1990); 7.5 µM and 1.6 min¹, respectively, for L-754,394 (Sahali-Sahly et al., 1996); 59 µM and 0.16 min¹, respectively, for 6', 7'-dihydroxybergamottin (Schmiedlin-Ren et al., 1997); and 7.7 µM and 0.3 min¹, respectively, for bergamottin (He et al., 1998). The K_I value is also comparable with the apparent K_m values for the hydroxylation of the 17-propynyl group of mifepristone, which are 9.9 and 4.1 µM for human liver microsomes and recombinant CYP-3A4, respectively (Jang et al., 1996). The similar binding affinities for these two independent processes implies that the orientation of the mifepristone molecule in the active site of CYP-3A4 may be similar for both the oxidation of the carbon-carbon triple bond and the hydroxylation of the adjoining terminal carbon. It is important to note that the most of known acetylenes that inactivate CYPs are those with a terminal carbon-carbon triple bond. Several terminal 17-ethynyl steroids have been shown to inactivate CYP-3A4 (Guengerich, 1990). However, the position of the triple bond may not be critical for the inactivation of CYPs. A recent study by Alworth and coworkers (Foroozesh et al., 1997) and our present results demonstrate that internal acetylenic groups are also functional for inactivation of CYPs.

Because the reduced CO-binding spectrum was decreased by 76%, whereas approximately 81% of CYP-3A4 activity was lost following incubation with mifepristone, we investigated whether heme-adduct formation or heme fragmentation occurred during the metabolism of mifepristone by CYP-3A4 by using several approaches. N-heme adduct formation has been successfully detected by HPLC and UV-visible spectroscopy by looking for the characteristic absorption at 445 nm (He et al., 1996b). There was no loss of the normal heme optical spectrum nor generation of the modified heme peak when the TFA-butanone extract of mifepristone-inactivated CYP-3A4 was analyzed by HPLC. The peak containing the heme dissociated from the inactivated CYP-3A4 was almost identical in its retention time, shape, and area to that found in the NADPH control when the whole incubation mixture was analyzed by HPLC. Furthermore, UV-visible spectra of the inactivated enzyme did not show the characteristic absorption at 445 nm for heme adduct formation (He et al., 1996b). Therefore, heme modification does not appear to account for the inactivation of CYP-3A4 by mifepristone. Heme fragmentation is also ruled out by the observation that the Soret absorption of the inactivated CYP-3A4 did not decrease when compared with the controls. The reduction of CYP-3A4 appeared not to be a problem for measuring the CO-reduced spectrum of the inactivated CYP-3A4 because the CYP in the reconstituted mixture in the presence of NADPH was in the reduced state. Therefore, the marked decrease in the reduced CO-binding spectrum of the inactivated enzyme may be due to the fact that mifepristone is positioned close to the heme moiety as a result of covalent binding to an active site amino acid in such a way that it interferes with the interaction of CO with the ferrous heme. This phenomenon has been observed as a result of mechanism-based inactivation of CYPs by 8-methoxypsoralen and bergamottin (Labbe et al., 1989; Mays et al., 1990; He et al., 1998).

[³H]Mifepristone was shown to be covalently bound to the apoCYP-3A4 by HPLC and SDS-PAGE. The stoichiometry is approximately 1 mol of mifepristone bound per mole of CYP-3A4 inactivated. Covalent labeling of the apoCYPs has been shown to be the mechanism for inactivation of CYPs by terminal acetylenes such as 1-ethynylpyrene, 2-ethynylnaphthalene, and some other polycyclic arylacetylenes (Gan et al., 1984; Roberts et al., 1993; Yun et al., 1992), furan-containing compounds such as 8-methoxypsoralen, coriandrin, and bergamottin (Labbe et al., 1989; Cai et al., 1996; He et al., 1998), and sulfur-containing and halogenated compounds such as parathion and chloramphenicol (Halpert et al., 1980; Halpert, 1982). Recently, several modified peptides have been isolated that are thought to be involved in the active site. For example, in the investigations of the mechanism-based inactivation of CYP-2B1 by 2-ethynylnaphthalene and secobarbital, the modified peptide was identified in both cases to be the one that corresponds to helix I of CYP-101 and CYP-102 (Roberts et al., 1993, 1994; He et al. 1996a). This peptide contains the highly conserved threonine 302, which is believed to play a critical role in transferring protons and in the oxidation of the substrate. Mass spectrometry analysis of the peptide modified by 2-ethynylnaphthalene revealed that a species that corresponds to a naphthylacetic acid ester was added to the peptide (Roberts et al., 1994). The formation of ketene-derived species has been determined from the isolated protoporphrin IX adduct of CYPs inactivated by several terminal acetylenes (Ortiz de Montellano and Kunze, 1981; Kunze et al., 1983; Ortiz de Montellano and Correia, 1995; Roberts et al., 1998). It has been proposed that the ketene intermediate formed by 1,2-hydrogen shifts after addition of an oxygen to the triple bond could react with either the heme or protein moieties as well as undergo hydrolysis to the corresponding carboxylic acid. Similarly, 1,2-methyl rearrangement appeared to lead to the formation of a ketene intermediate when propynylaryl acetylenes were metabolized by CYPs, because the corresponding propionic acid metabolites could be detected (Foroozesh et al., 1997). We believe that mifepristone inactivates CYP-3A4 by a similar mechanism involving addition of reactive oxygen to the carbon-carbon triple bond to yield a highly reactive ketene intermediate through 1,2-methyl rearrangement. This ketene can then react with a nucleophilic residue at the enzyme active site. Heme modification versus protein labeling have been suggested to be controlled by the regiochemistry of oxygen addition to the internal or external carbon of the triple bond and the binding affinity of substrate to the enzyme (Chan et al., 1993). However, the terminal methyl group of the propynyl moiety may affect the geometric distance between the ketene intermediate and the nitrogen of the heme or the nucleophilic residues in such a way that it also affects heme or protein modification. Such a geometric effect has been seen in the mechanism-based inactivation of CYP-2B1 by 1-aminobenzotriazole and its derivatives. N-benzyl-1-aminobenzotriazole was shown to preferentially modify the protein whereas 1-aminobenzotriazole favors heme alkylation (Ortiz de Montellano and Correia, 1995, Kent et al., 1997).

Mifepristone was shown to inhibit the catalytic activity of CYP-3A more profoundly than the activities of the other CYPs tested. This is consistent with the findings that CYP-3A4 is the principle enzyme responsible for the metabolism of mifepristone (Jang et al., 1996). However, there appears to be little selectivity for inhibition of the catalytic activities of isoforms within the CYP-3A subfamily, because testosterone 6-hydroxylation was also inhibited by mifepristone in DEX-induced rat liver microsomes (presumably an activity of CYP-3A2). The inhibition of CYP 1A, 2B, and 2D6 activity by mifepristone was shown to involve competitive inhibition instead of mechanism-based inactivation, because more than 90% of the catalytic activities of the enzymes could be restored by extensive dilution of the reaction mixture.

The clinical significance of the inactivation of CYP-3A4 by mifepristone is that it would be expected to increase the bioavailability of several clinically used drugs metabolized by CYP-3A4 such as cyclosporine A, FK506, and dihydropyridines. The furanocumarin- derived mechanism-based inactivators in grapefruit juice have been shown to decrease the content of CYP-3A in enterocytes and consequently increase the bioavailability of several drugs (Lown et al., 1997; Schmiedlin-Ren et al., 1997). The potential for effects of mifepristone on the bioavailability of certain drugs may be strengthened by the finding that mifepristone is an inhibitor of P-glycoprotein, which is the other major determinant of the oral bioavailability of many drugs (Gruol et al., 1994; Lecureur et al., 1994). The coadministration of mifepristone with anticancer agents with lower bioavailability such as tamoxifen, vinblastine, vincristine, or taxol may be of interest, because mifepristone has also been shown to reverse P-glycoprotein-mediated drug resistance in vitro (Gruol et al., 1994). However, the potential for drug-drug interactions caused by the inhibition of the catalytic activity of CYP-3A4 by mifepristone should also be given careful consideration when mifepristone is administrated long term for the treatment of various cancers and Cushing's syndrome. Along these lines, it would be interesting to know whether mifepristone induces human CYP-3A4.

In summary, mifepristone has been shown to be a potent mechanism-based inactivator of human CYP-3A4. The mechanism of the inactivation was shown to involve irreversible modification of the apoprotein at the enzyme active site instead of heme adduct formation or heme fragmentation. It is suggested that mifepristone generates a highly reactive ketene species that reacts with a nucleophilic residue of CYP-3A4 by 1,2-methyl rearrangement, a process similar to that proposed for ethynyl acetylenes that inactivate CYPs via ketene intermediates formed by 1,2-hydrogen shifts (Ortiz de Montellano and Kunze, 1981; Foroozesh et al., 1997).