Chem. Res. Toxicol. 1994, 7, 191-198
191
S-Oxidative Cleavage of Farnesylcysteine and Farnesylcysteine Methyl Ester by the Flavin-Containing Monooxygenase Sang B. Park,? William N. Howald,t and John R. Cashman*ft Department of Pharmaceutical Chemistry, University of California, Sun Francisco, California 94143, and Department of Medicinal Chemistry, University of Washington, Seattle, Washington 98195 Received September 16, 199P
Posttranslational modification of proteins with a farnesyl or geranylgeranyl group appears to be crucial in the signal transduction of eukaryotic cells. For example, farnesylation of rusencoded proteins is a key process that apparently leads to membrane association of proteins that perform a function in cell growth-promoting activity. Although it has been suggested that prenylation of proteins may be an important regulatory mechanism, little is known about the mechanism whereby prenylated proteins are removed from the membrane. In our previous report [(1992) Chem. Res. Toxicol. 5, 193-2011, we showed that S-alkenylated cysteines and mercapturates of xenobiotics were S-oxygenated by the flavin-containing monooxygenase. The S-oxides were not indefinitely stable and rearranged or underwent elimination reactions that cleaved the C-S(0) bond. As a model for farnesylated proteins and peptides, the biotransformation of farnesylcysteine methyl ester was examined in the presence of pig liver microsomes. Two prominent products were formed: farnesyl methyl ester sulfoxide and farnesylcysteine, arising from the action of the flavin-containing monooxygenase and esterase of pig liver, respectively. Formation of farnesylcysteine methyl ester sulfoxide by the flavin-containing monooxygenase was stereoselective (i.e., 71.5% :28.5%, major to minor diastereomer) in good agreement with previously reported stereoselectivity studies of other related S-alkylcysteinecontaining compounds. That the stereoselectivity observed was due to S-oxygenation of the sulfur atom was verified in parallel chemical oxidation studies by using micellar electrokinetic capillary chromatography. Farnesylcysteine methyl ester was an excellent substrate for the flavin-containing monooxygenase, and the S-oxide product was confirmed by HPLC electrospray mass spectrometry. Also, farnesylcysteine was efficiently S-oxygenated in the presence of pig liver microsomes in a flavin-containing monooxygenase-dependent process. Farnesylcysteine methyl ester S-oxide was not indefinitely stable and decomposed presumably via an elimination reaction to produce carbon-sulfur bond cleavage products. It is possible that S-oxidative cleavage of farnesylcysteine methyl ester derivatives may have important implications for predicting regulation and homeostasis of prenylated proteins.
Introduction The posttranslational modification of proteins with isoprene residues such as a farnesyl or a geranylgeranyl group appears to be crucial in the signal transduction of eukaryotic cells (3-5). For example, farnesylation of rasencoded proteins is a key process that leads to membrane association of proteins that perform a function in cell growth-promotingactivity (5-8). Farnesylation of proteins is distinguishedfrom geranylgeranylationby the specificity of proteins that are prenylated (9,IO). In nature, proteins that are prenylated with the 20-carbon geranylgeranyl group constitute the predominant prenylated protein. The 1Bcarbon farnesyl group has been found as a posttranslational modification of fewer proteins including the rasencoded proteins (61,nuclear lamins (II), and the y-subunit of transducin (12, 13). Farnesylation andlor palmitoylation of the ras-encoded proteins appears to be a *Addre= correspondence to this author at the IGEN Research Institute, 130 6th Ave. N., Seattle, WA 98109. Telephone: (206) 44166&1;FAX: (206) 443-0685. t University of California. 8 University of Washington. 0 Abstract published in Advance ACS Abstracts, February 1, 1994.
specific prerequisite for oncogenic activity (9,14,15). The ras-encoded proteins prenylated with a geranylgeranyl group instead of a farnesyl group yielded only nononcogenic forms of ras proteins. The biochemical basis for the prenylation and membrane association of proteins is becoming more clearly understood. The isoprenoidmoiety has been shown to be covalently attached to the cysteine of the C-terminal tetrapeptide CAAX' by the action of farnesyltransferase via a cysteinyl thioether linkage (16). Next, the prenylated protein is hydrolyzed by a microsomal peptidase and then carboxymethylated by a microsomal methyltransferase to yield the final form of the prenylated protein containing a prenylated cysteine methyl ester moiety at the C-terminus (17-19). Isoprenylation of proteins appears to lead to proper protein localization and association with membranes. 1 Abbreviations: CAAX, cysteine-alanine-alanineother amino acids; FMO, flavin-containingmonooxygenase; FCM, famesylcysteine methyl ester;FCMSO,farnesylcysteinemethyl ester S-oxide;FC, farnesylcysteine; FCSO, farnesylcysteine S-oxide; EI, electron impact; CI, chemical ionization; LSIMS, liquid secondary ion masa spectrometry; ESI, electrospray ionization; DETAPAC, diethylenetriaminepentaaceticacid; FABMS,fast atom bombardment mass spectrometry.
0893-228~/94/2707-0191$04.50/00 1994 American Chemical Society
192 Chem. Res. Toxicol., Vol. 7, No. 2, 1994 R=
H
3 c CH,
H
e C CH3
H 2 CH3
-
O H R - 5 A C O O H
R-S,,+COOH NH2
FC
Park et al.
NH2
FCSO
Figure 1. Structures of S-tram,tram-farnesylated cysteine methyl ester (FCM) and the corresponding sulfoxide (FCMSO).
Even though it has been suggested that the prenylation of proteins may be an important regulatory mechanism in the signal transducing machinery by cycling the protein on and off the membrane, little is known about the dissociation mechanism of proteins from the membrane and the fate of prenylated proteins (4, 20). Chemically, the carbon-sulfur bond of farnesylated cysteine and cysteine-containing peptides may be cleaved through the formation of a sulfonium ion by treatment with methyl iodide (19,21).It is possible that S-oxidative biotransformation of the sulfur atom of farnesylated cysteine to form a sulfoxide could be the biological equivalent of formation of a methylsulfonium ion. In principle, sulfoxidation could lead to the formation of an unstable S-oxide and result in cleavage of the isoprenoid moiety from the protein or peptide. S-Oxidative cleavage of prenylated proteins could provide a mechanism to turn off the signal mediated by the prenylated proteins by leading to the dissociation of the protein from the membrane. Because the major microsomal monooxygenases that catalyze the formation of aliphatic sulfoxides are cytochromes P450 and the flavin-containing monooxygenase5 (FMO) that are found mainly in the rough endoplasmic reticulum where other processing (i.e., proteolysis and methylation) of prenylated proteins has been observed, we examined hepatic microsomes for S-oxygenase activity of S-farnesylated cysteine methyl ester as a model for more complex S-farnesylated cysteine-containing peptides. In our previous report, we showed that S-alkenylated cysteines and mercapturates were S-oxygenatedby FM012 and not by hepatic cytochromes P450 (22). S-Alkenylated cysteines and mercapturates arise in vivo from a variety of haloalkenes via glutathione S-conjugate formation, followed by metabolism to cysteine S-conjugates or mercapturates (23). Once formed, cysteine S-conjugates or mercapturates may be sulfoxidized in vivo (24). The S-alkenylated cysteine sulfoxides and mercapturate sulfoxides are not indefinitely stable and may undergo spontaneous rearrangements and/or elimination reactions (22). With this information in mind, we chose to investigate the chemical stability of the C-terminal amino acid of prenylated proteins because the C-terminal amino acid ZFMOl has also been designated FMO-A, and FM03 has been designated FMO-D(or form II), as described in our previous publications (Le., refs 1 and 2).
shares several common chemical features with the S-alkenylated conjugates or mercapturates previously studied in our laboratory. Both classes of compounds contain a cysteinyl backbone, an alkenylgroup attached to the sulfur atom, and an amide bond with a carbonyl group (Figure 1). We hypothesized that if farnesylcysteine methyl ester (FCM) was a substrate for a microsomal monooxygenase, the incipient sulfoxide would be unstable, which would lead to carbon-sulfur bond cleavage. The objective of this study was to determine the role of monooxygenases in the S-oxidative modification of farnesylcysteine methyl ester as a model for prenylated peptides and proteins. For this purpose, we synthesized and examined S-trans,trans-farnesylcysteinyl methyl ester (FCM) as asubstratefor S-oxidation and degradation using hepatic preparations. The results reported herein show that FCM and FC were efficiently S-oxygenated by pig liver FMO1, but not by pig liver microsomal cytochromes P450. FCM is also a substrate for pig liver esterase. The sulfoxides were not indefinitely stable and decomposed to several products under the mild conditions employed.
Experimental Procedures Chemicalsused in this study were of the highestpurity available from commercial sources. tram,tram-Famesyl bromide,cysteine methyl ester, cysteine, and hydrogen peroxide were purchased from Aldrich Chemical Co. (Milwaukee, WI). All of the compounds of the NADPH-generating system were obtained from Sigma Chemical Co. (St. Louis, MO). Aminobenzotriazolewas a gift of Professor Paul Ortiz de Montellano, University of California, San Francisco. All other reagents and buffers were of the highest purity available and were obtained from Fisher Chemical Co. (San Jose, CA). Instrumental Analysis. 'H-NMR spectra were recorded with a VXR or a General Electric spectrometer operating at 300 MHz. 'H-NMR chemical shift values are expressed in ppm downfield from tetramethylsilane. Electron impact (EI) and chemical ionization(CI) mass spectra were recorded with a VG70S spectrometer at 6 kV and a source temperature of 80 OC. Liquid secondaryion mass spectra (LSIMS) were recorded with a Kratos MS 50 spectrometer at 8 kV equipped with a cesium ion gun at the University of California, San Francisco. Electrospray ionization (ESI) mass spectrometry was done in the batch mode with a VG BioQmass spectrometer (VG Biotech, Cheshire,U.K.) at the University of California, San Francisco. Liquid chromatography/electrosprayionization mass spectrometry was carried out on a VG Trio 2000 quadrupole mass spectrometer (VG Biotech, Cheshire, U.K.) at the University of Washington equipped with an electrosprayionization source using nitrogen as a nebulizing gas (mass range 3000 daltons). Liquid chromatography experiments were performed on a LKB 2150 HPLC utilizing a LKB 2152 controller unit (LKB, Bromma, Sweden) fitted with a Valco C6W manual injector (Valco Instruments, Houston, TX). The chromatogram was developed at a flow rate of 200 pL/min using an eluent of water/acetonitrile/ ammonium hydroxide (20800.01 v/v). Analysis was monitored at 215 nm utilizing a Dynamax Microsorb 5-pm C-8 analytical reverse-phase column (2.0 mm X 25 cm, Rainin Corp., Alameda, CA). The liquid chromatography/mass spectrometer eluent was split 7:l for quantification. The instrument was controlled and data were acquired with a 80486 DX50 CPU driven by VG MassLynx Windows software. Capillary electrophoresis was done with a Beckman Model 2100 system (Beckman Inc., Palo Alto, CA) with the cathode on the injection side and the anode on the detection side. Six nanoliters of asample solution containing0.25 mg/mL of FCMSO (i.e., 1.5 pg) was introduced into a 50 pm X 27 cm silica capillary by vacuum injection. The electrophoresis was done with a run
S-Oxygenation of Farnesylcysteine Methyl Ester by FMO
Chem.Res. Toxicol., Vol. 7,No. 2, 1994 193
vinyl H), 6 5.35 (m, lH, vinyl H); FABMS (MH+ calcd 342.20) found MH+ 342.0. Measurement of the Decomposition Kinetics of FCM and FCMSO. Decomposition of FCM and FCMSO was measured by HPLC, and rates were determined on the basis of the disappearance of FCM and FCMSO. Stock solutions of FCM and FCMSO were prepared in dry methanol immediately before use, and 0.5 mL of the methanolic stock solutions was added to 0.5 mL of 50 mM potassium phosphate solution suoh that the final concentration of FCM or FCMSO was 400 pg/mL. The reactions were stirred at 25 "C or at 37 "C for at least 4 half-lives. The pH and ionic strength were held constant, and the buffer employed at pH 4.0, pH 6.0, pH 7.2, pH 8.4, and pH 9.6 was potassium phosphate. The reactions were monitored by HPLC as described below. Enzyme Preparationsand Assays. The incubationmedium for assays with highly purified FMO and microsomes from pig liver contained 0.05 M potassium phosphate buffer (pH = 8.4), 0.75 mM diethylenetriaminepentaacetic acid (DETAPAC), 0.4 mM NADP+, 0.4 mM glucose 6-phosphate, and 1IU of glucoseJ=4.8Hz,7.8Hz,C,H),63.70(s,3H,OCH3),65.05(m,2H,vinyl6-phosphatedehydrogenase. In the incubationsemployinghighly H), 6 5.19 (t, lH, J = 7.5 Hz, vinyl H); CIMS (calcd 339.2, Cl@H33purified FMO1, 10-50 pg of protein was used, and for the N02S)found 340.3 (MH+,loo), 205.2 (C15HzB,l2);ESIMS,found incubations with the microsomes, 0.2-0.5 mg of protein was used 339.5 (M+, 100). in a total volume of 0.25 mL. The reaction was initiated by the addition of substrate to an ice-cold solution of the incubation Synthesis of S-trans,trans-FarnesylcysteineMethyl mixture and incubated at 37 OC with constant shaking. The Ester S-Oxide (FCMSO).S-trans,trans-Farnesyl~steinemethreaction was stopped by the addition of 0.7 mL of cold yl ester (34 mg, 0.1 mmol) was dissolved in 1.7 mL of methanol. dichloromethane. The mixture was vigorously mixed, and the Hydrogen peroxide (30% w/v, 15pL, 0.13 mmol) was added, and organic layer was separated from the aqueous layer by a brief the mixture was stirred for 4 h at 25 OC. The reaction was centrifugation. After filtration through a 0.45-pm nylon filter quenched with 10 pL of dimethyl sulfoxide. The methanol was and evaporation, the extract was taken up in methanol for HPLC evaporated, and the crude product was partitioned between water analysis. The recovery of FCM and FCMSO from metabolic and dichloromethane. Evaporation of the dichloromethane incubation mixtures was greater than 93 % and 76 % ,respectively, fraction yielded S-trans,trans-farnesylcysteinemethyl ester as judged by HPLC. S-oxide in quantitative yield with greater than 95% purity as judged by HPLC. 1H-NMR (in CDC13) 6 1.58 (s,6H, vinyl CHa), The incubation medium for assays with highly purified pig 6 1.66 (s,3H, vinyl CHs), 6 1.74 (s,3H, vinyl CH3,6 1.90-2.15 (m, liver esterase (obtained from Sigma Chemical Co.) contained 8H, vinyl CH2), 6 2.75-2.96 (m, lH, C.$H), 6 3.04-3.18 (m, lH, 0.05 M potassium phosphate buffer (pH = 8.4), 0.75 mM CflzH),6 3.45-3.66 (m, 2H, SCHzC=C), 6 3.75 (s, 3H, OCH3), DETAPAC, and 8.7 pg of esterase in a total volume of 0.25 mL. 4.05-4.12 (m, lH, C,H), 6 5.05 (m, 2H, vinyl H), 5.19 (t, lH, J The pig liver esterase used had linoleic acid methyl esterase = 6Hz, vinyl H); ESIMS (calcd 355.2, C ~ ~ H S ~ Nfound O ~ S 356.1 ) activity of 345 nmol min-' (mg of protein)-' as determined by (MH+, loo), 237.8 (CisHzsS, 27). HPLC. For hydrolysis of FCM, the reaction was initiated by the addition of substrate to an ice-cold solution of the enzyme mixture Synthesis of S-trans,trans-Farnesylcysteine(FC). Cysand incubated at 37 OC with constant shaking for 10 min. The teine (127 mg, 1.0 mmol) was dissolved in 2 M ammonia in reaction was stopped by the addition of 0.8 mL of cold methanol (8 mL). Farnesyl bromide (300 mg, 1.0 mmol) was dichloromethane/2-propanol( 2 1 v/v). The mixture was thoradded dropwise to the reaction mixture at 0 "C. The mixture oughly mixed, and the organic fraction was separated from the was allowed to stir at 0 OC for 3 h. The solvent was evaporated aqueous layer by a brief centrifugation. After filtration through at reduced pressure, and the residue was partitioned between a 0.45-pm nylon filter and evaporation, the extract was taken up water and dichloromethane. The organic fraction was collected in methanol for HPLC analysis. The recovery of FC from and dried. The residue was dissolved in methanol (10 mL), and metabolic incubation mixtures was greater than 81% as judged the methanolicsolutionwas washed twice with 1volume of hexane by HPLC. and dried. The purity of the product was greater than 95% as The incubation of FC with pig liver microsomes was similar judged by HPLC: yield 314 mg (92%);'H-NMR (in CD30D) 6 1.60 (8, 6H, vinyl CHs), 6 1.67 (s, 6H, vinyl CHs), 6 1.72 (s, 3H, to the procedure described above for FCM. The reaction was vinyl CH3), 6 1.92-2.20 (m, 8H, vinyl CHz), 6 2.83 (dd, lH, J = stopped by the addition of 1 mL of cold dichloromethane/28.6 Hz, 14.5 Hz, Co'H), 6 3.18 (dd, lH, J = 3.6 Hz, 14.5 Hz, Co'H), propanol (2:l v/v). The mixture was thoroughly mixed, and the 6 3.30 (m, 2H, C,H2), 6 3.68 (dd, lH, J = 3.6 Hz, 8.6 Hz, C,H), organic fraction was separated from the aqueous layer by a brief 6 5.12 (m, 2H, vinyl H); 6 5.28 (t, lH, J = 6.75 Hz, vinyl H); centrifugation. After filtration through a 0.45-pm nylon filter FABMS (MH+ calcd 326.20) found MH+ 326.0. and evaporation, the extract was taken up in methanol for HPLC analysis. The recovery of FC and FCSO from metabolic Synthesis of S-trans,trans-FarnesylcysteineS-Oxide incubation mixtures was greater than 80% as judged by HPLC. (FCSO). S-trans,trans-Farnesylcysteine(5mg, 0.015mmol) was HPLC Analytical Methods. FCM and FCMSO were anadissolved in 2 mL of methanol. Hydrogen peroxide (30% w/v, lyzed by either a normal-phase or a reverse-phase HPLC system. 3 p L , 0.026 mmol) was added, and the mixture was stirred for 24 The reverse-phase HPLC analysis was performed on a Dynamax h at 25 "C. The reaction was quenched with 5 pL of dimethyl Microsorb C-18 (7.7 mm X 25 cm) analyticalreverse-phasecolumn sulfoxide. The methanol was evaporated, and the crude product was partitioned between water and dichloromethane. Evapoand quantified by a Beckman Model lOOA HPLC system interfaced to an HP-85B computer with an HP 1040A dioderation of the dichloromethane extract yielded S-trans,transfarnesylcysteineS-oxide with greater than 95 5% purity as judged array detector at 220 nm. The mobile phase consisted of an isocratic system of water/acetonitrile/ammonium hydroxide (20 by HPLC. 'H-NMR (in CD30D) 6 1.60 (s, 3H, vinyl CHs), 6 1.62 80:O.Ol v/v). The normal-phase HPLC analysis was performed (8, 3H, vinyl CHs), 6 1.67 (s, 3H, vinyl CH3), 6 1.78 (8, 3H, vinyl on an AXXIOM silica column (4.5 mm X 25 cm, Richard CH3), 6 1.87-2.20 (m, 8H, vinyl CH2) 6 3.04 (dd, lH, J = 9.2 Hz, 13.9 Hz, CB'H), 6 3.43 (dd, lH, J = 4.4 Hz, 13.9 Hz, CBH), 6 3.70 Scientific, Novato, CA) and quantified by an IBM Model 9533 HPLC system interfaced to an HP Model 3396 integrator with (d, 2H, J = 7.1 Hz, C,H*-S), 6 4.06 (m, lH, C,H), 6 5.12 (m, 2H, buffer that consisted of 100 mM borate (pH = 9.1) containing 100 mM sodium dodecyl sulfate saturated with j3-cyclodextrin. The eluent was monitored on-column by UV detection at 200 nm. The temperature of the column was maintained at 30 f 1 "C. The electrophoresiswas carried out under a constant voltage of 23 kV generating a current of approximately 90 pA. Synthesis of 9-trans,trans-FarnesylcysteineMethyl Ester (FCM). Cysteine methyl ester (142 mg, 1.0 mmol) was dissolvedin an anhydrous 2 M solution of ammonia in methanol (5 mL). Farnesyl bromide (0.28 mL, 1.0 mmol) was gradually added to the reaction mixture at 0 "C. The mixture was allowed to stir at 0 "C for 3 h. The solvent was evaporated at reduced pressure, and the residue was partitioned between dichloromethane and water. The dichloromethane layer was dried, and the crude product was chromatographed on a preparative silica gel TLC plate (Rf = 0.57,0.1%triethylamine/ethylacetate):yield 238 mg (67% ); 'H-NMR (in CDCl3) 6 1.56 (s, 9H, vinyl CHs), 6 1.63 (s,3H, vinyl CH3), 6 1.80-2.07 (m, 8H, vinyl CHz), 6 2.64 (dd, lH, J = 7.8 Hz, 13.5 Hz, Co'H), 6 2.84 (dd, lH, J = 4.8 Hz, 13.5 Hz, CpzH), 6 3.05-3.20 (m, 2H, SCH&=C), 6 3.60 (dd, lH,
194 Chem. Res. Toricol., Vol. 7,No. 2, 1994 an LC/9523 variable UV detector set at 220 nm (25).The mobile phase consisted of an isocratic system of hexane/2-propanol/ methanol/perchloricacid (80155:0.02 v/v). The retention volumes were 6.7 mL for FCM and 9.9 mL and 10.3 mL for the diastereomers of FCMSO which were separated on a normalphase silica gel column, and 9.8 mL for FC, 10.5 mL for FCMSO, and 13.7 mL for FCM which were separated on a C-18 column. For the hydrolysis reactions, FC and FCM were analyzed by a reverse-phase HPLC system. The reverse-phaseHPLC analysis was done with a Dynamax Microsorb C-18 analytical column as described above and quantified by an IBM Model 9533 HPLC interfaced to a HP Model 3396 integrator with an IBM LC/9523 variable UV detector set at 220 nm. The mobile phase consisted of an isocratic system of water/acetonitrile/trifluoroaceticacid (20800.02 v/v). The retention volume was 8.8 mL for FC, 6.5 mL for FCSO, and 12.3 mL for FCM. For incubationsof FC in the presence of pig liver microsomes, FC and FCSO were analyzed by the reverse-phase HPLC system described above. The mobile phase consisted of an isocratic system of water/acetonitrile/trifluoroaceticacid (20800.02 v/v). The retention volume was 8.5 mL for FC and 6.0 mL for FCSO.
Results Chemistry. FCM was synthesized with minor modifications by the method of Brown et al. (26). Treatment of FCM with hydrogen peroxide gave FCMSO in essentially quantitative yield. The aqueous hydrogen peroxide reaction yielded a racemic mixture of diastereomeric sul, / foxides. The purity of the product of the reaction of FCM with hydrogen peroxide was greater than 95% as judged by HPLC. No attempt was made to further purify FCMSO because the sulfoxide apparently decomposed to some extent during subsequent silica gel chromatographic purification procedures. The 'H-NMR spectrum of FCM was identical to the reported spectrum. In the lH-NMR spectrum of FCMSO, the proton resonances attached to the carbons adjacent to the sulfur atom were shifted significantly downfield due to the deshielding effect of the sulfoxide. The pattern of the other FCMSO proton resonances was similar to that of FCM. Obtaining a molecular ion by mass spectrometric techniques was difficult to achieve due to the lability of FCM and FCMSO, Several mass spectrometric techniques were attempted to obtain the exact mass of FCM. Of the mass spectrometry methods we used, only soft ionization methods worked for FCM and FCMSO. Chemical ionization and electrospray mass spectrometry gave the requisite molecular ion for FCM. In contrast, only electrospray and fast atom bombardment mass spectrometry gave spectra containing the molecular ion for FCMSO and FCSO, respectively. Other mass spectrometry techniques including EI, CI, direct CI, and LSIMS failed to give a molecular ion for FCMSO. Because we expected that the sulfonium ion characteristics of the sulfoxide of FCM would modify the stability of the compound, we measured the decomposition rates under various experimental conditions. The decomposition of FCMSO was dependent on the temperature and pH of the reaction (Table 1). The decomposition of FCMSO was faster under conditions of higher temperature or greater pH. The decomposition of FCMSO was also linearly dependent upon the hydroxide ion concentration (Figure 2)(i.e., correlation coefficients were 0.937at 25 OC and 0.971 at 37 OC for the decomposition reactions). FCMSO decomposed as much as 10-fold faster than FCM under identical reaction conditions (i.e., 22 X lO-"s-'versus 240 X lO-'s-l) (Table 1). HPLC analysis showed that
Park et al. Table 1. Kinetics of FCM aad FCMSO Decomposition at Various PES. substrate temp ("C) PH k x 107 (5-1) FCM 25 9.6 22 21 8.4 7.2 11 6.0 11 4.0 9 FCM 37 9.6 107 8.4 60 7.2 33 6.0 26 4.0 13 FCMSO 25 9.6 241 8.4 107 7.2 55 6.0 48 4.0 34 FCMSO 37 9.6 664 8.4 507 7.2 226 6.0 203 4.0 120 The reaction was followed by HPLC under pseudo-first-order reaction conditions as described in the Experimental Procedures.
-91
-1 0
=
-=
:-ll!
9'
A
/
-1 2 -1 3
3 4 5 6 7 8 9 1 0
PH Figure 2. Dependence on pH of the decomposition of FCMSO at 25 "C (0) and 37 O C (A).
numerous products were formed from the decomposition of either FCM or FCMSO. However, during the FCM decomposition studies, no detectable amount of FCMSO was formed as judged by HPLC analysis. The results suggested that FCM was sufficiently stable to autoxidation during the short time course of our in vitro metabolic studies to allow us to perform enzyme studies. The HPLC analysis of both the FCM and FCMSO decomposition reactions was performed primarily using a silica gel normal-phase HPLC system. In some cases, a reverse-phase C-18HPLC column was also used to separate the HPLC peaks corresponding to FCMand FCMSO from the several products that arose from the decomposition reaction mixtures. The kinetic constants determined for product formation were essentially the same as those for loss of starting materials. However, we have reported only kinetic constants for loss of starting materials. As anticipated from batch electrospray mms spectrometry experiments, the HPLC electrospray mass spectra of FCM showed an intense molecular ion a t mlz 340 (100% ) and a minor ion at mlz 326 (31%). HPLC electrospray mass spectra of FCMSO showed an intense molecular ion at mlz 356 (100%) and a minor ion at mlz 312 (32%). To confirm the chemical identity of the products arising from aqueous decomposition of FCMSO (Table 1and Figure 2),the major products from dichloromethane/2-propanol extracts of decomposition reactions were analyzed by inline HPLC electrospray mass spectrometry. FCSO, with a characteristic molecular ion of mlz 342,was one of the products observed. In addition, a very nonpolar material
S - Oxygenation of Farnesylcysteine Methyl Ester by FMO
Chem. Res. Toxicol., Vol. 7, No. 2, 1994 195
Table 2. Effects of Various Conditions and Inhibitors on the Metabolism of FCM and FC in the Presence of Pig Liver Microsomes metabolite formation [nmol min-1 (mg of protein)-'] description FC FCMSO FCSO' 25.6 f 8.9 7.1 f 2.9 3.9 f 0.6 complete' >0.2 25.7 f 8.7
