Stereospecificity of microsomal and purified epoxide hydrase from rat liver. Hydration of arene oxides of polycyclic hydrocarbons.

The stereospecific metabolism of racemic benzo[aIpyrene 4,5-, 7,8-, and 9,10-oxide by rat liver microsomes or highly purified epoxide hydrase has been examined. The enantiomerit purity of the metabolically formed benzolalpyrene 7,8and 9,10-dihydrodiol is relatively low (8% and 22%, respectively) whereas metabolically formed benzo[alpyrene 4,5dihydrodiol appears to be highly enriched in the (-)-enanComer (78% enantiomeric purity). The low optical purity of benzo[alpyrene 7,Sdihydrodiol appears to be due to the ability of epoxide hydrase to act upon position 8 of both optical enantiomers of benzo[alpyrene 7,Soxide with almost equal ease. Metabolism of racemic benzolalpyrene 7,8-oxide in l80-enriched water indicated that this substrate was hydrated almost exclusively (>98%) at position 8 by the membrane-bound or highly purified enzyme, whereas racemic benzo[a]pyrene 4,5-oxide was attacked by water with almost equal ease at positions 4 and 5 of the molecule. Other polycyclic hydrocarbon K region dihydrodiols, such as phenanthrene 9,10-dihydrodiol and benz[alanthracene 5,6-dihydrodiol, are formed with less stereospecificity (26% and 42% optical purity, respectively) than is benzo[aIpyrene 4,5-dihydrodiol. The above observations indicate that the degree of stereospecific hydration of arene oxides by epoxide hydrase depends on the substrate studied. The type and extent of specificity appears to be governed by both steric and electronic factors associated with the substrate itself.

The above observations indicate that the degree of stereospecific hydration of arene oxides by epoxide hydrase depends on the substrate studied. The type and extent of specificity appears to be governed by both steric and electronic factors associated with the substrate itself.
The liver microsomal cytochrome P-450-containing monooxygenase system and epoxide hydrase play important roles in determining the biological fate of many environmental pollutants and toxic compounds. For many of these chemicals such as the polycyclic aromatic hydrocarbons, biologically active arene oxides are the initial products formed by the microsomal cytochrome P-450 system (1, 2). Certain of these arene oxides have toxic, mutagenic, and carcinogenic activity (l-6). The microsomal enzyme epoxide hydrase plays a pivotal role in the * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked 'hduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. metabolism of these intermediate arene oxides to dihydrodiols via the tram addition of water (1). The dihydrodiols formed are usually biologically inactive per se, but some of these compounds are further metabolized by the cytochrome P-450dependent monooxygenase system to highly mutagenic and carcinogenic diol epoxides (7-10) which are poor substrates for epoxide hydrase (7, 11-13). Thus, epoxide hydrase plays a central role in both the inactivation and activation of polycyclic aromatic hydrocarbons to mutagenic and carcinogenic metabolites.
Both the cytochrome P-450 system and epoxide hydrase show varying and often very high degrees of stereospecificity in their metabolism of different substrates. For example, liver microsomes from 3-methylcholanthrene-treated rats oxidize the (-)-enantiomer of BP 7,8-dihydrodioll primarily (86%) to BP 7,8-diol-9,10-epoxide with the 7-OH group and the oxirane oxygen having trans stereochemistry (14, 151, whereas under similar conditions the ( + )-enantiomer is almost exclusively (97%) oxidized to BP 7,8-diol-9,10-epoxide in which the 7-OH group and oxirane oxygen have cis stereochemistry (14). The 4,5-, 7,8-, and 9,10-dihydrodiols formed metabolically from benzo[a]pyrene were found to have high optical purity (14).2 However, BP 7,8-dihydrodiol formed from the racemic BP 7,8oxide was of very low optical purity (14) indicating low stereo- However, optical purity of BP 4,5-and 9,10-dihydrodiol was not determined, although the (-)-enantiomers of BP 4,5-and 9,10dihydrodiol were shown to be formed to a greater extent than the ( + I-enantiomers. specificity of rat liver epoxide hydrase for this substrate. Since the BP 7,&3-dihydrodiol formed from benzo[alpyrene by rat liver is of high optical purity, the cytochrome P-450-dependent monooxygenase system must form BP 7,8-oxide with high optical purity. Epoxide hydrase from rabbit liver is highly stereospecific toward the 9,10-oxide of phenanthrene, forming 9,10-dihydrodiol with high optical purity (60 to 70%) but less stereospecific toward benzene oxide (optical purity of the dihydrodiol -50%) and naphthalene 1,2-oxide (optical purity of the dihydrodiol -3O-50%) (17). Epoxide hydrase from rat liver catalyzes the hydration of cis stilbene oxide to give the corresponding dihydrodiol of high optical purity (18). Since the stereospecific metabolism of chemicals by the cytochrome P-450 system and epoxide hydrase can play a critical role in the expression of the biological activity of these chemicals, we have studied the stereochemical course of hydration of several carcinogenic and noncarcinogenic arene oxides. (2 to 4 mg of protein), tritiated substrate (400 nmol), and water to reach a final volume of 1.00 ml. Reactions were initiated by addition of the substrate and the reaction mixture was incubated at 37" for 2 to 10 min after which 3.0 ml of ethyl acetate was added to the incubation mixture and the product as well as the unreacted arene oxide were extracted into the organic phase. The ethyl acetate layer was separated by centrifugation and dried (anhydrous Na,SO,), and the solvent was evaporated with a stream of dry nitrogen,

MATERIALS AND METHODS
The residue was dissolved in a small volume of tetrahydrofuran (-50 ~1) and the dihydrodiols were isolated by HPLC. With purified epoxide hydrase, the experimental conditions were similar except that highly purified epoxide hydrase (30 pg) and dilauroyl phosphatidylcholine (600 Fg) were used instead of microsomes.
Optical Purity of Dihydrodiols -"H-labeled dihydrodiols obtained from the reactions of the corresponding arene oxides with epoxide hydrase were diluted with a large excess (-1 mg) of the same unlabeled, racemic dihydrodiol as carrier and were dissolved in 100 /rl of pyridine.
An excess of MTPA-Cl was then added and the reaction mixture was allowed to stand under N, at 4" for 12-15 h. The pairs of diastereomeric diesters which formed were separated by HPLC and the radioactivity associated with each diastereomer was determined.
Since the recovery of each member of the pair of diastercomers for each dihydrodiol was identical as determined by peak area, the enantiomeric purity of the enzymatically formed "H-labeled dihydrodiol could be calculated directly from the radioactivity in each peak. The concept is illustrated below: --. ^. In most cases, incubation conditions were selected to limit the conversion of the arene oxides to dihydrodiols to less than 20% of the added substrate.
Enzyme Preparations -Immature (50 to 60 g) male rats of the Long-Evans strain were treated with phenobarbital (75 mg/kg per day) or 3-methylcholanthrene (  all these samples were performed by the usual method (22) except that peak heights were determined by a computer.

RESULTS
The enantiomeric purity of benzolalpyrene dihydrodiols, obtained by hydration of the corresponding racemic arene oxides with liver microsomal epoxide hydrase, are given in Table II. A typical HPLC trace of the separation of the diastereomers from which the optical purity of the corresponding dihydrodiol was computed is shown in Fig. 2. The enantiomerit purity of the metabolically formed BP 7,8-and 9,10-dihydrodiols is relatively low (8% and 22%, respectively) whereas BP 4,Sdihydrodiol appears to be highly enriched in the (-Ienantiomer (78% enantiomeric purity). The low enantiomeric purity of the BP 7,8-and 9,10-dihydrodiols could be due to any of several reasons. For example, the enzyme epoxide hydrase may lack specificity for positions 7 and 8 of the molecule or it may have comparable preference for both enantiomeric forms of the substrates. These possibilties were investigated when  tained 7% l8O, indicating that 'sO-enriched water in the incubation medium had attacked positions 4 and 5 of BP 4,5-oxide at relative rates of 3:2. The fact that the total IsO in the two phenol acetates (17%) does not equal that in the starting dihydrodiol (33%) suggests that some exchange of the dihydrodiol with solvent water had occurred during the course of the dehydration. However, comparable incorporation of '*O in positions 4 and 5 of racemic BP 4,5-oxide suggests that attack of water occurs at both positions to about the same extent. Moreover, it was also observed that when 30% or 75% of the racemic BP 4,5-oxide was metabolized to the corresponding dihydrodiol, similar optical purity (84% and 80%, respectively) was obtained. These results indicate that epoxide hydrase can utilize both enantiomers of BP 4,5-oxide as substrates and that it directs the attack of water at position 4 of one enantiomer and position 5 of the other enantiomer, predominantly only one of the chiral centers, giving rise to BP 4JGdihydrodiol of high optical purity (Fig. 3).
The optical purity of BP 4,5-, 7,8-, and 9,10-dihydrodiol formed from the metabolism of benzolalpyrene by liver microsomes from rats treated with 3-methylcholanthrene were described earlier (14)' and are listed in Table II. All three dihydrodiols formed by this route have very high optical purity C-92%, cf. Ref. 14). This result, along with the fact that epoxide hydrase forms dihydrodiols of low optical purity from racemic BP 7,8-oxide and racemic BP 9,10-oxide, suggests that these two oxides must be formed with high optical purity from BP by the monooxygenase system. Since the high optical purity of BP 4,5-dihydrodiol, formed by the microsomal metabolism of BP, could have resulted from the high stereospecificity of epoxide hdyrase toward the racemic BP 4,5-oxide, it is not known whether or not the monooxygenase system metabolizes BP to the 4,5-oxide with high stereoselectivity.
Previous studies have indicated that the properties of rat liver epoxide hydrase are unaltered by pretreatment of the animals with either phenobarbital or 3-methylcholanthrene (19). However, the stereospecificity of the enzyme in microsomes obtained from induced or control animals has not been examined. Hence, the enantiomeric purities of BP 4,5-and 7,8dihydrodiol obtained from the hydration of the corresponding arene oxide by microsomes from control, phenobarbital-, and 3-methylcholanthrene-pretreated rats as well as with purified epoxide hydrase from rats treated with phenobarbital were determined. The optical purity of the dihydrodiols obtained is compared in Table III. The enantiomeric purity of BP 4,5dihydrodiol formed from racemic BP 4,5-oxide by epoxide hydrase from different microsomal sources and the homogeneous enzyme is identical, indicating that the high stereospecificity of the enzyme towards racemic BP 4,5-oxide is unaltered after pretreatment of rats with either phenobarbital or 3-methylcholanthrene.
The hydration of racemic BP 7,8-oxide to BP 7,8-dihydrodiol occurred with low stereospecificity when epoxide hydrase from several sources was used (Table III).
BP 4,5-dihydrodiol, formed from the corresponding racemic oxide, has much higher optical purity than the other two non-K region dihydrodiols (Tables II and III). It was of interest to determine if the high stereospecificity of epoxide hydrase was associated with the hydration of all K region arene oxides. The results in Table IV compare the enantiomeric purity of the K region dihydrodiols of benzolalpyrene, benzcalanthracene and phenanthrene obtained from the corresponding racemic arene oxides. A steady decrease in the enantiomeric purity of the product was observed as the size of the substrate decreased. The enantiomeric purities of the metabolically formed K region dihydrodiols of benzo[a]pyrene, benzlalanthracene, and phenanthrene were 78%, 42%, and 26%, respectively (Table  IV). DISCUSSION The stereospecific metabolism of polycyclic hydrocarbon arene oxides by rat liver microsomes or by highly purified epoxide hydrase has been examined. Several racemic K region and ( Signs of phenanthrene 9,10-dihydrodiol enantiomers were determined by comparing retention times of the two diastereomeric MTPA diesters with that of an MTPA diester of authentic (-)-9S, lOS-enantiomer of phenanthrene 9,10-dihydrodiol.