Determinants of protein modification versus heme alkylation

Jan 1, 1993 - Kelly A. Regal, Michael L. Schrag, Ute M. Kent, Larry C. Wienkers, and ... Ginny Primrose, Zuyu Guo, L. Chastine Bell, William L. Alwort...
0 downloads 0 Views 1MB Size
Chem. Res. Toricol. 1993,6, 30-45

38

Articles Determinants of Protein Modification versus Heme Alkylation: Inactivation of Cytochrome P450 1Al by 1-Ethynylpyrene and Phenylacetylene William K. Chan, Zhihua Sui, and Paul R. Ortiz de Montellano* Department of Pharmaceutical Chemistry, School of Pharmacy, and Liver Center, University of California, San Francisco, California 94143-0446 Received July 20, 1992

Reaction of cytochrome P450 enzymes with arylacetylenes results in heme N-alkylation [e.g., Komives, E. A., and Ortiz de Montellano, P. R., (1985) J.Biol. Chem. 260,3330-33363 and/or protein modification [e.g., Gan, L.4. L., Acebo, A. L. and Alworth, W. L. (1984) Biochemistry 23,3827-38361. To clarify the factors that determine whether heme or protein alkylation occurs, we have investigated the cytochrome P450 1Al-catalyzed oxidation of 1-ethynylpyrene (1-EP) and phenylacetylene (PA). Cytochrome P450 1Al in microsomesfrom jhaphthoflavone-induced rats is inactivated in a time- and NADPH-dependent manner by 1-EP and PA. Parallel loss of the heme chromophore is observed with PA but not with 1-EP, although partial heme chromophore loss is observed when the purified, reconstituted enzyme is inactivated by either agent. Product analysis shows that 1-EP and PA are oxidized to, respectively, (1'-pyreny1)acetic and phenylacetic acids. In contrast to the inactivation of cytochrome P450 2B1 by PA, no isotope effect is observed on enzyme inactivation or metabolite formation when the acetylenic hydrogen is replaced by deuterium in either 1-EP or PA. Inactivation of cytochrome P450 1Al by 1-EP results in covalent binding of 0.8-0.9 equiv (relative to total cytochrome P450 content) of the inhibitor to the microsomal protein. The results demonstrate that a single isozyme can be inactivated, depending on the structure of the arylacetylene, by heme or protein alkylation. show that 1-EP binds to the enzyme with >2000 times Spectroscopic binding constants (K,) greater affinity that PA. This large difference in binding affinity is a probable factor in the heme versus protein specificity of the inactivation reaction.

Introduction The cytochrome P450-catalyzed oxidation of terminal acetylenes converts them to acetic acid derivatives and results in concomitant inactivation of the enzyme (1-3). Acetic acid derivatives have been explicitly identified in the metabolism of phenylacetylenes (4, 5 ) , p-biphenylacetylenes (6-11), 2-ethynylnaphthalene (121, and 10undecynoic acid (13). Studies with I2H1-and [l3C1-labeled phenylacetylene establish that the triple bond is enzymatically oxidized to an intermediate, undoubtedly a ketene, in which the acetylenic hydrogen has migrated quantitatively to the adjacent carbon (9, 10). The observed acetic acid metabolites are produced by uncatalyzed hydration of the ketene, as shown: ArCECD ArCD=C=O

-

ArCD=C=O

+ H,O

-

ArCHDC0,H

Studies with 1 8 0 2 show that the oxygen of the ketene produced from biphenylacetylene derives from molecular oxygen and is therefore introduced by transfer of the enzymatically activated ferryl oxygen to the terminal carbon of the acetylenic group (4). Replacement of the

* Author to whom correspondence should be addressed.

acetylenic hydrogen of PA' (4,5) and biphenylacetylene (4,5,11) with deuterium results in a kinetic isotope effect on product formation ( k ~ / kbetween ~) 1.4 and 1.8. This is a large isotope effect for a process in which the hydrogen migrates through a bent rather than linear transition state (14). A linear free energy correlation with p = -2.2 for formation of acetic acid derivatives from substituted phenylacetylenesshowsthat the transition state is sensitive to electron withdrawal (5). Two mechanisms for the inactivation of cytochrome P450 by acetylenes have strong experimentalsupport. One is N-alkylation of the prosthetic heme group by a transient species generated when the ferryl oxygen is added to the internal carbon of the triple bond (1-3). This mechanism results in loss of the heme chromophore due to formation of an adduct in which the acetylene is bound to a nitrogen of the porphyrin as a 2-oxoalkyl group (Fe-N represents the iron and one of the porphyrin nitrogens):

-

Fe-N + R C s C H Fe-N-CH2COR N-Alkyl heme adducts have been isolated and fully characterized from rats treated with ethyne, propyne, octyne, and ethchlorvynol(5,15-17). The oxygen incorporated intq the porphyrin N-alkyl groups obtained with Abbreviations: heme, iron protoporphyrin IX regardless of the iron oxidation and ligation state; 1-EP,1-ethynylpyrene;PA, phenylacetylene; BNF, 8-naphthoflavone; DLPC, dilauroylphosphatidylcholine.

0 1993 American Chemical Society 0893-228~/93/2706-0038$04.00/0

Cytochrome P450 Inactivation by Arylacetylenes

\ PA

Q-9yp H

\

\

\

/

/

/

1-EP

F i g u r e 1. Structures of PA and 1-EP and proposed mechanisms for their conversion to the corresponding arylacetic acid metabolites.

terminal olefins derives from molecular oxygen and is therefore presumably the ferryl oxygen (15). Deuterium substitution and electron-withdrawing substituents, in contrast to their effect on metabolite formation, do not significantly influence the rate of enzyme inactivation (4, 5 )* The second mechanism for inactivation of cytochrome P450 by terminal acetylenes involves modification of the protein. Inactivation of liver microsomal cytochrome P450 by polycyclic arylacetylenes results in greater loss of catalytic activity than can be accounted for by the fractional loss of the prosthetic heme group that is observed (18, 19). Detailed studies with 2-ethynylnaphthalene indicate that inactivation of cytochromes P450 1A12and 1A2 occurs with little loss of the chromophore and with substoichiometric (-0.7 equiv) binding of the radiolabeled substrate to the cytochrome P450 protein (12). Likewise, the inactivation of purified, reconstituted lauric acid o-hydroxylase (cytochrome P450 4A1) by 10-undecynoic acid occurs without detectable loss of the heme chromophore and apparent substoichiometric binding of radiolabeled substrate to the protein (13). These inactivation reactions clearly involve the protein rather than the prosthetic heme group even though the nature of the inactivation reaction remains unclear. The most attractive mechanism involvesreaction of the ketene metabolite with an active site nucleophile:

-

RCH=C=O + protein-XH protein-X-C(O)CH,R It is not known what factors determine whether heme or protein modification occurs. Addition of the ferryl oxygen to the terminal carbon to give the ketene occurs in both situations, but there is no evidence for addition to the internal carbon in the absence of heme alkylation. This suggests that the inactivation reaction is controlled, in part, by the regiochemistry of oxygen addition. The strong specificity of lauric acid w-hydroxylase for the thermodynamically disfavored terminal carbon appears, for example, to rule out heme alkylation in that system (13). Protein modification, on the other hand, may reflect the presence or absence of active site nucleophiles suitable for reaction with the ketene metabolite. We report here an investigation of the inactivation of cytochrome P 4 5 0 1 A l by l-ethynylpyrene (1-EP) and phenylacetylene (PA) (Figure 1)that sheds some light on this matter.

Experimental Procedures Materials a n d Methods. PA was obtained from Aldrich (Milwaukee, WI). 1-EP was synthesized as reported by Gan et The cytochrome P450 nomenclature used is that proposed by Nebert et al. (20).

Chem. Res. Toxicol., Vol. 6, No. 1, 1993 39 al. (18). NADPH and catalase were purchased from Sigma (St. Louis, MO). 7-Ethoxyresorufin and resorufin were from Pierce (Rockford, IL). [3H]1-EP (43 500 cpm/nmol) was synthesized by tritium exchange a t the National Tritium Labeling Facility (Berkeley, CA) by Prof. Chin-Tzu Peng (21). The radiolabeled product was shown to be >90% pure by thin-layer chromatog raphy. Cytochrome P450 studies were carried out on an Aminco DW2000 spectrophotometer. Routine ultraviolet spectra were recorded on a Hewlett Packard Model 8450A diode array spectrophotometer. A Perkin Elmer P E 650-10s fluorescence spectrophotometer was used to assay 7-ethoxyresorufin O-deethylation. Mass spectra were recorded on a Kratos MS-25 instrument. NMR spectra were obtained in CDC13on a Varian 300-MHz instrument. NMR peaks are given in parts per million relative to tetramethylsilane. Deionized, glass-distilled water was used for all the biological work. (1'-Pyreny1)acetic Acid. (1'-Pyreny1)acetic acid, which was required as a standard, was synthesized by reduction of l-pyrenecarboxaldehyde to the alcohol, conversion of the alcohol to the bromide, displacement of the bromide by cyanide, and hydrolysis of the nitrile group. The first step, preparation of l-(h,droxymethyl)pyrene, was carried out by the procedure of Johnson and Rickborn (22). Lithium aluminium hydride (0.23 g, 6 mmol) was added slowly a t -78 "C to a solution of l-pyrenecarboxaldehyde (1.4 g, 6 mmol) in dry T H F (10 mL). The mixture was allowed to warm to room temperature and was stirred for 1 h before it was poured onto ice, extracted with diethyl ether, and dried over anhydrous magnesium sulfate. After evaporation of the solvent, the residue was purified by flash column chromatography (silica gel, hexane/ethyl acetate = 73). l-(Hydroxymethy1)pyrenewas obtained as a white solid (1.27 g, 90%): mp 124-125 "C [from hexane; lit. mp 123-124 "C ( 2 3 ) l ; IR (KBr) 3288, 3050, 2894, 1589, and 1434 cm-I; lH-NMR (300 MHz, CDC13) 6 2.00 (br s, 1

H,OH),5.24(~,2H,CH~),and7.8&8.25ppm(m,9H,aromatic); 13C-NMR(75MHz, CDC13) 663.6 (CHz), 122.8,124.6,124.8,125.1, 125.8, 127.3, 127.7, 128.6, 130.6, 131.1, and 133.6 ppm (aryl C); MS m/z 232 (M+, loo), 215 (M+ - OH, 74). HRMS: Calcd for C17H120, 232.0888; found, 232.0876. A solution of l-(hydroxymethy1)pyrene (300 mg, 1.29 mmol), carbon tetrabromide (729 mg, 2.2 mmol), and triphenylphosphine (575mg, 2.2 mmol) in dry acetonitrile (5mL) was stirred a t room temperature for 16h. Silica gel was added to the reaction mixture, and the solvent .was evaporated. The residue was purified immediately by flash column chromatography (silica gel, hexane/ ethyl acetate = 9:l). l-(Bromomethy1)pyrenewas thus obtained as a white solid (279 mg, 73%): mp 134-136 "C; after recrystallization from hexane, mp 144-145 OC [lit.mp 145-147 "C ( 2 4 ) l ; IR (KBr) 3041,1450, and 1204 cm-l; 'H-NMR (300 MHz, CDCl3) 6 5.24 (s, 2 H, CHz), and 7.99-8.39 ppm (m, 9 H, aromatic); 13CNMR (75 MHz, CDCl3) 6 32.1 (CHz), 122.7, 124.5, 124.7, 125.0, 125.5,126.2,127.2,127.6,127.9,128.1,128.2,129.0, 130.4,130.7, 131.1, and 131.8 ppm (aryl carbons); MS m/z 296 (M + 2, 7 9 , 294 (M+,841, and 215 (M - Br, 100). HRMS: Calcd for C17HllBr, 294.0044; found, 294.0021. A mixture of l-(bromomethy1)pyrene (100 mg, 0.339 mmol) and potassium cyanide (44 mg, 0.68 mmol, dried overnight a t 90 "C under vacuum) in dry DMSO (5 mL) was stirred a t room temperature for 1h and then a t 50 "C for 20 min. After cooling, water was added. The mixture was extracted with CH2C12 and dried over magnesium sulfate. After evaporation of the solvent, the residue was purified by flash column chromatography (silica gel, hexane/ethyl acetate = 73). (1'-Pyreny1)acetonitrile was obtained as a slightly yellow solid (49 mg, 60%): mp 111-112 O C (from hexane and dichloromethane); IR (NaC1) 3050,2984,2295, and 1425 cm-1; 'H-NMR (300 MHz, CDC13) 6 4.29 (8, 2 H, CHz), and 7.94-8.19 ppm (m, 9 H, aromatic); W - N M R (75 MHz, CDCl3) 6 22.1 (CHz), 117.9, 121.4, 122.7, 124.3,124.8,124.9, 125.6,125.8, 126.3,127.2,127.9,128.3,128.7,130.5,13l.l,and 131.3ppm(aryl carbons); MS mlz 241 (M+, loo), 214 (M- CN, 10). HRMS: Calcd for ClsHI1N, 241.0891; found, 241.0879. A mixture of (1'-pyreny1)acetonitrile (24 mg, 0.1 mmol) and potassium hydroxide (500 mg) in water (1mL) and ethanol (0.5

40 Chem. Res. Toricol., Vol. 6, No. I , 1993 mL) was heated under reflux for 45 min. The mixture was cooled to room temperature, acidified with 1 N HC1, extracted with ether, and dried over anhydrous magnesium sulfate. After evaporation of the solvent the residue was purified by flash column chromatography (silica gel, hexaneiethyl acetateiacetic acid = 7:3:0.1). (1'-Pyrenybacetic acid was obtained as a white solid (16 mg, 61 % ): mp 220 "C [from toluene; lit. mp 222.5-225 "C (25)l; IR (NaCl) 3080,1680, and 1450 cm-l; 'H-NMR (300 MHz, CDC13) 6 4.39 (s, 2 H, CHz) and 7.93-8.25 ppm (m, 9 H, aromatic); 13CNMR (75 MHz, CDC13) 6 38.9 (CHZ), 123.0, 124.7, 124.9, 125.0, 125.2, 125.4, 126.0, 126.5, 127.3, 127.4, 128.1, 128.4, 129.5, 130.7, 131.0, and 131.3 ppm; MS m/z 260 (M+, 27), 215 (M - COOH, 100). HRMS: Calcd for C18H1z02, 260.0837; found, 260.0816. Deuterated 1-EP a n d PA. A solution of 1-EP (139 mg, 0.6 mmol) in 3 mL of very dry tetrahydrofuran was stirred a t 0 OC under an argon atmosphere. Ethylmagnesium bromide (1 mL of a 2.0 M solution in tetrahydrofuran, 3.3 equiv) was added slowly by syringe. The reaction mixture was then removed from the ice bath and allowed to reach room temperature. The mixture, which turned from yellow to brown upon addition of the ethylmagnesium bromide, became green and formed a precipitate. After stirring for 3 h, the reaction was recooled in an ice bath. A syringe which had been prerinsed with 0.5 mL of deuterium oxide was used to inject 0.5 mL of deuterium oxide into the stirred reaction mixture. The reaction turned brown. A small amount of anhydrous sodium sulfate was added to the mixture to dry it before the mixture was filtered. Removal of the solvent gave 171.5 mg of a brown solid that was purified by flash silica gel chromatography with diethyl ether as the solvent. Solvent removal and drying over P205under high vacuum gave 139 mg (0.6 mmol, 100% yield) of [2'-2H11-EP. FT-IR showed almost quantitative displacement of the acetylenic band from 3297 cm-l (C-H) to 2580 cm-l (C-D). lH NMR (500 MHz) showed no more than a trace of the signal a t 3.63 ppm attributable to the acetylenic proton. Complete deuteration was confirmed by observation of a mass spectrometric molecular ion peak a t m/z 227 rather than 226. Essentially the same procedure was used to label PA. Enzymes. Microsomes were obtained from Sprague-Dawley male rats (150-200 g) (Bantin-Kingman, Fremont, CA) that had been injected intraperitoneally with PNF in corn oil (50 mgikg, 10 mg/mL corn oil) once a day for 4 days. Microsomes used for the purification of cytochrome P450 1A1, however, were obtained from 5 week old Long Evans male rats (Bantin-Kingman, Fremont, CA) after similar treatment with a 40 mgikg daily dose of PNF. The rats were decapitated 24 h after the final injection of PNF, and the livers were removed and homogenized in 20 mL of buffer (150 mM KCl, 50 mM/Tris-HC1, pH 7.4)iliver. Centrifugation a t lOOOOg for 20 min followed by centrifugation of the supernatant at lOOOOOg for 1 h (4 "C) gave a microsomal pellet that was resuspended in 150 mM KCl/lO mM EDTA and recentrifuged a t lOOOOOg for 1 h. The final microsomal pellet was resuspended in 250 mM sucrose (approximately 3 mliliver), and the suspension was divided into 1-mL aliquots that were stored at -86 "C until used. The microsomes from SpragueDawley and Long Evans rats contained respectively 1.04 and 1.22 nmol of cytochrome P450img of protein. Cytochrome P450 1Al was purified to homogeneity by the method of Ryan et al. (26, 27). The specific content of the preparation was 10.2 nmolimg. Cytochrome P450 reductase, purified as described by Dutton et al. (28), had a cytochrome c reducing activity of 3.13 pmol of cytochrome c reduced/(nmol of cytochrome P450 reductaseimin) at 25 "C. Protein concentrations were determined by the Lowry procedure using bovine serum albumin as the standard (29). Cytochrome P450 1Al Activity Assay. NADPH (1 mM) was added a t 25 "C to a 1-mL incubation mixture containing rat liver microsomal cytochrome P450 ( 5 pM), 3 mM MgClz, and an arylacetylene (1-EP, 120 pM, or PA, 10mM) in 50 mM potassium phosphate buffer (pH 7.4). A t desired time points, a 0.1-mL aliquot was transferred to a tube containing 1 mL of 50 mM potassium phosphate buffer (pH 7.4) containing 1mM NADPH, 20 mM 7-ethoxyresorufin, and 3 mM MgC12. After incubation

Chan et al. a t 25 OC for 10 min, 3 mL of cold acetone was added, and the solution was mixed for 10 s and was then placed on ice. The precipitated protein was sedimented by centrifugation, and the amount of resorufin in the supernatant was quantitated by its fluorescence (emission X = 585 nm, excitation X = 535 nm) relative to the emission of standard samples (30). Standards were prepared by adding 5-10 pmol of resorufin rather than 7-ethoxyresorufin to the enzyme mixture as described above and quenching immediately with acetone. The base line was obtained with an assay mixture to which 7-ethoxyresorufinbut no resorufin had been added. Experiments with purified, reconstituted cytochrome P450 1 A l were carried out similarly except that 1 nmol of purified cytochrome P450 1A1, 2 nmol of cytochrome P450 reductase, and 40 pg of DLPC replaced the liver microsomes in the incubation mixture and a 5- rather than 10-min incubation period was used for the 7-ethoxyresorufin 0-deethylation assay. Soret Loss. Cytochrome P450 l A l ( 1nmol), cytochrome P450 reductase (2 nmol), and DLPC (40 pg) were allowed to stand a t room temperature for 5 min before taking up the mixture in 50 mM potassium phosphate buffer (pH 7.4) containing 3 mM MgC12. Catalase (1 unit/mL) was added to some incubations to suppress the contribution of HzOzt o Soret loss. Arylacetylenes in acetonitrile were added to the following final concentrations: l-EP, 40 pM, and PA, 10 mM. The reference cuvette received an equal amount of' acetonitrile. In work with microsomes, the reconstituted cytochrome P450 was replaced by rat liver microsomes (final P450 concentration 10 KM). The incubations (25 "C) were started by adding NADPH (1mM). At the desired time points, 100-pL aliquots were transferred to sample and reference cuvettes containing 1 mg of dithionite in 0.4 mL of potassium phosphate buffer (pH 7.4). The sample tube was immediately bubbled with carbon monoxide for 15 s, and the difference spectrum was recorded on a Aminco DW2000 instrument. Formation of (H'-Pyreny1)acetic Acid. A 1-mL incubation mixture containing 1.0nmol of liver microsomal cytochrome P450 from PNF-induced rats, 3 mM MgC12, and 120 pM 1-EP in 50 mM potassium phosphate buffer (pH 7.4) was warmed t o 25 "C. NADPH (final Concentration 1 mM) was then added to initiate the reaction. At time points between 0.5 and 30.0 min, a 100-pL aliquot was quickly transferred to 1mL of ice-cold 10% HC1 and the mixture was stirred vigorously for 10 s before placing it on ice. In some incubations, deuterated 1-EP was used as the substrate. The quenched aliquots were extracted thrice with 2 mL of diethyl ether, and the combined extracts were combined with 2 mL of a diethyl ether solution of diazomethane and allowed to stand a t 25 "C in a closed vessel for 40 min. The ether was then allowed to evaporate before 0.5mL of acetonitrile was added. The resulting solution was filtered and a 25-pL aliquot was analyzed by HPLC on a Dupont Zorbax ODS column (4.6 X 250 mm) eluted from 0 to 10 min with 70% (v/v) acetonitrileiwater, from 1Oto 15minwithalineargradientrisingto100%acetonitrile, and from 15 to 25 min with acetonitrile. The column flow rate was 1 m l i m i n , and the column effluent was monitored at 342 nm. The retention times of methyl (1'-pyreny1)acetate and 1-EP in this system were 15.9 and 23.4 min, respectively. Spectroscopic Binding Constants. A 1-mL suspension of liver microsomes from @NF-inducedrats in 50 mM potassium phosphate buffer (pH 7.4), containing 1nmol of cytochrome P450, was added to each of a pair of matched quartz cuvettes. Alternatively, 1 nmol of purified cytochrome P450 1Al in 1mL of the same buffer was added to both cuvettes. Increasing amounts of PA (1-5 pL of a 5 M solution in acetonitrile) or 1-EP (2.5-25 p L of a 10 mM solution in acetonitrile) were then added with vigorous mixing t o the sample cuvette, and an equivalent amount of the solvent was added to the reference cuvette, and the difference spectra were recorded on an Aminco DW2000 spectrophotometer. Covalent B i n d i n g of 1-EP t o Microsomal Cytochrome P450. Two methods were utilized to quantitate NADPH-

Chem. Res. Toxicol., Vol. 6, No. 1, 1993 41

Cytochrome P450 Inactivation by Arylacetylenes

dependent covalent binding of [3H]1-EPto rat liver microsomal cytochrome P450 Method A. NADPH (0.5 pmol) was added to a 0.5-mL incubation of BNF-induced rat liver microsomes in 50 mM potassium phosphate buffer (pH 7.4) containing 0.5 nmol of cytochrome P450 and 120 pM PHIl-EP, and the mixture was incubated at 25 "C for 70 min. NADPH was omitted from otherwiseidentical controlincubations. A 10-foldvolume (5 mL) of 5% sulfuric acid in methanol was added to quench the incubation, and the mixture was mixed vigorously for 10 8. The protein pellet obtained after centrifugation at 3000 rpm for 10 min was washed three or more times by resuspension in 5 mL of 5 % sulfuric acid in methanol and recentrifugation. The washes were continued until there was less than 300 cpm/0.5 mL of supernatant. The final protein pellet was dissolved in 0.6 mL of 1 N NaOH, and the resulting homogeneous sample was subjected to liquid scintillation counting. In order to avoid fluorescenceinterference, it was necessary to count less than 0.1 mL of the sample. Method B. NADPH (1pmol)was added to a 1-mL incubation mixture in 50mM potassium phosphate buffer (pH7.4) containing rat liver microsomal cytochromeP450 (5pM enzyme)and 1EP (120pM). Control incubationswere initiated by addingbuffer rather than NADPH. After incubation for 70 min at 25 O C , unlabeled 1-EP (12 mM, a 100-fold excess over [3H]1-EP)was added and the sample was shaken vigorously for 1min before it was loaded onto a long (70- X 1.5-cm),freshly packed Sephadex G-25 (Sigma) column. The protein was eluted from the column with 50 mM potassium phosphate buffer (pH 7.4). The protein eluted in fractions 11-17 when 4-mL fractions were collected and counted.

Results Inactivation of Cytochrome P450 1Al by Arylacetylenes. Incubation of liver microsomes from BNFinduced rats at 25 "Cwith 40 p M 1-EPresults in time- and NADPH-dependent loss of 7-ethoxyresorufin O-deethylation activity. 7-Ethoxyresorufin 0-deethylation is a selective indicator of the catalytic activity of cytochrome P450 1Al (31). A semilog plot of the activity remaining as a function of time gives a straight line from which it can be calculated that the t1p for inactivation is 5.2 min (Figure 2). Similar experiments with purified, reconstituted cytochrome P450 1Al show that the purified enzyme is inactivated by 40 p M 1-EPwith t 1 / 2 = 2.6 min (not shown). Incubation of liver microsomes from BNF-induced rats with P A shows that it also inactivates cytochrome P450 1Al (Figure 2). The t1p for inactivation by 10 mM PA is 3.7 min. Reconstituted cytochrome P450 1Al is likewise inactivated with t1p = 2.3 min by 10 mM PA. Destruction of the Cytochrome P450 Chromophore by Arylacetylenes. Incubation of rat liver microsomes from BNF-induced rata with 120 pM 1-EPdoes not cause detectable loss of the total cytochrome P450 chromophore, as determined by difference spectrophotometric quantitation of the reduced carbon monoxide complex (Figure 2). In order to estimate the cytochrome P450 1Al chromophore loss directly caused by 1-EP, it is necessary to include catalase in the incubation because a significant fraction of the enzyme chromophore is otherwise destroyed by reaction with hydrogen peroxide generated by uncoupling of the reconstituted enzyme (32). In the presence of catalase, under conditions where no loss is observed in the absence of 1-EP,up to 38% of the Soret absorbance of the reconstituted enzyme is destroyed by 1-EP in a time- and NADPH-dependent manner. The purified and microsomal enzymes thus respond differently to 1-EPwith respect to chromophore loss.

."

1

.-.-e M

I

a. 1-EP

e

E

M

10

.-e

2e

-

0

10

20

I

b. PA

30 40 50 time (min)

60

70

1

I

1

c

0

5

.1

.-P

3 .

s i

e

.01 0

2

4 6 time (min)

8

10

Figure 2. Time dependence of the loss of rat liver microsomal 7-ethoxyresorufin 0-deethylation activity ( 0 )and cytochrome P450 chromophore (A)caused by (a) 1-EP and (b) PA. The chromophore loss is corrected for enzyme that is resistant to inactivation by taking the chromophore present after 15min ns the zero value. PA causes significant, time-dependent loss of microsomal cytochrome P450. The decrease in the absorption of the ferrous carbon monoxide complex levels off after approximately 15min, at which point -65% of the initial chromophore has been lost. The residual chromophore represents the residual chromophore of inactivated enzyme and cytochrome P450 isozymes not susceptible to inactivation by PA. The decline in the Soret absorbance, after correction for the residual cytochrome P450 chromophore, indicates that the chromophore is destroyed with a halflife of 4.2 min (Figure 2). This half-life is very similar to that determined by measuring the rate of loss of 7-ethoxyresorufin 0-deethylation activity (3.7 min) (Figure 2). Inactivation of cytochrome P450 1A1, the enzyme that turns over 7-ethoxyresorufin, by PA thus involves modifications of the heme group. Incubation of purified, reconstituted cytochrome P450 1Al with PA also causes chromophore loss, although chromophore loss levels off after 6 0 4 % of the chromophore is lost. Spectroscopic Binding Constants for 1-EPand PA. for the binding of 1-EP The spectroscopic constants (K,) and PA to BNF-induced rat liver microsomes and purified cytochrome P450 1Al have been determined in order to compare the affinities of 1-EP and PA for the enzyme. The K,values for the binding of 1-EPto the microsomes and the purified enzyme are 13and 0.13 p M , respectively. Tighter binding of 1-EPto the purified enzyme is readily explained by the fact that partitioning between water and the active site favors the active site to a greater extent than partitioning between water, the microsomal membrane, and the active site. The K,values for the binding of PA to the membrane-bound and purified enzyme, in contrast, are 29 OOO and >29 OOO pM. 1-EPthus binds to purified cytochrome P450 1Al with 200 OOO times higher affinity, and to the microsomal form of the enzyme with -2000 times higher affinity, than PA.

-

42 Chem. Res. Toxicol., Vol. 6, No. 1, 1993

1

.-

h 3

.-+ 3

m

a

Chan et al. Table I. Isotope Effects on Inactivation of Cytochrome P450 1Al and Spectroscopic Binding Constants for PA and 1-EP 1-EP PA

I-EP D.EP

~~

a*

microsomal reconstituted

0 v1

B

@

1

0

1

2 3 time (min)

5

4

I

1 J

1

2

3

5

4

time (min)

A

0

1

2 3 time (min)

I.EP D-EP

4

5

Figure 3. Ratesof inactivationof cytochrome P450 1 A l by 1-EP (A)and deuterated 1-EP ( 0 )as measured by 7-ethoxyresorufin O-deethylation: (a) purified enzyme and (b) PNF-induced rat liver microsomes. (c) Rates of (1'-pyreny1)aceticacid production in incubations of 1-EP (A)and deuterated 1-EP ( 0 )with PNFinduced rat liver microsomes. Cytochrome P450 1 A l activity is expressed as nmol of resorufin formediminlnmol of total cytochrome P450 at 25 O C . The amount of product formed is expressed as nmol of (1'-pyreny1)acetic acidinmol of total cytochrome P450 at 25 "C.

1-EP and PA Metabolites. Incubation of 1-EP with 6"-induced rat liver microsomes yields (1'-pyreny1)acetic acid as the primary detectable metabolite (Figure 1).The acid metabolite was characterized by electronic absorption spectroscopic and HPLC comparisons with an authentic standard after esterification of the carboxylic acid group of both samples with diazomethane. Similar experiments confirm that PA is oxidized to phenylacetic acid by BNFinduced rat liver microsomes, as previously demonstrated for oxidation of P A by phenobarbital-pretreated rat liver microsomes or reconstituted, purified cytochrome P450 2B1 (Figure 1) ( 4 , 5 ) . Isotope Effects on Enzyme Inactivation and Metabolite Formation. Replacement of the acetylenic hydrogen in 1-EPby deuterium does not significantly alter the rate of inactivation of purified, reconstituted cytochrome P450 1 A l or of the ethoxyresorufin O-deethylase activity catalyzed by it in BNF-induced rat liver microsomes (Figure 3, Table I). The ratio of the rates of inactivation of the purified enzyme is thus kHlkD = 1.08, a value that is not significantly different from 1.00. The corresponding ratio for inactivation of the O-deethylase

kdkD 1.03 1.08

K,(PM) 13 0.13

kdkD 1.01 nd@

Ks(PM) >29OOO 29 OOO

nd = not determined.

activity of the liver microsomes is even lower ( k ~ l =k ~ 1.03). These results clearly indicate that the acetylenic C-H bond is not broken in the rate-limiting step@)of the process that results (result) in enzyme inactivation. Previous studies have demonstrated that oxidation of PA to phenylacetic acid by cytochrome P450 2B1 is subject to an isotope effect of 1.7-1.8 when the acetylenic hydrogen is replaced by deuterium ( 4 , 5 ) . Examination of the formation of (1'-pyreny1)acetic acid in incubations of cytochrome P450 1 A l with 1-EPor ita ethynyl-deuterated analogue, however, shows that the metabolite is formed at essentially the same rate in both instances ( k ~ l =k ~ 1.03) (Figure 3). Formation of the metabolite therefore does not involve migration of the acetylenic proton in the rate-determiningstep even though ita migration is required to explain retention of the deuterium in the acetic acid metabolite. Quantitation of the formation of (1'-pyreny1)acetic acid and loss of cytochrome P450yields a metabolite: inactivation partition ratio of 43 for 1-EP and 36 for its deuterated analog. The absence of a detectable isotope effect for 1-EP on either enzyme inactivation or metabolite formation, compared to the isotope effect observed for cytochrome P450 2B1-catalyzed formation of phenylacetic acid from PA (4, 5 ) ,could reflect a difference in the oxidation mechanisms or differences in the interactions of the two arylacetylenes with the enzyme. In an effort to differentiate these alternatives, we have determined whether the oxidation of PA by cytochrome P450 1 A l is subject to an isotope effect. In agreement with earlier studies on the inactivation of cytochrome P450 2B1, cytochrome P450 1 A l is inactivated at the same rate by PA and its ethynyldeuterated derivative ( k d k D = 1.01) (Figure 4). However, in contrast to the work with cytochrome P450 2B1, the formation of phenylacetic acid is not subject to more than a marginal isotope effect ( k ~ l k= ~1-09) (Figure 4). Comparison of the amount of metabolite formed and enzyme lost yields partition ratios of 5 and 4, respectively, for PA and its deuterated analogue. The inactivation is thus much more efficient for PA than 1-EP, for which the partition ratio is approximately 40. Covalent Binding of 1-EP. Inactivation of the 'I-ethoxyresorufin O-deethylase activity of hepatic microsomes from PNF-induced rata by radiolabeled 1-EP is accompanied by NADPH-dependent covalent binding of 0.80.9 nmol of the inactivating agenthmol of total microsomal cytochrome P450 (Table 11). The extent of binding of 1-EP was determined after washing the microsomes with 5% sulfuric acid in methanol to remove noncovalently bound radioactivity. A variety of other solventa,including ethyl acetate, diethyl ether, 20 3'% trichloroacetic acid, and acetic acid in ethyl acetate were tried but were not found to efficiently remove nonspecifically bound material. The total radioactivity determined after washing the microsomes repeatedly with 5% sulfuric acid in methanol was corrected for the background radioactivity obtained

Cytochrome P450 Inactivation by Arylacetylenes

I

1-J

.1

0

.n lV

2

4

6 8 time (min)

1

0

I

0

1

2

I

1

2

3 4 time (min)

5

6

Figure 4. Rates of (a) inactivation of microsomal cytochrome P450 1Al and (b) formation of phenylacetic acid in incubations of PNF-inducedrat liver microsomes with PA (A)and deuterated PA ( 0 ) . Cytochrome P450 1Al activity is expressed as nmol of resorufin formed/(min.nmoloftotal cytochrome P450)at 25 O C . The amount of product formed is expressed as nmol of phenylacetic acid/nmol of total cytochrome P450 at 25 “C. Table 11. Covalent Binding of 1-EPto BNF-Induced Rat Liver Microsomal Protein radiolabeled protein (nmol/nmol of total P450) NADPH +NADPH -NADPH dependent precipitation assay (n = 6) 0.98 0.06 0.92 Sephadex G-25(n = 1) 1.31 0.53 0.78

by analogous workup of parallel incubations to which no NADPH was added. The sulfuric acid washes remove the heme group from cytochrome P450, so that the residual radioactivity represents material covalently bound to the protein rather than the heme group. Covalent binding of 1-EPhas also been demonstrated by passing the microsomes through a long Sephadex G-25 column after incubation with radiolabeled 1-EP. This method requires addition of a large excess of unlabeled 1-EP to the microsomes prior to passage through the column to help displace nonspecificallybound radiolabeled 1-EP. This methqd indicates that 0.8 nmol of radiolabel is bound in an NADPH-dependent manner per nanomole of microsomal cytochrome P450.

Discussion The inactivation of cytochrome P450 by arylacetylenes generally involves heme or protein alkylation by reactive species produced by catalytic oxidation of the triple bond (1-3). Earlier work demonstrated that phenylacetylenes inactivate microsomal and purified cytochrome P450 2B1 by alkylating the prosthetic heme group ( 4 , 5 ) ,whereas 1-ethynylpyrene decreases the cytochrome P450 1 A l activity of hepatic microsomesfrom BNF-induced rats with little apparent loss of the cytochrome P450 chromophore (18, 19). This suggests that cytochrome P450 1 A l is inactivated by protein alkylation. The mechanism of the

Chem. Res. Toricol., Vol. 6 , No. I, 1993 43

inactivation reaction (Le., heme or protein alkylation) therefore may be isozyme specific. To examine this possibility, we have first confirmed that 1-EPdoes, indeed, inactivate microsomal cytochrome P450 1 A l by a mechanism that does not detectably involve heme modification (Figure 2). The inference that inactivation of cytochrome P450 1Al involves covalent binding of catalyticallyactivated 1-EPto the protein is supported by the observation that inactivation is associated with covalent binding of 0.84.9nmol of radiolabeled substrate per nanomole of microsomal cytochrome P450 (Table 11). This corresponds to slightly more than 1equiv per nanomole of cytochrome P450 1Al because this isozyme accounts for 7@75% of the cytochrome P450 content of hepatic microsomes from PNF-induced rats (33). Surprisingly, inactivation of purified, reconstituted cytochrome P450 1A1, in contrast to inactivation of the microsomal enzyme, occurs with some chromophore loss (Figure 2). The reason for the difference between the microsomal and reconstituted enzyme is not obvious but may involve changes in the heme environment caused by separation of the protein from its lipid environment. The catalytic properties of at least some cytochrome P450 enzymes have been shown to be sensitive to their membrane environment (34). Alternatively, the chromophore loss observed with the purified enzyme could be due to reaction with peroxide formed by uncoupled turnover of the enzyme. Although catalase is added to prevent this process, it is conceivable that sufficient hydrogen peroxide escapes detoxification by catalase to cause partial inactivation of the enzyme. In contrast to 1-EP, inactivation of microsomal cytochrome P450 1 A l or the purified, reconstituted enzyme by PA involves destruction of the prosthetic heme group (Figure 2). Thus, the half-livesfor loss of the chromophore (4.2 min) and microsomal 7-ethoxyresorufin 0-deethylase activity (3.7 min) are nearly the same. Chromophore loss is also observed with the purified enzyme. A residual chromophore absorbance is observed with both the fully inactivated microsomal and purified enzymes. The residual absorbance is partially due to the alkylated prosthetic group but could also be due to a small amount of protein inactivation by a mechanism other than heme alkylation or, in the microsomal case, to isozymes that do not oxidize PA or 7-ethoxyresorufin. The absorption of the alkylated prosthetic group was found in earlier studies of the inactivation of cytochrome P450 2B1 by PA to cause an apparent leveling of the chromophore loss (5). Despite the differences in the inactivation mechanisms, the only detectable metabolites of PA and 1-EP involving oxidation of the triple bond are, respectively, phenylacetic acid and (1’-pyreny1)acetic acid (Figure 1). This is consistent with the earlier findings that phenylacetic acid is the only detectable PA metabolite produced by cytochrome P450 2B1 ( 4 , 5 ) and (2’-napthyl)acetic acid is the only 2-ethynylnaphthalene metabolite produced by cytochrome P450 1A2 (12). Inactivation of the enzyme by both PA and 1-EP thus appears to involve oxidation of the triple bond, although the different inactivation mechanisms imply differences in the triple bond oxidation. As already noted, heme alkylation by acetylenic agents involves delivery of the ferry1oxygen to the internal carbon of the triple bond and alkylation of the heme by the terminal carbon (1,3). This is presumably the mechanism involved in inactivation of cytochrome P450 1 A l by PA,

44 Chem. Res. Toxicol., Vol. 6, No. 1, 1993

although the resulting heme adduct has not actually been isolated. In contrast, formation of the acetic acid metabolites requires delivery of the oxygen to the terminal carbon to generate a ketene that is converted to the acetic acid derivative by addition of water (8-10). The ratio of metabolite formation to inactivation is therefore determined for PA by the extent to which the ferryl oxygen is delivered to the internal versus external carbon of the triple bond. Alkylation of the protein skeleton of cytochrome P450 1 A l by catalytically activated 1-EP, as previously argued for the inactivation of cytochrome P450 4A1 by 10undecynoic acid (13), is likely to involve competition between the protein and water for the ketene intermediate. The formation of (1’-ppeny1)aceticacid clearly shows that the ketene is formed by catalytic oxidation of 1-EP. Furthermore, the stoichiometry of protein alkylation is approximately that expected if inactivation is caused by binding of one activated molecule of 1-EP to the protein. This covalent binding involves the protein rather than the heme group because the chromophore is not lost in the incubation and the radioactivity is retained through washing procedures that remove the prosthetic heme group. The partition ratios for inactivation by PA and 1-EP are of interest in this context. PA, with a partition ratio between metabolite formation and inactivation of 4-5, is a much more efficient inactivating agent than 1-EP, for which the partition ratio is 10-fold higher. The low partition ratio for heme alkylation by PA, if the assumption is made that the ketene derived from it would inactivate the enzyme with a partition ratio as high as that for 1-EP, clearly explains why PA inactivates the enzyme primarily by heme rather than protein alkylation. Heme alkylation is not expected to be sensitive to replacement of the acetylenic hydrogen by deuterium because the carbon-deuterium bond is not broken. In accord with this, no isotope effect was found earlier on either chromophoreor activity loss during the inactivation of cytochrome P450 2B1 by PA (5). Likewise, no isotope effect is observed here on inactivation of cytochromeP450 1Al by PA. The formation of phenylacetic acid from PA, however, requires migration of the acetylenic hydrogen to the vicinal carbon and was found in the reaction catalyzed by cytochrome P450 2B1 to be subject to a large isotope effect (4,5). However, no isotope effect is observed in the formation of phenylacetic acid by cytochrome P450 1 A l (Figure 4). Oxygen transfer to the triple bond is therefore not the rate-limiting step in the oxidation of PA by cytochrome P450 1 A l because hydrogen migration and oxygen transfer are concerted events ( 4 , 5 ) . The absence of an isotope effect on metabolite formation or enzyme inactivation is readily explained, despite the fact that both processes depend on ketene formation, if the rate-limiting step occurs before or after ferryl oxygen transfer to 1-EP. The present resultssupport the view that the mechanism of inactivation of a given cytochrome P450 enzyme by an arylacetylene depends on the ratio of delivery of the ferryl oxygen to the internal (heme alkylation) versus terminal (protein alkylation) carbon of the triple bond. The first of these processes appears to be highly efficient in that no metabolite from delivery of the oxygen to the internal carbon is detected other than the heme adduct itself. Delivery of the oxygen to the terminal carbon, on the other hand, is less efficient as an inactivation mechanism in that considerable acetic acid metabolite is formed. This

Chan et al.

suggests that the ketene is hydrolyzed, either within the active site or after diffusion out of it, at a rate that competes effectively with reaction with a protein nucleophile. The modification of the protein that inactivates it may not occur in the active site if the ketene readily diffuses out of the active site. The factors that determine whether oxidation of an acetylene results in oxygen addition to the internal or terminal carbons are thus critical determinants of the inactivation mechanism. The much higher affinity of 1-EP (K,= 13 pM) than PA (K,= >29 OOO pM) for the enzyme suggests that restrictions in the mobility of the substrate caused by high-affinity binding, and therefore of the accessible orientations of the triple bond with respect to the ferryl oxygen, contribute to the nature of the inactivation event.

Acknowledgment. We thank Prof. Chin-Tzu Peng for kindly carrying out the radiolabelingof 1-EPand Dr. David R. Dutton for help in purifying cytochrome P450 1Al. This work was supported by Grant GM 25515 from the National Institutes of Health. Support services were provided by the Liver Core Center (R. Ockner, Director), funded by Grant P-30 DK 26743, and the Biomedical, Bioorganic Mass Spectrometry Facility of the University of California, San Francisco (A. Burlingame, Director), supported by Grant RR 01614.

References Ortiz de Montellano, P. R. (1985) Alkenes and alkynes. In Bioactivation offloreign Compounds (Anders, M. W., Ed.) pp 121155, Academic Press, New York. Ortiz de Montellano, P. R. (1986)Oxygen activation and transfer. In Cytochrome P450 Structure, Mechanism, and Biochemistry (Ortiz de Montellano, P. R., Ed.) pp 217-271, Plenum Press, New York. Ortiz de Montellano, P. R., and Reich, N. 0. (1986) Inhibition of cytochrome P-450 enzymes. In Cytochrome P450 Structure, Mechanism, and Biochemistry (Ortiz de Montellano, P. R., Ed.) pp 273-314, Plenum Press, New York. Ortiz de Montellano, P. R., and Komives, E. A. (1985) Branchpoint for heme alkylation and metabolite formation in the oxidation of arylacetylenes by cytochrome P-450. J.Biol. Chem. 260,3330-3336. Komives, E. A., and Ortiz de Montellano, P. R. (1987) Mechanism of oxidation of r bonds by cytochrome P-450. Electronic requirementa of the transition state in the turnover of phenylacetylenes. J.Biol. Chem. 262, 9793-9802. Wade, A,, Symons, A. M., Martin, L., and Parke, D. V. (1979) Metabolic oxidation of the ethynyl group in 4-ethynylbiphenyl. Biochem. J. 184, 509-517. Sullivan, H. R., Roffey, H. R., and McMahon, R. E. (1979) Biotransformation of 4’-ethynyl-2-fluorobiphenylin the rat. In oitro and in vivo studies. Drug Metab. Dispos. 7, 76-80. Wade, A., Symons, A. M., Martin, L., and Parke, D. V. (1980) The metabolic oxidation of the ethynyl group in 4-ethynylbiphenyl in vitro. Biochem. J. 188, 867-872. Ortiz de Montellano, P. R., and Kunze, K. L. (1980) Occurrence of a 1,2-shift during enzymatic and chemical oxidation of a terminal acetylene. J. Am. Chem. SOC. 98, 7373-7375. Ortiz de Montellano, P. R., and Kunze, K. L. (1981) Shift of the acetylenic hydrogen during chemical and enzymatic oxidation of the biphenylacetylene triple bond. Arch. Biochem. Biophys. 209, 710-712. McMahon, R. E., Turner, J. C., Whitaker, G. W., and Sullivan, H. R. (1981) Deuterium-isotope effect in the biotransformation of 4-ethynylbiphenyle to 4-biphenylacetic acids by rat hepatic microsomes. Biochem. Biophys. Res. Commun. 99,662-7. Hammons, G. J., Alworth,W. L., Hopkins, N. E., Guengerich, F. P., and Kadlubar, F. F. (1989) 2-Ethynylnaphthalene 88 a mechanismbaaed inactivator of the cytochrome P-450 catalyzed N-oxidation of 2-naphthylamine. Chem. Res. Toxicol. 2, 367-374. CaJacob, C. A., Chan, W. K., Shephard,E., and Ortiz de Montellano, P. R. (1988) The catalytic site of rat hepatic acid w-hydroxylase. Protein versus prosthetic heme alkylation in the w-hydroxylationof acetylenic fatty acids. J. Biol. Chem. 263, 18640-18649.

Cytochrome P450 Znactioation by Arylacetylenes (14) O'Ferrall, R. A. M. (1970) Model calculations of hydrogen isotope effects for nonlinear transition states. J . Chem. SOC.,Sect. B , 785790. (15) Kunze, K. L., Mangold, B. L. K., Wheeler, C., Beilan, H. S., and Ortiz de Montellano, P. R. (1983)The cytochrome P-450 active site. Regiospecificity of prosthetic heme alkylation by olefins and acetylenes. J . Biol. Chem. 258, 4202-4207. (16) Ortiz de Montellano, P. R., and Kunze, K. L. (1981) Cytochrome P-450 inactivation: structure of the prosthetic heme adduct with propyne. Biochemistry 20, 7266-7271. (17) Ortiz de Montellano, P. R., Beilan, H. S.,and Mathews, J. M. (1982) Alkylation of the prosthetic heme in cytochrome P-450 during oxidative metabolism of the sedative-hypnotic ethchlorvynol. J . Med. Chem. 25, 1174-1179. (18) Gan, L.-S. L., Acebo, A. L., and Alworth, W. L. (1984) 1-Ethynylpyrene,a suicide inhibitor of cytochrome P-450 dependent benzo[alpyrene hydroxylase activity in liver microsomes. Biochemistry 23,3827-3836. (19) Gan, L.4. L., Lu, J.-Y., Hershkowitz, D. M., and Alworth, W. L. (1985)Effecta of acetylenic and olefinic pyrenes upon cytochrome P-450 dependent benzo[a]pyrene hydroxylase activity in liver microsomes. Biochem. Biophys. Res. Commun. 129, 591-596. (20) Nebert, D. W., Nelson, D. R., Coon, M. J., Estabrook, R. W., Feyereisen, R., Fujii-Kuriyama, Y., Gonzalez,F. J., Guengerich, F. P., Gunsalus, I. C., Johnson, E. F., Loper, J. C., Sato, R., Waterman, M. R., and Waxman, D. J. (1991)The P450 Superfamily: Update on New Seauences. Gene MaDDinE. and Recommended Nomenclature. DNA Cell.' Biol. 10, i-i4.-' (21) Hiltunen. J.. Pew. C. T.. and Yana. Z. C. (1988) Tritium labeling . of simple 7-membered ring compounds. J . Labelled Compd. Radiopharm. 28, 543-554. (22) Johnson, M. R., and Rickborn, B. (1970) Sodium borohydride reduction of conjugated aldehydes and ketones. J. Org. Chem. 35, 1041-1045. (23) Bachmann, W. E., and Carmack, M. (1941) Methyl derivatives of 3,4-benzpyrene. The Willgerodt reaction on some 3-acylpyrenes. J . Am. Chem. SOC. 63, 2494-2500. (24) Akiyama, S., Nakasuji, K., and Nakagawa, M. (1971) Linear conjugated systems bearing aromatic terminal groups. IV. Bull. Chem. SOC.Jpn. 44, 2231-2236.

Chem. Res. Tonicol., Vol. 6, No. 1, 1993 46 (25) Klassen, S. E., Daub, G. H., andVanderJagt, D. L. (1983)Carbon-13 labeled benzo[alpyrenes and derivatives. 4. Labeling the 7-10 positions. J . Org. Chem. 48, 4361-4366. (26) Ryan, D. E., Thomas, P. E., Korzeniowski,D., and Levin, W. (1979) Separation and characterization of highly purified forms of liver microsomalcytochromeP-450from rata treated with polychlorinated biphenyls,phenobarbital, and 3-methylcholanthrene. J . Bid. Chem. 254, 1365-1374. (27) Ryan, D. E., Thomas, P. E., and Levin, W. (1982)Purification and characterization of a minor form of hepatic microsomal cytochrome P-450 from rata treated with polychlorinated biphenyls. Arch. Biochem. Biophys. 216, 272-288. (28) Dutton, D. R., McMillen, S. K., Sonderfan, A. J., Thomas, P. E., and Parkinson, A. (1987) Studies on the rate-determining factor in testosterone hydroxylation by rat liver microsomal cytochrome P-450 evidence against cytochrome P-450 isozyme-isozyme interactions. Arch. Biochem. Biophys. 255, 316-328. (29) Lowry, 0. H., Rosebrough, N. J., Fan, A. L., and Randel, R. J. (1951) Protein measurement with the Folin phenol reagent. Biochemistry 193, 265-275. (30) Burke, M. D., and Mayer, R. T. (1974) Ethoxyresorufin: direct fluorimetric assay of a microsomal 0-dealkylation which is preferentially inducible by 3-methylcholanthrene. Drug Metab. Dispos. 2, 583-588. (31) Guengerich, F. P. (1987)Enzymologyof rat liver cytochromes P-450. In Mammalian Cytochrome P-450 (Guengerich, F. P., Ed.) Vol. 1, pp 1-54, CRC Press, Boca Raton, FL. (32) Archakov, A. I., and Bachmanova, G. I. (1990) Cytochrome P-450 and Active Oxygen, pp 129-138, Taylor & Francis, New York. (33) Guengerich, F. P., Wang, P., and Davidson, N. K. (1982) Estimation of isozymes of microsomal cytochrome P-450 in rata, rabbits, and human using immunochemicalstaining coupled with sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Biochemistry 21,16981706. (34) Christou, M., Mitchell, M. J., Aoyama, T., Gelboin, H. V., Gonzalez, F. J., and Jefcoate, C. R. (1992)Selective suppression of the catalytic activity of cDNA-expressed cytochrome P4502B1toward polycyclic hydrocarbons in the microsomal membrane: modification of this effect by specific amino acid substitutions. Biochemistry 31,28352841.