Chem. Res. Toxicol. 1996, 9, 871-874
871
The Mechanism of Oxidation of Allylic Alcohols to r,β-Unsaturated Ketones by Cytochrome P450† Giuseppe Bellucci,‡ Cinzia Chiappe,*,‡ Laura Pucci,§ and Pier Giovanni Gervasi*,§ Dipartimento di Chimica Bioorganica, Universita` di Pisa, via Bonanno 33, 56126 Pisa, Italy, and Istituto di Mutagenesi e Differenziamento, CNR, via Svezia 10, 56124 Pisa, Italy Received January 10, 1996X
The oxidation of cyclohex-2-en-1-ol, a simple model substrate for allylic alcohols, is catalyzed by several P450 isoenzymes and leads exclusively to cyclohex-2-en-1-one. No double bond epoxidation or C(4) hydroxylation have been observed. The large primary kinetic isotope effect measured using [2H]-1-cyclohex-2-en-1-ol is consistent with an at least partially rate limiting breaking of the C(1)-H bond. The mass spectrometric analysis of cyclohex-2-en-1-one obtained from [18O]cyclohex-2-en-1-ol has established that a gem-diol intermediate is involved, even if a dual hydrogen abstraction pathway may contribute to the reaction.
Introduction
Materials and Methods
Cytochrome P450s are a family of isoenzymes able to oxidize a broad variety of endogenous and exogenous substrates playing a crucial role in the metabolism of xenobiotics (1). Olefinic compounds are usually oxidized by cytochrome P450s to the corresponding epoxides (2), while alcoholic functions can be converted to carbonyl groups (3). The latter biotransformation is particularly prominent for allylic alcohols, where double bond epoxidation appears to be hampered in favor of alcoholic group oxidation. This is particularly important from the toxicological point of view, since epoxides are electrophilically reactive compounds able to react with the cellular macromolecules producing toxic, mutagenic, and carcinogenic effects (4). A typical example of allylic alcohol oxidation is provided by the metabolism of trans-R,R,4trimethyl-5-hydroxy-3-cyclohexene-1-methanol (sobrerol), a mucoregulatory drug, whose main metabolite is the corresponding R,β-unsaturated ketone (5). No double bond epoxidation of sobrerol was detected, in spite of the fact that 1-methylcyclohexene, a model compound for sobrerol lacking the allylic hydroxyl group, was extensively epoxidated by liver microsomal preparations from control and induced rats (6).
Chemicals. Dexamethasone (DEX), β-naphtoflavone (β-NF), phenobarbital (PB), and pyrazole (PZ) were obtained from common commercial sources. Cyclohex-2-en-1-ol (1) was prepared by refluxing 3-bromocyclohexene (2 g) in 7.5 mL of tetrahydrofuran-water (2:1) containing Na2CO3 (1.4 g) for 4 h, followed by distillation (bp 164-165 °C). [18O]-1-Cyclohexen2-en-1-ol ([18O]-1) was prepared on a 10-fold reduced scale using the same procedure, employing commercial (Aldrich) [18O]water, followed by Kugelrohr short-path distillation. [2H]-1-Cyclohex2-en-1-ol ([2H]-1, >95% [2H], by NMR) was obtained by reduction of commercial cyclohex-2-en-1-one (2) with an equimolar amount of LiAlD4 in anhydrous ethyl ether at room temperature. cis2,3-Epoxycyclohexanol and trans-2,3-epoxycyclohexanol were prepared by epoxidation of 1 with m-chloroperbenzoic acid in dichloromethane. Cyclohex-2-ene-1,4-diol was prepared as previously reported (7). Animals and Microsomal Preparations. Male SpragueDawley rats were purchased from Charles River, Germany. They were treated for 3 days with PB ip (80 mg/kg daily), DEX (40 mg/kg daily), β-NF (40 mg/kg daily), or PZ (200 mg/kg daily), and microsomes were obtained from the liver as previously described (8). Microsomal protein concentrations were assayed by using the method of Lowry et al. (9); the total cytochrome P450 concentration was measured according to Omura and Sato (10). Enzymatic Activity and Kinetic Parameters. 1-Cyclohex-2-en-1-ol oxidase activity was usually measured as follows. Incubation mixtures (2 mL), containing 100 mM K/phosphate buffer, pH 7.4, 2 mg of hepatic microsomal protein, and an NADPH-generating system, consisting of 0.5 mM NADP+, 5 mM glucose 6-phosphate, and 0.5 unit/mL glucose 6-phosphate dehydrogenase, were incubated for 5 min, and the reactions were initiated by the addition of a proper amount of 1 or its [2H]-1 or [18O] isotopomers. After 15 min, NaCl was added to precipitate the microsomal proteins. The reaction products were extracted with ethyl acetate (2 × 5 mL) and quantified by GLC (OV 225 column, 95 °C) after addition of appropriate amounts of an ethyl acetate stock solution of cyclohexanone as an internal standard. The substrate concentrations ranged from 1 to 50 mM. Concentrations of 1 higher than 50 mM were not used as they may cause microsomal enzyme denaturation. 1-Cyclohex2-en-1-ol oxidase activity was also measured in reconstituted systems by using purified 2B1, 2E1, 1A1, and NADPH cytochrome P450 reductase obtained as previously described (8, 11), and 2C11 purified from control male SD rats according to Fugita et al. (12). The reconstituted incubation mixtures (0.5 mL) contained 0.2 nmol of P450, 0.6 nmol of P450 reductase, 15 µg of dilau-
In order to focus on the factors affecting the competition between double bond epoxidation and hydroxyl group oxidation and to try to identify the involved P450s, as well as to clarify the mechanistic aspects of this reaction, we undertook a study of the oxidative biotransformation of cyclohex-2-en-1-ol (1) by rat liver P450s. The use as sources of P450s of microsomal preparations obtained after pretreatment of rats with P450 selective inducers, and the use of some purified P450s (1A1, 2E1, 2B1, 2C11), has shown that 1 is susceptible of oxidation by some P450 isoforms only at its OH group. In addition, the use of 18O or 2H labeled substrate has allowed us to formulate a mechanistic rationalization of the investigated reaction. † This paper is dedicated to the memory of Professor Giuseppe Bellucci (d. March 3, 1996). ‡ Universita ` di Pisa. § Istituto di Mutagenesi e Differenziamento, CNR. X Abstract published in Advance ACS Abstracts, June 15, 1996.
S0893-228x(96)00005-7 CCC: $12.00
© 1996 American Chemical Society
872
Chem. Res. Toxicol., Vol. 9, No. 5, 1996
roylphosphatidylcholine, and 1 in 0.1 M K/phosphate buffer, pH 7.4. The reactions were carried out at 37 °C for 15 min after the addition of 1 mM NADPH, and the products were extracted and analyzed as described above. In all cases, the oxidation of 1 to 2 followed simple MichaelisMenten kinetics. With control microsomes it was linear up to 20 min and 1 mg/mL of microsomal protein. The apparent kinetic parameters Vmax and Km were determined from Lineweaver-Burk plots of 2 formation data obtained from incubations containing eight different substrate concentrations. Control experiments carried out incubating cis-2,3-epoxycyclohexanol, trans-2,3-epoxycyclohexanol, and cyclohex-4-ene-1,4diol for the same time as 1 showed that these compounds were completely recovered. Mass Spectral Determinations. The ratio between the [16O] and [18O] isotopomers of 1 was determined by GC-MS using a Perkin-Elmer 8200 QMass 910 equipped with a DB1 column (70 °C, retention time 4.46 min), on the basis of the M - 1, M, and M + 1 peaks at m/z 97, 98, and 99 (for the unlabeled) and at 99, 100 and 101 (for the [18O] labeled compound), as well as on the basis of the M - 15 peaks at m/z 83 and 85. Both determinations gave an 80:20 ratio of the labeled to the unlabeled alcohol. The product of the incubation of a sample of this alcohol (5 mg) with a microsomal preparation (5 mL) obtained from PB pretreated animals was subjected to GC-MS analysis under the above conditions (retention time of 2, 5.1 min). A 50:50 ratio of the [16O] and [18O] isotopomers was determined on the basis of the molecular ion peaks at m/z 96 and 98. This isotopomer ratio remained exactly unchanged when the oxidation product was left in the experimental conditions for a much longer time, showing that oxygen exchange on the ketone with H2O did not occur at an appreciable rate.
Results Microsomal Oxidation of 1-Cyclohex-2-en-1-ol (1). The allylic alcohol 1 was oxidized by microsomal enzymes in the presence of an NADPH-generating system only at the hydroxyl group, giving the corresponding ketone 2. Neither products of epoxidation of the double bond (cisand trans-2,3-epoxycyclohexanols or triols arising from their hydrolysis) nor products of allylic hydroxylation at C(4) of 1 (cyclohex-2-ene-1,4-diol) were detected by GLC of the extracts of the incubation mixtures. That these products were actually not formed was confirmed by control experiments showing that they were stable in the incubation conditions.
The formation of 2 followed monophasic MichaelisMenten kinetics and required NADPH, whereas NADH was rather ineffective as a source of electrons. The oxidation also required, as typical for P450-dependent reactions, molecular oxygen and was inhibited by CO, whereas SOD and mannitol, scavengers of O2•- and •OH, respectively, did not inhibit the formation of 2, indicating that the above radicals are not involved in the oxidation of 1. Effect of Pretreatments with P450 Inducers and Use of Purified P450 1A1, 2B1, 2E1, and 2C11. Pretreatments of rats with β-NF, PZ, PB, and DEX, respective classical inducers of P450 1A1/2, 2E1, 2B1/2, and 3A1/2 (13), significantly decreased, compared to the control values, both the apparent Km and the Vmax for the
Bellucci et al. Table 1. Values of Apparent Kinetic Constants for the Oxidation of 1 by Microsomes from Control and Treated Rats and by P450 2C11, 2B1, 2E1, and 1A1a microsomes or purified P450 control β-NF PZ PB DEX P450 2C11 P450 2B1 P450 2E1 P450 1A1
Km (mM) 79 ( 10 29 ( 3** 22 ( 8** 17 ( 4** 9 ( 4** 72 ( 11 20 ( 5 NDb NDb
Vmax [nmol min-1 Vmax [nmol min-1 (mg of protein)-1] (nmol of P450)-1] 28 ( 5 33 ( 4 22 ( 3 17 ( 5 9 ( 3**
51 ( 9 27 ( 4* 32 ( 5* 19 ( 6** 11 ( 4** 55 ( 8 11 ( 3 3.2 ( 0.7c 2.1 ( 0.5c
a The reported values are averages ( SD of three determinations performed with different microsomal preparations. b ND ) not determined. c Value obtained at 20 mM concentration of 1. (** and *) Significantly different from the control microsomes by Student’s t-test: **P
