Differential Stereoselectivity of Cytochromes P-450b and P-45Oc in ...

THEJOURNAL OF BIOLOGICAL CHEMISTRY

Vol. 260, No. 18, Issue of August 25, pp. 10226-10235,1985 Printed in U.S. A.

Differential Stereoselectivity of Cytochromes P-450b and P-45Oc in the Formationof Naphthalene and Anthracene 1,a-Oxides

(Received for publication, October 25, 1984)

Peter J. van Bladeren$, Jane M. Sayer$, Dene E. Ryan& PaulE. Thomas§, Wayne Levine, and Donald M. Jerina$ From the $Laboratory of Bioorganic Chemistry, Section onOxidation Mechanisms, National Institute of Arthritis, Diabetes, and Digestive and Kidney Diseases, National Institutes of Health, Bethesda, Maryland 20205 and the §Laboratory of Experimental Carcinogenesis and Metabolism, Roche Znstitute of Molecular Biology, NutZey, New Jersey 07110

As is the case for cytochrome P-450c, arene 1,Zoxides have been identified as initial metabolites when naphthalene andanthracene are oxidized by cytochrome P-450b ina highly purified, reconstitutedsystem. Overall rates of metabolism by cytochrome P450b are >%fold and >50-fold lower than therespective rates of metabolism by cytochrome P-45Oc. For both hydrocarbons,the(-)-(lS,ZR)-oxide predominates (74%) withcytochrome P-450b as the terminal oxidant, based on trapping the labile arene oxides as N-acetyl-L-cysteine S-conjugates of known absolute to configuration. This resultis in marked contrast data obtained with cytochrome P-45Oc wherethe (+)(lR,ZS)-oxides predominate (73->95%). Intheabsence of added epoxide hydrolase, the metabolically formed arene oxides rapidly isomerize to phenols. Addition of increasing amounts of epoxide hydrolase to the incubation medium results in the formation of trans-1,Z-dihydrodiols at the expense of phenols from the common arene oxide intermediates. Evaluation of the kinetic parameters ( K , and kcat)for the hydration of the (+)- and (-)-enantiomersof both arene oxides by epoxide hydrolase has indicated that the (+)-(lR,ZS)enantiomersexhibitlower values of K , (-1 PM) whereas the values of kcatare similar for both enantiomers of a given arene oxide. These parameters have allowed construction of a mathematical model which predicts the enantiomer composition of the dihydrodiols formed from naphthalene in reconstituted systems containing specific epoxide hydrolase concentrations. The data reported argue against a selective functional coupling mechanism between cytochrome P450cand epoxide hydrolase in the metabolism of naphthalene and anthraceneto the 1,Z-dihydrodiols.

Although evidence exists for alternate pathways for the metabolism of aromatic hydrocarbons to phenols by monooxygenase enzymes (1-3), arene oxides are well established as the preponderant if not exclusive metabolic intermediates in the formation of phenols and dihydrodiols from many aromatic substrates (4). Whereas phenols arise via spontaneous isomerization of arene oxides (5), the formation of dihydrodiols generally requires catalysis by epoxide hydrolase (6).

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Early studies on the liver microsomal metabolism of naphthalene suggested that certain isozymes of cytochrome P-450 might be more tightly coupled with epoxide hydrolase than others (7). In particular, results with intact microsomes and partiallypurified enzymes revealed a higher ratio of trans1,2-dihydrodiol to naphthol formed after 3-methylcholanthrene treatmentof rats compared to phenobarbital treatment (8). Since these early observations, which suggested an apparent selective coupling of epoxide hydrolase with cytochrome P-448 ( P - ~ ~ O Cother ) , results have been obtained which argue against a differential functional coupling mechanism between epoxide hydrolase andthe various cytochromes P-450. Detailed metabolism studies of a series of polycyclic hydrocarbons other than naphthalene have not revealed any patternof substantially higher ratios of dihydrodiols to phenols formed with microsomes prepared from rats treated with 3-methylcholanthrene compared to untreated or phenobarbital-treated rats (cf. Ref. 9). In addition, epoxide hydrolase is regulated independently from the cytochrome P450 isozymes. For example, cytochrome P-45Oc is induced d50-fold by 3-methylcholanthrene treatment while epoxide hydrolase is induced less than 2-fold (10). After treatment of rats with a single dose of Aroclor-1254, maximum induction of cytochrome P-45Oc occurs within 2 days, while epoxide hydrolase levels continue to increase during the next 6-7 days (11).Since the intact microsomal membrane architecture is not required for the proposed coupling mechanism, based on reconstitution studies with synthetic lipid (8,12), theincrease in the dihydrodiol/phenol ratio observed with epoxide hydrolase and cytochrome P-45Oc relative to other cytochromes P450 could be due to a physical association between the enzymes. However, antibodies prepared against cytochrome P450c or epoxide hydrolase precipitate the antigen of immunization from solubilized microsomes without any detectable cross-contamination of the other enzyme (10, 13). Despite the above evidence against a coupling mechanism, Kaminsky et al. (14) have re-evaluated the metabolism of naphthalene to its trans-1,2-dihydrodiol and naphthol with homogeneous cytochromes P-450 and epoxide hydrolase and have concluded that cytochrome P-450 BNF-B (P-45Oc) is inherently more tightly associated functionally with epoxide hydrolase than is cytochrome P-450 PB-B (P-450b). They further proposed thatthis functional coupling has major implications in toxicology and chemical carcinogenesis. It seemed to us that a more likely alternative to the putative coupling mechanism might be an enantioselectivity by epoxide hydrolase toward the enantiomers of naphthalene 1,2-

10226

Stereoselectivity of Cytochromes P-450 and Epoxide Hydrolase

10227

Physics 3500B) on an Applied Sciences Adsorbosphere ODS column (0.94 x 10 cm) eluted with 50% methanol in water at a flow rate of 3.6 ml/min. Fractions were collected every 0.3 min and assayed for radioactivity in Hydrofluor scintillation mixture. For the methylated naphthalene conjugates, initial purification was by HPLC on the above column. Three 75-rl aliquots were injected; the column was eluted with 25% methanol in water at a flow rate of3.6 ml/min. A single fraction containing all the conjugates was collected. The solvents were removed in vacuo, and the residues were dissolved in 75 pl of methanol which was then analyzed under the same HPLC conditions. Fractions were collected every 0.2 min and analyzed for radioactivity in Hydrofluor scintillation mixture. The identification of the four possible conjugates (major adducts (+)-A and(-)-A resulting fr6m allylic attack and minor adducts (+)B and (-)-B from benzylic attack; see Fig. 1) from racemic naphthalene and anthracene oxide, as well as themethod utilized to calculate the enantiomer composition of the metabolically formed arene oxides, has been described previously (19). MATERIALS ANDMETHODS Metabolism of Anthracene and Naphthalene by CytochromeP-450b or P-45Oc in the Presence of Epoxide Hydrolase-In a final volume of Radioactive Chemicals-[["CINaphthalene (5 mCi/mmol) was obtained from New England Nuclear. [3H]Anthracene(55.7 mCi/mmol) 1.0 ml of 0.1 M phosphate buffer (pH 7.4), 0.01-0.05 p~ cytochrome was prepared as previously described (19) and was purified by high P-45Oc or 0.2-1.0 p~ cytochrome P-450b, 0.01-0.5 pM epoxide hydroperformance liquid chromatography (HPLC') on a Waters Associates lase, 0.32-2.4 p~ microsomal NADPH-cytochrome c reductase, 16160 p~ dilauroylphosphatidylcholine,1 mM NADPH, 3 mM MgC12, Z-Module equipped with a silica cartridge (0.8 X 8 cm) which was and 50 p~ [3H]anthracene or [14C]naphthalene (added in 50 plof eluted with hexane (k' = 1.5). Under the HPLC conditions employed for metabolite quantitation (see below), the naphthalene and anthra- acetone) were incubated for 10 min at 37 "C. In blank incubations, the reductase was omitted. cene were both found to be >97.5% radiochemically pure. For incubations with anthracene as the substrate, reactions were Optically Active Compounds-Enantiomerically pure (+)-(1R,2S)and (-)-(lS,2R)-naphthalene and anthracene 1,2-oxides as well as terminated by addition of 2 ml of ethyl acetate/acetone (2:l). After the (+)-(lS,2S)- and (-)-(lR,2R)-naphthalene and anthracene 1,2- vortexing the mixture, the organic layer wasremoved, dried over dihydrodiols were prepared according to published procedures (cf. sodium sulfate, and concentrated in a stream of dry nitrogen. The residue was dissolved in 100 p1 of methanol for subsequent analysis Ref. 19). Enzymes-Cytochromes P-4506 and P-45Ocwere obtained from by HPLC. A 25-pl aliquot of each solution was injected on a DuPont the livers of immature male rats of the Long-Evans strain after prior Zorbax Golden series ODS column (0.62 X 10 cm), and the column treatment of the rats with Aroclor-1254 (20). Epoxide hydrolase (M, was eluted with a linear gradient (3.3%/min) of 50-100% methanol in water at a flow rate of 1.6 ml/min (unknown, 6.5 min; dihydrodiol, = 49,000) waspurified to homogeneity from isosafrole-treated rats of the same strain (21). Rat liver microsomal NADPH-cytochrome c 8 min; anthrol, 10.5 min). For reactions with naphthalene as the substrate, the incubation reductase was purified (22) to a specific activity of 39,800 units/mg of protein assayed at 22 "C. For the preparation used, the activity medium was saturated with sodium chloride, and metabolites were corresponded to 3,800 units/nmol, where the molar concentration extracted into 0.5ml of ethyl acetate/acetone (21). Due tothe volatility of naphthalene, this solution was not concentrated, and 40 was determined onthe basis of an extinction coefficient of 21.4 mM" cm" at 456 nm (23). Microsomes were prepared by established pl of each extract was injected directly onto a DuPont Zorbax ODS procedures (24). Monoclonal antibody (C8) against cytochrome P- column (0.94 X 25 cm) eluted with a linear gradient (6%/min) of 40100% methanol in water at a flow rate of2.8 ml/min (dihydrodiol, 450c was obtained as described (25). Trapping of Naphthalene and Anthracene l,%-Oxidesformed by 10.3 min; naphthol, 13.5 rnin). For both hydrocarbons, fractions of Cytochrome P-450b-For metabolism of [3H]anthracene, in a final column effluent were collected every 0.3-0.4 min and were analyzed volume of 1.7 ml,0.47 p M cytochrome P-4506, 0.7 p M microsomal for radioactivity in Hydrofluor scintillation mixture. Metabolism of Anthracene and Naphthalene by Liver Microsomes NADPH-cytochrome c reductase, 28 p~ dilauroylphosphatidylcholine, 0.6 mM NADPH, 3.5 mMMgC12, and 60 p~ [3H]anthracene from Rats Treated with Phenobarbital or 3-Methylcholanthrene-In a (added in 100 p1 of acetone) in 0.12 M phosphate buffer (pH 7.4) were final volume of 2.0 ml, 0.25-1.0 mg/ml microsomal protein (containincubated for 10 min at 37 "C. Incubations were terminated by ing 1.95 nmol of cytochrome P-450/mg of protein and 2.18 nmol of addition of0.3mlof 5 M NaOH, followed by 2 ml of dioxane cytochrome P-450/mg of protein for microsomes from phenobarbitalcontaining 2.5 mM unlabeled, racemic anthracene 1,2-oxide and 0.3 and 3-methylcholanthrene-treated rats,respectively), 1mM NADPH, ml of a solution of 150 mg/ml N-acetyl-L-cysteine in water at a pH 3 mM MgCL, and 50 p M [3H]anthracene or ["Clnaphthalene (added of7.3. The resulting solution was stored for 2 h a t 37 "C and then in 100 plof acetone) in 0.1 M phosphate buffer (pH 7.4) wereincubated extracted with 2 X 2 mlof diethyl ether. The aqueous phase was for 10 min at 37 "C. Incubations with heat-inactivated microsomes lyophilized to dryness and kept a t -25 "C until the time of analysis. were used as blanks. Reactions were terminated, worked up, and For incubations with ["Clnaphthalene, an identical procedure was analyzed as described above for incubations with the purified enfollowed, except that an incubation timeof 5 min rather than 10 min zymes. was used, and 2 ml of a 7.5 mM solution of racemic carrier naphthalene Enantiomer Composition of Metabolically Formed Dihydrodiokoxide in dioxane was added after base treatment of the incubation To the remainder of each of the above extracts used for analysis of mixtures. In blank incubations, the reductase was omitted. metabolism by cytochrome P-450b or cytochrome P-45Oc inthe Residues of the above incubations were each dissolved in 0.5 ml of presence of epoxide hydrolase or by liver microsomes was added 0.35methanol containing 5 drops of acetic acid and were treated with 0.50mgof racemic, unlabeled dihydrodiol. The resulting solutions sufficient ethereal diazomethane to cause the yellow color to persist. were concentrated in a stream of air, and theresidues were dissolved Solvents were removed in a streamof air, and theresidues were taken in 100 pl of ethyl acetate. The dihydrodiols were isolated by preparup in 250 pl of methanol. ative HPLC on a DuPont Zorbax SIL column (0.94 X 25 cm) eluted For the methylated anthracene conjugates, 50-250 p1 of the above with ethyl acetate/hexane (1:l)at a flow rate of 7 ml/min; k' = 3 for methanol solution was applied to a silica Sep-Pak column (Waters naphthalene 1,2-dihydrodioland k' = 3.5 for anthracene 1,2-dihydroAssociates), which was then eluted with 10 ml of ethyl acetate. This diol. The isolated dihydrodiols were esterified with the acid chloride procedure removes most of the N-acetylcysteine methyl ester such (40 mg) of (-)-a-methoxy-a-trifluoromethylphenylaceticacid in pyrthat subsequent HPLC of the adducts is greatly improved. The idine (100 pl). After 6 h at ambient temperature, 50 pl of methanol solvent was evaporated, and the residue was taken up in 100 pl of was added to the reaction mixture and 25-50plof the resulting methanol. This solution (75 pl) was analyzed by HPLC (Spectra- solution was injected onto a DuPont Zorbax ODS column (0.46 x 25 cm). The column was eluted (2.0 ml/min) with 20% water in methanol l The abbreviation used is: HPLC, high performance liquid chrofor the naphthalene esters (a= 1.25, the (lR,ZR)-diastereomer elutes matography. first) and 15% water in methanol for the anthracene esters (a= 1.50,

oxide formed in different ratios by specific isozymes of cytochrome P-450. In attemptingto develop a model for the steric requirements of the catalytic binding site of cytochrome P-45Oc (15), we (16, 17) and others (18) have devised trapping techniques for establishing the enantiomerratio of metabolically formed arene oxides. Through application of such trapping techniques (19) to the arene 1,2-oxides formed from naphthalene and anthracene by cytochromes P-450b and P-45Oc and through kinetic analysis of the epoxide hydrolase catalyzed hydration of the enantiomeric naphthalene and anthracene 1,2-oxides, we are now able to conclude that there is no evidence for a selective functional coupling mechanism between epoxide hydrolase and cytochrome P-45Oc.

10228


the (lR,2R)-diastereomer elutes first). Radioactivity in the effluent was determined using a FLO ONE model HS radioactive flow detector (Radiomatic Instrumentand Chemical Co. Inc., Tampa, FL) equipped with a 2.5-ml detector cell, using Hydrofluor scintillation mixture at a mixture flowof 6 ml/min. The presence of carrier esters in the column effluent was monitored at 262 and 300 nm for naphthalene and anthracene metabolites, respectively. Hydration of (+)- and (-)-Naphthalene and Anthracene 1,2-Oxides by Purified Epoxide Hydrolase-All enzymatic reactions were carried out in 0.1 M sodium phosphate buffer (pH 7.4). Reactions were initiated by addition of the substrates (in acetonitrile) to buffered enzyme solutions, to give final acetonitrile concentrations of 0.2% ((+)-naphthaleneoxide),5.0% ((-)-naphthalene oxide), or 2.0% ((+)and (-)-anthracene oxides). Enzyme concentrations used are given in Table I. Because of large differences in the kinetic parameters for these four substrates as well as differences in the spectral characteristics of reactants and products in the anthracene and naphthalene series, different methods were required forfollowing the rates of reaction, as described in detail below. Kinetics of the hydrolysis of (+)-naphthalene oxide were followed by HPLC after quenching aliquots of the reaction mixture with mercaptoethanol (cf. Ref. 26), which traps unreacted substrate as a thioether (or as a co-chromatographic pair of isomeric thioethers). After initiation of reaction, 200-pI portions of a 3.0-ml reaction mixture were quenched a t appropriate time intervals with 40 pl of a solution containing 0.7 M mercaptoethanol and 0.66 M sodium hydroxide. After 5-15 min, the thiolate-containing reaction mixtures were neutralized by the addition of30 p1 of 1 M hydrochloric acid containing between 10 and 30 p~ p-nitrophenylethanol as aninternal standard for the chromatographic analyses. Samples were kept at 0 "C prior to analysis or frozen (-27 "C) if overnight storage was necessary. Samples (80 p l ) were analyzed by HPLC on a DuPont Zorbax ODS column (0.46 X 25 cm) eluted with a linear gradient of methanol in water at a flow rate of 1.0 ml/min and a rate of gradient change of 2%/min. For an experiment using 13 p~ substrate, the methanollwater gradient was begun a t 35% methanol, and the peak area corresponding to the dihydrodiol (retention time, 11.2 min) and unreacted substrate (as thethioether; retention time, 12.4 min) were quantified (relative to internal standard) by integration of the signal at 280 nm. At a lower initial substrate concentration (2.6 p M ) , the gradient was begun at 30% methanol (to reduce interference by an impurity), and the dihydrodiol (retention time, 13.5 min) was quantified by integration of the signal at 260 nm. Mixtures of known concentrations of the dihydrodiol with the internal standard were used for calibration. The enzymatic reaction was followed to completion, and values of K, and Vmax were calculated by fitting the differential Equations 1 and 2, where [SI= substrate concentration, [ P I = dihydrodiol concentration, and is the pseudo-first-order rate constant (determined independently) for nonenzymatic isomerization to naphthol, to the observed data for dihydrodiol formation, using the curve-fitting program MLAB (27). diPl/dt = V,,ISI/(& d[Sl/dt = -V,..[SI/(Km

+ [SI)

(1)

+ [SI)- MSI

(2)

In the case of (-)-naphthalene oxide, analogous attempts todetermine kinetic parameters from the complete time course of the enzymatic reaction indicated an apparent dependence of K,,, on the initial substrate concentration, such that larger values of K,,, were required to fit the data at high initial substrate concentrations. This effect may result from apremature slowing of the rate as a result of inhibition of the enzyme by either the dihydrodiol ornaphthol. Consequently, for this substrate, kinetic constants were determined from experiments in which 100-plaliquots of 300-pl reaction mixtures containing 9-370 p~ substrate were quenched with mercaptoethanol after 30 and 60 s, and initial rates were determined from the linear increase in dihydrodiol concentration measured chromatographically as above (detection a t 280 nm). For the anthracene oxide enantiomers, the enzymatic reaction was followed to compietion by measuring the decrease in absorbance a t 310 nm. Since at this wavelength the absorbance change is virtually identical upon formation of phenol and dihydrodiol, the observed absorbance change is directly related to disappearance of substrate, and kinetic constants could be calculated using Equation 2. Experiments a t initial substrate concentrations of1.4-17.2 pM (+)-oxide and 7-22 PM (-)-oxide gave internally consistent values of K, and

V, indicative of no significant product inhibition a t the higher substrate concentrations. RESULTS

Enantiomer Composition of the Naphthalene and Anthracene 1,Z-Oxides Formed by Cytochrome P-4506 Based on Trapping Experiments-Previously, a procedure was developed to determine the enantiomeric purity of the naphthalene and anthracene 1,2-oxides formed by homogeneous cytochrome P-45Oc (19). A similar experiment was performed in the presentstudy with homogeneous cytochrome P-4506 in a reconstituted system. Thus, after incubation of both radioactive hydrocarbons with cytochrome P-450b, racemic carrier arene 1,2-oxides were added and trapped as N-acetyl-L-cysteine adducts. These adducts were then converted to their methyl esters with diazornethane for analysis by HPLC. Due to the fact that the (+)- and (-)-enantiomer of each arene oxide can undergo both allylic (type A adduct, major) and benzylic (type B adduct, minor) trans opening with the thiol as shown in Fig. 1, a total of four adducts is possible from each racemic arene oxide. These were previously identified (19), and therelative amounts of each of the minor and major components were determined by reactions with the optically pure, syntheticoxides. Their separations on a reverse phase ODS column are shown in Fig. 2. Acceptable yields of radioactive adducts (-4% of the metabolically formed arene oxides) were obtained by incubating naphthalene for 5 min and anthracene for 10 min with 0.47 p~ cytochrome P-4506 in the reconstituted system. Anthracene is metabolized -20-fold slower than naphthalene by cytochrome P-450b (cf. Table XI, described below). The radioactivity measured in each of the HPLC peaks after incubation and trappingis shown in Fig. 2. After calculation of the amount of each adduct as previously described (19), 74% of the 1,2-oxides formed from both hydrocarbons OH

SR

(+)-(lR, 2S)-OXIDE

\ "")

FIG. 1. Structures of the two possible pairs of diastereomeric adducts obtainable on reaction of racemic naphthalene or anthracene 1,a-oxides with a chiral thiol nucleophile (RSH). Minor chemical pathways are designated by dashed arrows. The actual diastereomer separations reported in the present study were achieved with the methyl esters of N-acetyl-L-cysteine adducts. The designations (+)-A, (+)-B, (-)-A, and (-)-B for the adducts represent products from allylic attack (A) or benzylic attack ( B ) on the (+)- and (-)-arene oxides and are not meant to imply signs of rotation of these adducts. Incontrast, the signs of rotation and absolute configurations for the arene oxides are as designated for both naphthalene and anthracene 1,2-oxides.


10229

oxides of both hydrocarbons, the (+)-(lR,2S)-enantiomer exhibits the lower value of K,. Experiments at 37 "C with the anthracene oxide enantiomers gave values of K, that were within experimental errorof those observed at 25 "C, andkcat values that were 2.2 f 0.2 times greater than thosemeasured at 25 "C, for both enantiomers. Thiseffect of temperature on kcatis similar to that observed by Armstrong etal. (28) for the hydration of phenanthrene 9,lO-oxide. An experiment at 37 "C with (+)-anthracene oxide (6 p ~ and ) epoxide hydrolase (0.32 p ~ in) the presenceof 0.02 p~ cytochrome P-450c, 0.32 p~ reductase, and 31 I . ~ Mdilauroylphosphatidylcholine gave values of K , and kea, that were identical t o those observed with epoxide hydrolase in the absence of the reconstituted system. Similarly, in the presence of the reconstituted system and 0.32 p~ epoxidehydrolase inhibited with100 p M 1,2epoxy-3,3,3-trichloropropane, the observed pseudo-first-order rateconstant, b, for nonenzymaticdisappearance of (+)anthracene oxide was only -10% less than this rate constant in phosphatebuffer alone. Thus, the protein and lipid present in the reconstituted system have no marked effect on the kinetics of the enzymatic or nonenzymatic reactions at moderate concentrationsof anthracene oxide. Metabolism of Naphthalene and Anthraceneby Cytochrome P-450b in the Presenceof Epoxide Hydr~lase-['~C]Naphthalene and [3H]anthracene were each incubated with a highly purifiedmonooxygenase systemreconstituted with cytochrome P-4506 or P-45Oc and one concentration of epoxide hydrolase (Table 11). Naphthalene is a fairly good substrate for cytochrome P-450b, with arate of metabolism of -7 nmol of products formed per nmol of cytochrome P-450bjmin. Naphthol and the 1,2-dihydrodiol, in a ratio of 40:60, were TIME (mid the only detected products. Of the dihydrodiol formed, 60% FIG. 2. N-Acetyl-L-cysteineS-adducts chemically formed was the (-)-(lR,ZR)-enantiomer. Results of a parallel incufrom racemic naphthalene and anthracene l,a-oxides (top bation using cytochrome P-45Oc are also given in Table 11. traces, detected by UV at 254nm) and the oxides formed from The rateat which naphthalene is metabolized by cytochrome ['Hlanthracene and [14C]naphthaleneby cytochrome P-4506 P-45Oc is 3-4-fold higher than the rate observed for cyto(lower traces, cpm). Chromatographyconditions are described chrome P-4506. Compared to theincubations withcytounder "Materials and Methods." chrome P-450b, both the amount of dihydrodiol relative to naphthol and the percentage of the (-)-(lR,2R)-enantiomer TABLE I (92%) in thedihydrodiol are substantially increased. Kinetic constants for hydrolysis of the enantiomeric naphthaleneand Anthracene isa poor substrate for cytochrome P-450b,with anthracene oxides by epoxide hydrolase at 25 "C and pH 7.4. a rate of metabolism of-0.2 nmol of products formed per Enzyme K"l L t nmol of cytochrome P-450bjmin. This necessitated the ofuse Substrate Concentration (+ S.D.) (-C S.D.) a high concentration of cytochromeP-450b (1 p ~ in) the PM rnin" PM incubation medium in order to obtain a reasonable conversion Naphthalene of substrate. Because of the necessarily high cytochrome P 1" 2.2" (+)-(lR,ZS)-Oxide 0.64 450b concentration, the dependence of the reaction on the 12 3.6 (-)-(lS,ZR)-Oxide 1.06 concentration of dilauroylphosphatidylcholine was examined Anthracene (Table 111). The lowest lipid concentration (0.01 mgjml, 16 1.0 F 0.3' 29 f '5 (+)-(lR,ZS)-Oxide 0.32 7.1 -C 1.6' 17 k 0.6* (-)-(lS,ZR)-Oxide 0.32 p ~ gave ) the highest turnover. This concentration was used since productratiosas well as a Average of two experiments, which gave K , = 0.6 PM and k , = inallfurtherexperiments 2.1 min", and K,,, = 1.4 W M and Lt= 2.3 min", respectively. enantiomer composition were unaffected by variation inlipid 'Average of four experiments. concentration. Three productswere formed from anthracene: an unknown, which previously was suggested (19) to be the 10-hydrogens of anthracene was determinedto be the (-)-(lS,2R)-enantiomer. This result result of an exchangeof the 9- and medium, anthrol, and is in marked contrast to data obtained with cytochrome P-45Oc with some component in the incubation (19) where the (+)-(lR,2S)-oxides of both hydrocarbons pre- the 1,2-dihydrodiol, in a ratio of 5:7:88, respectively. Of the dihydrodiol formed, -40% was the (-)-(lR,2R)-enantiorner. domtnate (75-95%). Hydration of and (-)-Anthracene and Naphthalene 1,Z- The results of a parallel experiment withcytochrome P-45Oc Oxides by Epoxide Hydrolase-Kinetic constants for the hy- are also given in Table 11. Differences both with respect to dration of (+)- and (-)-naphthalene and anthracene oxides product ratio (no anthrol formed by cytochrome P-450c, but by purified epoxide hydrolase at 25 "C are given in Table I. 7% by cytochrome P-4506) and enantiomer composition of Althoughthevalues of K, forthe two enantiomers of the dihydrodiol (99% (-)-(lR,2R)-enantiomer with cytonaphthalene 1,2-oxide differed by factor a of -12, kcatfor both chrome P-450c, but only 38-39% with cytochrome P-450b) enantiomers was similar. Analogous results were obtained are apparent. The rate of metabolism catalyzedby cytochrome with the (+)- and (-)-oxides of anthracene. For the arene P-45Oc is -70-fold higher than the rate observed with cytoNAPHTHALENE

(+)-

ANTHRACENE


10230

TABLEI1 Metabolism of anthracene and naphthalene by a reconstituted system containing cytochromeP-4506 or P-45Oc in the presenceof epoxide hydrolase Incubations of 1.0 ml were run as described under “Materials andMethods,” in the presence of 23.4 pg/ml (0.48 p ~ epoxide ) hydrolase, 31 pM dilauroylphosphatidylcholine,and 0.39 pM NADPH-cytochrome c reductase, except for the incubation which contained 1.0 nmol of cytochrome P-4506, where 0.32 p~ epoxide hydrolase and 2.4 p~ reductase were used. In all cases, >90% of the total radioactivity before substrate emerged from the ODS column in the designated peaks. Data for cytochrome P-45Oc were taken from a previous study (19). Individual metabolites as % of total metabolites

Cytochrome P-450

Unknown

Dihydrodiol

Phenol

Conversion (rate)”

Enantiomer composition of the 1,Z-dihydrodiol (1R,ZR)

(m2.9

40.0 7.3 (7.3) 42.4 11.5 (5.6) 38.8 61.2 61.1 (4.0)

61 63 66

39 37 34

14.6 85.4 5.2 (25.5) 12.7 8.9 (21.9) 11.6 18.9 (18.9)

93 93 91

7 7 9

0.6 (0.16) 4.0 (0.20)

39 38

61 62

%

Naphthalene Cytochrome P-4506 (nmol/ml) 0.05 0.10 0.20 Cytochrome P-45Oc (nmol/ml) 0.01 0.02 0.05 Anthracene Cytochrome P-4506 (nmol/ml) 0.2 1.0 Cytochrome P-45Oc (nmol/ml) 0.01 0.02 0.05

60.0 57.6

87.3 88.4

8.3 4.9

91.7 88.3

6.8

24.3

75.7 3.1 (15.9) 76.4 23.6 5.4 (13.5) 99 25.5 7.7 74.5 (7.7) Rate is expressed as nanomoles of substrate metabolized per nanomole of cytochrome P-450/minute.

1

TABLEI11 Metabolism of anthracene by a reconstituted system containing cytochromeP-4506 and epoxide hydrolase with varying amounts of lipid and with antibody against cytochromeP-45Oc Incubations of 1.0 ml were run asdescribed under “Materials and Methods,” in the presence of 15.6 pg/ml (0.32 p ~ epoxide ) hydrolase and 2.4 p~ NADPH-cytochrome c reductase. In all cases, >90% of the total radioactivity before substrate emerged from the ODS column in the designated peaks. The purified cytochrome P-450b was preincubated for 5 min at 22 “C with0.1 mg/ml monoclonal antibody against cytochrome P-45Oc (MAb-C8) in the antibody experiments. Cytochrome P-4506 (1.0 nmol/ml)

Individual metabolites as % of total metabolites

Lipid

Unknown

Dihydrodiol

Conversion (rate)” Phenol

Preparation 1 + (MAb-C8) - (MAb-C8) Preparation 2 5.6 + (MAb-C8)94.4 - (MAb-C8)

(1R,ZR)

(1S2S)

%

mdml

Preparation 1

Enantiomer composition of the 1.2-dihydrodiol

0.01 0.02 0.05 0.10

4.2 4.9 5.6 5.7

89.7 88.3 88.0 3.7 86.0

6.1 6.8 6.4 8.3

4.7 (0.24) 4.0 (0.20) (0.18) 2.4 (0.12)

41 38 36 36

59 62 64 64

0.01 0.01

1.9 9.2

95.1 85.3

3.0 5.5

4.1 (0.21) 3.3 (0.17)

42 40

58 60

0.01 0.013.0

1.25.0

93.8

3.3 (0.17) (0.15)

38 39

62 61

Rate is expressed as nanomoles of substrate metabolized per nanomole of cytochrome P-450/minute.

chrome P-450b. To exclude the possibility that even a trace amount of cytochrome P-45Oc is present in the cytochrome P-450b preparation, an experiment was performed with two separate preparations of cytochrome P-4506 in the presence of a monoclonal antibody against cytochrome P-45Oc. As shown in Table 111, addition of the antibody does not affect the rate of metabolism, product distribution, or enantiomer composition of the 1,2-dihydrodiol observed from incubations containing cytochrome P-450b. The monoclonal antibody (C8) has previously been shown to inhibit completely the catalytic activity of purified cytochrome P-45Oc (25). Effect of Varying Concentrations of Epoxide Hydrolase on the Metabolism of Naphthaleneand Anthracene byCyto-

chromes P-4506 and P-450c”Radioactive naphthalene was incubated separately with cytochrome P-45Oc (0.03 PM) and with cytochrome P-450b (0.10 PM) such that conversion of substrate (cf. Table 11) by either cytochrome was -10%. Varying amounts of purified epoxide hydrolase (0.01-0.50 nmol; Fig.3) were added to these incubations, and therelative amounts of the two possible products derived from the metabolic intermediatenaphthalene 1,2-oxide were determined; the dihydrodiol from epoxide hydrolase catalyzed trans addition of water and the naphthol from spontaneous rearrangement (see “Discussion”). For cytochrome P-450b (Fig. 3A), the relative amount of 1,Z-dihydrodiol formed went from 7% of the total metabolites at the lowest epoxide hydrolase con-


'ool"-l

J

0

n

A

t

't

0

0 n

Pn

1oc

1

olo"-

10231

K 0

\

> r

y

50

I n

75

0

a

o!

ap

K

F Y

ap

0

0.25

0.50

0 0.500

0.25

50 0.50

I

~

I1

0.25

EPOXIDE HYDROLASE (nmoI/ml) FIG. 3. Effect of varying amounts of epoxide hydrolase on the distribution of dihydrodiol and phenol metabolites ( A and B ) and on the 1,2-dihydrodiol enantiomer composition (0formed from naphthalene by cytochrome P-45Ob (0)and P-45Oc (U). Incubations of 1.0 ml were run as described under "Materials and

Methods" in the presence of 0.39 p~ NADPH-cytochrome c reductase and 31 p~ dilauroylphosphatidylcholine. Concentrations of cytochrome P-450 used were 0.03 pM for cytochrome P-45Oc and 0.10 F M for cytochrome P450b. The solid lines represent theoretical curves calculated as described in the "Appendix," using the kinetic constants for epoxide hydrolase given in Table I. The broken lines indicate the effect on the calculated curves of increasing katand decreasing K,,, for (+)-naphthalene oxide by factors of 1.2 and 1.8, respectively, with all other Darameters held constant. and illustrate the uncertaintv that is present in the kinetic model due to inaccuracies icf. Table I) in the measured kinetic constants. centration (0.01 p ~ to) 55% at thehighest epoxide hydrolase goes a dramatic change from 82 to 40% as the (-)-(lR,2R)concentration (0.50 p ~ ) Over . the same range of epoxide enantiomer (Fig. 4C). Concurrentwiththe increase in the percentage of 1,2hydrolase concentrations, the relative amount of 1,2-dihydrodiol formed in incubations with cytochrome P-45Oc went up dihydrodiol with increasing amountsof epoxide hydrolase, an from 10 to 81% (cf. Fig. 3B). The enantiomer composition of apparent increase was observed in the total conversion of the 1,2-dihydrodiol formed in these experiments is shown in anthracene (from 11 to 22%)by cytochrome P-45Oc. FurtherFig. 3C. At the lowest epoxide hydrolase concentration, the more, the observed rates of metabolism by cytochrome P-45Oc (-)-( lR,2R)-dihydrodiol predominates both for cytochrome showed -2-fold variations among experiments. Some of this P-450b and P-45Oc (81 and 95%, respectively). On increasing variation may be due to incomplete recovery of anthrol. In was formed fromthe areneoxide the epoxide hydrolase concentration, the relative amount of control experiments, anthrol (-)-( lR,2R)-dihydrodiol decreases in incubations with cyto- in the absence of epoxide hydrolase, and its recovery upon chrome P-450b, until at the highest concentration of epoxide extraction and evaporation(cf. "Materials andMethods") was -75% of the recovery upon direct injection of the buffered hydrolase only62% of the 1,2-dihydrodiol consists of the (-)(lR,BR)-enantiomer. This is in contrast to the incubations aqueous reaction mixture on the chromatographic column. with cytochrome P-45Oc where the 1,2-dihydrodiol still conMetabolism of Naphthalene and Anthracene by Liver Microsomes from Rats Treatedwith Phenobarbital or3-Methylcholsists of a very high amount of the (-)-(lR,2R)-enantiomer (92%) evenat the highest epoxide hydrolase concentration. anthrene-Metabolism of naphthalene and anthracene was Similar experiments were performed with [3H]anthracene also examined with liver microsomes from 3-methylcholanas shown in Fig. 4. Concentrations of cytochromeP-450 threne- and phenobarbital-treated rats (Table IV). For purrequired to obtain the necessary conversions (at least 4-5% poses of comparison of the microsomal data with that obmetabolism) were 0.02 p~ for cytochrome P-45Oc and 1.0 p~ tainedfromthereconstitutedsystems,itisimportantto for cytochrome P-4506(cf. Table 11).On additionof increasing recognize that in themicrosomes used in this study (i) cytoamounts of epoxide hydrolase (0.01-0.20 nmol; Fig. 4) to chrome P-450b constitutes -2% and cytochrome P-45Oc conincubations with cytochrome P-450b or P-450c, the amount stitutes >75% of the total hepaticmicrosomal cytochrome P of 1,2-dihydrodiol as a per cent of total metabolites increased 450 after3-methylcholanthrenetreatmentand(ii) cytofrom 72 to 95% and from 85 to loo%, respectively. Compari- chrome P-450b constitutes >55% and cytochrome P-45Oc son of Figs. 3A and 4A illustrates that smaller amounts of constitutes -1% of the total hepatic microsomal cytochrome epoxide hydrolase were necessary to obtain high percentages P-450 after phenobarbital treatment(10). of dihydrodiol from anthracene, consistent with the fact that The rate at which both microsomal preparations metabolize the anthraceneoxides are better substrates for epoxide hydro- naphthalene is very similar: -2.5 nmol of products/nmol of lase than the naphthalene oxides (Table I). Whereas 399% cytochrome P-450/min. Both the ratio of naphthol to 1,2of the anthracene 1,2-dihydrodiol formed by cytochrome P- dihydrodiol(-15:85) and the enantiomeric composition of the 450c and epoxide hydrolase consists of the (-)-(lR,ZR)-en- dihydrodiol (92% (lR,2R)-enantiomer) arepractically identiantiomer over the entire range of epoxide hydrolase concen- cal when metabolisms of naphthalene by liver microsomes trations, the enantiomer composition of the 1,2-dihydrodiol from 3-methylcholanthrene-treated rats and by the purified formed by cytochrome P-450b and epoxide hydrolase under- system reconstituted with cytochrome P-45Oc and epoxide

-

Stereoselectivity of Cytochromes P-450Epoxide andHydrolase

10232

1oc

P450c

J

0 0

* 0 0

50

I

n

a cy. a

-

ap

Y

I 0 0

0.10

I

0

0.20

ap

B 0.10

C1

I

0.20

u 0

0.20

0.10

EPOXIDE HYDROLASE (nmol/ml) FIG.4. Effect of varying amounts of epoxide hydrolase on the distribution of dihydrodiol and phenol metabolites ( A and I?) and the 1,2-dihydrodiol enantiomer composition (C) formed from anthracene

(m.

by cytochrome P-4506 (0)and cytochrome P-45Oc Incubations of 1.0 ml were run as described under “Materials and Methods” in the presence of 0.32 p~ NADPH-cytochrome c reductase and 31 p~ dilauroylphosphatidylcholinefor cytochrome P-45Oc (0.02 p ~ and ) 2.4 1M reductase and16 p~ lipid for cytochromeP-4506 (1.0 p ~ ) .The lines represent theoretical curves calculated as described in the “Appendix.” Symbols represent data corrected for 75% recovery of anthrol (see text).

TABLEIV Metabolism of anthracene and naphthalene by liver microsomes from phenobarbital- or 3-methylcholanthrenetreated rats Incubations of 2.0 ml were run as described under “Materials and Methods.” The designations PB and MC refer to liver microsomes from phenobarbital- and 3-methylcholanthrene-treated rats, respectively. Substrate

Microsones

Individual Enantiomer metabolites as % of total metabolites Unknownb

Anthracene

composition of the 1,2-dihydrodiol

Dihydrodiol

Phenol

0.5, PB 1.0, PB 2.0, PB

53.4 64.7 77.0

46.6 35.3 23.0

26.0 (2.67) 39.6 (2.03) 59.8 (1.53)

66 59 50

34 41 50

0.5, MC 1.0, MC

81.1 85.3

18.9 14.7

26.4 (2.42) 38.8 (1.78)

93 90

7 10

mgprotein

Naphthalene

Conversion (rate)’ (1.9.2s) %

(1R2.R)

0.5, PB 1.0, PB 2.0, PB

6.3 6.9 12.8

86.8 86.9 83.0

6.9 6.2 4.2

10.0 (1.03) 18.7 (0.96) 47.0 (1.21)

71 67 65

30 33 35

0.5, MC 1.0,MC

23.5 26.6

76.2 73.2

0.3 0.2

39.1 (3.59) 73.8 (3.39)

99 99

1 1

Rate is expressed as nanomolesof substrate metabolized per nanomole of cytochrome P-450/minute. The nature and possible origin of this unknown, which does not appear to be derived from anthracene, have been discussed ina previous study (19). hydrolase are compared. The product studies are consistent with nearly all of the metabolism of naphthalene occurring via cytochrome P-45Oc in thesemicrosomes. Since cytochrome P-450b metabolizes naphthalene at only one-third the rateof cytochrome P-45Oc (Table 11),the similar rates of metabolism by the two microsomal preparations suggest that cytochromes P-450 other than P-450b contribute significantly to the metabolism of naphthalene with liver microsomes from phenobarbital-treated rats. Over a 4-fold range in the amount of microsomal protein from the livers of phenobarbital-treated rats, the rateof metabolism of naphthalene (nmolof products/ nmol of cytochrome P-450/min)decreases by about one-half, and the amount of naphthol decreases about 50% (as a percentage of total metabolites) along with a concomitant in-

crease in 1,2-dihydrodiol. Thisresultisconsistentwith a protein concentration-dependent decrease in the rateof oxidation (nmol of products/nmol of cytochrome P-450/min) while the rate of hydration by epoxidehydrolase remains unchanged. The decrease (from 66 to 50%) in the fraction of dihydrodiol as the (lR,2R)-enantiomer with increasing amounts of livermicrosomal proteinfromphenobarbitaltreated rats may be due to differential changes in rate for the cytochromes P-450 and epoxidehydrolase withincreasing microsomal protein. Qualitative similarities exist when the metabolism of anthracene is compared with theabove results for naphthalene (Table IV). In the anthracenecase, liver microsomes from 3methylcholanthrene-treated rats are about %fold more effi-


10233

cient than those from phenobarbital-treated rats when compared per nanomole of cytochrome P-450; yet cytochrome P 450c is a t least 80-fold more effective than cytochrome P450b in the reconstituted system (Tables TI and IV). The argument for the presence of cytochromes P-450 other than P-450b and P-45Oc which contributesubstantially tothe metabolism of anthracene by liver microsomes from phenobarbital-treated rats is much more convincing than in the naphthalene case. The modest amounts of anthrol (s8%, Tables 11-IV) formed under all incubation conditions are consistent with the finding (Table I) that both enantiomers of anthracene 1,2-oxide are quite good substrates for epoxide hydrolase.

lase when its substrate is predominantly (+)-(1R,2S)naphthalene oxide (generated by cytochrome P-45Oc) than when its substrate is predominantly the (-)-(lS,BR)-oxide (generated by cytochrome P-450b). Hence, no coupling between the enzymes is required to account for the different dihydrodiol/phenol ratios observed with epoxide hydrolase and the two cytochromes P-450. (ii) At very low concentrations of epoxide hydrolase, the ratio of (R,R)-to (S,S)-dihydrodiols formed by the sequential action of cytochrome P-450 and epoxide hydrolase is determined by the enantiomer ratio of the arene oxides generated by the specific cytochrome P 450 isozyme, the regioselectivity of the epoxide hydrolase reaction, and the relative values of k,.JKm for the (+)- and (-)-arene oxides. (iii) At infinitely high epoxide hydrolase DISCUSSION concentrations, the reaction of both enantiomeric oxides Both naphthalene and anthracene l,2-oxides are unstable should proceed enzymatically to completion, giving dihydrometabolic intermediates which undergo rapid, spontaneous diol with no competition from naphthol formation, and thus isomerization (t%= 2.8 and 1.2 min, respectively) to phenols the ratio of (R,R)-to (S,S)-dihydrodiols should depend only in the incubation medium (0.10 M phosphate (pH 7.4), 37 "C) on the enantiomer ratio of the arene oxides formed and the of the reconstituted system. The enantioselectivity of their regioselectivity of epoxide hydrolase, but not on the relative formation and further conversion to optically active trans- rates of hydration of the two enantiomeric substrates. Thus, 1,2-dihydrodiolsis dependentupon the stereoselectivity of the the fraction of (R,R)-dihydrodiol formed should be relatively specific cytochrome P-450 utilized as illustrated in Fig. 5. large at low epoxide hydrolase concentrations where the (+)is preferentially converted to Unlike cytochrome P-450c, which favors the formation of the (lR,2S)-oxide (higher kCat/Km) (+)-( 1R,2S)-oxides from both hydrocarbons (19), cytochrome (R,R)-dihydrodiol, whereas at higher epoxide hydrolase conP-450b favors the formation of the (-)-(lS,BR)-oxides. The centrations, the fraction of (R,R)-dihydrodiolshould decrease, enantiomer composition of the 1,2-dihydrodiol formed from approaching as a limit the value given by Equation 3, either hydrocarbon and theratio of dihydrodiol to phenol are Fraction (R,R) = a + (1 - a)p (3) thus dependent upon (i) the specific cytochrome P-450 utilized as well as its concentration, (ii) the amount of epoxide where (Y = the fraction of (+)-(lR,BS)-arene oxide generated hydrolase and its kinetic parameters(Table I) as well as by the cytochrome P-450 and fi = the fraction of (-)-(lS,2R)regiospecificity for hydration of the arene oxides present in oxide that reacts with epoxide hydrolase atthe benzylic the medium, and (iii) the spontaneous isomerization rate of carbon to give (1R,2R)-dihydrodiol.These limiting values for the arene oxide to phenol. Given the data in the present and the fraction of (R,R)-dihydrodiol formed at highepoxide previous (19) studies, predictions of the ratio of phenol to hydrolase concentrations are0.56 and 0.84 for the cytochrome dihydrodiol and enantiomer composition of the dihydrodiol P-4506 ( a = 0.26) and cytochrome P-45Oc ( a = 0.73) systems, are possible. The two hydrocarbons will betreated separately. respectively. The effect of increasing epoxide hydrolase concentrations For the sequential metabolism of naphthalene by cytochrome P-450b or P-45Oc and epoxide hydrolase: (i) A t low on both the fraction of dihydrodiol relative to naphthol and concentrations of epoxide hydrolase andarene oxide, the the fraction of (R,R)-dihydrodiol formed from napthalene in epoxide hydrolase-catalyzed reaction of (+)-( 1R,2S)- the presence of reconstituted systems containing cytochrome 2.2 FM" min"; Table I) will P-45Oc or P-450b (Fig. 3) agree with the above qualitative naphthalene oxide (kcat/& compete more favorably with nonenzymatic rearrangement to predictions. Results obtained with intact microsomesfrom naphthol than will the epoxide hydrolase-catalyzed reaction phenobarbital-treated rats (cf. Table IV) are also consistent with a competition between nonenzymatic rearrangement and of (-)-(lS,2R)-naphthalene oxide (kCat/Km 0.3 p ~ "min"; Table I). Thus, a higher dihydrodiol/phenol ratio should be enantioselective enzymatic hydrolysis of the arene oxides observed at low-to-moderate concentrations of epoxide hydro- formed. In this system, increasing microsomal protein con-

-

-

P45oc (295%)

NAPHTHALENE

OH

P450b

(60%)

0

EH

(299%)

1

ANTHRACENE

P450b 174%)

FIG. 5. Stereoselective formation of arene oxidesfrom naphthalene and anthracene by cytochromes

P-450band P-46Oc. The percentages shown represent enantioselectivity for the particular enzymatic steps. (-)(1S,2R)-Naphthalene 1,2-oxideis the only non-K-region arene oxide presently known whichis hydrated by epoxide hydrolase ( E H ) at both the allylic and benzylic positions of the epoxide group (19).

10234

Stereoselectivity of Cytochromes P-450Epoxide andHydrolase

centration has the effect of increasing the concentration of epoxide hydrolase without a proportionalincrease in the rate of conversion of the hydrocarbon to thearene oxide. Thus, at low microsomal protein (low epoxide hydrolase) concentrations, nonenzymatic naphthol formation competes favorably with enzymatic hydrolysis for the poor substrate, (-)-(lS,2R)naphthalene oxide, whereas at high microsomal protein concentrations, more of this oxide undergoes enzymatic hydrolysis to give an increased percentage of the (lS,2S)-dihydrodiol product. Quantitative treatment of the sequential enzymatic reactions in the reconstituted system requires the consideration of possible buildup of naphthalene oxide intermediates that are notconverted either to naphthol or to dihydrodiols during the reaction period. These naphthalene oxides would be analyzed as naphthol after workup and chromatography. Furthermore, if buildup is significant, one enantiomeric oxide could possibly act as a competitive inhibitor of the reaction of epoxide hydrolase with the other. A quantitative model that takes these factors into account, described in detail in the "Appendix," was used to calculate the theoretical curves shown in Fig.3. In thereconstituted system, agreement of the experimental data with the calculated values is very good considering the number of rate parameters required by the model and the possible experimental errors in each. Small positive deviations of the experimentally measured fraction of dihydrodiol from the theoretical lines of Fig. 3, A and B , may result from errors in the constants used in the model calculation, or they may reflect a slight retardation of phenol formation from the arene oxide when generated in situin the presence of microsomal protein and lipid. In any case, similar deviations are observed with both cytochromes P-450b and P 450c, and they thus provide no evidence for any preferential interaction of cytochrome P-45Oc with epoxide hydrolase. The enantiomer composition of the dihydrodiol formed in each case (Fig. 3C) is in excellent agreement with our predictions based on the stereochemical course of the cytochrome P-450 reactions and thekinetic constants for the reactions of epoxide hydrolase with each enantiomer. A similar quantitative treatment of the sequential reactions of the cytochromes P-450 and epoxide hydrolase with anthracene is shown in Fig. 4. In this case, the agreement between the experimental data and theoretical curves for the fraction of diol relative to phenol is much less satisfactory, although the data for the enantiomer composition of the dihydrodiol product are in agreement with prediction. Because of the likelihood of small losses of anthrol upon workup of the reaction mixtures (see "Results") the experimental data were corrected, assuming that only 75% of the anthrol formed was recovered in the analyses. This correction is small and does theoretical not significantly improve the fit of the data to the lines of Fig. 4, A and B . These substantial deviations from the theoretical curves are observed with both cytochromes P 450b and P-45Oc and are indeed more pronounced in the system containing cytochrome P-450b. Hence, selective coupling of cytochrome P-45Oc with epoxide hydrolase cannot provide an explanation for these deviations, although mechanisms whereby the arene oxides formed by the action of both cytochromes P-450 are protected from nonenzymatic rearrangement while being transferred to epoxide hydrolase are not excluded. In summary, arene 1,2-oxides are the predominant if not exclusive primary oxidative metabolites formed from naphthalene and anthracene by cytochrome P-450b and P 450c based on the present kinetic analysis. In previous studies, we have attempted to map the steric requirements of the

catalytic binding site of cytochrome P-45Oc based upon the absolute configurations of metabolites formed by this enzyme (cf. Refs. 15, 29, 30, and references therein). Although cytochromes P-450b and P-45Oc are isozymes, the preferential formation of (-)-(lS,ZR)-arene oxides (74%) from naphthalene and anthracene by cytochrome P-450b indicates that its binding site is markedly different from that of cytochrome P450c. APPENDIX

Kinetic Model forthe Sequential Reactions of Cytochromes P450 and Epoxide Hydrolase and Naphthalene and Anthracene Equations 4 and 5 describe the rates of formation and loss of the (+)-(1R,2S)- (I1) and (-)-(lS,BR)-arene oxides (Iz) at any given epoxide hydrolase concentration [ E ] ; d[Ill/dt = au - kY[EI[I11/([I11 + Km1 (1 + [IzI/K~z)) - MI11 d[Izlldt = (1 - a b - k~"t[EI[IzI/([Izl

+ Krnz(1 + [IlI/Kml)) - ko[Izl

(4)

(5)

where a = the fraction of hydrocarbon converted to (+)(lR,2S)-arene oxide (I1);u = the rate (kM/min) of conversion of hydrocarbon to both arene oxide enantiomers; k',", K m l , kYt, and Km2are the appropriate kinetic parameters for reaction of epoxide hydrolase with the (+)-(lR,2S)-arene oxide (I1) and (-)-(1S,2R)-arene oxide (Iz),respectively; and ko is the rate constant for nonenzymatic conversion of arene oxide to phenol. Rates of dihydrodiol formation from each arene oxide enantiomer are given by Equations 6 and 7. d[Pll/dt = kYt [EI[III/([III + KrnI(1+ [I~l/Kmd)

(6)

d[P~l/dt= kYt [EI[IzI/([IzI+ K m Z ( 1 + [11I/Kml))

(7)

The (1 + [ I ] / K m term ) in the denominator of each equation is introduced because of the assumption that each enantiomer can act as a competitive inhibitor of the other. Finally, the rate of phenol ( P 3 )formation is given by Equation 8. d[P,l/dt = MI11

+ [IZI)

(8)

Solution of differential Equations 4-8 at t = 10 min for any value of [ E ] ,using the program MLAB (27), gives values of 11, 12,PI, Pz,and P3 from which the fraction dihydrodiol and the fraction of dihydrodiol as the (R,R)-enantiomer may be calculated using Equations 9 and 10.

+ phenol) = (PI + PZ)/(Pl + Pz + Ps + I1 + 12)

Fraction dihydrodiol = dihydrodiol/(dihydrodiol

(9)

Fraction (R,R)-dihydrodiol= (R,R)-dihydrodiol/(total dihydrodiol) =

(10)

(P1+ PPZ)/(Pl+Pz)

where 0 = the fraction of (-)-(lS,BR)-oxide (Iz) thatundergoes benzylic attack to give (R,R)-dihydrodiol. Since the (+)(lR,2S)-oxide (I1) givesexclusively (R,R)-dihydrodiol, the coefficient of P I is 1.0. Values of K,,, for epoxide hydrolase used in the calculation were those given in Table I, and the values of kc., at 25 "C (Table I) were multiplied by 2.2 to give approximate values at 37 "C (see "Results"). Values of ko measured at 37 "C are 0.25 and 0.58 min" for naphthalene and anthracene oxides, respectively. Values of u were determined from the observed rates of metabolism of naphthalene and anthracene. This rate was equal to 0.6 WM min" for both cytochromes P-450b and P-45Oc with naphthaleneassubstrate (cf. Table 11). For

Stereoselectivity of Cytochromes P-450Epoxide andHydrolase anthracene as substrate, the values used were 0.18 ~ L min" M for cytochrome P-450b and 0.54 p~ min" for cytochrome P450c. These represent average values for the observed metabolic rates at different epoxide hydrolase concentrations. Use of the highest and lowest observed values did not significantly change the calculated curves. REFERENCES 1. Tomaszewski, J. E., Jerina, D. M., and Daly, J. W. (1975) Bio-

chemistry 14,2024-2031 2. Billings, R. E., and McMahon, R. E. (1978) Mol. Pharmucol. 14, 145-154 3. Hanzlik, R. P., Hogberg, K., and Judson, C. M. (1984) Biochemistry 23,3048-3055 4. Jerina, D. M., and Daly, J. W. (1974) Science 185,573-582 5. Boyd, D. R., and Jerina, D. M. (1984) in Small Ring Heterocycles 42 (Hassner, A., ed) Part 3, pp. 197-282, John Wiley & Sons, Inc., New York 6. Lu, A. Y. H., Jerina, D. M., and Levin, W. (1977) J. Biol. Chem. 252,3715-3723 7. Oesch, F., and Daly, J. (1972) Biochem. Biophys. Res. Commun. 46,1713-1720 8. Dansette, P. M., Yagi, H., Jerina, D. M., Daly, J. W., Levin, W., Lu, A. Y. H., Kuntzman, R., and Conney, A. H. (1974) Arch. Biochem. Biophys. 164, 511-517 9. Levin, W., Wood, A., Chang, R., Ryan, D., Thomas, P., Yagi, H., Thakker, D., Vyas, K., Boyd, C., Chu, S.-Y., Conney, A., and Jerina, D. (1982) Drug Metab. Reu. 13, 555-580 10. Thomas, P. E., Reik, L. M., Ryan, D. E., and Levin, W. (1981) J. Biol. Chem. 256,1044-1052 11. Parkinson, A., Thomas, P. E., Ryan, D. E., Reik, L. M., Safe, S. H., Robertson, L. W., and Levin, W. (1983) Arch.Biochem. Biophys. 225,203-215 12. Oesch, F., Jerina, D. M., Daly, J. W., Lu, A.Y. H., Kuntzman, R., and Conney, A. H. (1972) Arch.Biochem.Biophys. 1 5 3 , 62-67 13. Parkinson, A., Thomas, P. E., Ryan, D. E., and Levin, W. (1983) Arch. Biochem. Biophys. 225, 216-236 14. Kaminsky, L. S., Kennedy, M. W., and Guengerich, F. P. (1981)

10235

J.Biol. Chem. 256,6359-6362 15. Jerina, D. M., Michaud, D. P., Feldmann, R. J., Armstrong, R. N., Vyas, K. P., Thakker, D. R., Yagi, H., Thomas, P. E., Ryan, D. E., and Levin, W. (1982) in Microsomes, Drug Oxidations, and Drug Toxicity (Sato, R., and Kato, R., eds) pp. 195-201, Japan Scientific Societies Press, Tokyo 16. Armstrong, R. N., Levin, W., Ryan, D. E., Thomas, P. E., Mah, H. D., and Jerina, D. M. (1981) Biochem. Biophys. Res. Commun. 1 0 0 , 1077-1084 17. van Bladeren, P. J., Armstrong, R. N., Cobb, D., Thakker, D. R., Ryan, D. E., Thomas, P. E., Sharma, N. D., Boyd, D. R., Levin, W., and Jerina, D. M. (1982) Biochem. Bwphys. Res. Commun. 106,602-609 18. Hernandez, O., Walker, M., Cox, R. H., Foureman, G. L., Smith, B. R., and Bend, J. R. (1980) Biochem. Biophys. Res. Commun. 96, 1494-1502 19. van Bladeren, P. J., Vyas, K. P., Sayer, J . M., Ryan, D.E., Thomas, P. E., Levin, W., and Jerina, D.M. (1984) J. Biol. Chem. 259,8966-8973 20. Ryan, D. E., Thomas, P. E., and Levin, W. (1982)Arch. Biochem. Biophys. 2 1 6 , 272-288 21. Ryan, D. E., Thomas, P. E., and Levin, W. (1980) J.Biol. Chem. 2 5 5 , 7941-7955 22. Yasukochi, Y., and Masters, B. S. S. (1976) J. Biol. Chem. 251, 5337-5344 23. French, J . S., and Coon, M. J. (1979) Arch. Biochem. Biophys. 195,565-577 24. Lu, A.Y.H., and Levin, W. (1972) Biochem.Biophys.Res. Commun. 46,1334-1339 25. Thomas, P. E., Reik, L. M., Ryan, D. E., and Levin, W. (1984) J. Biol. Chem. 259,3890-3899 26. Sayer, J. M., Yagi, H., Croisy-Delcey, M., andJerina, D.M. (1981) J.A m . Chem. SOC.103,4970-4972 27. Knott, G. D. (1979) Comput. Programs Biomed. 10, 271-280 28. Armstrong, R. N., Levin, W., and Jerina, D. M. (1980) J. B i d . Chem. 255,4698-4705 29. Vyas, K. P., Shibata, T., Highet, R. J., Yeh, H. J., Thomas, P. E., Ryan, D. E., Levin, W., and Jerina, D. M. (1983) J. Bid. Chem. 258,5649-5659 30. Vyas, K. P., van Bladeren, P. J., Thakker, D. R., Yagi, H., Sayer, J. M., Levin, W., and Jerina,D. M. (1983) Mol. Phuramcol.2 4 , 115-123