Enzymatic characterization of the polynuclear aromatic hydrocarbons activating rat-liver preparations used in the mutagenicity test of Ames

Enzymatic characterization of the polynuclear aromatic hydrocarbons activating rat-liver preparations used in the mutagenicity test of Ames

Mutation Research, 125 (1984) 123-133 123 Elsevier MTR 03796 Enzymatic characterization of the polynuclear aromatic hydrocarbons activating rat-liv...

867KB Sizes 0 Downloads 33 Views

Mutation Research, 125 (1984) 123-133

123

Elsevier MTR 03796

Enzymatic characterization of the polynuclear aromatic hydrocarbons activating rat-liver preparations used in the mutagenicity test of Ames M. Tauc 1, M. Hermann 1, P.M. Dansette 2 and J.P. Vandecasteele 1 I Direction Environnement et Biologie POtrolibre, Institut Franqais du P~trole, 92506 Rueil-Malmaison, and 2 Ecole Normale Sup~rieure, 24 rue Lhomond, 75231 Paris (France)

(Received 2 March 1983) (Revision received 27 July 1983) (Accepted 8 August 1983)

Summary (1) Stability studies were performed on the mono-oxygenase system involved, in particular, in the activation of polynuclear aromatic hydrocarbons (PAHs) present in rat-liver preparations used in the Ames mutagenicity test. The results indicated a good stability of the spectral response of the cytochrome-P-450 system, but a much lower stability of its enzymatic activities measured with various substrates, thus showing the inadequacy of the spectral response to characterize the PAH mono-oxygenase activity of the preparations. Epoxid¢ hydrolase activity was found to be stable. (2) Various mono-oxygenase activities were measured in preparations induced with phenobarbital, 3-methylcholanthrene or Aroclor 1254. The ac.tivities of two enzymes, benzo[a]pyrene hydroxylase and ethoxyresorufin-O-dealkylase, were found suitable to characterize the capacity of the preparations to metabolize PAH to mutagens. (3) The efficiency of the same preparations to promote the mutagenicity of benzo[a]pyrene and aflatoxin Bt in the Ames test was determined. There was an excellent general correlation between the efficiencies for mutagenic activation of the preparations and the two enzymatic activities mentioned above. (4) Determination of ethoxyresorufin-O-dealkylase (or benzo[a]pyrene hydroxylase) and benzo[a]pyrene 4,5-oxide hydrolase activities is proposed for characterizing the rat-liver preparations used in the Ames test.

Among polycyclic aromatic hydrocarbons (PAHs), whose carcinogenic properties have been known for a long time, benzo[a]pyrene (B[a]P) is one of the most and best characterized carcinogenic agents, as well as a good indicator of PAH produced by combustion processes. Actually, many individual PAHs are carcinogenic. In recent years, Ames and co-workers have reported the correlation between mutagenicity and carcinogenicity (McCann et al., 1975; McCann and Ames, 1976) and made possible the use of a simple bacterial mutagerticity test for the detection of carcinogenic 0027-5107/84/$03.00 © 1984 Elsevier Science Publishers 1~.v.

compounds (Ames et al., 1973, 1975). In the case of PAH, however, metabolic activation into mutagenic intermediates of the compounds being tested has to be brought about by incubation in the presence of rat-liver homogenates. These homogenates contain the enzymes involved in PAH metabolism, cytochromes P-450 and P-448, a complex array of isoenzymes capable of oxygenating a variety of xenobiotics, and in particular, of epoxidizing PAH (Conney, 1967) and epoxide hydrolase which hydrolyses the arene oxides formed from PAH into the corresponding dihydrodiols (Oesch,

124

1973). Although epoxides are generally recognized as the ultimate mutagens/carcinogens derived from PAH (Ames et al., 1972; Grover and Sims, 1973; Blobstein et al., 1976), the whole process whose physiological role is the detoxication of xenobiotic substances, is complex because PAH can be epoxidized in several locations and give rise to a variety of mutagenic intermediates (Capdevila et al., 1975; King et al., 1976; Levin et al., 1976). This activation step is only poorly controlled in the Salmonella/mammalian microsome test of Ames since even carefully defined procedures for obtaining induced rat-liver homogenates were una~91e to yield extracts with completely reproducible activation properties. The development of reliable methods to determine the enzymatic activities involved in PAH activation in the homogenates used was an obvious first step towards the definition of standardized enzyme preparations. As part of our studies on the mutagenicity of petroleum fractions and other PAH-containing preparations (Hermann et al., 1980a, b; Hermann, 1981), the present work aimed at improving the control of the activation step of the Ames mutagenicity test, mainly by defining methods for the determination of both types of enzymatic activity involved in PAH activation: cytochromes P-450 and P-448 and epoxide hydrolase which are contained in rat-liver homogenates used in the standard Ames test (Ames et al., 1975). Special attention was given to the initial oxygenation step because of the variety of isoenzymes of overlapping specificity contained in rat-liver homogenates and of the substrates undergoing oxygenation by these enzymes. Materials

and methods

Enzyme preparations Rat livers were prepared by the "Drpartement Recherche et Essais Biologiques Stallergenes" (D.R.E.B.S.), Joinville, France. They were obtained from male Sprague-Dawley rats weighing 220-240 g that had been induced with one of the following substances: Aroclor 1254, phenobarbital and 3-methylcholanthrene. Aroclor 1254 was dissolved in corn oil and administered as a single intraperitoneal injection (500 mg/kg) 5 days before the rats were killed, as described by Ames et

al. (1975). Phenobarbital was distributed in drinking water (solution at 0.1%) during the 5 days before the rats were killed. Intraperitoneal injection of 3-methylcholanthrene (50 mg/kg body wt.) dissolved in corn oil was performed 2 days before the rats were killed. Livers from at least 6 rats (8 for 3-methylcholanthrene-induced rats) were used in each group of experiments. Homogenates were prepared according to Ames et al. (1975) and yielded, after centrifugation at 9000 × g, the supernatant fractions ($9) used. Microsomes were obtained from a 100 000 × g centrifugation of $9 fractions. The pellets were washed and resuspended in 5 mM potassium phosphate buffer, pH 7.4, containing 20% (w/v) glycerol and 0.1 mM EDTA. These operations were performed at 4°C, and the microsomal fractions were stored at - 80°C.

Determination of enzyme activities Ethoxycoumarin O-dealkylase and ethoxyresorufin O-dealkylase activities were measured at 37°C with an "Eppendorf" fluorimeter equipped respectively for these assays with 317-366 and 546 nm primary filters and with 470-3000 and 560-3000 nm secondary filters. The incubation medium contained, in a final volume of 2 ml: 100 mM Tri~-HcI buffer, pH 7.6; 60/zM NaDPH; 15 mM MgC12; 0.1 mM 7-ethoxycoumarin or 0.1 mM ethoxyresorufin. These methods were derived from those of Ullrich and Weber for ethoxycoumarin (Ullrich and Weber, 1972; Prough et al., 1978) and of Burke and Mayer for ethoxyresorufin (Prough et al., 1978; Burke and Mayer, 1974). Formaldehyde production in the N-demethylase assay of aminopyrine and benzphetamine was determined according to the method of Nash (1953). The incubation medium used for aminopyrine has been described by Werringloer (1978). For benzphetamine, the incubation medium contained, in a final volume of 6 ml: 100 mM potassium phosphate buffer, pH 7.4; 1 mM benzphetamine; 0.25 mM NADPH; 2.5 mM 5'-AMP; 10 mM MgC12; 3 mM glucose 6-phosphate; 0.5 unit glucose-6-phosphate dehydrogenase. For determination of lauric acid to-hydroxylase activity, the method used was adapted from Marchal et al. (1982). The incubation medium contained, in a final volume of 1 ml: 100 mM potassium phos-

125 phate buffer, pH 7.4; 1 mM NADPH; 123 #M [14C]lauric acid (0.195 #Ci/#mole). The amount of microsomal suspension used was 1-1.5 mg protein. The reaction was stopped after 20 min at 30°C by 50 #1 of 50% H2SO4. All metabolites were extracted with 4 ml of ether. The lighter fraction was recovered after freezing the aqueous phase in an acetone-dry ice mixture. The ether was evaporated, and the dry residue was taken up with 200 ~1 ether and chromatographed on thin-layer plates of silica gel 60 (0.25 mm thickness) from Merck, Darmstadt, F.R.G. Chromatograms were developed on a length of 15 cm with hexane-ether-acetic acid (60 : 39 : 1 (v/v/v)), Radioactive spots corresponding to lauric acid and to ~-hydroxylauric acid were detected by autoradiography, and their activities were determined by liquid scintillation by using 4 g / l 2,5-diphenyloxazole and 0.5 g/1 p-bis(O-methylstyryl)benzene dissolved in toluene as scintillation mixture. Calculation of enzymatic activity was done as described by Marchal et al. (1982). For benzo[a]pyrene hydroxylase activity a method derived from that used by Jerina et al. (1977) for epoxide hydrolase activity was devised. The incubation medium, contained in a final volume of 80 F1, was: 70 mM potassium phosphate buffer, pH 7.4; 1.5 mM NADPH; 92 FM [14C]benzo[a]pyrene (2.17/~Ci/#mole) in acetone. The amount of microsomal suspension used had 100-150 #g protein. The reaction was performed at 37°C for 6 min and stopped by 40 #1 of tetrahydrofuran. Of the incubation medium, 60 F1 was chromatographed on thin-layer plates of silica gel (LK6D from Whatman) with hexane/acetone (80 : 20 (v/v)) as solvent. This solvent was selected as it allowed a good separation between benzo[a]pyrene and its more polar metabolites which were themselves resolved into a series of radioactive spots after detection by autoradiography. The activities contained in the spots corresponding to benzo[a]pyrene on the one hand and in its metabolites taken together on the other hand were measured and expressed as for lauric acid. Epoxide hydrolase activity was determined according to the method described by Jerina et al. (1977). NADPH cytochrome c reductase activity was determined according to Vermillion and Coon (1978).

Spectral evaluation of cytochrome P-450 was done according to Omura and Sato (1964a, b). Protein content of the microsomal fraction was determined by the biuret reaction (Layne, 1957) after solubilization with cholic acid.

Mutagenicity assays The Salmonella typhimurium strain TA98 was used throughout this work and was obtained through the courtesy of Prof. B.N. Ames (Berkeley, CA, U.S.A.). Strain TA98 was checked in each experiment for amp r, rfa and uvrB. Rfa was checked on McConkey medium plates on which rfa strains are growth deficient because of bile salts. The gal character was used as an indicator of the presence of uvrB (since gal and uvrB are included in the same deletion). The gal character was checked on EMB gal plates. Mutagenesis assays were carried out by following closely the published procedure of Ames et al. (1975), except that 2.5 ml instead of 2 ml of top agar were added to Vogel-Bonner plates. The efficiency of reversion to histidine prototrophy of the bacterial culture without $9 mix was checked in each experiment with 2-nitrofluorene (1 and 2 #g per plate). The efficiency of the rat-liver preparations for mutagenic activation was checked by measuring the mutagenicity of B[a]P and aflatoxin B1. The carrier solvent used was dimethyl sulphoxide. Dose-response curves were plotted for each amount of $9 tested, and the slope of the linear portion of each curve was taken as expressing the mutagenicity of the compound under the test conditions used.

Chemicals Chemicals were obtained as follows: aflatoxin B~, benzo[a]pyrene, ethoxycoumarin and 4-dimethylaminoantipyrine (aminopyrine), from Aldrich; ethoxyresorufin from Pierce; [1-14C]lauric acid (15-30 mCi/mmole) and [7,10-14C]benzo[a]pyrene (5-15 mCi/mmole) from Amersham. Benzphetamine was a gift from J. Dayan (Laboratoire National de la Santr, Paris). [3n]Benzo[a]pyrene-4,5-oxide and benzo[a]pyrene-4,5dihydrodiol were synthesized as described elsewhere (Jerina et al., 1977).

126 Results

Stability of the enzymes involved in the metabolism of l, Ai4 The Salmonella/mammalian-microsome mutagenicity test involves an activation step characterized by a long incubation period. This step is presently poorly controlled as the stability of the enzymes involved (mono-oxygenase system and epoxide hydrolase) over such a long period is not known. It seemed useful to clarify this problem first. Accordingly, the stability of cytochrome P450 was studied in the 9000 × g centrifugation supernatant fraction ($9) of Aroclor-induced ratliver homogenates, that is, the preparation used in the mutagenicity test developed by Ames et al. (1975). The stability of the mono-oxygenase system was determined in two ways: the stability of the enzymatic activity and the stability of the absorption spectra at 450 nm. The results are shown in Fig. 1. The stabilities were established at two temperatures, 4 and 25°C, with aminopyrine as substrate. The enzymatic activity exhibited a poor stability at 25°C, 50% being lost in 24 h. At 4°C, the stability was better, the half-life being 5 days. A possible explanation of % of

ontrol o

o

this loss of activity could be the action of proteolytic enzymes. However, when proteolysis inhibitors (0.1 m M phenylmethyl sulphonyl fluoride, plus 1 m M EDTA) were added, no improvement of enzymatic stability was observed. On the contrary, the spectral stability was excellent. There was a marked increase of the spectral response with respect to the control, at 4°C as well as at 25°C. We have no explanation for this puzzling behaviour. In addition, the absorption spectra did not reveal any increase of cytochrome P-420, a denatured form of cytochrome P-450. On the whole, however, the good stability of the spectral response was in agreement with generally held views on this matter, b u t the low stability of enzymatic activity on which data in the literature are scarce could possibly be attributed to the use of a particular substrate and then could reflect the inactivation of only one of the cytochrome P-450 isoenzymes. This does not seem to be so, as similar results were obtained when ethoxycoumarin was used as the substrate. Further experiments were conducted with the microsomal fraction, as this preparation was judged more suitable for the work on enzyme characterization. The stability of the enzymatic activities of the three main enzymes participating in the metabolism of PAH (mono-oxygenase, epoxide

% Activity 100

100

50

m

10

20Time (days)

Fig. 1. Stability of rat-liver cytochrome P-450. Comparison between the stability of the spectral response at 4°C (©) and 25oc (A) and the stability of the enzymatic activity for aminopyrine at 4°C (O) and 25°C (A) with Aroclor-induced rat-liver $9. The incubation medium for the N-demethylase assay of aminopyrine contained: 100 mM KH2PO4, pH 7.4; 5 mM 4-dimethylaminoantipyrine; 0.25 mM NADPH; 2 mM 5'-AMP; 10 mM MgCI2; 3 mM glucose 6-phosphate; 0.5 unit glucose-6-phosphatedehydrogenase.

10

~

/__

5

10

I

I___

15 20 Time (days)

Fig. 2. Stability of the enzymes involved in the metabolism of polycyclic aromatic hydrocarbons. The experiments were performed at 25°C with Aroclor-induce,d microsomal fraction containing 20% (w/v) glycerol. Mono-oxygenase activity ([3) was measured with 123 /~M [14C]lauric acid as substrate, NADPH cytochromec reductase (A) with 74 #M cytochromec, and epoxide hydrolase (O) with 97 /~M [3H]benzo{a]pyrene 4,5-oxide.

127

1972; Lu and West, 1972; Burke and Mayer, 1975). Phenobarbital induces cytochrome P-450, a cytochrome with an absorption maximum at 450 rim, preferentially metabolizing substrates like benzphetamine and aminopyrine, whereas Aroclor induces both cytochromes (Ryan et al., 1977). The results are summarized in Table 1. Considering the efficiency of induction of the total amount of cytochrome P-450, the best inductions were obtained with Aroclor and phenobarbital. Apparently, treatment with methylcholanthrene produced only a moderate increase of the amount of cytochrome P-450. The enzymatic activity of these preparations for the various substrates used shows that the various inducers utilized induced different enzymatic activities. However, benzphetamine has been described as being preferentially metabolized by the species of cytochrome absorbing at 450 nm (Ryan et al., 1977), but the present results indicate that this substrate is not really specific for the phenobarbital-induced species. Ethoxycoumarin exhibits some specificity for the methylcholanthrene-induced cytochrome, but can also be metabolized by the phenobarbital-induced cytochrome P-450. The Aroclor-induced preparations showed no specificity for a particular substrate. This is expected, because Aroclor is known to induce the two main cytochrome species (Ryan et al., 1979). An interesting result was obtained when lauric acid was used as substrate. Lauric acid was metabolized by both cytochrome species (Table 1),

hydrolase and NADPH cytochrome c reductase) was measured. 20% (v/v) glycerol was added as stabilizer, and the preparation was incubated at 25°C. Compounds used as substrates of monooxygenase included ethoxycoumarin, ethoxyresorufin, aminopyrine, benzphetamine and lauric acid. As detailed later, the metabolism of some of these involves different mono-oxygenase isoenzymes, but similar results were obtained for all of them. From lauric acid, illustrated in Fig. 2, the loss of enzymatic activity was 50% in 8 days. These results confirm the known stabilizing effect of glycerol, but also indicate that a number of different cytochrome P-450 isoenzymes have similar stabilities. In the same conditions, the loss of activity of NADH cytochrome c reductase was 50% in 11 days, whereas epoxide hydrolase activity was unchanged after 20 days (Fig. 2).

Characterization of the mono-oxygenase system by its enzymatic activities To characterize the enzymatic activity correlated to the epoxidation of PAH, we attempted to identify substrates specific for the isoenzymes involved. Six substrates were tested by using rat-liver microsomal fractions induced with the following substances: Aroclor 1254 (a polychlorinated biphenyl), 3-methylcholanthrene and phenobarbital. It is known that 3-methylcholanthrene induces cytochrome P-448, a cytochrome with an absorption maximum at 448 nm (Sladeck and Mannering, 1966; Alvares et al., 1967) preferentially metabolizing PAH and ethoxyresorufin (Lu et al., 1971, TABLE 1

EFFECT OF INDUCTION WITH VARIOUS COMPOUNDS ON THE MONO-OXYGENASE ACTIVITY WITH VARIOUS SUBSTRATES With each substrate, the activity is expressed in nmoles/min/mg of protein (a) and in nmoles/min/nmole of cytochrome P-450 (b) determined according to Omura and Sato (1964a, b). Inducer

None (control) Arocior 1254 Methylcholanthrene Phenobarbital

Total protein of microsomal suspension (mg/ml) 10.4 14.3 11.8 13.6

Mono-oxygenase activity with

Specific content of cytochrome P-450 (nmoles/mg protein)

ethoxycoumarin

ethoxyresorufin

benzol a ]pyrene

benzphetamine

lauric acid

a

b

a

b

a

b

a

b

a

b

0.42 2.45 0.6 1.53

0.49 11.4 4.3 2

1.16 4.67 7.20 1.31

0.36 55.6 26.6 0.74

0.85 22.8 44.4 0.48

0.08 2.23 1.40 0.30

0,18 0,91 2.33 0.19

16 42 13.2 55.5

38.1 17.1 21.9 36.2

1.26 0.78 2.15 1.55

3 0.32 3.6 1

128 but with Aroclor-induced enzymes, it was evidently poorly metabolized because the measured activity was less than that of the non-induced control. This is remarkable, and one can ask whether the two Aroclor-induced enzymes are the same as those induced respectively by methylcholanthrene and phenobarbital (Ryan et al., 1979), or if a minor species of cytochrome P-450 would be very active for this particular hydroxylation. One of the most interesting substrates, considering the aim of the present work, is ethoxyresorufin. The activity of this substrate in the methylcholanthrene-induced preparations was 50 times that of the non-induced preparations, in agreement with the results of Burke and Mayer (1975). Table 1 indicates that, because methylcholanthrene did not induce the synthesis of substantially higher total amounts of cytochrome P-450, there was selective induction of a particular isoenzyme with a high specificity for ethoxyresorufin. Similar results were obtained with benzo[a]pyrene (Table 1), a PAH used as a reference compound in the Ames mutagenicity test.

Induction of other enzymes involved in the metabolism of PAH As already discussed, the use of the Ames test with PAH requires an activation step in the presence of induced rat-liver homogenates. Several enzyme systems such as glutathione S-transferase, which is known to carry out conjugation of arene oxides, are present in $9 and participate in PAH metabolism. However, a particular point of inter-

est for us was that the mutagenicity level is probably dependent on the ratio of the enzymatic activities of the mono-oxygenase system (in this case cytochrome P-448) and epoxide hydrolase, two enzymes exerting antagonistic actions (Bentley et al., 1977; Oesch et al., 1977). Therefore, it appeared important to consider the effect of various inducers on the activity of epoxide hydrolase. Induction of the N A D P H cytochrome c reductase, another enzyme participating in the process by making possible the electron transfer from N A D P H to cytochromes P-450 or P-448, was also studied. Table 2 shows the activities of these two enzymes in comparison with the induction of cytochrome P-450 by substances cited above. N A D P H cytochrome c reductase, apparently, was co-induced with cytochrome P-450. We expected this result because the two enzymes are functionally interdependent, although it has been reported that N A D P H cytochrome c reductase is not co-induced with cytochrome P-450 in rat liver (Gnosspelius et al., 1969-1970; Glauman, 1970). Epoxide hydrolase was induced by phenobarbital as reported previously (Bresnick et al., 1977; Schassmann and Oesch, 1978), and also by Aroclor. Several metabolites produced by cytochrome P-450 can serve as substrates for epoxide hydrolase (Jerina and Daly, 1974). We have confirmed this. Thus, the amount of epoxide hydrolase present in the rat-liver preparations used is an additional parameter in the Ames test (Bentley et al., 1977; Oesch et al., 1977).

TABLE 2 INDUCTION BY AROCLOR 1254, 3-METHYLCHOLANTHRENEAND PHENOBARBITALOF MICROSOMAL EPOXIDE HYDROLASE AND NADPH CYTOCHROMEc REDUCTASE The specific content of cytochrome P-450, obtained by spectral determination, is given for comparison. (a) Activity expressed in nmoles of [3H]benzo[a]pyrene-4,5-dihydrodiol/min/mg protein at 37°C. (b) Activity expressed in nmoles of cytochrome c redticed/min/mg protein at 25~C. Inducer

None (control*) Aroclor 1254 3-Methylcholanthrene Phenobarbital

Specificcontent of cytochromeP-450 (nmoles/mg protein)

Enzymaticactivityof epoxidehydrolase (a)

cyt. P-450reductase (b)

0.3 1.46 0.42 2.9

7 10 8.4 22

122 201 154 400

129

Comparison between enzymatic and biological activities of various induced preparations As discussed above, ethoxyresorufin seems to be a specific substrate for the cytochrome responsible for PAH activation. To confirm this hypothesis, we performed mutagenicity tests using benzo[a]pyrene and aflatoxin B 1 a s mutagenic agents and rat-liver $9 induced by various treatments as activating preparations. For each of these mutagens, a series of dose-response curves was obtained, as described by Ames et al. (1975), each curve being obtained with a given amount of $9 enzyme preparation. The slope of the linear portion of each curve, which expresses the mutagenicity of the compound in the test conditions used, was plotted against the amount of $9 used for the activation step. The results presented in Fig. 3 show that the mutagenicity responses of benzo[a]pyrene and aflatoxin B~ to the amount of activating enzymes used were different. However, for both mutagens, the methylcholanthrene-induced preparations gave the highest mutagenicity values. Non-induced or phenobarbital-induced preparations gave the lowest values. Intermediate values were obtained with the enzymatic system induced by Aroclor. We selected the mutagenicity values obtained with 50 #1 of each enzymatic preparation to express the efficiency for mutagenic activation of these preparations for comparison with their enzymatic activities for benzo[a]pyrene and ethoxyresorufin. The results presented in Table 3 illustrate the close

®

~1500

i

®

10000

E 1000

[3

5000

500

50

50

100 AmHnt0f S, perplatelldl

lon

AmountOl $, per platelldl

Fig. 3. Efficiency for mutagenic activation of various $9 preparations. This efficiency was determined for benzo{a]pyrene (A) and aflatoxin B1 (B). For each amount of enzyme tested, a dose-response curve was obtained, the amounts of mutagen having a range of 0 - 3 #g per plate for benzo[a]pyrene and 0-0.5 #g per plate for aflatoxin Br The slope of the linear portion of each curve, which is defined as the mutagenicity of the compound in the presence of the enzyme preparation added and is expressed as the number of revertants per #g of compound, is plotted versus the amount of enzyme preparation used. $9 liver homogenate was from non-induced rats (A), and from rats induced with phenobarbital (e), Aroclor 1254 (O) or 3-methylcholanthrene (to).

correlation observed between the efficiency of the preparations for mutagenic activation and their enzymatic activities with the two substrates utilized, the three biological properties examined being highest with methylcholanthrene-indueed

TABLE 3 COMPARISON BETWEEN THE EFFICIENCY FOR MUTAGENIC ACTIVATION OF VARIOUS $9 PREPARATIONS WITH THEIR MONO-OXYGENASE ACTIVITIES FOR ETHOXYRESORUFIN AND BENZO[a]PYRENE The mutagenicity of each compound (slope of the linear portion of the dose-response curve expressed as the amount of revertants per t~g of compound) for an amount of $9 of 50 #1 per plate has been taken to represent the efficiency for mutagenic activation of benzo[a]pyrene and aflatoxin B l of the $9 preparation studied (a). Enzymatic activities, determined as for Table 1, are expressed in nmoles of p r o d u c t / m i n / m g protein (b) or in nmoles of product/nmole of cytochrome P-450 (c). Inducer

None (control) Aroclor 1254 3-Methylcholanthrene Phenobarbital

Efficiency for activation of

Mono-oxygenase activity with

benzo{a]pyrene

aflatoxin B a

ethoxyresorufin

benzo[a]pyrene

(a)

(a)

b

c

b

c

100 410 1020 50

1600 22 000 63 000 14000

0.36 55.6 26.6 0.74

0.85 22.8 44.4 0.48

0.08 2.23 1.4 0.3

0.18 0.91 2.33 0.19

130 preparations, intermediate with those induced by Aroclor and lowest with those induced with phenobarbital. The intermediate result obtained with Aroclor, which has been reported the best inducer for the detection of various carcinogens (Ames et al., 1975), is noteworthy. Discussion

The main result of our studies of cytochrome P-450 stability has been to show the lack of correlation between the spectral signal and the enzymatic activity. To our knowledge, this lack of correlation had not yet been reported and, in fact, the stability of the spectral signal led various authors to postulate a similar stability for the enzymatic activity, and consequently, to purify the enzymatic system at room temperature (Ryan et al., 1979; Warner et al., 1978). It is possible that the limited enzymatic stability in these conditions could account for some of the difficulties in re-associating the mono-oxygenase system into an active system after purification, any loss of enzymatic activity being undetected when reliance on spectral determinations alone is made to follow the purification procedure. The quantitative knowledge of enzymatic activity is especially necessary when purified enzymes are used in the Ames mutagenicity test. From our results, it is clear that spectral determinations are not suitable for characterizing the activity of the monooxygenase system. Also, the absence from the spectral response of the peak at 420 nm characterizing cytochrome P-420 cannot be considered as indicating the presence of an active monooxygenase system. So, to characterize preparations used in the Ames test, it is necessary to determine directly the enzymatic activity of the monooxygenase system. The use of a PAH, such as benzo[a]pyrene, for the determination of the enzymatic epoxidation activity of the preparations gives values that are well correlated with the efficiency of these preparations for mutagenic activation, but this assay measures an overall result of a set of complex reactions. Actually, in the rnicrosomal fraction, the presence of the mono-oxygenase system and of epoxide hydrolase leads to a complex process whereby the initial benzo[a]pyrene oxidation product after hydrolysis to dihydrodiols or transposition to phe-

nols can again be epoxidized by cytochrome P-450 (Capdevila et al., 1975; King et al., 1976). This complexity is avoided when a non-PAH substrate, ethoxyresorufin, is used. Enzyme induction studies have shown that this substrate is representative of the PAH epoxidation activity of the preparation. In addition, the enzymatic test with this substrate is highly sensitive and easy to handle. Finally, the results obtained from the mutagenicity test, in which benzo[a]pyrene and aflatoxin B1 were used as mutagenic agents, show that an excellent correlation also holds between this enzymatic activity and the efficiency of mutagenic activation. Hence, the measurement of enzymatic activity for ethoxyresorufin makes it possible to determine the active cytochrome P-448 content in rat-liver homogenate and to standardize the activation step of PAH in preparations for the mutagenicity tests. The induction studies have shown that epoxide hydrolase is co-induced, as well as NADPH cytochrome c reductase, with the mono-oxygenase system, by the various drugs used (Bentley and Oesch, 1978). On the other hand, the efficiency of mutagenic activation of the preparations is dependent on the levels of the two antagonistic induced enzyme systems, mono-oxygenase and epoxide hydrolase (Lu and Miwa, 1980), and probably on the presence of other microsomal components of ratliver $9 (Malaveille et al., 1979). The different shapes exhibited by the mutagenicity-response curves of benzo[a]pyrene and aflatoxin B1 to the addition of activating enzymes (Fig. 3) are worth comment in this respect. Both compounds are apparently activated by the same oxygenase species induced by methylcholanthrene. Methylcholanthrene therefore appears as the inducer to be used in studies of aflatoxin B~ activation rather than phenobarbital as is usual. However, the mutagenicity response of benzo[a]pyrene presents a maximum for a defined amount of added enzyme and then decreases, whereas that of aflatoxin B~ does not. The existence of a maximum in the curve of benzo[a]pyrene has already been reported (Ames et al., 1975; Bentley and Oesch, 1978; Malaveille et al., 1979) and possible explanations have been proposed (Malaveille et al., 1979). One may suggest another plausible explanation which is the

131

antagonistic action of epoxide hydrolase towards that of mono-oxygenase, since epoxide hydrolase is involved in the metabolism of PAH but not in that of aflatoxin Br These considerations lead to the conclusion that, along with the determination of the monooxygenase activity with ethoxyresorufin, the determination of epoxide hydrolase by the radiometric assay (Jerina et al., 1977; Dansette, 1980) would improve the control and the interpretation of the mutagenicity test of Ames. However, full understanding of the respective effects of these two enzymatic activities on PAH activation and achievement of complete reproducibility of the activation step in the Ames test require the ability to use controlled amounts of each of them. This can only be obtained by separating and recombining these two enzymatic activities as already shown by several authors (Bentley and Oesch, 1978). For this purpose, we experimented with a simple purification procedure performed at 4°C which segregates the three main enzymes (cytochromes P-450 and P-448, epoxide hydrolase and NADPH cytochrome c reductase). The microsomal fraction from Aroclor-treated rats was solubilized at low ionic strength (5 mM KH2PO4) in the presence of 0.5% cholate and 0.2% Emulgen 911 from Kao-Atlas. Cholate was then removed by treatment through a Dowex AG 1 × 2 column and the eluate was chromatographed through a DEAE-cellulose column. The enzymatic fractions obtained after elution with a KC1 gradient were treated with Biobeads SM 2 resin, from Biorad. This procedure did not produce pure enzymes (which was not the goal intended), but fractions in which the three enzymes were well separated. The cytochrome fractions contained most of the cytochrome P-450 isoenzymes as deduced from the high yield of purification (60% in spectral estimation) in contrast to the low yield (4%) obtained by others in more complete procedures yielding highly purified enzymes (Jerina and Daly, 1974). The mono-oxygenase system could be re-associated in an active form after this treatment. We are now experimenting on the use of wellZchecked mixtures of mono-oxygenase and epoxide hydrolase previously separated by the above method in mutagenicity tests. The efficiency for mutagenic activation of these preparations,

containing most of the multiple oxygenase isoenzymes, should be interesting for comparison with that of mixtures of completely purified enzymes, which have already been used by others (Levin et al., 1977). References Alvares, A.P., G. Schilling, W. Levin and R. Kuntzman (1967) Induction of CO-binding pigments in liver microsomes by phenobarbital and 3-methylcholanthrene, Biochem. Biophys. Res. Commun., 29, 521. Ames, B.N., P. Sims and P.L. Grover (1972) Epoxides of carcinogenic polycyclic hydrocarbons are frameshift mutagens, Science, 176, 47. Ames, B.N., W.E. Durston, E. Yamasaki and F.D. Lee (1973) Carcinogens are mutagens, Simple test system combining liver homogenates for activation and bacteria for detection, Proc. Natl. Acad. Sci (U.S.A.), 70, 2281. Ames, B.N., J. McCann and E. Yamasaki (1975) Methods for detecting carcinogens and mutagens with the Salmonella/ mammalian-microsome mutagenicity test, Mutation Res., 31,347. Bentley, P., and F. Oesch (1978) Enzymes involved in activation and inactivation of carcinogens and mutagens, in: H. Returner, H.M. Bolt, P. Bannasch and H. Popper (Eds.), Primary Liver Tumors, Falk Symposium, 25, MTP Press, p. 239. Bentley, P., F. Oesch and R. Glatt (1977) Dual role of epoxide hydrase in both activation and inactivation of benzo[a]pyrene, Arch. Toxicol., 39, 65. Blobstein, S.H., I.B. Weinstein, P. Dansette, H. Yagi and D.M. Jerina (1976) Binding of K- and non-K-region arene oxides and phenols of polycyclic hydrocarbons to polyguanylic acid, Cancer Res., 36, 1293. Bresnick, E., H. Mukhtar, T.A. Stoming, P.M. Dansette and D.M. Jerina (1977) Effect of phenobarbital and 3-methylcholanthrene administration on epoxide hydrase in liver microsomes, Biochem. Pharmacol., 26, 891. Burke, M.D., and R.T. Mayer (1974) Ethoxyresorufin, Direct fluorimetric assay of a microsomal O-dealkylation which is preferentially inducible by 3-methylcholanthrene, Drug Metab. Disp., 2, 583. Burke, M.D., and R.T. Mayer (1975) Inherent specificities of purified cytochromes P-450 and P-448 towards biphenyl hydroxylation and ethoxyresorufin deethylation, Drug Metab. Disp., 3, 245. Capdevila, J., B. Jernstrom, H. Vadi and S. Orrenius (1975) Cytochrome P-450 linked activation of 3-hydrobenzo[a]pyrene, Biochem. Biophys. Res. Commun., 65, 894. Conney, A.H. (1967) Pharmacological implications of microsomal enzyme induction, Pharmacol. Rev., 19, 317. Dansette, P.M. (1980) Epoxide hydrolase microsomale, Ann. Biol. Clin., 38, 25. Glauman, H. (1970) Chemical and enzymatic composition of microsomal subfractions from rat liver after treatment with

132 phenobarbital and 3-methylcholanthrene, Chem.-Biol. Interact., 2, 369. Gnosspelius, Y., H. Thor and S. Orrenius (1969-1970) Comparative study on the effects of phenobarbital and 3,4-benzopyrene on the hydroxylating enzyme system of rat liver microsomes, Chem.-Biol. Interact., 1, 125. Grover, P.L., and P. Sims (1973) O-Dealkylation of 7ethoxycoumarin by liver microsomes; direct fluorimetric test, Biochem. Pharmacol., 22, 661. Hermann, M. (1981) Synergistic effects of individual polycyclic aromatic hydrocarbons on the mutagenicity of their mixtures, Mutation Res., 90, 399. Hermann, M., O. Chaude, N. Weill, H. Bedouelle and M. Hofnung (1980a) Adaptation of the Salmonella/mammalian microsome test to the determination of the mutagenic properties of mineral oils, Mutation Res., 77, 327. Hermann, M., J.P. Durand, J.M. Charpentier, O. Chaude, M. Hofnung, N. Petroff, J.P. Vandecasteele and N. Weill (1980b) Correlations of mutagenic activity with polynuclear aromatic hydrocarbons content of various mineral oils, in: A. Bj~rseth and A.J. Dennis (Eds.), Polynuclear Aromatic Hydrocarbons: Chemistry and Biological Effects, Battelle Press, Columbus, OH, pp. 899. Jerina, D.M., and J.N. Daly (1974) Arene oxides: a new aspect of drug metabolism, Science, 185, 573. Jerina, D.M., P.M. Dansette, A.Y.H. Lu and W. Levin (1977) Hepatic microsomal epoxide hydrase: a sensitive radiomettic assay for hydration of arene oxides of carcinogenic aromatic hydrocarbons, Moi. Pharmacol., 13, 342. King, H.W.S., M.H. Thompson and P. Brookes (1976) The role of 9-hydroxybenzo[a]pyrene in the microsome mediated binding of benzo[a]pyrene to DNA, Int. J. Cancer, 18, 339. Layne, E. (1957) Spectrophotometric and turbidimetric methods for measuring proteins, Methods Enzymol., 3, 447. Levin, W., A.W. Wood, H. Yagi, D.M. Jerina and A.H. Conney (1976) ( + )-trans-7,8-Dihydrobenzo[a ]pyrene: a potent skin carcinogen when applied topically to mice, Proc. Natl. Acad. Sci. (U.S.A.), 73, 3867. Levin, W., A.W. Wood, A.Y.H. Lu, D. Ryan, S. West, A.H. Conney, D.R. Thakker, H. Yagi and D.M. Jerina (1977) Role of purified cytochrome P-448 and epoxide hydrase in the activation and detoxification of benzo[a]pyrene, in: Drug Metabolism Concepts, A.C.S. Symposium Series No. 44, American Chemical Society, Washington, DC, p. 99. Lu, A.Y.H., and G.T. Miwa (1980) Molecular properties and biological functions of microsomal epoxide hydrase, Annu. Rev. Pharmacol. Toxicol., 20, 513. Lu, A.Y.H., and S.B. West (1972) Reconstituted liver microsomal enzyme system that hydroxylates drugs, other foreign compounds, and endogenous substrates, III. Properties of the reconstituted 3,4-benzopyrene hydroxylase system, Mol. Pharmacoi., 8, 490. Lu, A.Y., R. Kuntzman, S. West and A.H. Conney (1971) Reconstituted liver microsomal enzyme system that hydroxylates drugs, other foreign compounds, and endogenous substrates, I. Determination of substrate specificity by the cytochrome P-450 and P-448 fractions, Biochem. Biophys. Res. Commun., 42, 1220.

Lu, A.Y.H., R. Kuntzman, S.B. West, M. Jacobson and A.H. Conney (1972) Reconstituted liver microsomal enzyme system that hydroxylates drugs, other foreign compounds, and endogenous substrates, II. Role of the cytochrome P-450 and P-448 fractions in drug and steroid hydroxylations, J. Biol. Chem., 247, 1727. Malaveille, C., T. Kuroki, G. Braun, A. Hautefeuille, A.M. Camus and H. Bartsch (1979) Some factors determining the concentration of liver proteins for optimal mutagenicity of chemicals in the Salmonella/microsome assay, Mutation Res., 63, 245. Marchal, R., M. Metche and J.P. Vandecasteele (1982) Hydroxylation of alkanes and fatty acids in Saccharomycopsis lipolytica, Evidence for the involvement of cytochrome P-450, J. Gen. Microbiol., 128, 1125. McCann, J., and B. Ames (1976) Detection of carcinogens as mutagens in the Salmonella/microsome test: assay of 300 chemicals: discussion, Proc. Natl. Acad. Sci. (U.S.A.), 73, 950. McCann, J., E. Choi, E. Yamasaki and B. Ames (1975) Detection of carcinogens as mutagens in the Salmonella/ microsome test: assay of 300 chemicals, Proc. Natl. Acad. Sci. (U.S.A.), 72, 5135. Nash, T. (1953) Colorimetric estimation of formaldehyde by means of the Hantzsh reaction, Biochem. J., 55, 416. Oesch, F. (1973) Carbon monoxide-binding pigment of liver microsomes, I. Evidence for its hemoprotein nature, Xenobiotica, 3, 305. Oesch, F., D. Raphael, H. Schwind and H.R. Glatt (1977) Species differences in activating and inactivating enzymes related to the control of mutagenic metabolites, Arch. Toxicol., 39, 97. Omura, T., and E. Sato (1964a) Carbon monoxide-binding pigment of liver microsomes, I. Evidence for its hemoprorein nature, J. Biol. Chem., 239, 2370. Omura, T., and E. Sato (1964b) Carbon monoxide-binding pigment of liver microsomes, II. Solubilization, purification and properties, J. Biol. Chem., 239, 2379. Prough, R.A., M.D. Burke and R.T. Mayer (1978) Direct fluorimetric methods for measuring mixed-function oxidase activity, Methods Enzymol., 52, 372. Ryan, D.E., P.E. Thomas and W. Levin (1977) Properties of purified liver microsomal cytochrome P-450 from rats treated with the polychlorinated biphenyl mixture Aroclor 1254, Mol. Pharmacol., 12, 521. Ryan, D.E., P.E. Thomas, D. Korzeniowski and W. Levin (1979) Separation and characterization of highly purified forms of liver microsomal cytochrome P-450 from rats treated with polychlorinated biphenyls, phenobarbital and 3-methylcholanthrene, J. Biol. Chem., 254, 1365. Schassmann, H.U., and F. Oesch (1978) Trans stilbene oxide: a selective inducer of rat liver epoxide hydratase, Mol. Pharmacol., 14, 834. Sladeck, N.E., and G.J. Mannering (1966) Evidence for a new P-450 hemoprotein in hepatic microsomes from methylcholanthrene-treated rats, Biochem. Biophys. Res. Commun., 22, 668. Ullrich, V., and P. Weber (1972) O-Dealkylation of 7-

133 ethoxycoumarin by liver microsomes direct fluorimetric test, Hoppe-Seyler's Z. Physiol. Chem., 353, 1171. Vermillion, J., and M.J. Coon (1978) Purified liver microsomal NADPH-cytochrome P-450 reductase, J. Biol. Chem., 253, 2694. Warner, M., M.V. Lamarca and A.H. Neims (1978) Chromato-

graphic and electrophoretic heterogenicity of the cytochrome P-450 solubilized from untreated rat liver, Drug Metab. Disp., 6, 353. Werringloer, J. (1978) Assay of formaldehyde generated during microsomal oxidation reactions, Methods Enzymol., 52, 297.