Biotransformation of 1,2-Dihydronaphthalene and 1,2

Oct 1, 1993 - 3700 AJ Zeist, The Netherlands. Received November 30,199.P. Both 1,2-dihydronaphthalene and 1,2-dihydroanthracene were hydroxylated at ...
1 downloads 0 Views 736KB Size
Download PDF
Chem. Res. Toxicol. 1993,6, 808-812

808

Biotransformation of 1,2-Dihydronaphthalene and 1,2-Dihydroanthraceneby Rat Liver Microsomes and Purified Cytochromes P-450.Formation of Arene Hydrates of Naphthalene and Anthracene Derek R. Boyd,’?? Narain D. Sharma,t Rajiv Agarwal,? R. Austin S. McMordie,t Jos G. M. Bessems,t Ben van Ommen,$ and Peter J. van Bladerent School of Chemistry, The Queen’s University of Belfast, Belfast BT9 5AG, U.K., and Department of Biological Toxicology, TNO Toxicology and Nutrition Institute, P.O. Box 360, 3700 AJ Zeist, The Netherlands Received November 30,199.P Both 1,2-dihydronaphthalene and 1,2-dihydroanthracene were hydroxylated at the benzylic (1-) or the allylic (2-) position by rat liver microsomes and purified cytochrome P-450 enzymes to yield “arene hydrates”. Two other classes of metabolites were formed, the dehydrogenation products naphthalene and anthracene, and trans-1,2-dihydroxy-1,2,3,4-tetrahydronaphthalene and its anthracene analog as products of the classical epoxide pathway. Regioselectivity (hydroxylation a t benzylic or allylic positions) and stereoselectivity (hydroxylation a t pro-R or p r o 3 hydrogen atoms) during metabolism of dihydroarenes to yield arene hydrates were found to be dependent upon the nature of the inducing agents used during pretreatment of the rats and thus the level of particular P-450 enzymes. This selectivity was more pronounced for anthracene than for naphthalene. Naphthalene and anthracene were formed enzymatically by direct dehydrogenation of the dihydro compounds rather than by dehydration of the arene hydrate metabolites. A general mechanism involving the intermediacy of benzylic and resonancestabilized allylic carbon radicals can account for the formation of both enzyme-catalyzed hydroxylation (arene hydrate) and dehydrogenation (arene) metabolites of dihydroarene substrates.

Introduction Early reports of the metabolism of naphthalene (1-51, 1,2-dihydronaphthalene (51, and anthracene (3, 6 ) in animals have led to the proposal that arene hydrates of naphthalene (lA, 1B) and anthracene (2A, 2B) are transient metabolites. Although no direct evidence was obtained for the presence of these arene hydrate metabolites, a glucosiduronic acid derivative of arene hydrate 1A was isolated and characterized (5). The evidence for arene hydrate metabolite formation from the polycyclic , arenes naphthalene (143,anthracene ( 3 , 6 )phenanthrene (3), and quinoline (7, 8) was mainly based upon the liberation of the parent arene upon acidification of the animal urine. Stronger evidence for arene hydrate formation was obtained when 9,10-dihydrobenzo[elpyrene was incubated with liver microsomes (9). In the latter report both spectral (UV and MS) and chromatographic data, allied to chemical evidence, i.e., the facile acidcatalyzed dehydration to yield benzo[e]pyrene, strongly supported the original conclusion (5) that a dihydroarene may be metabolized in a liver microsomal system to an arene hydrate. Unfortunately, neither the structure nor the stereochemistry of the arene hydrate formed from 9,10-dihydrobenzo[e]pyrene was determined (9). Liver microsomal metabolism of the polycyclic aromatic hydrocarbon cyclopenta[cd]ppene has recently been reported to yield a hydrate metabolite (IO). The latter

* Author to whom correspondence should be addressed.

The Queen’e University of Belfast. TNO Toxicology and Nutrition Institute. * Abstract published in Advance ACS Abstracts, October 1, 1993.

t 2

metabolite cannot however be considered as a genuine arene hydrate since hydration occurs at the alkene (3,4) bond rather than at an arene bond. The recent isolation, structure determination, and configurational assignment of the naphthalene arene hydrates lA, l B , and 1C as metabolites produced by hydroxylation of 1,2-dihydronaphthalene(1A; ref 11)and 1,4-dihydronaphthalene (lB, 1C; ref 12) using growing cultures of a mutant strain of the bacteriumPseudomonas putida prompted the present search for unequivocal direct evidence of arene hydrate formation by liver microsomal enzymes. This study was also facilitated by the development of a range of simple synthetic routes to both racemic and optically pure arene hydrates (12). Recent kinetic studies (13)on the dehydration of arene hydrates l A , lB, and 1C and other arene hydrates including 2A and 2B (14) have shown that they ought to be relatively stable under the aqueous conditions used for liver microsomal biotransformation studies (pH 7.4). Thus, the half-lives of arene hydrates l A , lB, lC, 2A, and 2B were estimated to be 57.5, 11.2, 0.4, 403.0 and 95.4 days respectively at pH 7.4 and 25 OC. The present report describes the successful detection, isolation, and structure stereochemistry assignment of arene hydrates as microsomal metabolites of dihydroarenes.

Experimental Procedures 1,2-Dihydronaphthaleneand 1,2-dihydroanthracenewere obtained by dehydration of the correspondingalcohols (l-hydroxy1,2,3,4-tetrahydronaphthalene and l-hydroxy-1,2,3,4-tetrahydroanthracene) using a catalytic amount of p-toluenesulfonic

0893-228x/93/2706-0808$04.00/00 1993 American Chemical Society

Oxidative Biotransformation of l,2-Dihydroarenes acid. l-Hydroxy-l,2-dihydronaphthalene (1A), 2-hydroxy-1,2dihydronaphthalene (lB), l-hydroxy-l,4-dihydronaphthalene (lC),l-hydroxy-l,2-dihydroanthracene (2A),and 2-hydroxy-l,2dihydroanthracene (2B) were all synthesized in both racemic and enantiomerically pure forms using reported methods (12). Samples of 1,2-epoxy-1,2,3,4-tetrahydronaphthalene (LD), 1,2epoxy-1,2,3,4-tetrahydroanthracene(2D), trans-l,2-dihydroxy1,2,3,4-tetrahydronaphthalene(lE), and trans-1,2-dihydroxy1,2,3,4-tetrahydroanthracene(2E) were available from earlier studies (15). (-)-2-Methoxy-2-(trifluoromethyl)phenylacetic acid (MTPA acid)lwas obtained from the Aldrich Chemical Co. Pregnenolone 16a-carbonitrile was a gift from Upjohn Diagnostics, Ede, The Netherlands. Preparation of Microsomes. Male Wistar rata were treated with one of the following compounds: phenobarbital (PB; 7 days, 0.1 % in drinking water), 3-methylcholanthrene (MC; three daily ip injections of 30 mg/kg of body weight), isosafrole (ISF; four daily ip injections of 150 mg/kg of body weight), isoniazide (ISN; 0.1 % in drinking water for 10 days), dexemathasone ( D E X four daily ip injections of 300 mg/kg of body weight, dissolved in olive oil), and pregnenolone 16a-carbonitrile (PCN; four daily ip injections of 300 mg/kg of body weight). Microsomes were prepared as described previously (16).The cytochromes P-450 1Al and 2B1 were purified essentially according to Ryan et al. (17). Both preparations were homogeneous by SDS gel electrophoresis and had specific concentrations of 9.2 and 17.3 nmol of P-450/mg of protein, respectively. NADPH-cytochrome P-450 reductase was purified from livers of PB-induced male rata using DEAE cellulose and 2',5'-ADP Sepharose affinity chromatography (18). The preparation had a specific activity toward reduced cytochrome c of 38 nmol min-1 (mg of protein)-'. Microsomal Incubations. Microsomal incubations were performed using a combination of microsomal protein (0.25 mg), NADPH (1mM), and MgCl2 (3 mM) in a total volume of 0.5 mL of 0.1 M potassium phosphate buffer (pH 7.4). The substrates were added a t t = 0 (e.g., 1,2-dihydronaphthalene in acetone, not exceeding 25 pL), and a 5-min incubation was performed a t 37 "C. The reaction was terminated by addition of 0.5 mL of methanol containing triethylamine (0.1% ), and the precipitated protein was removed by centrifugation (3000g). Incubations with Purified Cytochromes P-450. Purified cytochrome P-450 (0.1 nmol) was incubated with 0.3 nmol of reductase and 10 pg of dilauroylphosphatidylcholinein 0.1 M potassium phosphate buffer (pH 7.4) in a finalvolume of 0.5 mL. All other conditions were identical to the microsomal incubations. MicrosomalIncubations with [180]H20. In order to assess the origin of the oxygen incorporated into the arene hydrates, 1,2-dihydronaphthaleneand 1,2-dihydroanthracene(both 1mM) were incubated with liver microsomes from PB-treated rats using the conditions as described above. ["W]H20 (Sigma, St. Louis, MO) was added to an l80isotopic abundance of 32 % . Quantitative HPLC Analysis of Metabolites. The supernatant of the microsomal incubation (50-100 pL) was subjected to direct HPLC separation, using an analytical Hypersil ODS (250 x 4.3 mm) reversed-phase column (Chrompack,Middelburg, The Netherlands), eluted with a gradient of 40-90% methanolwater in 10 min with a flow rate of 1mL/min. UV detection a t 250 nm, with peak area integration using Nelson analytical software, resulted in quantification of the arene hydrate and arene metabolites. External standards were used to provide reference. The stability of all metabolites during workup and separation procedures was experimentally confirmed. k' values for metabolites from 1,2-dihydronaphthalene were 2.47 (trans1,2-dihydroxy-1,2,3,4-tetrahydronaphthalene, lE), 3.28 (arene hydrate, lA), 3.31 (arene hydrate, lB), 4.47 (naphthalene), and 1 Abbreviations: MTPA acid, (-)-2-methoxy-2-(trifluoromethyl)phenylacetic acid; PB, phenobarbital, MC, 3-methylcholanthrene; ISF,isosafrol; ISN, isoniazid; DEX, dexamethasone; PCN, pregnenolone 16acarbonitrile; SDS, sodium dodecyl sulfate; CSP, chiral stationary phase; ee, enantiomeric exess.

Chem. Res. Toxicol., Vol. 6, No. 6, 1993 809 4.68 (1,2-dihydronaphthalene). The HPLC conditions used allowed only a qualitative approach to the detection of trans1,2-dihydroxy-1,2,3,4-tetrahydronaphthalene (1E) since the extinction coefficient was a factor of 60 less than the values for arene hydrates (1A and 1B) a t 250 nm. By use of similar HPLC conditions, the metabolites of 1,2-dihydroanthracene were found to have the following k' values: 3.54, trans-l,2-dihydroxy-l,2,3,4tetrahydroanthracene (2E);4.50, arene hydrate 2A; 4.70, arene hydrate 2B; 9.16, anthracene; and 9.88,1,2-dihydroanthracene. K, and V,, values were calculated using the kinetic analysis program EZ-FIT (19).

Separation, Structure, and Stereochemistry Assignments. With 1,2-dihydronaphthalene, alarger scale microsomal (induced by P B or MC) incubation was conducted in a manner similar to that described above, but using a volume of 100 mL with a protein concentration of 1 mg/mL. The l,2-dihydronaphthalene substrate (0.25-0.5 mM) was introduced as a 10 mM solution in methanol. The reaction was initiated by addition of 1 mmol of NADPH with a second addition of NADPH after 10 min. After a 20-min incubation a t 37 OC, the extraction was performed using ethyl acetate (2 X 100 mL). The extract was dried with NazSOd, concentrated under a stream of dry nitrogen, dissolved in CDCla, and examined by 500-MHz lH-NMR analysis (GeneralElectric GN-OM500). The 'H-NMFt spectrum indicated the presence of trans-l,2-dihydroxy-1,2,3,4-tetrahydronaphthalene (1E) [shifts 6 3.79-3.87 (m, 2-H), 4.58 (d, J1,2= 7.2 Hz, 1-H)] and the arene hydrates 1A [shifts 6 4.67-4.79 (lH, m, 1-H), 5.866.00 (lH, m, 3-H)] and 1B [shifts 6 4.40-4.47 (lH, m, 2-H), 6.12 (lH, dd, 3-H),6.56 (lH, d, 4-H)I as majormetabolites. Tentative evidence for the presence of 1,2-epoxy-l,2,3,44etrahydronaphthalene, naphthalene, and 1-naphthol as minor metabolites was also obtained from the lH-NMR spectrum of the mixture. Enzyme-catalyzed oxidation of 1,2-dihydroanthracene was carried out on a 50-mL scale, with other conditions as described for 1,2-dihydronaphthalene, using both PB- and MC-induced microsomes and pure monooxygenases (P-450 1Al and P-450 2B1) to yield arene hydrates 2A and 2B and trans-diol 2E as the major metabolites. A small proportion of the total metabolites from 1,a-dihydronaphthalene was treated with bis(trimethylsily1)trifluoroacetamide in pyridine solvent prior to GLC-MS analysis using aVG 12-250instrument linked toaPDP11/23PLUSdatasystem. GLC-MS analysis of the derivatized mixture of metabolites was carried out using a 25-m BP1 capillary column a t 175 OC, and this again confirmed the presence of trans-diol (1E) and arene hydrates (1A and 1B) as major metabolites. Using the normal or SIM mode and standard reference compounds, 1,a-epoxy1,2,3,4-tetrahydronaphthalene(lD),naphthalene, 1-hydroxy-1,4dihydronaphthalene (E),and 1-naphthol were found as very minor metabolites of 1,2-dihydronaphthalene (using PB-induced microsomes). Similar silylation procedures were adopted in the GLC-MS analysis of the products obtained by enzyme-catalyzed oxidation of 1,2-dihydroanthracene. The incubations with [Wlwater were also analyzed using this procedure. Thus, using liver microsomes from both PB- and MC-induced rata and pure monooxygenase enzymes (P-450 1Al and P-450 2B1), in conjunction with a H P GH-5890 Model GLC-MS instrument linked to a MSD-5970 detector and a HP-ULTRA-1 capillary column, the presence of trans-diol 2E and arene hydrates 2A and 2B as major metabolites was confirmed. Since the total yield of metabolites from 1,2dihydroanthracene was significantly lower than that found using 1,2-dihydronaphthalene as substrate for microsomal oxidations, unequivocal evidence of the formation of the tetrahydroepoxide (2D),arene hydrate ( 2 0 , or anthracene could not be obtained using the GLC-MS method. However, with the reversed-phase HPLC system used for quantitation of the metabolites described above, anthracene could be identified by the retention time and the UV spectrum. Normal-phase HPLC analysis of the crude microsomal products using a Microsorb SIL (150 X 4.6 mm, 5 pm) column and

810 Chem. Res. Toxicol., Vol. 6, No. 6, 1993 Table I. Relative Yields,. Enantiomeric Excess (% em),* and Absolute Configurations" of the Arene Hydrate (lA, lB, 2A, 2B) and trans-Dihydrodiol (lE, 2E) Metabolites inducer metabolite re1 yield % ee abs config 70 1s MC 1A 7 MC 1B 18 27 2R 54 lR, 2R MC 1E 15 PB 1A 12 27 1s PB 1B 28 46 2R PB 1E 60 4 lR, 2R 98 1s MC 2A 13 47 2R MC 2B 12 d d MC 2E 75 PB 2A 1 0 1SIR PB 2B 22 48 2R PB 2E 77 d d P450 2B1 2A 4 98 1R P450 2B1 2B 16 53 2R P450 2B1 2E 80 d d P450 1Al 2A 22 98 1s P450 1Al 2B 12 e 2R P450 1Al 2E 66 d d a Average values based on 'H-NMR, GLC-MS, and normal-phase HPLC data (15%). Yields are related to the s u m of A, B, and E. By direct CSP HPLC analysis (lA, lB, 2A, 2B) or by indirect determination of the diMTPA esters (1E). From direct comparison with authentic samples of established absolute configuration. d Insufficient sample for accurate analysis. e Excess of 2R enantiomer observed,but insufficientsample available for accurate determination.

*

propan-2-01 (0.3%) in hexane as eluant at a flow rate of 1.0 mL/ min showed the separation of arene hydrates 1A (k' = 9.0), 1B (k' = 11.3), 2A (k' = 2.4), and 2B (k' = 3.4). The trans-diols 1E (k' = 4.1)and2E (k' = 1.7) weresubsequentlyelutedusingpropan2-01 ( 5 % ) in hexane (0.7 mL/min). Peaks containing the arene hydrates lA, 2A, lB, and 2B were collected, and these samples were then used for enantiomeric analysis using a chiral stationary phase (CSP) HPLC column (Chiralcel OB, 250 x 4.6 mm, 10% propan-2-01 in hexane as eluant). By use of a flow rate of 0.5 mL/min, the arene hydrates 1A and 1Bwere each separated into 2A and 2B were also found enantiomers (1A (~1.32;1B (~1.11). to be separable into enantiomers using the same CSP-HPLC (~1.52).While the enantiomer separation conditions (2A (~1.12;2B of arene hydrates lA, lB, 2A, and 2B was satisfactory, the Chiralcel OB column used was found to have a short lifetime, despite using only recommended solvents (hexane-2-propanol). The absolute configurations and optical purities of arene hydrate metabolites (lA, l B , 2A, 2B) were determined by comparison with homochiral reference samples which were available from separate studies (11, 12). The HPLC analysis of the arene hydrates involved use of a UV detector operating a t 254 nm, and the ratio obtained was corrected for small differences in extinction coefficients between arene hydrates: lA, 7.2 X 103; lB, 9.4 X 103; 2A, 2.8 X lo4; and 2B, 2.7 X lo4. The stereochemistry of the dihydrodiol metabolite 1E was determined by converting it to the di-MTPA esters, separation of the diastereomers on a Zorbax Si1 HPLC column (80 X 6.2 mm, 3 Mm, 2.5% diethyl ether in hexane as eluant (k'(1E) = 8.1 and 8.7) followed by comparison (HPLC) with an authentic sample of the di-MTPA ester of (lR,2R)-trans-1,2-dihydroxy1,2,3,4-tetrahydronaphthalene.

Results and Discussion Based upon 'H-NMR, GLC-MS, and HPLC analyses, three types of metabolites resulted from the microsomal oxidation of both 1,2-dihydronaphthalene and 1,2-dihydroanthracene: (1)the classical trans-dihydrodiols (lE, 2E) derived from epoxides, (2) the arene hydrates (lA, lB, 2A, 2B), and (3) the formal dehydrogenation products, naphthalene and anthracene (Table I). Their formation was consistent with cytochrome P-450 catalyzed

Boyd et al.

hydroxylation and epoxidation mechanisms; Le., the formation of metabolites was linear with time and protein concentration within the values used and was dependent upon NADPH. Carbon monoxide decreased the rate of formation of metabolites by ca. 50%. The formation of trans-1,2-dihydroxy-1,2,3,4-tetrahydronaphthalene (1E) as the major biotransformation product of 1,2-dihydronaphthalene probably results from a cytochrome P-450 catalyzed epoxidation to yield 1,2epoxy-l,2,3,4-tetrahydronaphthalene(1D) followed by hydration. Using microsomes, the formation of the diol may either be spontaneous or catalyzed by epoxide hydrolase. Since a diol (2E) was also formed from 1,2dihydroanthracene using highly purified cytochrome P-450, it seems likely that spontaneous hydrolysis occurs. Evidence for the intermediacy of an epoxide metabolite (1D) during the formation of trans-diol 1E was confined to GLC-MS analysis where only trace quantities were detected. The trans-diol metabolite (1E)was purified by HPLC but could not be separated into enantiomers using the Chiralcel OB-CSP HPLC system. Formation of the di-a-methoxy-a- (trifluoromethy1)phenylaceticacid (diMTPA) ester derivatives of the trans-diol metabolite lE, followed by separation on a Microsorb SIL column provided a measure of the diastereoisomeric excess (and thus the enantiomeric excess) of the trans-diol 1E. An excess of the 1R,2R configuration was observed when 1,2dihydronaphthalene was oxidized to yield trans-diol 1E using PB-induced (4% ee) and MC-induced microsomes (54% ee, Table I). Using PB microsomes, hydroxylation at the allylic position to yield arene hydrate 1B (or 2B) appeared to be preferred to benzylic hydroxylation for 1,2-dihydronaphthalene (Table I). In contrast, using MC microsomes, the arene hydrate of anthracene (2A)resulting from benzylic hydroxylation of 1,2-dihydroanthracene was the major isomer observed. Although for 1,2-dihydronaphthalene, with MC microsomes benzylic hydroxylation is not the major pathway, the same trend was found as observed for L2-dihydroanthracene (Table I). Using microsomes from rats induced with a range of known inducers of cytochrome P-450, the ratio of maximal ratesofcatalysis, Vm=(lB)/Vm=(lA)(1.6-22.2) and Vm=(2B)/Vm=(2A)(0.5-24.6), showed a wide range of values according to the nature of the inducing agent (Table 11). The largest V,, ratio (i.e., favoring allylic hydroxylation) was found for control, PB, and ISN (induction of P-450 2B1, P-450 2B2, and P-450 3A1 enzymes) induced microsomes. The smallest Vm, ratio (favoring benzylic hydroxylation) was observed for MC (inducer of cytochrome P-450 1Al and 1A2 enzymes) and ISF (inducer of P-450 1Al and 1A2 enzymes). The apparent Kmvalues for the formation of the arene hydrates of naphthalene (lA, 1B) were similar (ca. 0.14 mM) for all inducers except for microsomes from PBtreated rats. Kmvalues for the formation of 2A and 2B were generallysmaller (ca. 0.02-0.13 mM) than those found for 1A and 1B (Table 11). The absolute yields of arene hydrates of anthracene (2A, 2B) formed from 1,2-dihydroanthracene were generally lower than those found for the arene hydrate metabolites of 1,2-dihydronaphthalene. On the basis of the data in Table I, it would appear that the arene hydrate metabolites obtained from MC-induced microsomal incubations generally had higher ee values

Chem. Res. Toxicol., Vol. 6, No. 6, 1993 811

Oxidative Biotransformation of 1,2-Dihydroarenes

inducer CC PBc

MCc ISP IS" PCNC

CC PBC MCC ISFC ISNC PCNC DEXC

Table 11. Metabolism of 1f-Dihydronaphthalene and lf-Dihydroanthracene by Rat Liver Microsomes arene hydrate 1A arene hydrate 1B naphthalene Km" V-b Km' V-b V-(lB)/V-(W Km' V-b 0.25 f 0.02 0.11 f 0.03 0.22 f 0.05 1.2 f 0.16 0.13 f 0.02 1.6 f 0.10 0.17 f 0.03 2.5 f 0.19 0.15 f 0.03 0.33 f 0.03 0.68 f 0.04 0.19 f 0.03 arene hydrate 2A d d

0.04 f 0.01 0.07 i 0.03 0.02 f 0.01 0.09 f 0.04 0.09 f 0.04

0.17 f 0.07 0.15 f 0.02 1.4 f 0.10 2.6 f 0.28 0.18 f 0.02 0.54 f 0.06 0.56 f 0.08

5.6 f 0.21 0.13 f 0.01 0.21 f 0.03 24.2 f 1.8 0.13 f 0.02 2.7 f 0.14 0.16 f 0.04 4.1 f 0.37 0.11 f 0.01 4.7 f 0.21 2.8 f 0.09 0.17 f 0.01 arene hydrate 2B 0.05 f 0.01 0.06 f 0.02 0.06 f 0.03 0.13 f 0.06 0.05 f 0.01 0.10 f 0.03 0.11 f 0.05

2.1 f 0.16 3.7 f 0.27 0.82 f 0.10 1.2 f 0.16 2.3 f 0.14 1.2 & 0.10 1.1 f 0.15

22.2 20.7 1.7 1.6 14.2 4.1

12.5 24.6 0.56 0.45 12.6 2.3 2.0

0.30 f 0.02 5.1 f 0.14 0.63 f 0.16 17.3 f 3.3 0.31 f 0.03 6.1 f 0.28 0.34 f 0.07 7.3 f 0.62 0.15 f 0.02 3.9 f 0.22 9.0 f 1.11 1.39 f 0.26 anthracene 0.02 f 0.03 0.07 f 0.02 0.06 f 0.03 0.05 f 0.02 0.03 f 0.03 d d

0.49 f 0.14 0.76 f 0.06 0.40 f 0.06 0.41 f 0.16 0.69 f 0.07 d d

0 The Michaelis constant (K,) is expressed in millimolar. The specific activity (V-) is expressed as nanomoles of product formed per minute per milligram of microsomal protein. All incubations were performed in a 0.5-mL volume with 0.5 mg of microsomal protein for 5 min at 37 "C. Quantification wa8 achieved by direct injection onto reversed-phase HPLC of 50 pL of the incubate after stopping the reaction with one volume of MeOH and precipitating the protein by centrifugation. C, control; PB, phenobarbital; MC, 3-methylcholanthrene; ISF, isosafrole; ISN, isoniazid; PCN, pregnenolone 16a-carbonitrile; DEX, dexamethasone. d Values were too low for accurate determination.

(27-98% ee) relative to those obtained using PB-induced microsomes (0-48% ee). A marked preference for the 1s configuration during benzylic hydroxylation of 1,2-dihydronaphthalene (arene hydrate lA, 70% ee) and 1,2dihydroanthracene (arene hydrate 2A, 98% ee) was observed using MC-induced microsomes. A modest excess of the (2R) enantiomers (27-48% ee) was consistently observed for the arene hydrates obtained by allylic hydroxylation (lB, 2B) using both PB- and MC-induced microsomes. The origin of the oxygen which is incorporated to form the arene hydrates was determined in microsomal incubations of 1,2-dihydronaphthalene and 1,2-dihydroanthracene containing POIHz0. GC-MS analysis showed no incorporation of l8O in arene hydrates, since the M + 2 intensities of both derivatized arene hydrates, as obtained from the incubations, were identical to the derivatized reference compounds. The trans-dihydrodiol 2E detected in the incubation with 1,2-dihydroanthracene showed a ca. 35 % incorporation of l 8 0 , as indicated by the presence of an M + 2 intensity at 360 Da of the TMS derivative of 2E with a ca. 70% intensity relative to the molecular ion at 358 Da. Thus, the oxygen incorporated in the dihydrodiols is derived from water, while the oxygen of the arene hydrates originates from molecular oxygen. The presence of a very minor proportion (