Characterization of the cytosolic epoxide hydrolase-catalyzed hydration products from 9,10:12,13-diepoxy stearic esters.

In a previous report we hypothesized that diepoxy fatty methylesters are metabolized to tetraols and/or tetrahydrofurandiols through an epoxydiol intermediate. In this study, p-nitrophenyldiepoxystearate was incubated with affinity-purified liver cytosolic epoxide hydrolase and product formation was monitored by reverse phase HPLC. The diepoxystearate was converted to the corresponding 9,10,12,13-tetraol using a concentrated enzyme (greater than or equal to 100 micrograms/ml). When lower concentration of the enzyme was used, simultaneous elevation of 9,10-epoxy-12,13-dihydroxy and 12,13-epoxy-9,10-dihydroxystearate along with disappearance of tetraol was observed. The epoxydiols were intermediates which could be isolated and cyclized quantitatively to form two chromatographically distinct tetrahydrofurandiols (A with a low Rf value and B with a high Rf value on TLC). Gas chromatographic analysis on a cyclodex-beta capillary column revealed that each compound was composed of two different isomers. The structure of these isomers was 9(12)-oxy-10,13-dihydroxystearate and 10(13)-oxy-9,12-dihydroxystearate using mass spectrometry. Stereochemistry of the aliphatic chain across the tetrahydrofuran moiety was determined by nuclear Overhauser effect spectroscopy. Chemically and enzymatically generated tetrahydrofurandiols had similar retention time on GC and HPLC, and identical mass spectra using the electron impact mode.     

Hydration of an 18O epoxide by a cytosolic epoxide hydrolase from mouse liver

The mechanism of enzymatic epoxide hydration by a cytosolic or 100,000 g soluble mammalian liver enzyme (in contrast to the microsomal enzymes) was examined by monitoring ¹⁸O distribution following chemical and enzymatic hydrations of ¹⁶O or ¹⁸O epoxide labeled (±) 1-(4′-ethylphenoxy)-3, 7-dimethyl-6, 7-epoxyoctane. Acid catalyzed hydration of the ¹⁸O epoxide in ¹⁶O water, and hydration of the ¹⁶O epoxide in ¹⁸O water, indicated that attack by water was predominantly on the tertiary carbon (C-7). Enzymatic epoxide hydration led to attack predominantly on secondary carbon (C-6). These data are consistent with water attacking as a nucleophile in the enzymatic reaction.     

Purification and characterization of rat-liver cytosolic epoxide hydrolase

Rat liver cytosolic epoxide hydrolase has been purified and characterized. The enzyme was purified from tiadenol-induced rat liver 540-fold with respect to trans-stilbene oxide as a substrate. Similar purification was obtained with the substrates trans-beta-ethyl styrene oxide and styrene 7,8-oxide, the specific activities decreasing in the order trans-beta-ethyl styrene oxide greater than styrene 7,8-oxide greater than trans-stilbene oxide. The enzyme exerts highest activity at pH 7.4 Km and Vmax of the pure enzyme for trans-stilbene oxide were 1.7 microM and 205 nmol x min-1 x mg protein-1 respectively. With trans-stilbene oxide as a substrate, the inhibition by organic solvents (2.5% by vol.) increased in the order ethanol less than methanol less than acetone less than isopropanol = N,N-dimethyl formamide less than acetonitrile less than tetrahydrofuran. The native enzyme, with a molecular mass of 120 kDa, consists of two 61-kDa subunits. Digestion of rat liver cytosolic and microsomal epoxide hydrolase by three proteases resulted in markedly different peptide maps. Western-blot analysis with antiserum against rat liver cytosolic epoxide hydrolase revealed a single band with the purified enzyme, and with liver cytosol from control and clofibrate-induced rats. No cross-reactivity was observed with purified rat microsomal epoxide hydrolase or microsomes. A positive reaction at the same molecular mass was obtained with liver cytosol of mouse, guinea pig, Syrian hamster and New Zealand white rabbit but not with that of green monkey.     

The regulation of cytosolic epoxide hydrolase in mice

The concentration of cytosolic epoxide hydrolase in untreated and clofibrate-treated mouse liver extracts was estimated by immunoblotting. Clofibrate treatment of mice was found to increase liver cytosolic epoxide hydrolase concentration by two fold, showing that the increase in cytosolic epoxide hydrolase in mouse liver after clofibrate treatment is primarily due to induction. The induced and uninduced cytosolic epoxide hydrolase, and epoxide hydrolase in the cytosolic and mitochondrial fractions were compared and found to be identical or very similar. Cytosolic epoxide hydrolases in kidney and liver were similar in molecular weight and antigenic properties.     

The interaction of cytosolic epoxide hydrolase with chiral epoxides

• 1.1. The kinetic parameters of the cytosolic epoxide hydrolase were examined with two sets of spectrophotometric substrates. The (2S,3S)- and (2R,3R)-enantiomers of 4-nitrophenyl trans-2,3-epoxy-3-phenylpropyl carbonate had a K_mof 33 and 68 μm and a V_max of 16 and 27 μmol/min/mg, respectively. With the (2S,3S)- and (2R,3R)- enantiomers of 4-nitrophenyl trans-2,3-epoxy-3-(4-nitrophenyl)propyl carbonate, cytosolic epoxide hydrolase had a K_M of 8.0 and 15 μM and a V_max of 7.8 and 5.0 μmol/min/mg, respectively.
• 2.2. Glycidyl 4-nitrobenzoate had the lowest I₅₀ of the compounds tested in the glycidyl 4-nitrobenzoate series (I₅₀= 140 μM). The I₅₀ of the (2R)-enantiomer was 3.7-fold higher. The inhibitor with the lowest i₅₀ in the glycidol series, and the lowest I₅₀ of any compound tested, was (2S,3S)-3-(4-nitrophenyl)glycidol (I₅₀ = 13.0μM). It also showed the greatest difference in I₅₀ between the enantiomers (330-fold).
• 3.3. All enantiomers of glycidyl 4-nitrobenzoates and _trans-3-phenylglycidols gave differential inhibition of cytosolic epoxide hydrolase. However, neither the (S,S)-/(S)- or (R,R)-/(R)-enantiomer always had the lower I₅₀.
• 4.4. Addition of one or more methyl groups to either enantiomer of glycidyl 4-nitrobenzoate resulted in increased I₅₀. However, addition of a methyl group to C₂ of either enantiomer of 3-phenylglycidol resulted in a decreased I₅₀. Finally, when the hydroxyl group of trans-3-(4-nitrophenyl)glycidol was esterified the I₅₀ of the (2S,3S)- but not the (2R,3R)-enantiomer increased.

     

Regio- and enantioselectivity of soybean fatty acid epoxide hydrolase.

Soluble epoxide hydrolase purified from soybean catalyzes trans-addition of water across the oxirane ring of cis-9,10-epoxystearic acid with inversion of configuration at the attacked carbon, yielding threo-9,10-dihydroxystearic acid. Kinetic analyses of the progress curves, obtained at low substrate concentrations (i.e. [S] much less than Km), and determination of the enantiomeric excess of the residual substrate by chiral-phase high-performance liquid chromatography at different reaction times, indicate that the epoxide hydrolase hydrates preferentially cis-9R, 10S-epoxystearic acid (V/Km ratio, approximately 20). Interestingly, this enantiomer is obtained by epoxidation of oleic acid catalyzed by peroxygenase, a hydroperoxide-dependent oxidase, we have previously described in soybean (Blée, E., and Schuber, F. (1990) J.Biol. Chem. 265, 12887-12894). For the epoxide hydrolase to show high enantioselectivity there must be a free carboxylic acid functionality on the substrate which probably influences its positioning within the active site. This selectivity, which in principle can be used for kinetic resolution of the cis-9,10-epoxystearic acid enantiomers, is much reduced with methyl cis-9,10-epoxystearate. 18O-Labeling experiments indicate that water attacks both cis-9,10-epoxystearic acid enantiomers on the oxirane carbon which has the S-chirality. Results show that soybean epoxide hydrolase produces exclusively threo-9R,10R-dihydroxystearic acid, i.e. a naturally occurring metabolite in higher plants. cis-9,10-Epoxy-18-hydroxystearic acid, a cutin monomer, was a poorer substrate of the epoxide hydrolase than 9,10-epoxystearic acid (V/Km ratio for the preferred enantiomers, approximately 19). From a physiological point of view, peroxygenase and this newly described epoxide hydrolase could be responsible, in vivo, for the biosynthesis of a class of oxygenated fatty acid compounds known to be involved in cutin monomers production and in plant defense mechanisms.     

Uptake of long-chain fatty acid methyl esters by mammalian cells

Albumin-bound long-chain fatty acid methyl esters (ME) were taken up and utilized by Ehrlich ascites tumor cells and slices of rat heart, liver, and kidney. Much more ME than albumin was taken up by the tumor cells, indicating that ME dissociated from the carrier protein during their uptake. 70-80% of the radioactivity associated with the cells after 1 min of incubation at 37 degrees C remained as ME. The results of studies with metabolic inhibitors and glucose suggest that uptake of ME is an energy-independent process. Changes in incubation medium pH between 7.8 and 6.5 did not markedly alter uptake of ME. Cells incubated with FFA and methanol did not synthesize ME. These findings indicate that ME are taken up intact, and they suggest that the presence of an anionic carboxyl group is not essential for the binding of a long-chain aliphatic hydrocarbon to a mammalian cell. When incubation with labeled ME was continued for 1 hr, increasing amounts of radioactivity were recovered in FFA, phospholipids, neutral lipid esters, and CO(2). ME radioactivity associated with the cells after a brief initial incubation was released in the form of ME and FFA when the cells were incubated subsequently in a medium containing albumin. If the second incubation medium contained no albumin, most of the ME radioactivity initially associated with the cells was incorporated into phospholipids, neutral lipid esters, and CO(2). These results suggest that much of the ME which is taken up, is hydrolyzed to FFA, and that the fatty acids derived from ME are available for further metabolism.     

Genetic factors that regulate cytosolic epoxide hydrolase activity in normal human lymphocytes.

To determine whether genetic mechanisms control large variations in cytosolic epoxide hydrolase (cEH) activity of unstimulated lymphocytes from normal human subjects, cEH activity was measured in: a) 6 sets of monozygotic (MZ) twins and 6 sets of dizygotic (DZ) twins; b) 100 unrelated male subjects and c) 6 families. The twin study revealed predominantly genetic control (H2/1 = 0.95). Variability was markedly less within MZ (intrapair variance = 0.25) than DZ twins (intrapair variance = 6.33). In 100 unrelated male subjects the extent of interindividual variation was 11-fold. Unimodal distribution of values among 99 subjects encompassed a six-fold range. One outlier with very high activity clearly stood apart. Using the whole distribution curve we phenotyped members of 6 families. In the outlier's family, analysis of 3 generations suggested autosomal dominant transmission of high cEH activity. Analysis of the other 5 families and 12 sets of twins, all from the large unimodal distribution, was consistent with either monogenic or polygenic control of variations within this mode.     

Enantioselectivity of the hydrolysis of linoleic acid monoepoxides catalyzed by soybean fatty acid epoxide hydrolase.

Soybean epoxide hydrolase efficiently catalyzes the hydration of the two positional isomers of linoleic acid monoepoxides into their corresponding vic-diols. Kinetic analysis of the progress curves, obtained at low substrate concentrations (i.e. [So] much less than Km), and analysis of the residual substrates by chiral-phase HPLC, indicate that the hydrolase is highly enantioselective, i.e. cis-9R,10S-epoxy-12(Z)-octadecenoic and cis-12R,13S-epoxy-9(Z)-octadecenoic acids are preferentially hydrolyzed (the enantioselectivity ratios are 15 and 28, respectively). Importantly, these two enantiomers are the one formed preponderantly by epoxidation of linoleic acid by peroxygenase, a hydroperoxide-dependent oxidase we have previously described in soybean (Blée, E., and Schuber, F., Biochem. Biophys. Res. Commun. (1990) 173, 1354-1360).     

Affinity purification of cytosolic epoxide hydrolase using derivatized epoxy-activated Sepharose gels

Improved affinity chromatography procedures for the purification of cytosolic epoxide hydrolase are described. An earlier affinity purification method using immobilized 7-methoxycitronellyl thiol (MCT) sporadically produced final enzyme preparations containing major impurities. To eliminate these impurities, we tested alternate ligands, spacer arms, and ligand concentrations. A series of alkyl and aryl thiols coupled to epoxy-activated Sepharose were found to exhibit markedly different binding characteristics as compared with commercially available alkyl- and aryl-Sepharose gels. Using one of these new matrices, benzylthio-Sepharose, cytosolic epoxide hydrolase from mouse liver was purified over 100-fold, appeared homogeneous by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, and was obtained with 60-90% recovery of enzyme activity. The impurities previously observed with the MCT-Sepharose procedure were reduced or eliminated by using an MCT ligand concentration of 5 microequivalents per gram or less. MCT-Sepharose and benzylthio-Sepharose provide rapid and convenient one-step procedures for obtaining purified cytosolic epoxide hydrolase from numerous species and tissues.     

Selective inhibition of cytosolic epoxide hydrolase activity in vitro by compounds that inhibit catalase

The ability of a number of known inhibitors of catalase activity to affect cytosolic and microsomal epoxide hydrolase activities in vitro, measured as enzymatic trans-stilbene oxide hydrolysis and styrene oxide hydrolysis, respectively, was investigated. Catalase and cytosolic epoxide hydrolase activities are inhibited by hydroxylated metabolites of 2-amino-4,5-diphenylthiazole (DPT). The metabolite hydroxylated on the 4-phenyl ring (4OH-DPT) and the metabolite hydroxylated on both phenyl rings (4,5-DIOH-DPT) are potent inhibitors of both enzymes; the metabolite hydroxylated on the 5-phenyl ring (5OH-DPT) is less potent. Unmetabolized DPT has no effect on either enzyme. 4OH-DPT inhibits, but 5OH-DPT enhances, microsomal epoxide hydrolase activity. 4,5-DIOH-DPT and DPT have no effect on this enzyme. Other compounds that inhibit both catalase and cytosolic epoxide hydrolase activities, but do not inhibit microsomal epoxide hydrolase activity, are nordihydroguaiaretic acid and 2-aminothiazole. Microsomal epoxide hydrolase activity is enhanced by 2-aminothiazole and levamisole in vitro. Thus these inhibitors of catalase are selective epoxide hydrolase inhibitors in that they inhibit cytosolic epoxide hydrolase activity in vitro, but have either no effect on, or increase the activity of, microsomal epoxide hydrolase in vitro. Conversely, the selective cytosolic epoxide hydrolase inhibitors 4-phenylchalcone oxide and 4'-phenylchalcone oxide do not inhibit catalase activity, nor does trichloropropene oxide, a selective microsomal epoxide hydrolase inhibitor.     

Purification of human liver cytosolic epoxide hydrolase and comparison to the microsomal enzyme

Epoxide hydrolase (EC 3.3.2.3) was purified to electrophoretic homogeneity from human liver cytosol by using hydrolytic activity toward trans-8-ethylstyrene 7,8-oxide (TESO) as an assay. The overall purification was 400-fold. The purified enzyme has an apparent monomeric molecular weight of 58 000, significantly greater than the 50 000 found for human (or rat) liver microsomal epoxide hydrolase or for another TESO-hydrolyzing enzyme also isolated from human liver cytosol. Purified cytosolic TESO hydrolase catalyzes the hydrolysis of cis-8-ethylstyrene 7,8-oxide 10 times more rapidly than does the microsomal enzyme, catalyzes the hydrolysis of TESO and trans-stilbene oxide as rapidly as the microsomal enzyme, but catalyzes the hydrolysis of styrene 7,8-oxide, p-nitrostyrene 7,8-oxide, and naphthalene 1,2-oxide much less effectively than does the microsomal enzyme. Purified cytosolic TESO hydrolase does not hydrolyze benzo[a]pyrene 4,5-oxide, a substrate for the microsomal enzyme. The activities of the purified enzymes can explain the specific activities observed with subcellular fractions. Anti-human liver microsomal epoxide hydrolase did not recognize cytosolic TESO hydrolase in purified form or in cytosol, as judged by double-diffusion immunoprecipitin analysis, precipitation of enzymatic activity, and immunoelectrophoretic techniques. Cytosolic TESO hydrolase and microsomal epoxide hydrolase were also distinguished by peptide mapping. The results provide evidence that physically different forms of epoxide hydrolase exist in different subcellular fractions and can have markedly different substrate specificities.     

Cyclohexene long-chain fatty acid esters from Uvaria dulcis (Dunal)

Two new cyclohexene long-chain fatty acid esters, namely Dulcisenes A and B, were isolated from the twigs of the Uvaria dulcis together with seven known compounds, uvarigranol E, (−)-zeylenol, ellipeiopsol B, 5,7-dihydroxyflavone, 8-hydroxy-5,7-dimethoxyflavanone, lupeol, and benzyl benzoate. The structures of the isolated compounds were elucidated by spectroscopic and mass-spectrometric analyses, including 1D, 2D NMR and HR TOF MS. Several of these metabolites were tested for cytotoxicity against HepG2, A549, S102, HuCCA-1, HeLa, MDA-MB-231, T47D, HL-60, and P388 cell lines.     

Biochemical characterization of a variant form of cytosolic epoxide hydrolase induced by parental exposure to N-ethyl-N-nitrosourea

1. ENU4 mice express a protein variant originally detected in a CBF1 mouse sired by a C57BL/6 mouse exposed to N-ethyl-N-nitrosourea. It appears to be an isolelectric point variant of cytosolic epoxide hydrolase. Affinity purified cytosolic epoxide hydrolase from ENU4 mice has a pI of approximately 5.1 compared to 5.6 in other mouse strains.2. Clofibrate induced cytosolic epoxide hydrolase to similar levels in five strains of mice. However, CBF1 and ENU4 mice were more sensitive to the induction of palmitoyl CoA oxidase activity.3. Except for isoelectric point, the physico- and immunochemical properties of cytosolic epoxide hydrolase from ENU4 mice were similar to those of the other mouse strains. Substrate specificities for five of six substrates tested were also similar.     

Separation of long-chain fatty acid esters of retinol by high-performance liquid chromatography

Individual long-chain fatty acid esters of retinol can be resolved by high-performance liquid chromatography using an octyl- or phenyl-substituted reverse-phase column and mixtures of acetonitrile with water as mobile phase. This simple procedure provides good resolution of biologically important retinyl esters including retinyl palmitate and retinyl oleate. Using an isocratic elution system, it is shown that nine synthetic esters of retinol, ranging in fatty acyl chain length from 12 to 20 carbons, each elute with a unique elution volume and produce an absorbance signal at 340 nm proportional to molar concentration. The method is suitable for analysis of various esters of retinol in biological samples including lymph chylomicrons and blood plasma. The octyl-substituted reverse-phase column can also be used to separate more polar neutral retinoids including retinol and retinaldehyde.     

Leukotriene A4. Enzymatic conversion into 5,6-dihydroxy-7,9,11,14-eicosatetraenoic acid by mouse liver cytosolic epoxide hydrolase

Mouse liver homogenates transformed leukotriene A4 into a 5,6-dihydroxy-7,9,11,14-eicosatetraenoic acid. This novel enzymatic metabolite of leukotriene A4 was characterized by physical means including ultraviolet spectroscopy, high performance liquid chromatography, and gas chromatography-mass spectrometry. After subcellular fractionation, the enzymatic activity was mostly recovered in the 105,000 X g supernatant and 20,000 X g pellet. Heat treatment (80 degrees C, 10 min) or digestion with a proteolytic enzyme abolished the enzymatic activity in the high speed supernatant. A purified cytosolic epoxide hydrolase from mouse liver also transformed leukotriene A4 into a 5,6-dihydroxyeicosatetraenoic acid with the same physico-chemical characteristics as the compound formed in crude cytosol, but not into leukotriene B4, a compound previously reported to be formed in liver cytosol (Haeggstr?m, J., R?dmark, O., and Fitzpatrick, F.A. (1985) Biochim. Biophys. Acta 835, 378-384). These findings suggest a role for leukotriene A4 as an endogenous substrate for cytosolic epoxide hydrolase, an enzyme earlier characterized by xenobiotic substrates. Furthermore, they indicate that leukotriene A4 hydrolase in liver cytosol is a distinct enzyme, separate from previously described forms of epoxide hydrolases in liver.     

Characterization of the epoxide hydrolase from an epichlorohydrin-degrading Pseudomonas sp.

An epoxide hydrolase was purified to homogeneity from the epichlorohydrin-utilizing bacterium Pseudomonas sp. strain AD1. The enzyme was found to be a monomeric protein with a molecular mass of 35 kDa. With epichlorohydrin as the substrate, the enzyme followed Michaelis-Menten kinetics with a Km value of 0.3 mM and a Vmax of 34 mumol.min-1.mg protein-1. The epoxide hydrolase catalyzed the hydrolysis of several epoxides, including epichlorohydrin, epibromohydrin, epoxyoctane and styrene epoxide. With all chiral compounds tested, both stereoisomers were converted. Amino acid sequencing of cyanogen bromide-generated peptides did not yield sequences with similarities to other known proteins.     

Lupeol long-chain fatty acid esters and other triterpenoid constituents from Plumeria obtusifolia

From the bark of Plumeria obtusifolia was isolated a series of lupeol fatty esters with the carbon numbers 16, 18,20 and 21–28 in the fatty acid part. Furthermore, lupeol, lupeol acetate, sitosterol, stigmasterol and campesterol were also identified.     

Immobilization of the epoxide hydrolase from Aspergillus niger

Three methods for the immobilization of the epoxide hydrolase from the fungus Aspergillus niger were tested. The highest immobilization yield (90%) and retention of activity (65%) were obtained by adsorption onto DEAE-cellulose compared to adsorption onto hydrophobic porous polypropylene and covalent linkage using Eupergit resin. The enzymatic properties of the immobilized enzyme were similar to those of the free enzyme with respect to the effect of temperature and pH on both activity and stability as well as the effect of solvent (DMF) on activity. The kinetic parameters were affected leading to lower K _M(app) and higher Vm _(app).