Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The cytochromes P450 comprise a large family of microsomal heme-containing monooxygenases that are involved in the metabolism of a wide variety of xenobiotics including drugs, pesticides, environmental pollutants, and carcinogens, as well as endogenous compounds such as steroids, retinoids, and fatty acids. The catalytic mechanism appears to be common to all P450s and involves a two-electron reduction of molecular oxygen to form a reactive oxygen species and water (Porter and Coon, 1991; Rendic and Di Carlo, 1997).

17-Ethynylestradiol (EE), the major estrogenic component of many oral contraceptives, can be metabolized by P450 enzymes in various animal species and humans (Bolt, 1979). Drugs, hormones, insecticide synergists, carcinogens, or dietary constituents that influence the expression and activity of various P450s can modulate the efficacy and side effects of EE (Bolt and Kassel, 1976; Guengerich, 1990a,b; He et al., 1998b). Studies using microsomes from phenobarbital-induced rats have demonstrated that the metabolism of EE results in a decrease in the cytochrome P450 content determined by the reduced CO difference spectrum and leads to the formation of the green pigments obtained from N-alkylated porphyrins (White, 1978; Ortiz de Montellano et al., 1979; Blakey and White, 1986). In addition, the mechanism-based inactivation of human liver microsomal P450s by EE with the loss of the P450-CO spectrum during incubation with NADPH has been shown (Guengerich, 1988). The EE-mediated inactivation in both rat and human liver microsomal P450s was postulated to be due to the metabolism of the acetylenic moiety of EE and the ensuing modification of the heme moiety of the P450s. However, the details of the kinetic parameters for the inactivation and modification of the apoprotein by an EE-derived metabolite were not reported (Guengerich, 1990a).

Cytochrome P450 3A4 (3A4) is the most abundant P450 isoform in human liver, has very broad substrate specificity, and is believed to be responsible for the metabolism of more than 60% of all clinically relevant drugs including contraceptive steroids (Guengerich, 1995). The substrates for 3A4 can be inactivators as well as inducers; thus during therapy with multiple drugs, drug-drug interactions may result in problems of clinical significance. 3A4 has been shown to be the principal catalyst involved in the oxidation of EE. In a series of human liver microsomes, the rate of the 2-hydroxylation of EE correlated well with both the rate of nifedipine oxidation and immunochemically determined 3A4 (Guengerich, 1988). We have used purified 3A4 reconstituted with phospholipids, NADPH-cytochrome P450 reductase (reductase), and catalase to characterize the EE-dependent inactivation of the testosterone 6-hydroxylase activity of 3A4. The following parameters were measured: the values for the concentration of EE required to give the half-maximal rate of inactivation (K_I); the maximal rate constant of inactivation at saturating concentrations of EE (k_inact); the time required for half the P450 to be inactivated at saturating concentrations of EE (t_1/2); the partition ratio; the UV-visible spectrum; the reduced CO difference spectrum; the effect of an alternate substrate on the loss of catalytic activity; and the stoichiometry and specificity of EE binding.

The reactive intermediates of acetylenic compounds formed by several isoforms of P450 have been known to alkylate the prosthetic heme group as well as to bind covalently to the protein (Ortiz de Montellano and Correia, 1995). Studies with P450 2B1 demonstrated that 2-ethylnylnaphthalene predominantly inactivates P450 2B1 through modification of the apoprotein, whereas phenylacetylene inactivates P450 2B1 via N-alkylation of heme (Ortiz de Montellano and Komives, 1985; Roberts et al., 1993). In this report, radiolabeled EE was employed in studies using SDS-PAGE and reversed-phase HPLC analysis to investigate the targets for EE-mediated 3A4 modification. To elucidate the identity of the reactive metabolite(s) involved in the heme and apoprotein modification and the mechanism(s) for inactivation by EE, the metabolites of EE were resolved by HPLC and further characterized by GC-MS.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. Cholic acid, catalase, NADPH, glutathione, testosterone, 6-hydroxytestosterone, 6-hydroxytestosterone, estradiol, estrone, and EE were purchased from Sigma-Aldrich (St. Louis, MO). 2-, 4-, and 16-hydroxyestradiol were obtained from Steraloids Inc. (Newport, RI). Synthesis of 17-carboxyestradiol was described in the previous study (Kent et al., 2002). 2-Hydroxyethynylestradiol (2-OH-EE) was a generous gift from Dr. William Slikker (Department of Health and Human Services, Food and Drug Administration, Jefferson, AR). 17-[6,7-³H]Ethynylestradiol (46.2 Ci/mmol) with radiochemical purity of 99% was obtained from Amersham Biosciences (Piscataway, NJ). All other chemicals and solvents were of the highest purity from commercial sources.

Purification of Enzyme. Both 3A4 and reductase were expressed in Escherichia coli and purified to homogeneity as described (Hanna et al., 1998; He et al., 1999).

Enzyme Assay and Inactivation. For the primary reaction mixture, 0.5 nmol of 3A4 was reconstituted with 60 µg of a mixture (1:1:1) of L--dilauroyl-phosphocholine, L--dioleyl-sn-glycero-3-phosphocholine, and L--phosphatidylserine, 200 µg of recrystallized sodium cholate, 1 nmol of reductase, 100 U of catalase, and 2 mM glutathione in 1 ml of assay buffer containing 50 mM HEPES (pH 7.5), 20% glycerol, 30 mM MgCl₂, and 0.5 mM EDTA. In studies of the time- and concentration-dependent inactivation by EE, the reactions were initiated by the addition of 1 mM NADPH to the primary reaction mixture, or the same volume of water was added as a control and the reaction mixtures were incubated at 37°C for the times indicated. The secondary reactions were started by transferring 50 µl of the primary reaction mixtures to 950 µl of assay buffer containing 200 µM testosterone and 200 µM NADPH. Incubations were carried out at 37°C for 20 min and then the reactions were terminated by the addition of 2 ml of ethyl acetate. The internal standard, 6-hydroxytestosterone, was added and the products were extracted into the organic phase. The major metabolite, 6-hydroxytestosterone, and the internal standard were quantified after separation by HPLC as described previously (He et al., 1999). The activity of 3A4 in the reconstituted system was 14.8 ± 1.2 nmol of 6-hydroxytestosterone formed per minute per nanomole of P450.

Spectral Analysis. After incubating the primary reaction mixtures with 50 µM EE in the presence (inactivated sample) or absence of NADPH (control sample), the absolute spectra and reduced CO difference spectra of 0.25 nmol of 3A4 were determined by scanning from 400 to 600 nm on an SLM-AMINCO 3000 spectrophotometer (Omura and Sato, 1964). In addition, 50 µl aliquots of the reaction mixtures were removed for the determination of the testosterone 6-hydroxylation activity. To test the irreversibility of inactivation, the control and inactivated samples were dialyzed overnight at 4°C against 1 liter of assay buffer and then reanalyzed both for enzymatic activity and reduced CO difference spectra.

Substrate Protection. The inactivation of 3A4 by EE in the presence or absence of substrate was investigated by adding an 8- and 16-fold molar excess of testosterone over EE to the primary reaction mixture. At the end of the incubation time, aliquots were removed for the determination of 3A4 activity remaining as described above.

Partition Ratio. EE at concentrations of 1 to 125 µM was added to the primary reaction mixtures containing 0.5 µM 3A4. After adding 1 mM NADPH, the reaction mixtures were incubated for 1 h to allow the inactivation to reach completion. Aliquots were removed and assayed for catalytic activity remaining as previously described (Silverman, 1996).

Stoichiometry and Specificity of Binding. Following incubation with 50 µM radiolabeled EE in the primary reaction mixtures for 1 h, 1-ml aliquots of the control (NADPH) and inactivated (+NADPH) samples were mixed with 10 mg of bovine serum albumin and precipitated by adding a 5-fold volume of a 5% solution of sulfuric acid in methanol according to the method described by Ortiz de Montellano and coworkers (Ortiz de Montellano, 1991; Chan et al., 1993). The precipitates were collected by centrifugation, and the resulting pellets were washed five times with the same solvent until the radioactivity in the supernatant was essentially at background level. The final pellets were dissolved in 1 N NaOH, incubated at 60°C for 1 h and neutralized with HCl prior to liquid scintillation counting using Econo-Safe counting cocktail (Research Products International Corp., Mount Prospect, IL). For SDS-PAGE analysis, two sets of control and inactivated samples were resolved on 8% polyacrylamide gels. After staining with Coomassie Blue, one set of gels was photographed, and the protein bands were excised and dissolved in H₂O₂ at 60°C for 3 h followed by liquid scintillation counting to determine the protein-associated radioactivity. The other set of gels was treated with universal autoradiograph enhancer (PerkinElmer Life Sciences, Boston, MA) and then dried on 3-mm chromatography paper. The dried gels were exposed to Kodak BioMax MS film (Eastman Kodak, Rochester, NY) at 80°C for 2 weeks before developing.

HPLC Analysis. The control (NADPH) and inactivated (+NADPH) samples that had been incubated with [³H]EE were dialyzed extensively to remove noncovalently bound radioactivity. The dialyzed samples were analyzed by HPLC on a C4 protein and peptide column (4.6 × 250 mm, 300 Å; VYDAC, Hesperia, CA) with a solvent system consisting of solvent A (0.1% trifluoroacetic acid in water) and solvent B (95% acetonitrile and 0.1% trifluoroacetic acid) using a linear gradient from 35% B to 80% B over 45 min with a flow rate of 1 ml/min. The eluate was monitored at 220 nm for protein, at 280 nm for EE and protein, and at 405 nm for heme. Fractions were collected and the radioactivity was determined by liquid scintillation counting. Additionally, we used 10% trifluoroacetic acid/butanone to extract noncovalently bound heme or other components from proteins in the radiolabeled control and inactivated samples. The organic phases were dried and analyzed by HPLC as described above.

Electrospray Ion Mass Spectrometry. The major fractions from HPLC analysis that corresponded to the radiolabeled peaks were collected and analyzed by electrospray mass spectrometry using a VG/Fisons Platform single-quadrupole spectrometer (Beverly, MA). These analyses were performed at the University of Michigan Protein and Carbohydrate Structure Facility.

Metabolism of EE. Following incubation with 100 µM [³H]EE for 1 h in a 1-ml reaction mixture, the control (NADPH) and inactivated (+NADPH) samples were extracted with 4 ml of CH₂Cl₂ and dried under N₂ according to a method previously described (Guengerich, 1990a). The metabolites were separated by HPLC on a Varian Microsorb-MV C₁₈ column (5 µm, 4.6 × 250 mm; Palo Alto, CA) with a solvent system consisting of solvent A (0.1% acetic acid in water) and solvent B (70% acetonitrile, 29.9% methanol, and 0.1% acetic acid) using linear gradients from 30% B to 50% B over 3 min, then to 60% B within the next 12 min, and then to 95% B for an additional 10 min at a flow rate of 1.2 ml/min, and the eluate was monitored at 280 nm. The retention times of the metabolites were compared with those of authentic standards. The major metabolite containing fractions (A-D) were collected, dried, derivatized, and analyzed by GC-MS at the Michigan State University Mass Spectrometry Facility. Each sample was incubated with 5 µl of redistilled pyridine and 20 µl of N,O-bis[trimethylsilyl]trifluoroacetamide containing 1% trimethylchlorosilane for 30 min at 70°C. Each sample was chromatographed on a 30-m DB1 fused silica capillary column (0.32 mm i.d., 0.25 µm film coating; J&W Scientific, Folsom, CA) with a temperature gradient of 10°C per minute from 80-320°C and analyzed over m/z range from 45 to 750 on a JEOL JMS AX-505 double focusing mass spectrometer (Tokyo, Japan) coupled to a Hewlett-Packard 5890J gas chromatograph (Palo Alto, CA) via a heated interface.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Inactivation of 3A4 by EE. The time course for the inactivation of 3A4 by various concentrations of EE is shown in Fig. 1. The loss of activity followed pseudo first order kinetics. Linear regression analysis of the time course data was used to determine the initial rate constants for inactivation (k_obs) at various concentrations of EE. From the double-reciprocal plot (inset) of the values for k_obs and the concentration of EE, the k_inact was determined to be 0.04 min¹, the K_I was 18 µM, and the t_1/2 was 16 min.

View larger version (20K): [in this window] [in a new window]   Fig. 1.   Time- and concentration-dependent inactivation of testosterone 6-hydroxylation activity by EE. P450 3A4 was first incubated with 0 (), 5 µM (), 10 µM (), 20 µM (), and 50 µM () EE in the reconstituted system, and aliquots (50 µl) were removed at the time points indicated and assayed for residual activity as described under Materials and Methods. The inset shows the double reciprocal plot of the initial rate of inactivation of testosterone 6-hydroxylation activity as a function of the inactivator concentration. Each point shown represents the mean from four separate experiments that do not differ by more than 6%.

Changes in Absolute and Reduced CO Difference Spectra of EE-Inactivated 3A4. When 3A4 was incubated with 50 µM EE for 30 min, the enzymatic activity of the NADPH-treated samples decreased to 33 ± 2% (n = 7), the absorbance at 415 nm of the absolute spectrum decreased to 65 ± 4% (n = 5), and the reduced CO difference spectrum decreased to 30 ± 8% (n = 5) compared with the control samples incubated with EE in the absence of NADPH. Representative data for the changes in the absolute and reduced CO difference spectra are illustrated in Fig. 2, A and B, respectively. The loss of the spectrally detectable cytochrome P450 at 450 nm was accompanied by a concomitant loss of the heme spectrum at 415 nm in the EE-inactivated 3A4. The removal of free EE by extensive dialysis did not lead to a significant recovery of either the catalytic activity, the absolute or the reduced CO difference spectra of the inactivated 3A4 (data not shown). Thus, the inactivation of 3A4 by EE could not be reversed by dialysis.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Absolute (A) and reduced CO spectra (B) for control or NADPH-inactivated P450 3A4. P450 3A4 was incubated with 50 µM EE in the reconstituted system at 37°C for 30 min in the absence (a) or presence (b) of NADPH as described under Materials and Methods.

Substrate Protection. Simultaneous incubation of P450 3A4 with 50 µM EE in the presence of an 8- or 16-fold molar excess of testosterone over EE in the primary reaction mixture reduced the ability of EE to inactivate 3A4 in a time- and concentration-dependent manner (Fig. 3). These results indicate that testosterone competes with EE for metabolism by 3A4 and thereby protects against inactivation.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Substrate protection of P450 3A4 against inactivation by 50 µM EE. The reaction mixtures were as described under Materials and Methods. Portions of the samples were removed from the primary reaction mixture at the time points indicated and assayed for catalytic activity remaining. The molar ratios of substrates were no testosterone or EE (), testosterone/EE = 16:1 (), testosterone/EE = 8:1 (), and EE only ().

Determination of the Partition Ratio. P450 3A4 was incubated with various concentrations of EE, and the inactivation was allowed to progress for 1 h until it was essentially complete. The percentage of activity remaining was plotted as a function of the molar ratio of EE to 3A4. The turnover number was estimated from the intercept of the linear regression line obtained from lower ratios of EE to 3A4 with the straight line derived from the higher ratios of EE to 3A4 as described previously (Silverman, 1996). With this method, we estimated a partition ratio of ~50 as illustrated in Fig. 4. A partition ratio of ~120 for EE has previously been reported in human liver microsomes (Guengerich, 1988).

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Loss of P450 3A4 activity as a function of the ratio of EE to 3A4. P450 3A4 was incubated with various concentrations of EE for 1 h until the inactivation was essentially complete. The extrapolated partition ratio was estimated from the intercept of the linear regression line from the lower ratios and the straight line obtained from higher ratios.

Stoichiometry of Binding. The amount of EE that covalently bound to the apoprotein was determined by precipating the proteins from reaction mixtures incubated with [³H]EE in the presence or absence of NADPH followed by extensively washing the protein pellets with 5% sulfuric acid in methanol. After subtracting the residual radioactivity of EE in the control samples from the inactivated samples, a stoichiometry of approximately 1.3 nmol of EE metabolite bound per nanomole of 3A4 inactivated was obtained.

SDS-PAGE Analysis. After incubating the 3A4 in the reconstituted system with 50 µM radiolabeled EE, the proteins in the control (NADPH) and inactivated (+NADPH) samples were separated by SDS-PAGE and then stained with Coomassie Blue. One set of gels was photographed (Fig. 5A) and then the protein bands were excised and the radioactivity measured. After subtracting the radioactivity in the control samples from the inactivated samples, the relative radioactivities associated with the proteins in the inactivated samples were as follows: reductase = ~8%, catalase = ~1%, and P450 3A4 = ~91%. The other set of gels was dried and analyzed by autoradiography. As shown on Fig. 5B, [³H]EE was only observed in the NADPH-treated samples with the major amount of radioactivity associated with the 3A4 apoprotein and some radioactivity with reductase. These results demonstrate that the intensities of the protein components are similar in both the control and inactivated samples whereas the radioactivity derived from an EE reactive intermediate was predominantly associated with P450 3A4 in the inactivated sample. From the stoichiometry of binding, the autoradiography, and the radioactivity associated with the 3A4 apoprotein, we conclude that the EE reactive intermediate is covalently bound to the 3A4 apoprotein. The binding of a small amount of [³H]EE to reductase may reflect the ability of some of the reactive intermediate escaping from the 3A4 active site.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   SDS-PAGE separation of 3A4 apoprotein reductase and catalase in the reconstituted system after incubation with radiolabeled EE in the absence (minus) or presence (+) of NADPH. A, Coomassie Blue-stained gel. B, autoradiography of the dried gel.

HPLC Analysis. After extensively dialyzing the reconstituted systems, which had been exposed to radiolabeled EE, HPLC was used to separate the heme moiety from the apoprotein under acidic conditions. The elution profiles monitored at 405 nm showed that the heme eluted at ~18 min for both the inactivated (+NADPH) and control (NADPH) samples (Fig. 6A). The area under the EE-inactivated 3A4 heme peak was 48 ± 3% (n = 7) compared with the control and a small peak of modified heme, which exhibited maximum absorbance at 400 nm with photodiode-array detector, was apparent at ~28 min in the inactivated sample. As shown in Fig. 6B, the reductase eluted at ~36 min, and the 3A4 apoprotein eluted at ~43 min. The elution profile monitored at 280 nm (Fig. 6B) and 220 nm (data not shown) demonstrated that essentially all the reductase can be recovered from the column, whereas only 53 ± 3% (n = 7) of the 3A4 apoprotein in the NADPH-treated sample can be recovered compared with control samples. When the fractions were collected for liquid scintillation counting, several radiolabeled peaks were found in the NADPH-treated sample with no detectable radioactivity in the control sample (Fig. 6C). Radiolabeled EE was associated with the 3A4 apoprotein in the inactivated sample. However, the largest radiolabeled fraction corresponded to the peak eluted at ~31 min (Peak 31). "Peak 31" and the other three small peaks at 24 to 27 min were also apparent in the elution profile of the NADPH-treated sample monitored at 280 nm but were not seen in the control sample (Fig. 6B). HPLC analysis using a photodiode-array detector demonstrated that the absorption spectrum of Peak 31 exhibited a maximum at 280 nm (spectrum not shown). The lack of absorbance of Peak 31 around 400 nm indicates that this radiolabeled species is neither intact heme nor its tetrapyrrolic skeleton. Moreover, HPLC analysis of the trifluoroacetic acid/butanone extract revealed the presence of a major radiolabeled peak in the elution profile from the NADPH-treated sample, but not in the profile from the control sample (data not shown). These findings suggested that one or more radiolabeled products were formed from the reaction of metabolites of EE with heme fragments in the reconstituted system that were still associated with the protein after exhaustive dialysis and then could be dissociated from the apoprotein under acidic/organic conditions.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   HPLC elution profiles of the reconstituted system after incubating with radiolabeled EE in the presence or absence of NADPH. A, chromatograms monitored at 405 nm for intact heme. B, chromatograms monitored at 280 nm for apoprotein and EE. C, corresponding HPLC elution profile monitored by liquid scintillation counting to determine the fractions containing covalently bound EE. Heme eluted at ~18 min, reductase at ~36 min, and 3A4 apoprotein at ~43 min.

Electrospray Ion Mass Spectrometry. The fraction corresponding to Peak 31 was collected and analyzed using an electrospray mass spectrometer in the negative-ion mode. As shown in Fig. 7, there is one major peak with a mass (M  H) of 479 Da from the fraction in the NADPH-treated sample. No such peak could be detected by electrospray mass spectrometry from the fraction, which eluted at ~31 min in the control sample (data not shown). The chemical structure of this EE-related modified species remains to be established.

View larger version (10K): [in this window] [in a new window]   Fig. 7.   Electrospray mass spectrum (negative-ion mode) of Peak 31 from HPLC separation of the NADPH-treated sample. The fractions corresponding to Peak 31 in Fig. 6 were collected and subjected to electrospray mass spectrospray as described under Materials and Methods.

Metabolite Analysis. In humans, the primary route for the metabolism of EE involves hydroxylation at the 2 position, but hydroxylation at the 4, 6, and 16 positions also occurs. There is also evidence for the oxidation of the ethynyl triple bond, deethylation and D-homoanulation (Bolt, 1979). As shown in Fig. 8, four distinct radiolabeled metabolites, peaks A, B, C, and D, were produced from the metabolism of EE (peak E) by 3A4 in the presence of NADPH. These four peaks were not observed in the HPLC elution profile of control samples (data not shown). The major metabolite (D) eluted from the HPLC with the same retention time (18.5 min) as authentic 2-OH-EE standard. GC-MS analysis confirmed that D was 2-OH-EE since the GC retention time in the total ion chromatogram (TIC), the m/z of the molecular ion (528) as well as the m/z of the ion fragments were identical to those observed for the 2-OH-EE standard (data not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 8.   HPLC chromatogram of metabolites in CH2Cl2 extracts from the EE-inactivated samples. Experimental conditions were described under Materials and Methods. Peaks A, B, C, and D are the major metabolites and peak E is the parent EE.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Enzymatic Activity. The results presented here demonstrate that EE is an effective mechanism-based inactivator of 3A4. The inactivation of the 6-testosterone hydroxylation of 3A4 was time- and concentration-dependent and required NADPH. The inactivation exhibited pseudo first order kinetics with a K_I = 18 µM, a k_inact = 0.04 min¹ and a t_1/2 = 16 min. In addition, the inactivation was irreversible by dialysis and the rate of inactivation decreased in the presence of the alternate substrate testosterone. Although the inclusion of cytochrome b₅ in the reconstituted mixture stimulated the rate of inactivation up to 2-fold, the residual activity of the inactivated samples was very close regardless of the presence or absence of b₅ (data not shown). Since the absolute spectrum of the b₅ heme shifted from 415 to 424 nm in the presence of reductase and the heme moieties of both b₅ and 3A4 coeluted in the HPLC analysis under acidic conditions, we eliminated b₅ from our studies on the mechanism of EE-mediated inactivation of P450 3A4 to avoid interference from the b₅ heme moiety.

Covalent Binding of EE to 3A4 Apoprotein. Modification of the 3A4 apoprotein was demonstrated by the following: 1) SDS-PAGE separation of the proteins in the reconstituted system demonstrated that most of the counts from radiolabeled EE were associated with the 3A4 protein (Fig. 5B); 2) stoichiometry of 1.3 nmol of EE bound per nmol of 3A4 inactivated was determined; 3) HPLC analysis indicated that radiolabeled EE associated with the fraction corresponding to 3A4 apoprotein in the NADPH-treated sample (Fig. 6C, retention time at 43 min); and 4) the amount of heme loss detected in the UV-visible spectrum and HPLC analysis was much less than the loss of spectrally detectable P450, and the loss in catalytic activity indicated that heme modification alone was not responsible for all the EE-dependent inactivation. The inactivated 3A4 apoprotein was fully recovered during the SDS-PAGE separation (Fig. 5A), whereas only ~53% was recovered by HPLC separation of the reconstituted system compared with the control sample (Fig. 6B). Modification of 3A4 by cumene hydroperoxide, mifepristone, or bergamottin is believed to result in the exposure of hydrophobic groups on the P450 protein, so that it cannot elute from reversed-phase HPLC column (He et al., 1998a,b, 1999). Taken together, our results suggest that the amount of radiolabeled EE associated with 3A4 apoprotein is underestimated from the reversed-phase HPLC analysis.

Heme Destruction and Formation of EE-Conjugated Compounds. Based on the loss of spectrally detectable P450, the formation of an N-substituted porphyrin was postulated to be involved in the inactivation of human P450s by EE (Guengerich, 1988). We report here that 3A4 inactivation by EE results not only in the loss of the reduced CO difference spectrum at 450 nm, but also in the loss of the absolute spectral absorbance at 415 nm and the loss of intact heme, as determined by HPLC, with concomitant generation of several EE-radiolabeled species. The formation of a small fraction of modified heme adduct, observed in the NADPH-treated sample, was in agreement with the formation of green pigment in rat hepatocyte suspensions after incubation with EE (Blakey and White, 1986). The EE-radiolabeled species are very stable and can be recovered even after dialysis at 4°C for 4 days. The major radiolabeled species, Peak 31, having a mass of 479 would appear to be an EE-alkylated heme fragment that does not irreversibly bind to 3A4 apoprotein based on the following observations: 1) the maximal absorbance at 280 nm rather than 400 nm suggests that the EE-mediated inactivation apparently ruptures the intact heme chromophore or that the EE-labeled heme has a short half-life and is degraded to heme fragments; 2) EE and heme fragments exhibit absorbance at 280 nm (Schaefer et al., 1985; Correia et al., 1987; Guengerich, 1988); 3) the difference in the mass of the Peak 31 (479 Da) and EE (296 Da) is consistent with a monopyrrole fragment; and 4) the EE-labeled moiety can be dissociated from the 3A4 apoprotein under organic/acidic conditions. Therefore, we conclude that reaction of the EE metabolites with prosthetic heme results in the formation of not only the modified heme, but also the destruction or degradation of the intact heme with concomitant generation of EE-modified heme fragment adducts. The chemical structure(s) and the nature of adduction of these modified species remain to be established.

Oxidation of Acetylenic Compounds. Studies by Ortiz de Montellano and coworkers on the inactivation of P450 by acetylenic compounds have led to the suggestion that the differential inactivation by modification of the prosthetic heme group versus the apoprotein depends primarily on the delivery of the ferryl oxygen to the internal carbon (heme alkylation) or to the external carbon (protein alkylation) of the acetylene moiety (Kunze et al., 1983; Komives and Ortiz de Montellano, 1987; Chan et al., 1993). Phenylacetylene attacks the prosthetic heme in both P450 2B1 and P450 1A1, whereas 2-ethynylnaphthalene attacks the P450 2B1 apoprotein and 1-ethynylpyrene attacks the P450 1A1 apoprotein (Chan et al., 1993; Roberts et al., 1993; Ortiz de Montellano and Correia, 1995). These results demonstrate that a single isozyme can be inactivated by alkylation of either the heme or the protein depending on the structure of the arylacetylene. The loss of catalytic activity and the formation of heme adducts in rabbit P450 2E1 modified by ethynyl compounds is much greater than in a mutant of P450 2E1 where threonine 303 was replaced with alanine (Roberts et al., 1998). The studies reported here demonstrate that EE can modify both the heme and the apoprotein of P450 3A4, whereas EE only modified the apoprotein in P450 2B1 and P450 2B6 (Kent et al., 2002). These results support the hypothesis that the metabolic activation of a single ethynyl compound can result in different reactivities toward heme versus apoprotein with different P450 isozymes.

Inactivation of 3A4. Three pathways for the mechanism-based inactivation of P450 isozymes have been characterized: 1) covalent modification of the apoprotein; 2) alkylation of the heme moiety; and 3) covalent binding of the modified heme to the apoprotein (Osawa and Pohl, 1989; Ortiz de Montellano and Correia, 1995). Our results provide evidence for the occurrence of pathways 1 and 2 in the EE-mediated inactivation of 3A4. Using [³H] and [¹⁴C]heme-labeled P450, a wide variety of suicide substrates have been shown to inactivate P450 by the destruction of heme and covalent attachment of the modified heme to the apoprotein as demonstrated by the isolation of active site heme-modified peptides from P450s 2B1 and P450 3A4 (Guengerich, 1986; Correia et al., 1987; Osawa and Pohl, 1989; Yao et al., 1993; He et al., 1998a). Ortiz de Montellano and Correia (1995) have proposed that the 3A4 active site is particularly susceptible to inactivation by heme-protein adduct formation. It should be pointed out that during the oxidation of norethindrone, an acetylenic steroid, by rat liver microsomes, a considerable portion of the heme is destroyed and is irreversibly attached to the P450 apoprotein (Davies et al., 1986). Since we were only able to recover half the 3A4 P450 in the inactivated samples, we cannot rule out the possibility that the cross-linked EE-modified heme or heme fragments were lost in the HPLC column.