Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Metabolism studies, both in vivo and in vitro, have mainly used mice or rats as the experimental animal model. However, it is difficult to extrapolate to humans from data obtained with rodents, essentially because of differences among species in many physiological functions. Accordingly, nonhuman primates, which are genetically closer to humans than are rodents, could be better species for comparative studies. The pharmaceutical industry often includes monkeys in preclinical and toxicological studies.

Among enzymes involved in xenobiotic metabolism, the most important system is represented by the P450¹ superfamily (Parkinson, 1996), which has been studied mostly in rodents and humans and less in monkeys (Nelson et al., 1996). To date, P450 isoforms closely related to the P450 1A (Komori et al., 1992a), 2A (Ohmori et al., 1993a), 2B (Ohmori et al., 1993b), 2C (Ohi et al., 1989; Ohmori et al., 1994), 2D (Wu et al., 1993), and 3A (Ohmori et al., 1993a; Dalet-Beluche et al., 1992) subfamilies have been purified from the livers of some monkey species and their biochemical properties have been compared with those of the corresponding human and rodent enzymes. On the other hand, although the sequence of the P450 2E1 gene from cynomolgus monkeys has been reported and found to be similar (90%) to the human orthologue (Komori et al., 1992b), the function of the 2E1 protein has not been described. Because P450 2E1 has a considerable influence on the metabolism of low-molecular weight xenobiotics and the metabolic activation of several hepatotoxins, including benzene, halothane, and carbon tetrachloride (Guengerich et al., 1991), interest in the development of model substrates and inhibitors of this enzyme in monkeys appears warranted.

The rate of demethylation of DMN, determined with substrate concentrations of 1-4 mM, is generally accepted as a selective activity to assay the contribution of 2E1 in rats, rabbits, and humans (Tu and Yang, 1983; Thomas et al., 1987; Yang et al., 1990; Raucy et al., 1987). Some authors (Stevens et al., 1993; Sharer et al., 1995) also used this activity, with DMN at concentrations of 1 mM, to probe 2E1 in monkey species. However, a selective DMNd role for the monkey 2E1 isoform cannot be assumed, because interspecies differences in 2E1 activities toward the same substrate have been reported (Mikalsen et al., 1991; Puccini et al., 1992). In general, the modification of a few key amino acids in the sequences of orthologous P450 isozymes may result in different substrate specificities in different species. The case of mouse P450 2A4 and 2A5, for example, is well known; the replacement of only one amino acid brings about a change of substrate specificity (Lindberg and Negishi, 1989).

Besides DMN demethylation, other reactions, such as the hydroxylation of aniline, pNP, and the muscle relaxant CLZ, have been frequently used as tools to screen for P450 2E1 activity in rats (Reinke et al., 1985; Carriere et al., 1993), rabbits (Raucy et al., 1987), and humans (Peter et al., 1990; Girre et al., 1994; Tassaneeyakul et al., 1993); however, their relevance for 2E1 in monkeys has not been assessed. The aim of the present study was to identify, by using a large number of untreated male and female cynomolgus monkeys, the interindividual variability of 2E1 expression in liver and to establish which of the aforementioned oxidations are suitable for evaluation of P450 2E1 in this species.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. Nitrocellulose filters (0.45 µm), 4-chloro-1-naphthol, corticosterone, 16-hydroxytestosterone, 4-androstene-3,17-dione, -naphthoflavone, triacetyloleandromycin, metyrapone, erythromycin, DMN, 4MPy, and 8MP were purchased from Sigma Chemical Co. (St Louis, MO). 15-, 6-, 16-, and 2-hydroxy metabolites of testosterone were obtained from the Steroids Reference Collection (St. Louis, MO). Goat anti-rabbit IgG was purchased from Dako (Copenhagen, Denmark). Enzymes and coenzymes were obtained from Boehringer (Mannheim, Germany). Ethoxyresorufin was synthesized from resorufin by ethylation with ethyl iodide (Klotz et al., 1984). CLZ, its 6-hydroxy metabolite, and benzoxazolone were kindly provided by Prof. F. P. Guengerich (Vanderbilt University, Nashville, TN). All other chemicals and solvents were of analytical grade and were obtained from commercial sources.

Animals and Preparation of Microsomes. Mature male and female cynomolgus monkeys (Macaca fascicularis), 2-3 years old (Inveresk Research, Tranent, Scotland), were housed individually in stainless steel cages in a temperature-, humidity-, and light-controlled facility (70-72°F, 48-52% humidity, 12-hr light/dark cycle). The animals were allowed free access to water and food. They were killed by CO₂ asphyxia; the livers were collected and microsomes were prepared from single animals, as described previously (Longo et al., 1986). The washed microsomal pellets were resuspended in 100 mM phosphate buffer, 1 mM EDTA (pH 7.4), and stored at 80°C. Protein content was determined according to the method of Lowry et al. (1951), using bovine serum albumin as the standard.

Enzyme Assays. Hepatic P450 levels were measured according to the method of Omura and Sato (1964). Microsomal AnH was determined by measuring the formation of p-aminophenol, as described by Ko et al. (1987). Aminopyrine, DMN, and erythromycin demethylase activities were assayed by measuring the formation of formaldehyde (Tu and Yang, 1983). Ethoxyresorufin O-deethylase activity was determined by measuring the formation of resorufin, with a Perkin-Elmer spectrofluorimeter (Krijgsheld and Gram, 1984). pNPH activity was determined by measuring the formation of 4-nitrocatechol according to the method of Reinke and Moyer (1985). Testosterone hydroxylase was determined as reported previously (Longo et al., 1991), using a HPLC method described by Platt et al. (1989). CLZ6H was assayed by a HPLC method described by Peter et al. (1990), using benzoxazolone as the internal standard. All enzymatic activities were assayed under conditions of linearity with respect to protein and time.

Gel Electrophoresis and Immunoblotting. Sodium dodecyl sulfate-gel electrophoresis was performed using the discontinuous system of Laemmli (1970), with a 1.5-mm-thick gel and 3% and 7.5% acrylamide in the stacking and separation gels, respectively. Proteins were transferred from the slab gel to nitrocellulose filters using the method of Towbin et al. (1979). Immunodetection was performed using rabbit anti-rat P450 2E1 as the primary antibody and anti-rabbit IgG conjugated to horseradish peroxidase as the secondary antibody. The peroxidase activity was detected with chloro-1-naphthol and H₂O₂ as previously described (Puccini et al., 1992). The bands on the nitrocellulose membranes were quantified with a laser densitometer (Ultrascan 2202; LKB).

Preparation of Antibodies. Female New Zealand white rabbits were immunized, as described previously (Kaminsky et al., 1981), with the purified rat P450 2E1 antigen. Preimmune serum was collected before the injections. For the first immunization, about 0.2 mg of antigen was mixed with complete Freund's adjuvant. After 6 and 10 weeks, injections of 0.1 mg of antigen in incomplete Freund's adjuvant were administered. The immune serum was collected 10 days after the last injection. IgG fractions from preimmune serum and immune serum were purified by precipitation of non-IgG proteins with caprylic acid at pH 4.5, followed by precipitation of the IgG fraction with ammonium sulfate at pH 7.4 (McKinney and Parkinson, 1987). These antibodies recognized, in microsomal samples of rat or human liver, a single, electrophoretically distinct protein. The IgG, at a concentration of 10 mg/nmol P450, was able to inhibit by approximately 80% the AnH activity in hepatic microsomes from rats pretreated with acetone, a potent inducer of 2E1 (Menicagli et al., 1994).

Chemical Inhibition Assay. All of the inhibitors were dissolved in methanol, except metyrapone and 4MPy, which were dissolved in water. Because methanol could have an inhibitory effect, it was evaporated under N₂. To facilitate the redissolution of inhibitors, the residue was resuspended in assay buffer by sonication and then microsomes were added and vigorously vortex-mixed before addition of the other components.

Immunoprecipitation Assay. Microsomes (0.75 mg/ml) were mixed with different amounts of anti-2E1 antibodies (IgG fraction) or preimmune rabbit IgG in 100 mM potassium phosphate buffer (pH 7.4) and were preincubated for 20 min at 4°C. Cofactors and substrate were then added and incubated for 30 min at 37°C, under the conditions previously described.

Analysis of Data. The Michaelis-Menten parameters K_M and V_max were obtained by fitting kinetic data to one- or two-enzyme models by using a simple computer program (devised by Dr. R. Ambrosetti, Istituto di Chimica Quantistica, Consiglio Nazionale delle Ricerche, Pisa, Italy) designed for nonlinear, least-squares, regression analyses; the data were depicted as Eadie-Hofstee plots. Correlation coefficients were calculated by least-squares regression analysis of the raw data. Student's t test was used, and correlations were considered to be statistically significant at p < 0.05.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Immunodetection of P450 2E1 in Cynomolgus Monkey Liver Microsomes. The anti-rat 2E1 antibodies were used to probe immunoblots of cynomolgus liver microsomal preparations. As shown in fig. 1, the antibodies cross-reacted with two proteins in monkey liver, one of which had a lower molecular weight than purified rat 2E1. This staining pattern was observed for all of the 13 male and 12 female monkey microsomal samples examined. The staining intensity of the protein with the higher molecular weight varied considerably among the individual liver samples, whereas that of the protein with the lower molecular weight was very weak and did not exhibit appreciable variability. Thus, the immunoreactive protein with a molecular weight similar to that of rat 2E1 was believed to be the cynomolgus monkey orthologue of P450 2E1.

View larger version (49K): [in this window] [in a new window]   Fig. 1.   Western blot analysis of liver microsomes from untreated cynomolgus monkeys. Microsomes (20 µg of protein in each lane) were subjected to electrophoresis, followed by immunoblot analysis using polyclonal anti-rat P450 2E1 (see Materials and Methods). Lanes 1-3 and 5-9, male and female microsomal monkey samples, expressing the lowest and highest 2E1 contents; lane 4, 1 pmol of purified rat P450 2E1.

Kinetics of CLZ, pNP, and Aniline Hydroxylation and DMN Demethylation by Monkey Liver Microsomes. Microsomal preparations from five male and five female monkey liver samples, expressing the lowest and highest immunodetectable 2E1 levels, were used to obtain the apparent kinetic parameters V_max and K_M for the oxidation of possible 2E1 substrates, i.e. CLZ, aniline, pNP, and DMN. In the case of CLZ 6-hydroxylation, microsomes displayed monophasic Michaelis- Menten kinetics over the range of substrate concentrations examined (5-750 µM) (fig. 2A) and the activity was linear up to 30 min and microsomal protein concentrations of 0.5 mg/ml. As calculated from the least-squares regression analyses, the apparent K_M value was 77 ± 35 µM, whereas V_max varied between 1.3 and 5.8 nmol/min/mg of protein. No gender differences were observed in the K_M.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   Typical Eadie-Hofstee plots for CLZ 6-hydroxylation (A), pNP hydroxylation (B), and aniline hydroxylation (C) catalyzed by liver microsomes from monkey samples. The substrate concentrations ranged from 5 to 750 µM for CLZ, from 5 to 100 µM for pNP, and from 100 to 2000 µM for aniline. For aniline hydroxylation, points are experimentally determined values and the solid line is the computer-generated curve for the best fit for a two-enzyme model. Experimental details are described in Materials and Methods.

Determination of P450-Linked Monooxygenase Activities and Comparison with CLZ6H in Liver Microsomes from Male and Female Monkeys. Table 1 shows findings for total P450 estimations and the oxidative metabolism of various P450 substrates in 13 male and 12 female monkey liver microsomal samples. AnH, DMNd, pNPH, and CLZ6H activities were determined, because they are known to be 2E1-dependent activities in rodents or humans (Raucy et al., 1987; Peter et al., 1990; Tassaneeyakul et al., 1993). The P450 content and DMNd activity level were similar to published data (including the lack of gender differences) (Sharer et al., 1995; Bullock et al., 1995; Longo et al., 1992; Weaver et al., 1994); no comparison was possible for the AnH, CLZ6H, and pNPH activities, because they had not been previously determined.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Correlations between enzymatic activities and amounts of immunodetectable P450 2E1. Enzymatic activities are expressed as nanomoles per minute per milligram of protein, and P450 2E1 content is expressed in arbitrary units.

Effects of Anti-P450 2E1 Antibodies and P450 Chemical Inhibitors on the Hydroxylation of CLZ and pNP. Immunoinhibition experiments demonstrated that antibodies raised against rat P450 2E1 inhibited both CLZ 6-hydroxylation and pNP hydroxylation catalyzed by monkey liver microsomes (fig. 4). Preimmune IgG had no effect on these reactions. At the highest concentration of antibodies (10 mg IgG/nmol P450), the rates of hydroxylation of CLZ and pNP were reduced by 68 and 63%, respectively. When a liver microsomal sample with high CLZ6H activity (5.8 nmol/min/mg of protein) was used, inhibition of about 80% of the CLZ6H activity by anti-rat 2E1 (10 mg IgG/nmol P450) was observed. Although greater inhibition occurred in microsomes exhibiting the highest rate of CLZ 6-hydroxylation, there was a basal level of CLZ6H activity that was not inhibitable by these antibodies, possibly reflecting the contribution of other P450 isoforms to the catalysis of this reaction or a limited ability of these antibodies to inhibit the 2E1-dependent activities. To further assess the role of 2E1 in the hydroxylation of CLZ and pNP and the possible participation of other P450 isoforms, these reactions were carried out in the presence of a panel of selective chemical inhibitors (4MPy, 8MP, metyrapone, -naphthoflavone, and triacetyloleandromycin). 4MPy, which has been reported to be a potent inhibitor of P450 2E1 in rats and humans (Feierman and Cederbaum, 1987), inhibited both CLZ 6-hydroxylation and pNP hydroxylation in monkey liver microsomes, with IC₅₀ values of approximately 1.5 and 13 µM, respectively (fig. 5). Unexpectedly, 8MP, which has been reported to selectively inhibit P450 2A-dependent coumarin hydroxylase activity in mice and humans (Mäenpää et al., 1993), did not inhibit, without preincubation, this activity in monkey liver microsomes (IC₅₀ > 300 µM). In contrast, 8MP inhibited both CLZ6H and pNPH activities, with IC₅₀ values of 65 and 155 µM, respectively (fig. 5). On the other hand, triacetyloleandromycin, metyrapone, and -naphthoflavone, which are known to be selective inhibitors of the P450 3A, 2B, and 1A subfamilies, respectively (Murray and Reidy, 1990; Halpert et al., 1994; Namkung et al., 1988; Roos et al., 1993), did not affect or weakly affected both CLZ and pNP hydroxylations, having IC₅₀ values of >300 µM. Regarding the inhibition selectivity of 4MPy, it was observed that, even at a concentration of 500 µM, 4MPy was unable to inhibit the monkey liver microsomal hydroxylation of testosterone (data not shown), a compound known to be oxidizable at many positions, in rats, by several P450 isoforms but not 2E1 (Platt et al., 1989). Thus, 4MPy might be a selective inhibitor of 2E1 also in monkeys. These findings suggest that P450 2E1 contributes primarily to CLZ and pNP oxidation.

View larger version (11K): [in this window] [in a new window]   Fig. 4.   Effects of anti-rat P450 2E1 IgG (, ) or preimmune IgG () on CLZ6H () or pNPH (, ) activity in microsomes from untreated cynomolgus monkeys. The microsomes were preincubated with IgG fractions for 20 min, and then incubations were performed as described in Materials and Methods. Values, which are expressed as percentages of control activity (3.3 and 0.68 nmol/min/mg of protein for CLZ6H and pNPH, respectively), are means of two experiments.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Inhibitory effects of 4MPy (solid lines) and 8MP (dashed lines) on pNP hydroxylation () and CLZ 6-hydroxylation () in monkey liver microsomes. Each point represents the mean value from three microsomal samples. Bars indicate SD. The mean control activities were 4.5 and 0.71 nmol/min/mg of protein for CLZ6H and pNPH, respectively.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

In the present investigation, we established that a P450 protein immunochemically and catalytically similar to rat 2E1 was, as expected (Komori et al., 1992b; Stevens et al., 1993), present and steadily expressed in adult male and female cynomolgus monkeys. Its variability among several individuals, as determined by immunoblotting, was approximately 8-fold, but the 2E1 content did not vary significantly between male and female liver microsomes.

The data also indicated that monkey hepatic P450 2E1 is the enzyme predominantly responsible for the hydroxylation of pNP and CLZ but not for the oxidation of aniline or DMN. Evidence for this conclusion was derived from studies using monkey liver microsomes to assess the metabolism of the aforementioned substrates, correlation of activities and immunodetectable 2E1 contents, inhibition by chemicals, and immunoinhibition.

Monkey liver microsomes from 10 individuals metabolized monophasically and without gender differences both CLZ and pNP, but not DMN and aniline, the metabolism of which exhibited multiphasic enzyme kinetics. The multiphasic nature of the steady-state kinetics for the latter two compounds was not surprising, because DMNd, at substrate concentrations of 4 mM, was found to be dependent mainly on P450 2E1 and P450 2A5 and their levels of expression in mice (Camus et al., 1993) and because aniline has been reported to be hydroxylated by various P450 isoforms in rats (Funae and Imaoka, 1993).

Our kinetic study suggested that P450 2E1 alone may be the major enzyme involved in CLZ 6-hydroxylation in monkeys because, despite a >3-fold variation in V_max values among samples, the same K_M values (77 ± 35 µM) were determined. For pNPH as well, although the V_max values varied about 3-fold, the same K_M values (14 ± 7 µM) were observed, suggesting that this reaction may also be catalyzed mostly by 2E1. Of course, the existence of other P450 isoforms having K_M values close to those of 2E1 for CLZ6H and pNPH cannot be excluded. It should also be noted that the K_M values for these monooxygenases in monkeys are quite similar to those reported for humans [39 ± 7 µM for CLZ6H (Peter et al., 1990) and 30 ± 7 µM for pNPH (Tassaneeyakul et al., 1993)], further supporting the hypothesis that orthologous 2E1 forms may be the major enzymes responsible for these reactions in both species.

Some authors have presented compelling evidence that, in rats and humans (Carriere et al., 1993), P450 1A1 is also involved in the 6-hydroxylation of CLZ, although with a K_M higher than that of 2E1. In untreated cynomolgus monkeys, however, the contribution of this isoform to CLZ6H activity is ruled out because the liver of this species lacks detectable expression of either 1A1 or 1A2 proteins (Komori et al., 1992a; Bullock et al., 1995).

A further strong indication that pNP and CLZ hydroxylations are catalyzed by 2E1 is derived from the activity correlation experiments. The catalytic properties, toward various substrates, of a number of purified P450s (1A1, 2A, 2B, 2C, and 3A) from monkeys have been reported (Komori et al., 1992a; Ohmori et al., 1993a,b; Ohi et al., 1989), and we used those substrates as isoform markers in the present investigation. Our results revealed that a strong correlation (r = 0.75) was found only between CLZ6H and pNPH in 25 different monkey samples and that only these activities were strongly correlated with P450 2E1 contents measured immunochemically. Because of the multiphasic kinetics observed for DMNd and AnH, it was not surprising to find that AnH showed a weak correlation with the immunoblotted 2E1 protein levels, whereas DMNd was not correlated at all. Therefore, DMNd activity, in contrast to previous assumptions (Stevens et al., 1993; Sharer et al., 1995), does not appear to be appropriate for probing 2E1 in monkeys. AnH also is not the best activity for 2E1.

Additional evidence for primary involvement of 2E1 in the hydroxylation of CLZ and pNP was derived from inhibition experiments with anti-rat P450 2E1 antibodies and chemical inhibitors. 4MPy (a potent P450 2E1 inhibitor) and the antibodies were able to strongly inhibit CLZ6H and pNPH; however, a part of these activities (20-40%) were resistant to inhibition, suggesting that other constitutive P450 isoforms may be implicated in these oxidations in monkey liver.

A contribution to CLZ6H activity might be derived from the P450 3A subfamily, because P450 3A-dependent 6-hydroxytestosterone hydroxylation was found to correlate significantly with CLZ hydroxylation. Indeed, the involvement of P450(s) belonging to the 3A subfamily in the metabolism of CLZ at low substrate concentrations has been reported in control and pretreated rats (Jayyosi et al., 1995).

Our findings do not imply that pNP and CLZ may be used as model probe substrates for hepatic 2E1 activity in cynomolgus monkeys pretreated with inducing drugs. The oxidation of pNP to 4-nitrocatechol, for example, has been shown to be useful in a number of species treated with 2E1 inducers (Koop et al., 1989), but a lack of selectivity was observed in phenobarbital-pretreated rats; treatment with phenobarbital caused an increase in pNPH activity (Lucas et al., 1990) but did not enhance the levels of 2E1 protein (Thomas et al., 1987). Therefore, studies of the selectivity of these substrates for 2E1 in monkeys pretreated with the more common classic inducers of P450 isoforms should be performed. In particular, because the P450 1A1 and 1A2 proteins are largely inducible by polycyclic aromatic compounds in many species, including monkeys (Bullock et al., 1995), the role of these P450 isoforms in converting CLZ to its 6-hydroxy derivative should be assessed before this activity is used as a 2E1 probe in liver from monkeys treated with 1A1/2-inducing drugs.

Because, for several reasons (including the toxicity and the extensive biotransformation of many 2E1 substrates), only the 6-hydroxylation of CLZ has been widely used as a valuable probe to test P450 2E1 activity in vivo in humans (Girre et al., 1994), it seems reasonable, on the basis of our findings, that this assay might be adopted for use in monkeys. Naturally, validation of the sensitivity and selectivity of this assay for in vivo evaluation of the P450 2E1 content of cynomolgus monkeys requires further investigation.