Abstract
Fifteen xanthates with carbon chains of different lengths or substitutions, including the antiviral compound D609 (O-tricyclo[5.2.1.02,6]dec-9-yl-dithiocarbonate), were tested for their ability to inactivate cytochromes P-450 (P-450s) 2B1 and 2B6. All of the xanthates tested were found to inactivate P-450 2B1 in a time- and concentration-dependent manner. The rates of inactivation at 30°C ranged from 0.22 min−1 to 0.02 min−1. The concentrations required for half-maximal inactivation were between 2.4 and 69 μM. A general trend in the inactivation kinetics could be observed with an increasing chain length of the xanthates. Longer carbon chains resulted in slower rates of inactivation with longer half-times of inactivation and higher partition ratios. For P-450 2B1, the most effective inactivators were xanthates with substitutions of intermediate length. The best inactivator for P-450 2B1 was the C8 xanthate, with an inactivation potency (KI) of 2.4 μM, a rate of inactivation of 0.07 min−1, and a partition ratio of 4. Four xanthates were further examined for their effect on the 7-ethoxy-4-(trifluoromethyl)coumarin activity of P-450 2B6. The C8 xanthate was again the most effective inactivator, with aKI of 1 μM. Although theKI values were generally lower than those found with P-450 2B1, the rates of inactivation for P-450 2B6 with the various xanthates were 3- to 5-fold slower. In addition, the isozyme selectivity of xanthates was tested with P-450s 2E1, 1A1, 3A2, 3A4, 2C9, and 2D6. P-450 2E1 was inactivated by xanthates at concentrations 15- to 100-fold higher than those required to inactivate either P-450 2B1 or 2B6. P-450 1A1 was not inactivated by xanthates. However, all of the xanthates tested were able to inhibit the enzymatic activity of P-450 1A1 to a different extent, depending on the length of the xanthate carbon chain. Virtually no inactivation of P-450s 2D6 or 2C9 was seen, except that C8 and D609 were inhibitory at high concentrations (0.2–0.6 mM). None of the xanthates studied had any effect on the activities of P-450s 3A2 or 3A4.
Liver microsomal cytochromes P-450 (P-450s1; for nomenclature, see Nelson et al., 1996) are involved in the metabolism and detoxification of numerous drugs and carcinogens (Porter and Coon, 1991). The physiological substrates for many of the P-450 enzymes have not been identified but include compounds such as prostaglandins, steroids, retinoids, and fatty acids. P-450 enzymes are catalytically diverse and are able to use a wide variety of structurally distinct substrates (Porter and Coon, 1991). Studies with mechanism-based inactivators have been a valuable approach to elucidating P-450 substrate specificity and the structure of the active site (Ortiz de Montellano and Correia, 1983). Evaluating the mechanism of action of different P-450 isoforms is important to the understanding of toxicity or therapeutic failure caused by drug-drug interactions in pharmacotherapy and also for the future development of more selective and safer drugs. A number of the known inhibitors of the different P-450 isoforms are mechanism-based inactivators, such as l0-undecynoic (P-450 1A1; Gan et al., 1984), 1-ethynylpyrene (P-450 1A2; Yun et al., 1992), 9-ethynylphenanthrene (P-450 2B1; Roberts et al., 1995), diethyldithiocarbamate (P-450 2A6 and P-450 2E1; Hickman et al., 1998), orphenadrine (P-450 2B6; Heyn et al., 1996), sulfaphenazole (P-450 2C9; Hickman et al., 1998), tranylcypromine (P-450 2C19;Wrighton et al., 1993), quinidine (P-450 2D6; Hickman et al., 1998), and ketoconazole and troleandomycin (P-450 3A4; Hickman et al., 1998).
Various derivatives of dithiocarbonic acid known as xanthates have been found to be specific mechanism-based inactivators of P-450 in rat liver microsomes (Yanev et al., 1985) and in the reconstituted system (Kent et al., 1999). Ethylxanthate was found to inhibit CCl4 (Stoytchev et al., 1987) and pentobarbital metabolism (Yanev et al., 1982), as well as to protect rats from nitrosodiethylamine-induced liver carcinogenesis (Hadjiolov et al., 1988). The effect of xanthates on the liver metabolism of different xenobiotics occurred through a P-450-dependent formation of reactive metabolites. Our previous studies have shown that a phenobarbital-inducible isozyme(s) was mainly involved in the metabolic activation of xanthates (Yanev et al., 1982).
Extensive studies over the past 10 years have focused on other important biological properties of xanthate derivatives. Certain members of the xanthate family were shown to be potent antiviral (Sauer et al., 1984) and antitumor agents (Schick et al., 1989). It was postulated that xanthates exert these effects by their ability to selectively inhibit phospholipase C (Kiss and Tomono, 1995), to stimulate tumor necrosis factor release (Amtmann and Sauer, 1990), and to inhibit angiogenesis (Maragoudakis et al., 1990). Our observations with xanthates in the P-450 system indicate an additional important role for xanthates in drug metabolism. Studying the metabolism of xanthates and their interactions with the P-450 system in more detail could be very valuable for the success of their future therapeutic application.
In the present study, the inhibitory and inactivation potencies of potassium salts of different alkyl and aryl derivatives of xanthic acid on the enzymatic activities of various rat and human P-450s were tested.
Experimental Procedures
Materials.
Dilauroyl-l-α-phosphatidylcholine (DLPC), BSA, NADPH, and erythromycin were purchased from Sigma Chemical Co. (St. Louis, MO). 7-Ethoxy-4-(trifluoromethyl)coumarin (7-EFC) was obtained from Molecular Probes, Inc. (Eugene, OR), and 7-hydroxy-4-(trifluoromethyl)coumarin (HFC) was obtained from Enzyme Systems Products (Dublin, CA).O-tricyclo[5.2.1.02,6]dec-9-yl-dithiocarbonate (D609) was purchased from Sigma Chemical Co. Microsomes from human lymphoblastoid AHH-1 cells expressing P-450 2B6 were obtained from Gentest Corp. (Woburn, MA). Various xanthates (ROCS2K, where R corresponds to alkyl groups with C2, C3, C3iso, C3allyl, C4, C5, C8, C10, C12, C14, C16, C18, and C20 carbon atoms or a cyclohexene ring; for example, C2 = ethylxanthate = CH2CH2COS2−K+) were synthesized as previously described by Rao (1971). Purified rat liver P-450 1A1 was a kind gift from Dr. Andrew Parkinson (Kansas Univ. Med. Ctr., Kansas City, KS), and purified human liver P-450 3A4 was a gift from Dr. Kan He (University of Michigan, Ann Arbor, MI).Escherichia coli cells cotransformed with plasmids for reductase and P-450 2C9 or P-450 2D6 were obtained from Dr. Thomas Woolf (Warner-Lambert/Parke-Davis, Ann Arbor, MI).
Purification of P-450 and Reductase.
P-450 2B1 and reductase were purified from liver microsomes prepared from male Long-Evans rats (150–175 g) that were given 0.1% phenobarbital in their drinking water for 12 days as described by Saito and Strobel (1981) and Strobel and Dignam (1978), respectively. Rat NADPH-P-450 reductase was also purified after expression in bacteria as described previously (Hanna et al., 1998). P-450 2E1 was expressed in bacteria and purified as previously described (Larson et al., 1991;Kent et al., 1998). P-450 3A2 microsomes were obtained from the livers of rats treated with i.p. injections of 50 mg/kg pregnenolone-16-α-carbonitrile in corn oil for 3 days. Bacterial membranes coexpressing reductase and either P-450 2C9 or P-450 2D6 were prepared as described by Gillam et al. (1993).
Enzyme Reconstitution.
Purified P-450s 2B1, 1A1, and 2E1 were reconstituted in combination with reductase and lipid for 45 min at 4°C. The reconstituted system consisted of either 0.5 μM P-450 2B1, 0.5 μM P-450 1A1 and 0.5 μM reductase, or 1.5 μM P-450 2E1 and 1 μM reductase along with 200 μg of DLPC/ml and 110 U/ml catalase in 50 mM potassium phosphate buffer (pH 7.4).
P-450 3A4 (0.5 μM) was reconstituted with 20 μg of a 1:1:1 mixture of DLCP, dioleyl-l-α-sn-glycero-3-phosphocholine, and phosphatidylserine, 200 μg/ml cholic acid, 1 μM reductase, 0.5 μM cytochrome b5, 500 U catalase/ml, 2 μM glutathione, 30 mM MgCl2, 0.5 mM EDTA, and 20% glycerol in 50 mM HEPES buffer (pH 7.5) for 15 min at room temperature.
P-450 2B6 in microsomes from AHH-1 cells was diluted to 0.6 nmol of P-450/ml with 50 mM potassium phosphate buffer (pH 7.4).
Bacterial membranes containing coexpressed reductase and P-450 2D6 were diluted to 1.1 nmol of P-450/ml in 50 mM HEPES buffer (pH 7.4) containing 20% glycerol and 0.5 mM EDTA. Membranes containing coexpressed reductase and P-450 2C9 were diluted to 3.3 nmol of P-450/ml in 50 mM potassium phosphate buffer (pH 7.4).
Microsomal membranes containing P-450 3A2 were diluted to 6.3 nmol of P-450/ml in 50 mM potassium phosphate buffer (pH 7.4).
Enzyme Activity Assays.
The inactivation reactions contained the reconstituted enzyme mixtures or the diluted P-450s in membranes along with different concentrations of xanthates dissolved in water (except for C10–C20, which were dissolved in dimethyl sulfoxide). Control samples received an equal volume of water or dimethyl sulfoxide. NADPH (1.2 mM) was added to initiate the reactions, and incubations were performed at 30°C. This temperature was chosen to more accurately determine the rates of inactivation and because the rates of inactivation for P-450 2B1 with xanthates were not linear at 37°C. Aliquots were removed after increasing times of inactivation, and the remaining 7-EFC activity was determined in a secondary reaction mixture according to the method ofDeLuca et al. (1988). The amount of enzyme assayed and the secondary assay buffer conditions were adjusted for optimal activity for each isozyme. An aliquot of the inactivation mixture (5 pmol of P-450 2B1 in 10 μl, 5 pmol of P-450 1A1 in 10 μl, and 41 pmol P-450 2E1 in 25 μl) was added into a final volume of 1 ml of a secondary reaction mixture containing 0.1 mM 7-EFC, 40 μg BSA/ml, and 0.2 mM NADPH in 50 mM potassium phosphate buffer (pH 7.4; Buters et al., 1993). A total of 89 pmol of P-450 2D6 in 90 μl was transferred into the same secondary reaction mixture (a final volume of 1 ml) except that the buffer was replaced with 50 mM HEPES buffer (pH 7.4). For P-450 2B6 (6 pmol in 10 μl/1 ml secondary reaction mixture) and P-450 2C9 (110 pmol in 50 μl/1 ml secondary reaction mixture), the secondary reaction mixtures also contained 50 mM MgCl2. After incubating the secondary reaction mixtures at 30°C (5 min for P-450s 2B1 and 1A1, 10 min for P-450 2B6, 15 min for P-450s 2E1 and 2D6, and 30 min for P-450 2C9), the reactions were stopped by adding 334 μl of ice-cold acetonitrile (for P-450 2B1, 1A1, 2E1, and 2D6) or ice-cold 0.1 M Tris buffer (pH 9.0) containing 30% acetonitrile (for P-450 2C9). The fluorescent HFC product was measured on a SLM-Aminco model SPF-500C spectrofluorometer (SLM-Aminco, Urbana, IL) with excitation at 410 nm and emission at 510 nm. The rate of HFC formation was calculated from a standard curve.
The remaining P-450 3A2 or 3A4 activity was determined by adding an aliquot of the inactivation mixture (250 pmol of P-450 3A2 in 20 μl and 45 pmol of P-450 3A4 in 95 μl) to a secondary reaction mixture containing 2 mM erythromycin, 1 mM NADPH, and 100 mM MgCl2 in 50 mM HEPES buffer (pH 7.4) in a final volume of 500 μl. The samples were incubated at 30°C for 10 min (for P-450 3A2) or 30 min (for P-450 3A4). The reactions were stopped with ice-cold 60% trichloroacetic acid. Formaldehyde was determined to be in the supernatant according to the method described by Nash (1953). The amount of formaldehyde was quantified by fluorimetric detection with excitation at 410 nm and emission at 510 nm with a SLM-Aminco model SPF-500C spectrofluorometer (Belman, 1963; de Andrade et al., 1996).
The inhibitory potency of the xanthates on the 7-EFCO-de-ethylation activity of P-450s 2B1, 2B6, 2E1, 2D6, 2C9, 1A1, 3A4, and 3A2 was tested by incubating the P-450s along with various concentrations of xanthates and different concentrations of 7-EFC (from 1 to 100 μM).
The kinetic constants for P-450 inactivation and inhibition were calculated by linear and nonlinear regression, respectively, with Graphpad Prism software (San Diego, CA).
Partition Ratio.
For the estimation of the partition ratio, the P-450s were reconstituted as described above. Samples were incubated with different concentrations of xanthates and the reaction was allowed to go to completion. The residual activity was assayed in a secondary reaction mixture with 7-EFC as the substrate as described above. The partition ratio was determined from the intercept of the linear regression line derived from the lower xanthate concentrations with thex-axis (Silverman, 1996).
Results
Inactivation and Inhibition of P-450 2B1 by Xanthates.
The 7-EFC O-de-ethylation activity of P-450 2B1 was inactivated in a mechanism-based manner by all of the xanthates tested. A representative example, demonstrating both time- and concentration-dependent inactivation of P-450 2B1 with C8, is depicted in Fig. 1. The estimated partition ratio for the inactivation of P-450 2B1 with C8 was 4 (Fig.2). Of the xanthates tested, C8 was the most potent inactivator based on the concentration of C8 required for half-maximal inactivation (KI = 2.4 μM) and the low partition ratio of 4 (Table1). The KI of the compounds and the rate of inactivation (kinactivation) of P-450 2B1 were dependent on the alkyl- or aryl-chain length of the xanthate. With increasing numbers of carbon atoms (C14 and longer), the half-time of inactivation and the partition ratio increased (Table 1). The C3iso compound was an exception, showing both high inactivation and inhibition properties along with a low rate of inactivation and a high partition ratio. The inhibitory potency (Ki) of all xanthates for P-450 2B1 was determined. The Ki was noncompetitive in all instances and increased with increasing length of the alkyl chain (Table 1). The antiviral compound D609 showed intermediate inactivation and inhibitory potency.
Inactivation of P-450 2B6 by Xanthates.
A representative subset of xanthates was chosen to test its effect on the human P-450 2B6 7-EFC O-de-ethylation activity. P-450 2B6 was inactivated by all of the xanthates, but with a 2- to 3- fold higher potency as compared with P-450 2B1. However, the rates of inactivation for P-450 2B6 were 4- to 5-fold slower (Table2).
Inactivation of P-450 2E1 by Xanthates.
P-450 2E1 was also susceptible to inactivation, but the concentrations of C2, C8, D609, C14, or cyclohexylxanthate required for half-maximal inactivation were considerably higher (0.06–1 mM) than those observed for either P-450 2B1 or P-450 2B6 (Table3). C8 was again the best inactivator, but the KI was 25- or 60-fold higher than that observed with P-450 2B1 or 2B6, respectively.
Inhibition of P-450 1A1 by Xanthates.
No time-dependent inactivation of the P-450 1A1 activity was seen with any of the xanthates tested (data not shown). The concentration of short-chain xanthates required to observe an effect on the P-450 1A1 activity was in the micromolar range (Table4). Table 4 also illustrates the dramatic difference in the IC50 values for C2, C3, and C8 when P-450 2B1 was compared with P-450 1A1. The C14 and C20 derivatives strongly inhibited the reaction of P-450 1A1 and their interaction lacked the characteristics of mechanism-based inactivation (Fig.3).
Effect of Xanthates on P-450s 2D6, 2C9, 3A2, and 3A4.
Table 5 illustrates the effect of different xanthates on the activities of P-450 2D6 and 2C9. For all of the xanthates tested, little inactivation was observed with P-450 2D6 over the course of 15 min. At concentrations of 0.2 to 0.6 mM, an inhibition in enzymatic activity was seen that did not increase significantly within 15 min. With P-450 2C9, no time-dependent loss in enzymatic activity was observed for all of the xanthates tested. With C8 and C14, some inhibitory effect on the initial enzymatic activity of P-450 2D6 and 2C9 was observed when high concentrations of xanthate (0.2 mM and 0.05 mM, respectively) were tested. No inhibition or inactivation of P-450s 3A4 or 3A2 was observed with any of the xanthates tested (data not shown). The 100% (control) activities ranged from 5 to 7 nmol of formaldehyde formed min−1 nmol−1 for P-450 3A2 and from 1.5 to 2 nmol of formaldehyde formed min−1 nmol−1 for P-450 3A4.
Discussion
We have previously demonstrated that the NADPH-dependent P-450 activity of microsomes from phenobarbital-induced rats was susceptible to a time- and concentration-dependent inactivation by ethylxanthate (Yanev et al., 1985, 1989). With purified rat liver P-450 2B1 and human P-450 2B6 in the reconstituted system, we further demonstrated thatn-propylxanthate inactivated both isoforms in a mechanism-based manner, and that the inactivation was due primarily to a modification of the apoprotein (Kent et al., 1998). In the present study, we have extended our findings to include a large number of distinct xanthates. We have compared their inactivation and inhibitory potencies on the catalytic functions of different rat P-450s as well as P-450 isoforms that play a role in human drug metabolism. Our results indicated that xanthates preferentially inactivated P-450 2B isoforms in a time- and concentration-dependent manner. However, some inactivation of P-450 2E1 also was observed at significantly higher xanthate concentrations. All other P-450 isoform activities were either not affected (P-450s 3A2 and 3A4) or were inhibited to different degrees at high concentrations of xanthate. We found a correlation with the length of the alkyl chain and the potency of the inactivating ability of the xanthate. For P-450 2B, shorter substitutions resulted in a more effective inactivator based on a lowKI and a low turnover number. One possibility for the different metabolic rates of the various xanthates may be the stability of the bond between the first carbon atom of the alkyl chain and the oxygen (S.Y., I. Stanoeva, L. Shalamanova, I. Pajeva, B.P., unpublished observations).
With an increasing chain length (C12–C20), the xanthates were less effective as inactivators and became better inhibitors. P-450s 2B1, 2B6, and 1A1 were the most susceptible to inhibition by long-chain derivatives, with Ki values between 1 and 20 μM, followed by P-450s 2D6 and 2C9, withKi values between 5 and 200 μM. Virtually no inhibitory effect of xanthates on P-450 2E1 was seen. There appears to be at least two different events involved in the xanthate/P-450 interaction. In the first instance, the binding of the xanthate resulted in inhibition to a degree that correlated well with the increased lipophilicity of the derivatives (C14 and C20). The only exception was compound C3iso. The second event involved metabolism of the xanthate molecule. The resulting structural modification of the xanthate resulted in inactivation of the enzymatic activity with different rates and potency. The most effective inactivator was C8, with the lowest inactivation constant (2 μM for P-450 2B1 and 1 μM for P-450 2B6) and a low partition ratio. The antiviral compound D609 and cyclohexylxanthate exhibited intermediate inactivation rates, inactivation potency, partition ratio, and inhibitory activity. From the classical equation of Waley (1980) describing the interaction of a mechanism-based inactivator with the enzyme:
In summary, we conclude that xanthates are selective, mechanism-based inactivators of P-450s 2B1 and 2B6.
Acknowledgments
We thank Drs. Andrew Parkinson and Kan He for their kind gift of purified enzymes and Dr. Thomas Woolf for providing us with cells coexpressing reductase and P-450s. We also thank Danielle Mills for her help with the expression of P-450s in E. coli.
Footnotes
-
Send reprint requests to: Dr. Paul F. Hollenberg, Department of Pharmacology, Medical Science Research Bldg. III, 1150 West Medical Center Dr., Ann Arbor, Michigan. E-mail:phollen{at}umich.edu
-
This publication was supported in part by National Institutes of Health Grant CA 16954 (to P.F.H.) from the National Cancer Institute. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute.
- Abbreviations used are::
- P-450
- cytochrome P-450
- P-450 2B1
- major form of P-450 from microsomes of phenobarbital-induced rats
- reductase
- NADPH-P-450 reductase
- 7-EFC
- 7-ethoxy-4-(trifluoromethyl)coumarin
- HFC
- 7-hydroxy4-(trifluoromethyl)coumarin
- DLPC
- dilauroyl-l-a-phosphatidylcholine
- D609
- O-tricyclo[5.2.1.02,6]dec-9-yl-dithiocarbonate
- Received November 9, 1998.
- Accepted January 29, 1999.
- The American Society for Pharmacology and Experimental Therapeutics