Purification of hepatic microsomal cytochromes P-450 from β-naphthoflavone-treated Atlantic cod (Gadus morhua), a marine teleost fish

Biochimica et Biophysica Acta 840 (1985) 409-417 Elsevier

409

BBA22082

Purification of hepatic microsomal cytochromes P-450 from fl-naphthoflavone-treated Atlantic cod (Gadus morhua), a marine teleost fish Anders Goks~yr Department of Biochemistry, University of Bergen, Arstadveien 19, N-5000 Bergen (Norway) (Received November 14th, 1984) (Revised manuscript received April 15th, 1985)

Key words: Cytochrome P-450; 5,6-Benzoflavone;(Fish liver microsome)

Four isozymes of cytochrome P-450 were purified to varying degrees of homogeneity from liver microsomes of cod, a marine teleost fish. The cod were treated with fl-naphthoflavone by intraperitoneal injection, and liver microsomes were prepared by calcium aggregation. After solubilization of cytochromes P-450 with the zwitterionic detergent 3-[(3-cholamidopropyl)dimethylammoniol-l-propansulfonate, chromatography on PbenyI-Sepharose CL-4B, and subsequently on DEAE-Sepharose, resulted in two cytochrome P-450 fractions. These were further resolved on hydroxyapatite into a total of four fractions containing different isozymes of cytochromes P-450. One fraction, designated cod cytochrome P-450c, was electrophoretically homogeneous, was recovered in the highest yield and constituted the major form of the isozymes. The relative molecular mass of this form (58000) corresponds well with a protein band appearing in cod liver microsomes after treatment with ~]-naphthoflavone. Both cytochrome P-450c and a minor form called cytochrome P-450d (56 000) showed activity towards 7-ethoxyresorufin in a reconstituted system containing rat liver NADPH-cytochrome P-450 reductase and phospholipid. Differences between these two forms were observed in the rate and optimal pH for conversion of this substrate, and in optical properties. Rabbit antiserum to cod cytochrome P-450c did not show any cross-reactions with cod cytochrome P-450a ( M r 55 000) or cytochrome P-450d in Ouchterlony immunodiffusion, but gave a precipitin line of partial identity with cod cytochrome P-450b ( M r 54000), possibly as a result of contaminating cytochrome P-450c in this fraction.

Introduction

Cytochromes P-450 constitute a family of isozymes catalyzing reactions in the metabolism of a great variety of compounds, including endogenous substrates, drugs and environmental pollutants such as pesticides, polychlorinated biphenyls and polycyclic aromatic hydrocarbons [1]. Although this polysubstrate monooxygenase system is found

Abbreviations: Chaps, 3-[(3-cholamidopropyl)dimethylammonio]-l-propansulfonate; SDS, sodium dodecyl sulfate.

in the whole range of living organisms, most attention has been devoted to the study of these enzymes in laboratory mammals, and much progress has been made in the isolation and characterization of an increasing number of cytochrome P-450 isozymes from mammalian species [2-4]. The study of these isoenzymes in non-mammalian species is relevant to a number of questions, including the inducibility of various cytochrome P-450 forms, the evolutionary relationship between these enzyme systems and the response of organisms to pollutants in their environment. In this context the aquatic species are of particular

0304-4165/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

410

interest, because the sea is the major recipient of antropogenic wastes, and because many of these organisms, fish in particular, are important human food sources. In the fish monooxygenase system several interesting differences as compared to the mammalian system have been demonstrated [5]. Among these, the apparent unresponsiveness of fish cytochromes P-450 to the phenobarbital class of inducers has been the most prominent (reviewed in Refs. 6-8). Many fish species are, however, very responsive to induction of aryl hydrocarbon hydroxylase by polycyclic hydrocarbons and /3-naphthoflavone [6-10]. Measurement of this activity has been suggested as a practical biological monitor for marine petroleum pollution [11]. The characteristic hypochromic shift in the CO-reduced difference spectra to 447-448 nm observed with mammalian preparations after this type of induction, however, is absent in fish [7]. The metabolism of polycyclic aromatic hydrocarbons may either lead to successful detoxification, or to mutagenic or carcinogenic activation by the regiospecific formation of bay-region diolepoxide intermediates [12]. In vivo metabolic studies of a model polyaromatic hydrocarbon substrate, phenanthrene, have revealed distinct differences in regio-specific hydroxylations between bony fish (teleosts), and mammals, cartilaginous fish and other marine animals [13]. The literature on purification of cytochromes P-450 from fish sources is scarce. These enzymes have been purified only recently from two different species of bony fish: /3-naphthoflavone-treated rainbow trout (Salmo gairdneri) [14,15] and untreated scup (Stenotomus chrysops) [16]. An increased access to purified cytochrome P-450 isozymes from fish will enable us to achieve a better understanding of how this enzyme system works, for example, in relation to induction by xenobiotics in the aquatic or marine environment. Preparation of antibodies to induced isozymes may even provide us with a more powerful tool for monitoring marine pollution than the mere measurement of enzyme activities. I have resolved and partially purified multiple forms of cytochrome P-450 from /3-naphthoflavone-treated Atlantic cod, and antibodies have been prepared to the major form purified. The cod

is a marine teleost of great economic importance. This species is characterized by a very fatty liver, which might have implications on the effect of lipophilic xenobiotics on the hepatic microsomal cytochrome P-450 system. Parts of this work were presented at the 6th International Symposium on Microsomes and Drug Oxidations, August 1984. Brighton, U.K. [17]. Materials and Methods

Materials'. Phenyl-Sepharose CL-4B and DEAE-Sepharose CL-6B were purchased from Pharmacia Fine Chemicals AB. Hydroxyapatite (Bio-Gel HTP), agarose (standard low Mr) and SDS-polyacrylamide gel electrophoresis low molecular weight standards were obtained from Bio-Rad, and ethylchloroformate from British Drug Houses. Ultrafiltration equipment and Diaflo (PM 30) membranes were from Amicon. Purified rat liver NADPH-cytochrome P-450 reductase was a generous gift from Dr. Edward T. Morgan, Karolinska Institutet, Stockholm, Sweden. Purified 7-ethoxyresorufin was generously supplied by Dr. John J. Stegeman, Woods Hole Oceanographic Institution, MA, U.S.A. All other chemicals were of the highest commercial grade available, purchased from either Sigma or Merck. Preparation of detergents. Chaps was synthesized as described by Hjelmeland [18], and cochromatographed by thin-layer chromatography with commercial Chaps from Sigma. Sodium cholate was prepared from cholic acid (Sigma) by two recrystallizations from 96% ethanol, dissolution in 3-4 M NaOH and titration to pH 7.5 with 2 M acetic acid. Treatment offish. Cod ( Gadus morhua ) for these experiments were caught in the coastal areas near Bergen and kept in tanks with flowing sea water (7°C, 34% salinity) at the Institute of Marine Research in Bergen. Cod need higher doses of B-naphthoflavone to obtain a significant increase in the hepatic microsomal cytochrome P-450 content than other fish (A. Goksoyr, unpublished data). Thus, 500 mg /3-naphthoflavone/kg (250 mg/ml suspended in soybean oil) was administered intraperitoneally to 24 cod of both sexes (average weight, 400 g). The fish were maintained

411 in the tanks without food for 4 days after the injection, and then killed by a blow to the head. Preparation of microsomes. The excised livers were weighed, washed in ice-cold buffer A (10 mM Tris-HC1 (pH 7.4), 0.25 M sucrose) and homogenized in 5 vol. of the same buffer. To eliminate most of the fat it was necessary to centrifuge the homogenate two times at 6000 x g for 15 min followed by 10000 x g for 15 min. Microsomes were prepared from the post-mitochondrial supernatant by Ca2+-precipitation, as described by Schenkman and Cinti [19] with some modifications. A stock solution of 0.32 M CaC12 was added to the post-mitochondrial supernatant to a final concentration of 12 mM Ca2÷ (this was optimal with cod liver microsomes, as opposed to 8 mM with mammalian preparations [19]). The washed microsomal pellet was resuspended in buffer C (50 mM Tris-HC1 (pH 7.5), 20% glycerol, 1 mM EDTA and 1 mM dithiothreitol) to a protein concentration of around 25 mg/ml, and stored at - 80°C. Under these conditions the microsomal cytochrome P-450 content was stable for several months. All procedures involved in the preparation and solubilization of microsomes and the purification of cytochromes P-450 were carried out at 4°C. Solubilization and purification. Problems with the solubilization step have been reported in other studies with teleost fish [14,16,20], indicating that this is a general property of fish liver cytochromes P-450. Use of the zwitterionic detergent Chaps [14,18], resulted in the best yields with cod liver microsomes. Addition of cholate increased the yield somewhat, and was used in these experiments. Thawed microsomes (865 nmol cytochromes P-450, 4000 mg protein) were diluted with buffer D (0.2 M Tris-HC1 (pH 7.5), 20% glycerol, 1 mM EDTA and 1 mM dithiothreitol) to a protein concentration of 10 mg/ml, and Chaps and cholate were added drop-wise to a final concentration of 1.0 and 0.2%, respectively. The solution was stirred for 30 min and then centrifuged at 45 000 × g for 60 min, which was sufficient to precipitate unsolubilized, Ca2+-aggregated microsomes. The supernatant was concentrated and diluted 20-fold with buffer E (50 mM potassium phosphate (pH 7.5), 20% glycerol, 1 mM EDTA and 1 mM dithiothreitol) to a volume of 2200 ml. After dilution to

0.05% Chaps, solubilized cod liver cytochromes P-450 bound to the Phenyl-Sepharose CL-4B matrix. A Phenyl-Sepharose CL-4B column (2.5 × 20 cm) equilibrated with buffer E, was loaded with 700 ml from this solution before it showed signs of overloading. The column was washed with 150 ml buffer E containing 0.05% Chaps and 200 ml buffer E containing 0.075% cholate and 0.125% Lubrol PX, before elution with 0.5% Lubrol PX and 0.3% cholate in buffer E. The rest of the solubilized microsomes was loaded onto a second Phenyl-Sepharose column (2.5 x 42 cm) equilibrated with buffer E. This column was washed with 200 ml buffer E containing 0.05% Chaps and 250 ml buffer E containing 0.15% Lubrol PX, and eluted with buffer E containing 0.5% Lubrol PX and 0.3% cholate (700 ml). Fractions (7.5 ml) were collected, and the absorbances at 417 and 280 nm were measured. From the first column, only one large cytochrome P-450 peak eluted, but the wash procedure resulted in loss of cytochrome P-450 (seen as a broad, heme-absorbing, cytochrome P450-containing shoulder). In the second column, one minor cytochrome P-450 peak eluted with the second wash step and was added to the major peak. The cytochrome P-450 peaks from the two different Phenyl-Sepharose columns were pooled together before the next step, concentrated by ultrafiltration, and diluted 5-fold with 20% glycerol and 0.16% cholate. The sample (270 nmol cytochromes P-450, 390 mg protein) was applied to a DEAE-Sepharose column (1.5 × 40 cm) equilibrated with buffer F (10 mM potassium phosphate (pH 7.7), 20% glycerol, 0.1 mM EDTA, 0.1 mM dithiothreitol, 0.1% Lubrol PX and 0.2% cholate), washed with 150 ml buffer F, and eluted with a linear gradient of 0-0.5 M KC1 in buffer F (600 ml total volume). 10-ml fractions were collected and measured for A417and A28o. Fractions from the two cytochrome P-450 peaks were separately pooled and dialyzed twice against buffer G (buffer F at pH 7.5 and with 1 mM dithiothreitol). Fraction A was loaded onto a hydroxyapatite column (1.3 × 4 cm) equilibrated with buffer G, washed with 25 ml of the same buffer and eluted with a linear gradient of 10-125 mM potassium phosphate in buffer G (80 ml). DEAE fraction B was likewise loaded onto a hydroxyapatite column (0.8 x 16 cm), but washed with 20 mM phosphate

412 in buffer G (40 ml) and eluted with a linear 20-125 mM phosphate gradient in buffer G (100 ml). Fractions (4 ml) were collected and measured as before. The cytochrome P-450-containing fractions from each peak were pooled, treated with Amberlite XAD-2 and dialyzed against 10 mM Tris-acetate (pH 7.4), 20% glycerol and 1 mM EDTA. Apparently, this treatment resulted in aggregation and precipitation of some of the enzyme from solution. The pooled, detergent-depleted cytochrome P-450 fractions were stored at - 8 0 ° C . Assays of protein and enzymes. Cytochrome P450 was assayed by the dithionite-difference procedure of Matsubara et at. [21]. Purified cod cytochromes P-450 were also assayed by the method of Omura and Sato [22]. 7-Ethoxyresorufin O-deethylase was measured spectrophotometrically at room temperature [23]. Protein was determined by the method of Lowry et al. [24] with bovine serum albumin as the standard, or with the modifications of Dulley and Grieve [25]. All spectrophotometric measurements were performed on a Perkin-Elmer Model 554 dual-beam spectrophotometer calibrated with a holmium filter (Unicam).

Results

Reconstitution of 7-ethoxyresorufin O-deethylase activity. The standard reconstituted system con-

Purification of cod cytochromes P-450

taining 40 pmol purified cod cytochrome P-450, 0.5 U rat NADPH-cytochrome P-450 reductase, 5 /~g sonicated dilauroylphosphatidylcholine (DLPC) and 0.1 M potassium phosphate (pH 7.4) (optimal for cod P-450c), was preincubated at 4°C for 30 min. Purified 7-ethoxyresorufin dissolved in methanol was added to a concentration of 2 mM, and the reaction was started with the addition of 5 /~1 100 mM NADPH to a final reaction volume of 1 ml. The activity was measured as described above [23]. SDS-polyacrylamide gel electrophoresis. Discontinuous SDS-polyacrylamide gel electrophoresis was performed according to Laemmli [26], with 10% acrylamide separating slab gels. The gels were silver stained [27], with modifications of Allan and Lossius (Lossius, personal communication). Molecular weights were determined by comparing the relative mobilities of unknowns to those of the molecular weight standards in a semilog plot. Amino acid analysis. The amino acid composition of cod cytochrome P-450c was analyzed by the method of Vasstrand et al. [28]. Samples (ap-

The aggregation of liver microsomes with calcium is a rapid and convenient method to obtain microsomes for further purification, allowing large sample volumes to be centrifuged simultaneously. The yield of cytochrome P-450 in such microsomes from cod was slightly lower than in preparations obtained by high-speed centrifugation (80-90%). Due to a generally increased yield of protein in the calcium-aggregation procedure, the specific content of cytochrome P-450 (nmol cytochrome P-450/mg protein), however, was 30-50% lower than in parallel high-speed preparations (100000 x g or 40000 x g [30]) (results not shown). Thus, when studying specific content or activity, care should be taken to note the method of preparation. When solubilized microsomes were applied to the Phenyl-Sepharose column, problems were encountered in overloading the column, and a second column was used. Several wash procedures have been tested for the Phenyl-Sepharose, and the one presented for the large column in this scheme, gave the best results in our hands.

prox. 200 ~g) were precipitated with 10% trichloroacetic acid to remove glycerol, washed twice with acetone, dried and hydrolyzed in 6 M HC1 for 18 h at 110°C in sealed, evacuated tubes. In the method applied, the ninhydrin derivatives of the amino acids are measured at 570 nm, which gives no values for proline or tryptophan. Norleucine is used as an internal standard. Preparation of antibodies. Antibodies were raised to cod cytochrome P-450c in a pigmented French-Burgundy rabbit (female, 4 months old, locally bred) by multiple intradermal injections. Blood was collected from the ear vein, and antiserum was prepared by centrifugation (2500 x g for 10 rain) of the coagulated blood after a few hours at room temperature. Immunochemical procedures. Ouchterlony double-diffusion analysis was performed essentially according to Thomas et al. [29], in gels containing 1 M sodium glycine (pH 7.4), 0.08 M NaC1, 0.9% agarose and 1.0% Chaps. Developed and washed gels were stained with Coomassie brilliant blue R.

413 A.

r.-

/

2

3

4

5

7

6

8

9

10

11

31

Q5

to

.<

i o

ca ,<

?i

I 0

10

20

40

30

w

F ~ c t i o n no i

- 4

a. 12

/" /"

C

~

(3£

8 E 7

<" 04

0

10

20

30

4()

50

Froction no

Fig. 1. Elution profiles of DEAE-Sepharose fractions A (A) and B (B) on hydroxyapatite columns. Cytochrome P-450 fractions were pooled, dialyzed against buffer G (see text) and loaded onto hydroxyapatite columns previously equilibrated with the same buffer. (A) The column (1.3 x 4 cm) was washed with 25 ml buffer G and developed with a linear gradient of 10-125 mM phosphate in buffer G (80 ml). A280 (protein absorbance) (O) and A417 (heme absorbance) (O) were measured in 4-ml fractions. Cytochrome P-450 fractions were pooled as follows: cytochrome P-450a, fractions 16-18; cytochrome P-450b, fractions 21-23. (B) The column (0.8 × 16 cm) was washed with 20 mM phosphate in buffer G (40 ml) before elution started with a linear gradient of 20-125 mM phosphate in buffer G (100 ml). 3.5-ml fractions were measured as in (A). Cytochrome P-450 fractions were pooled as follows: cytochrome P-450ci (peak), fractions 35-36; cytochrome P-450cli (shoulders), fractions 34 and 37, and cytochrome P-450d, fractions 42-44.

Chromatography of the pooled peaks from the Phenyl-Sepharose columns on DEAE-Sepharose resulted in a resolution of two cytochrome P-450 peaks and cytochrome b 5. The P-450 fractions were further resolved on hydroxyapatite (Fig 1) to a total of four cytochrome P-450-containing fractions, denoted cytochromes P-450a, P-450b, P450c and P-450d after their elution from the hydroxyapatite columns. The specific contents of these fractions were 1.5, 5.1, 13.4 and 17.3 nmol

Fig. 2. A 10% SDS-polyacrylamide gel showing the different purification steps for cod cytochromes P-450. The wells contain: (1) microsomes; (2) solubilized microsomes; (3) Phenyl-Sepharose peak; (4) DEAE fraction A; (5) cytochrome P-450a; (6) cytochrome P-450b; (7) DEAE fraction B; (8) cytochrome P-450cI (peak fractions); (9) cytochrome P-450cII (shoulder fractions); (10) cytochrome P-450d; (11) molecular weight standards (only the 45, 31 and 21.5 kDa bands show properly). The gel was stained with silver essentially after Ref. 27.

cytochrome P-450/mg protein, respectively (Table

I). SDS-polyacrylamide gel electrophoresis of samples from the various purification steps (Fig. 2) revealed that cytochrome fractions P-450a and P-450b contained major protein bands at M r 55000 and 54000, respectively. Cod cytochrome P-450c is homogeneous as determined by silver staining, and has a relative molecular mass of 58000, the same mobility as a protein band appearing in cod liver microsomes after treatment with fl-naphthoflavone (A. Goksoyr, unpublished results). This form was recovered in the highest yield and obviously constitutes the major form in these microsomes. Cytochrome fraction P-450d, although having the highest specific content, apparently is weakly contaminated with cytochrome P-450c and some other polypeptide of slightly lower molecular weight. The relative molecular mass of cod cytochrome P-450d is 56000. A summary of the results from the purification of cod cytochromes P-450 is presented in Table I. The relative molecular mass of the different forms are presented in Table II together with their optical properties (see below).

414 TABLE I PURIFICATION OF HEPATIC MICROSOMAL CYTOCHROMES P-450 FROM /3-NAPHTHOFLAVONE-TREATED COD

( GADUS MORHUA) Fraction

Cyt. P-450 (nmol)

Protein (mg)

Microsomes Solubilized microsomes Phenyl-Sepharose eluate DEAE-Sepharose eluate fraction A fraction B Hydroxyapatite eluate of A fraction a (P-450a) fraction b (P-450b) Hydroxyapatite eluate of B fraction cl (P-450c) fraction cII (P-450c) fraction d (P-450d)

865 635 272

4 000 1 705 394

0.22 0.37 0.69

52 45

0.78 1.92

4.7 10.0

3.5 8.7

1.46 5.06

1.1 2.0

6.6 23.0

13.41 10.63 17.25

5.8 3.1 1.0

61.0 48.3 78.4

40.4 86.5 9.2 17.2

6.3 3.4

50.3 26.8 8.8

3.75 2.52 0.51

Specific content (nmol/mg)

Yield (%) 100 73 31

Purification (-fold) 1.0 1.7 3.1

Opticalproperties The absorption maximum of the CO-reduced difference spectra of all the cod cytochrome P-450 fractions was at or very close to 448 nm (Table II). For reasons unknown, cod cytochrome P-450a underwent rapid degradation to cytochrome P-420 during the spectrophotometric measurement. The oxidized absorbance maxima of the four isolated forms are presented in Table II. The Soret maxima around 415 nm suggest that all of these forms are low-spin hemoproteins [31]. Only cytochrome P450a showed a weak shoulder just below 400 nm, indicating the presence of some high-spin hemoprotein.

TABLE II OPTICAL PROPERTIES A N D RELATIVE MOLECULAR MASS (Mr) OF ISOLATED CYTOCHROMES P-450 FROM /3-NAPHTHOFLAVONE-TREATED COD Fig. 3. Ouchterlony double-diffusion analysis of purified cod cytochrome P-450 isozymes and liver microsomes using rabbit antiserum to cod cytochrome P-450c (center well). Cytochrome P-450c was placed in wells 1, 4 and 6 (1.5 gg), control and fl-naphthoflavone-treated microsomes (approx. 50 t~g each) in wells 2 and 3, respectively, cytochromes P-450a (13 g g), P-450b (4 /~g) and P-450d (1.2 gg) in wells 8, 7 and 5, respectively. These amounts correspond to approx. 20 pmol P-450 in each well. Gels, with the addition of 1.0% Chaps, were poured and developed as described by Thomas et al. [29].

Isoenzyme

Cytochrome Cytochrome Cytochrome Cytochrome

P-450a P-450b P-450c P-450d

Mr

X m a x (nm) CO-reduced

oxidized

55000 54000 58000 56000

447.5 447.5 448.0 448.0

418.5 a 414.5, 533, 567 416.5, 532, 568 415.5, 538, 566

a Not measured at higher wavelengths due to low degree of purity.

415 TABLE llI

TABLE IV

7-ETHOXYRESORUFIN O-DEETHYLASE ACTIVITY IN LIVER MICROSOMES FROM fl-NAPHTHOFLAVONETREATED COD AND IN ISOLATED, RECONSTITUTED COD CYTOCHROMES P-450

AMINO ACID COMPOSITION OF COD CYTOCHROME P-450c

Cod cytochromes P-450 were isolated and reconstituted with NADPH-cytochrome P-450 reductase (from rat) and phospholipid as described in Materials and Methods. 7-EROD, 7-ethoxyresorufin-O-deethylase, measured as nmol product

formed/min per nmol cytochromeP-450. -, not measured. 7-EROD

Microsomes P-450a P-450b P-450c P-450d

pH 7.4

pH 8.0

0.49 < 0.05 < 0.05 1.07 0.21

0.85 < 0.05 0.39

Reconstitution studies In the reconstituted system described above (Materials and Methods), only cod cytochromes P-450c and P-450d were active towards 7-ethoxyresorufin at pH 7.4. The activity of cytochrome P-450c at this pH was about 1 nmol product formed/min per nmol cytochrome P-450, an activity close to that found in intact microsomes from fl-naphthoflavone-induced cod at pH 8.0 (Table III). However, at this higher pH, only the cytochrome P-450d form showed any activity towards 7-ethoxyresorufin. The activity of cytochrome P450d at pH 8.0 was even higher than at pH 7.4, demonstrating differences in the pH optima of these two forms. Amino acid analysis of cytochrome P-450c The amino acid composition of cod cytochrome P-450c is presented in Table IV. Disregarding the probable contribution of proline and tryptophan (not measured), the percentage of hydrophobic residues for cytochrome P-450c was 41%. lmmunochemical analysis Double-diffusion immunoprecipitin analysis of antiserum raised against cod cytochrome P-450c showed a weak cross-reaction with cod cytochrome P-450b which gave a line of partial identity with cod cytochrome P-450c (Fig. 3). Cross-reactions were not visible with the other cod cyto-

-, not determined Amino acid

P-450c

Asx Thr Ser Glx Pro Gly Ala 1/2Cys Val Met Ileu Leu Tyr Phe His Lys Arg Trp

46 29 30 41 41 33 4 33 13 23 53 11 27 14 31 21

chrome P-450 forms (a and d) in this system. Solubilized microsomes from fl-naphthoflavonetreated cod revealed a precipitin line of identity to cytochrome P-450c, as would be expected since cytochrome P-450c is purified from these microsomes. In addition, solubilized liver microsomes from both untreated and fl-naphthoflavone-treated cod showed a line of identity to each other which was unrelated to cytochrome P-450c (Fig. 3). Discussion

A procedure which allows reproducible purification of cytochromes P-450 from fl-naphthoflavone-treated cod is presented in this report. Four isozyme fractions have been isolated and given names after their appearance from the hydroxyapatite column, since any other nomenclature system seems premature. Cod cytochrome P-450c constitutes about 70% of the recovered isozymes. Clearly, this isozyme must be the most important form in microsomes from fl-naphthoflavonetreated cod. The relative molecular mass of cytochrome P450c is 58 000, the same as that of a protein band appearing in cod liver microsomes after treatment

416

with/~-naphthoflavone (results not shown). Several groups have demonstrated the appearance of a protein band around 58 000 in liver microsomes of different fish species treated with inducers of the 3-methylcholanthrene class [32-35]. These findings suggest the presence of a common induction response in these species and in cod. Williams and collaborators, who have purified multiple cytochromes P-450 from rainbow trout, showed that there was a many-fold increase in a band cross-reacting with IgG to their trout cytochrome P-450 LM4b (M r 58000) after induction with fl-naphthoflavone, using Western blotting [36]. Similar investigations using antibodies to cod cytochrome P-450c are underway in our laboratory. Cod cytochrome P-450c showed substantial activity towards 7-ethoxyresorufin in a reconstituted system. The turnover number was comparable to that found in intact microsomes from /~-naphthoflavone-treated cod, but with a different pHoptimum (Table III). This difference can perhaps be assigned to the different conditions facing the enzyme in intact microsomes and reconstituted micelles. Cod cytochrome P-450d, a minor, but nearly homogeneous cytochrome P-450-fraction with a molecular mass of 56000, also showed activity towards 7-ethoxyresorufin in a reconstituted system. However, this form showed a higher pH optimum and a lower turnover number than P-450c (Table III). Further differences between cod cytochromes P-450c and P-450d were observed in their oxidized absorbance maxima (Table II) and their antigenic properties towards cod cytochrome P-450c antiserum (Fig. 3). In addition, a preliminary investigation of the amino acid composition of cytochrome P-450d revealed several differences as compared to the data presented on cytochrome P-450c (Table IV). Since cytochrome P-450d is slightly contaminated by cytochrome P-450c (see Fig. 2), these data have not been included. It could be seen, however, that cytochrome P-450d had proportionally larger amounts of glutamine/glutamate, alanine and arginine, while the content of methionine, leucine, phenylalanine and histidine was lower than in the cytochrome P-450c form. Since the cytochrome P-450c contamination would tend to decrease differences in composition, the results indicate real differences between the two

forms. Neither of the forms showed close similarities to the cytochrome P-450 forms purified by Williams and Buhler [15], the only published results from teleost fish so far. Cod cytochromes P-450a and P-450b were recovered in yields of 1.1 and 2.0%, respectively, and with low specific contents (Table I). They contain major protein bands with M r values of 55000 and 54000, respectively. Neither of these forms showed any activity towards 7-ethoxyresorufin in a reconstituted system, and they were distinguished from cytochrome P-450c by their optical properties. In the immunodiffusion analysis of the isolated isoenzymes, only cod cytochromes P-450b and P-450c cross-reacted with antiserum to cytochrome P-450c, with a line of partial identity between these forms (Fig. 3). It cannot, however, be excluded that this cross-reaction is the result of contaminating cytochrome P-450c in the cytochrome P-450b fraction. In the same Chaps-containing system, solubilized microsomes from untreated and /~-naphthoflavone-treated cod showed a line of antigenicity related to each other, but unrelated to cytochrome P-450c, in addition to the expected line of identity between treated microsomes and cytochrome P-450c. These unexpected results indicate the presence of highly antigenic, contaminating protein in the antigen preparation used for immunization. Such contaminations, however, are invisible with silver staining of polyacrylamide gels. As noted in other reports on the purification of cytochromes P-450 from bony fish [14,16,20], the solubilization step is particularly critical. With cod microsomal cytochromes P-450, as with rainbow trout [14], the zwitterionic detergent Chaps was the best choice. However, the yield from cod liver microsomes was occasionally as low as 75%, and the conditions for solubilization can obviously be further improved. In conclusion, I have isolated and purified multiple forms of cytochromes P-450 from /3-naphthoflavone-treated cod to varying degrees of purity. The main form in these microsomes, cod cytochrome P-450c, has been distinguished from the other forms by its molecular weight, enzymatic activity, optical properties and immunological reactivity. Studies in progress will aim at further

417

characterization of these forms, both enzymologically and immunochemically. We will also try to establish the role of these isoenzymes in the regiospecific hydroxylation of phenanthrene, which has been shown to be fundamentally different in bony fish as compared to other organisms [13]. Antibodies to purified fish cytochromes P-450 will obviously be a useful tool in answering questions related to marine toxicology, like inducibility of these enzyme systems by environmental pollutants, and the application of these systems in biological monitoring of pollution.

Acknowledgements I want to thank Dr. Edward T. Morgan for valuable discussions and for supplying the purified reductase. I am also grateful to Dr. John J. Stegeman for supplying purified 7-ethoxyresorufin, to Dr. Jan E. Solbakken for his help with keeping and injecting the fish, to Dr. Harald B. Jensen, Bente Hogh and Dr. Endre Vasstrand for performing the amino acid analysis, to Dr. Richard Fosse and to Dr. Leiv Klungsoyr. This work was supported in part by the Norwegian Marine Pollution Research and Monitoring Programme.

References 1 Nebert, D.W., Eisen, H.J., Negishi, M., Lang, M.A., Hjelmeland, L.M. and Okey, A.B. (1981) Annu. Rev. Pharmacol. Toxicol. 21,431-462 2 Johnson, E.F. (1979) Rev. Biochem. Toxicol. 2, 1-26 3 Lu, A.Y.H. and West, S.B. (1980) Pharmacol. Rev. 31, 277-295 4 Guengerich, F.P. and Martin, M.V. (1980) Arch. Biochem. Biophys. 205, 365-379 5 Buhler, D.R. and Rasmusson, M.E. (1968) Comp. Biochem. Physiol. 25, 223-239 6 Lech, J.J. and Bend, J.R. (1980) Environ. Health Persp. 34, 115-131 7 Stegeman, J.J. (1981) in Polycyclic Hydrocarbons and Cancer (Gelboin, H.V. and Ts'o, P.O.P., eds.), Vol. 3, pp, 1-60, Academic Press, New York 8 Lech, J.J., Vodicnik, M.J. and Elcombe, C.R. (1982) in Aquatic Toxicology (Weber, L.J., ed.), pp. 107-148, Raven Press, New York 9 Bend, J.R., Ball, L.M., Elmamlouk, T.H., James, M.O. and Philpot, R.M. (1979) in Pesticide and Xenobiotic Metabolism in Aquatic Organisms (Khan, M.A.Q., Lech, J.J. and Bend, J.R., eds.), pp. 297-318, American Chemical Society, Washington DC

10 Vodicnik, M.J. and Lech, J.J. (1982) in Cytochrome P-450, Biochemistry, Biophysics and Environmental Implications (Hietanen, E., Laitinen, M. and Hanninen, O., eds.), pp. 225-231, Elsevier, Amsterdam 11 Payne, J.F. (1976) Science 191,945-946 12 Jerina, D.M. and Daly, J.W. (1974) Science 185, 573-583 13 Solbakken, J.E. and Palmork, K.H. (1981) Comp. Biochem. Physiol. 70C, 21-26 14 Williams, D.E. and Buhler, D.R. (1982) Biochim. Biophys. Acta 717, 398-404 15 Williams, D.E. and Buhler, D.R. (1984) Biochem. Pharmacol. 33, 3743-3753 16 Klotz, A.V., Stegeman, J.J. and Walsh, C. (1983) Arch. Biochem. Biophys. 226, 578-592 17 Goksoyr, A. (1984) Xenobiotica 14, Suppl. 1, 58 18 Hjelmeland, L.M. (1980) Proc. Natl. Acad. Sci. USA 77, 6368-6370 19 Schenkman, J.B. and Cinti, D.L. (1978) Methods Enzymol. 52, 83-89 20 Arinc, E. and Adali, O. (1983) Comp. Biochem. Physiol. 76B, 653-662 21 Matsubara, T., Prough, R.A., Burke, M.D. and Estabrook, R.W. (1974) Cancer Res. 34, 2196-2203 22 Omura, T. and Sato, R. (1964) J. Biol. Chem. 239, 2370-2378 23 Klotz, A.V., Stegeman, J.J. and Walsh, C. (1984) Anal. Biochem. 140, 138-145 24 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265-275 25 Dulley, J.R. and Grieve, P.A. (1975) Anal. Biochem. 64, 136-141 26 Laemmli, U.K. (1970) Nature 227, 680-685 27 Marshall, T. and Lattner, A.L. (1981) Electrophoresis 2, 228-235 28 Vasstrand, E., Jensen, H.B. and Miron, T. (1980) Anal. Biochem. 105, 154-158 29 Thomas, P.E., Lu, A.Y.H., Ryan, D., West, S.B., Kawalek, J. and Levin, W. (1976) J. Biol. Chem. 251, 1385-1391 30 Stegeman, J.J. and Binder, R.L. (1979) Biochem. Pharmacol. 28, 1686-1688 31 Jefcoate, C.R.E. and Gaylor, J.L. (1969) Biochemistry 8, 3464-3472 32 Elcombe, C.R. and Lech, J.J. (1979) Toxicol. Appl. Pharmacol. 49, 437-450 33 Vodicnik, M.J., Elcombe, C.R. and Lech, J.J. (1981) Toxicol. Appl. Pharmacol. 59, 364-374 34 Foureman, G.L. and Bend, J.R. (1982) in Cytochrome P-450, Biochemistry, Biophysics and Environmental Implications (Hietanen, E., Laitinen, M. and H~nninen, O., eds.), pp. 233-244, Elsevier, Amsterdam 35 Gooch, J.W. and Matsumura, F. (1983) Toxicol. Appl. Pharmacol. 68, 380-391 36 Williams, D.E., Bender, R.C., Morrissey, M.T., Selivonchick, D.P. and Buhler, D.R. (1984) in Responses of Marine Organisms to Pollutants (Stegeman, J.J. and Heath, G.W., eds.), Marine Environ. Res. 14, 13-21, Elsevier Applied Science Publishers, U.K.