Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Multiple CYP in human liver catalyze the oxidation of a range of xenobiotic and endobiotic lipophilic chemicals. Phosphorothioate pesticides, like parathion, are of great economic importance in agriculture as cholinesterase inhibitors. CYP are important in the bioactivation of these pesticides by oxidative desulfuration to the corresponding phosphate esters, or oxons (Norman et al., 1974; Kamataki and Neal, 1976). In the process, the thiono-sulfur atom of the phosphorothioate is transferred to CYP, leading to amino acid modification and enzyme destruction (Norman et al., 1974; Kamataki and Neal, 1976; Halpert et al., 1980). Thus, CYP not only activate phosphorothioate pesticides, but are also targets for destruction by the resultant reactive metabolites.

Poisoning of individuals after exposure to phosphorothioates is common, especially in agricultural workers and in the third world. Present treatments involve management of toxicity including the use of pralidoxime, which reactivates the esterases that are acylated by phosphorothioate oxons (Karalliedde and Senanayake, 1988). However, serious toxic effects may arise from the use of certain drugs, including phenothiazine tranquilizers (Arterberry et al., 1962), in the treatment of parathion intoxication. That is, the exposure to parathion appeared responsible for the adverse effects of the subsequent supportive therapy. These findings are consistent with impairment of CYP-mediated drug elimination by phosphorothioate pesticides.

In vitro studies in rat liver have established that oxidative metabolism is decreased by phosphorothioates (Hunter and Neal, 1975; Morelli and Nakatsugawa, 1978). More recently it has emerged that CYP are not destroyed uniformly during parathion oxidation. Thus, the constitutive CYP 2C11 and 3A2 were more susceptible than 2A1 and 2A2 to deactivation (Butler and Murray, 1993). Oxidation reactions mediated by CYP2C6 (in control rat liver) and CYP2B1 (in microsomes from phenobarbital-induced rats) were very sensitive to inhibition, but not inactivation, by parathion. Our study sought to assess the susceptibilities of human hepatic CYP to inhibition and inactivation by parathion. It emerged that 3A4, the principal CYP in human liver, has an important role in parathion oxidation. This enzyme was also susceptible to inhibition and inactivation by phosphorothioate pesticides in in vitro studies. In view of the pivotal role that CYP3A4 has in the oxidative biotransformation of many drugs the inactivation of this enzyme has serious potential consequences for concurrent drug therapy.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. Parathion and methylparathion were gifts from Rhone Poulenc (Brisbane, Australia), chlorpyrifos was provided by Dow Chemicals Australia (Frenchs Forest, NSW), fenitrothion was from Sumitomo Chemicals (Osaka, Japan), malathion was supplied by Cyanamid Australia (Baulkham Hills, NSW) and azinphosmethyl was from Bayer Australia (Pymble, NSW). Diclofenac sodium and sulfaphenazole were kindly provided by Ciba-Geigy Australia (Pendle Hill, NSW). Paraoxon, 4-nitrophenol, 7-ethylresorufin (resorufin O-ethyl ether), 7-ethoxycoumarin, resorufin, 7-hydroxycoumarin, N-nitrosodimethylamine, triacetyloleandomycin, debrisoquine, -naphthoflavone, 4-methylpyrazole and general biochemicals were from Sigma Chemical Co. (St. Louis, MO). [¹⁴C] Testosterone (specific activity, 59 mCi/mmol), Hyperfilm-MP and ACS II were from Amersham Australia (North Ryde, NSW). 6-Hydroxytestosterone was from Steraloids (Wilton, NH). Tolbutamide, chlorpropamide and 4-hydroxymethyltolbutamide were generously provided by Hoechst Australia (Kingsgrove, NSW) and ketoconazole was from Janssen-Csilag (Lane Cove, NSW). Aniline was purchased from Ajax Chemicals (Sydney, Australia) and was redistilled from zinc dust before use. Analytical grade solvents and other chemicals were obtained from Ajax.

Cell microsomes containing cDNA-derived human CYP. Microsomal fractions prepared from human lymphoblastoid cell lines (AHH-1 TK⁺/) in which cDNA-derived human CYP (1A1, 1A2, 2A6, 2B6, 2C8, 2C9, 2C19, 2D6, 2E1 and 3A4) had been expressed were purchased from Gentest Corp. (Woburn, MA). Control microsomes were prepared from untransfected AHH-1 TK⁺/ cells. The protein contents of the preparations were 10 mg/ml; P450 contents were determined by the supplier (presented in table 3).

Preparation of human hepatic microsomes. Human liver was obtained as unwanted tissue from donor or recipient livers during transplantation or resection. Liver was obtained from the Queensland transplant program at Princess Alexandra Hospital, Brisbane, Australia or the National transplant center at Royal Prince Alfred Hospital, Sydney, Australia. After perfusion with ice cold ViaSpan solution (Belzer University of Wisconsin solution; Du Pont Pharmaceuticals, Wilmington, DE) the tissue was packed in ice and transported by air (if it originated in Brisbane) to the laboratory where it was frozen in liquid nitrogen. Washed microsomal fractions were prepared from individual liver samples and stored at 70°C suspended in 50 mM potassium phosphate buffer (pH 7.4) containing 20% glycerol and 1 mM EDTA (Murray et al., 1983). Microsomal protein was estimated by the method of Lowry et al. (1951) using bovine serum albumin as standard.

Monooxygenase assays. Testosterone hydroxylation was measured by previously described methods (Murray, 1992). Briefly, incubations (0.4 ml, 2.5 min, 37°C) contained the substrate (0.18 µCi; 50 µM, except in kinetic experiments where the concentration range was 10-200 µM testosterone), microsomal protein (0.15 mg) and an NADPH generating system (1 mM NADP, 4 mM glucose 6-phosphate and 1 U glucose 6-phosphate dehydrogenase). In inactivation experiments, reaction components except substrate were incubated with microsomes and NADPH (at 37°C) for varying periods and then removed to tubes containing ¹⁴C-testosterone; the reactions continued for 2.5 min. Products were extracted from incubations in chloroform and applied to TLC plates (Merck silica gel 60 F₂₅₄ type, heated at 100°C for 15 min before use). Plates were developed in dichloromethane:acetone, 4:1, air-dried and then developed in chloroform:ethyl acetate:ethanol, 4:1:0.7 (Waxman et al., 1983). Autoradiography on Hyperfilm-MP (Amersham Australia) was used to locate radioactive steroid metabolites and formation rates were determined by scintillation spectrometry.

High-performance liquid chromatography for the separation of parathion, paraoxon and 4-nitrophenol. The separation of parathion, paraoxon and 4-nitrophenol on Ultrasphere-Si (5 µm, 4.6 mm i.d × 25 cm; Beckman, San Ramon, CA) has been described previously (Butler and Murray, 1993). The mobile phase was dichloromethane: acetonitrile: acetic acid (93:7:0.02) (Sultatos et al., 1982), the detection wavelength was 254 nm and the flow rate was 1 ml/min. Retention times were: parathion (3.3 min), 4-nitrophenol (4.3 min), 4,4-dihydroxybiphenyl (6.5 min) and paraoxon (8.7 min). Metabolite peak areas were calculated on a Waters 730 Data Module and product formation was quantitated (using peak area ratios) from standard curves constructed with known quantities of the authentic metabolites.

Estimation of holo-P450 inactivation. Microsomal P450 content was determined as described by Omura and Sato (1964). Inactivation of P450 in individual human microsomal fractions (1 mg/ml) was determined by previously described methods (Butler and Murray, 1993). Here, parathion (25 µM) was added to the microsomes and NADPH (1 mM) was used to initiate the reaction; apparent P450 content was estimated after 10 min.

Statistics. Data are presented throughout as means ± S.E.M. or as median values. Differences between group means were detected by analysis of variance and Dunnett's q' test.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Individual variation in human liver microsomal parathion oxidation. Microsomal fractions from 27 individual human livers were available in this study. Most of the samples were obtained as excess tissue in the case of adult donors for transplantation of pediatric recipients. Other samples were obtained from the normal margin of tissue taken for biopsy during liver resection. As shown in table 1, the individual variation in parathion (250 µM) oxidation to paraoxon and 4-nitrophenol was considerable (range of total metabolite formation 1.72 to 18.33 nmol/mg protein/min; n = 23). The recent drug histories of the liver donors are indicated in table 1; HL24-26 also received no medication and the drug history of HL27 was unknown (additional samples used in the derivation of the data in table 4). There were few instances where the individual had been administered drugs that are known to interact with CYP (four received spironolactone, one received cimetidine and one received norfloxacin). Thus, it is likely that variation in parathion oxidation reflects the interindividual expression of CYP involved in these pathways.

Correlation analysis of parathion oxidation with other microsomal oxidations. As shown in table 2, in microsomal fractions from 11 individual livers, rates of CYP3A4-mediated testosterone 6-hydroxylation were well correlated with rates of parathion oxidation to paraoxon and 4-nitrophenol (and to the sum of the two metabolites) (fig. 1). The variances (r²) were 0.927 to 0.950, which indicates that the relationships between the activities were highly significant (P < .001). In contrast, correlations between parathion oxidation and tolbutamide hydroxylation, N,N-dimethylnitrosamine N-demethylation and 7-ethylresorufin O-deethylation did not attain statistical significance. However, rates of microsomal parathion oxidation to paraoxon and 7-ethylresorufin O-deethylation were correlated (P < .05). As anticipated, testosterone 6-hydroxylation was not correlated with microsomal oxidation rates of any of the other substrates.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Correlation of parathion oxidation with testosterone 6-hydroxylation in human hepatic microsomes.

Effects of chemicals on parathion oxidation in human liver. A series of chemicals that are known to interact as substrates or inhibitors with individual human CYP enzymes were evaluated for their effects on microsomal parathion oxidation. It is evident from the data in figure 2 that the most pronounced effects were produced by the inhibitors of CYP3A4. Thus, ketoconazole decreased parathion oxidation to around 30 and 20% of control when included in incubations at concentrations of 25 and 200 µM (fig. 2). The macrolide antibiotic triacetyloleandomycin (500 µM), also an inhibitor of CYP3A4, decreased the activity to around 70% of control. In contrast with these findings, 4-methylpyrazole (an inhibitor of CYP2E1), -naphthoflavone and 7-ethylresorufin (1A2 inhibitor and substrate, respectively), tolbutamide and diclofenac (2C9/10 substrates), sulfaphenazole (2C9 inhibitor), debrisoquine sulfate (2D6 substrate) and 7-ethoxycoumarin (substrate for several CYP, including 2A6 and 2B6) were essentially without effect.

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Effect of CYP-specific inhibitors and substrates on microsomal parathion metabolism (sum of paraoxon and 4-nitrophenol formation) in human liver. Results are presented as the mean ± S.E.M. of determinations in three individual human liver preparations.

Oxidation of parathion by individual cDNA-derived human CYP. To further investigate the participation of human CYP in parathion metabolism we examined the capacity of a range of microsomal fractions from lymphoblastoid cells to oxidize the pesticide (250 µM). It emerged from these studies that CYP 1A2, 2B6 and 3A4 supported parathion oxidation to paraoxon and 4-nitrophenol, whereas 1A1, 2A6, 2C8, 2C9, 2C19, 2D6 and 2E1 were much less active (table 3). Because numerous CYP enzymes oxidize 7-ethoxycoumarin (Chang et al., 1993), this activity was measured in each of the microsomal preparations for comparative purposes.

Inhibition and inactivation of CYP enzymes in human liver during microsomal parathion oxidation. The capacity of parathion to deactivate CYP was assessed in microsomal fractions from seven individual human livers. From the data in table 4 it is apparent that 19 ± 4% (range 8.0-40.2%) of the microsomal total P450 content was lost during parathion oxidation (final concentration 25 µM; incubation time 10 min). Subsequent studies evaluated the time-dependent inactivation of four major CYP-mediated substrate oxidations in human liver.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Time-dependent effects of preincubation of parathion with NADPH-supplemented human hepatic microsomes before substrate addition. Parathion (25 µM) was preincubated with human hepatic microsomes at 37°C for varying periods before transfer to vials containing () testosterone, () tolbutamide, () aniline or () 7-ethylresorufin. Reactions then continued for the standard assay times indicated in "Materials and Methods," after which reactions were terminated and product formation was estimated.

View larger version (8K): [in this window] [in a new window]   Fig. 4.   Kinetics of the inhibition of microsomal CYP3A4-mediated testosterone 6-hydroxylation in human liver by parathion in the absence of a preincubation step. A, Dixon plots at several testosterone concentrations: , 10 µM; , 20 µM; , 50 µM; , 100 µM; , 200 µM. B, Dixon plot slopes vs. reciprocal testosterone concentrations. Units are substrate and inhibitor concentrations, µM; V, nmol/min/mg of protein.

View larger version (8K): [in this window] [in a new window]   Fig. 5.   Kinetics of the inhibition of microsomal CYP3A4-mediated testosterone 6-hydroxylation in human liver by parathion after a 10-min preincubation step. Details of substrate and inhibitor concentrations as for figure 4.

View larger version (11K): [in this window] [in a new window]   Fig. 6.   Effect of parathion metabolism in NADPH-fortified human liver microsomes on the extent of inhibition of (A) CYP3A4-mediated testosterone 6-hydroxylation and (B) CYP2C9-mediated tolbutamide hydroxylation. () No preincubation and () 10 min preincubation before substrate addition.

View this table: [in this window] [in a new window]   TABLE 5 Inactivation of human hepatic microsomal P450 3A4-mediated testosterone 6-hydroxylation by phosphorothioate pesticides

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Our results indicate the major involvement of CYP3A4, and perhaps other members of the CYP3A subfamily, in the oxidative biotransformation of the phosphorothioate pesticide parathion in human liver. The evidence in support of this assertion can be summarized as follows: 1) that significant correlations were observed in individual human microsomal fractions between rates of CYP3A4-dependent testosterone 6-hydroxylation and parathion oxidation to paraoxon and 4-nitrophenol, 2) that the CYP3A4 inhibitors ketoconazole and triacetyloleandomycin inhibited parathion oxidation and 3) that cDNA-derived CYP3A4 was an effective catalyst of parathion oxidation.

During parathion oxidation rodent hepatic microsomal CYP undergo inactivation (Norman et al., 1974; Kamataki and Neal, 1976; Halpert et al., 1980). From our study a similar process occurs in human liver. Thus, extensive loss of holo-P450 was observed when parathion was incubated with NADPH-fortified human liver microsomes; this is consistent with loss of heme from the CYP molecule. In view of the role played by CYP3A4 in parathion biotransformation in human liver, subsequent studies evaluated whether this enzyme was inactivated during oxidation of the pesticide. The approach taken was to assess the time-dependence of the intensification of inhibition by parathion of CYP-specific activities. Although it was clear that CYP3A4 is a principal target for parathion-mediated destruction, it also appeared that at least two other human CYP, 2C9 and 1A2, are deactivated by parathion in NADPH-supplemented hepatic microsomes; CYP2E1 was refractory to inactivation. The possibilities were assessed that these effects may be secondary to the action of CYP3A4 or that nondestructive parathion metabolites are involved in CYP1A2 inhibition. It was found that the CYP3A4 inhibitor ketoconazole did not impair the loss of CYP1A2-dependent 7-ethylresorufin O-deethylation during parathion oxidation. Further, paraoxon and 4-nitrophenol were noninhibitory toward 7-ethylresorufin O-deethylation activity (not shown).

On first examination the findings that CYP 1A2 and 2C9 were deactivated during parathion oxidation appear to be inconsistent with the data in figure 2. These data demonstrate the lack of effect of substrates and inhibitors of CYP 2C and 1A on microsomal parathion metabolism. However, it is noteworthy that several cDNA-derived CYP were able to support the oxidation of the pesticide. In particular, CYP1A2 was an efficient catalyst, as were the minor CYP 2B6 and 2D6. Although CYP 2C8, 2C9 and 2C19 were poor catalysts of parathion oxidation, the levels of expression of these proteins in microsomes from lymphoblastoid cells were quite low (23-34 pmol CYP 2C/mg protein, compared with levels between 100-220 pmol/mg protein for most other CYP preparations; table 3 and Gentest Corp.). Thus, it remains a possibility that CYP2C may be minor or inefficient catalysts of parathion oxidation. Considered together, it is likely that CYP3A4 is the dominant catalyst involved in parathion biotransformation in human liver, but that in some cases other CYP are also capable of supporting the reaction. This is not unlike the situation with rat liver CYP2C13, which is a steroid 6-hydroxylase in its purified state, but does not participate in the reaction in heterogeneous microsomal fractions (Swinney et al., 1987).

Accurate estimation of the concentrations of parathion or other phosphorothioate pesticides that may be encountered in vivo is difficult. However, it has been suggested that the minimum acute lethal dose of parathion in adults could be as little as ~4 mg/kg (approximately 14 µmol/kg; Gosselin et al., 1984). Certainly, the number of reports of poisoning by phosphorothioates is large and there appears little doubt that concentrations of the pesticide sufficient for adverse effects in humans may result from oral or dermal contact. The potential consequences of CYP3A4 inactivation after intoxication of individuals with parathion are likely to be significant. In a recent report it was demonstrated that CYP3A protein represented ~30% or more of total spectrophotometric P450 in microsomal fractions from 60 individuals; in some cases this proportion exceeded 50% (Shimada et al., 1994). It is also clear that this enzyme is involved in the oxidative biotransformation of numerous drugs, including midazolam (Kronbach et al., 1989), alfentanil (Yun et al., 1992), lansoprazole (Pichard et al., 1995) and docetaxel (Marre et al., 1996) among others. The enzyme is also important in the activation of carcinogens and mutagens, such as aflatoxin B1 and sterigmatocystin (Shimada and Gungerich, 1989; Shimada et al., 1989). Inactivation of CYP3A4 would be expected to impair the capacity of the individual to eliminate these drugs and carcinogens, although the consequences of impaired carcinogen activation may be less predictable than the effects on drug pharmacokinetics. In the cases of the drugs that are metabolized extensively by CYP3A4, their accumulation in serum could lead to adverse reactions characterized by enhanced therapeutic effect. In some cases this may have serious consequences such as the severe hypotension and eventual death produced in a parathion-intoxicated farm worker by promazine (Arterberry et al., 1962). In this regard, it is noteworthy that the S-oxidation of chlorpromazine, a phenothiazine tranquilizer structurally similar to promazine, is mediated by CYP3A4 in human liver (Cashman et al., 1993).