A novel adrenocorticotropin-inducible cytochrome P450 from rat adrenal microsomes catalyzes polycyclic aromatic hydrocarbon metabolism.

0013-7227/91/1292-0970$03.00/0 Endocrinology Copyright © 1991 by The Endocrine Society

Vol. 129, No. 2 Printed in U.S.A.

A Novel Adrenocorticotropin-Inducible Cytochrome P450 from Rat Adrenal Microsomes Catalyzes Polycylic Aromatic Hydrocarbon Metabolism* STEPHANIE OTTO, CRAIG MARCUS, CHARLES PIDGEON, AND COLIN JEFCOATE Department of Pharmacology, University of Wisconsin Medical School (S.O., C.J.), Madison, Wisconsin 53706; and the Departments of Pharmacology and Toxicology (CM.) and Medicinal Chemistry and Pharmacognasy (C.P.), Purdue University School of Pharmacy and Pharmacal Sciences, West Lafayette, Indiana 47907

microsomes in a single step, using detergent elution from a new HPLC matrix consisting of monolayers of phosphatidylcholine covalently bound to a silica support. The resulting P450 preparation contains a single major (57K) band, constituting approximately 70% of the total protein (specific content, 2 nmol P450/ mg protein; turnover number, 1.5 nmol DMBA min"1). A rabbit polyclonal antibody raised against this preparation also recognizes a single ACTH-inducible 57K rat adrenal microsomal protein on immunoblots and dose-dependently inhibits DMBA metabolism in solubilized reconstituted rat adrenal microsomes. This 57K P450 is immunochemically distinct from rat P450s of the I, II, III, XVII, and XXI families, but it is immunochemically closely related to a 55K benz(a)anthracene-inducible P450 in the 10T1/2 mouse embryo fibroblast cell line. {Endocrinology 129: 970-982,1991)

ABSTRACT. 7,12-Dimethylbenz(a)anthracene (DMBA) causes massive ACTH-dependent necrosis of the rat adrenal cortex. This may be related to an ACTH-inducible adrenal microsomal cytochrome P450 that metabolizes polycyclic aromatic hydrocarbons (PAH). The proportions of major monooxygenated products of rat adrenal microsomal DMBA metabolism (DMBA-8,9-diol, DMBA-3,4-diol, and DMBA-phenols) differ significantly from that of P450IA1, the most active PAH-metabolizing P450 in rat liver microsomes. After hypophysectomy, both DMBA metabolic activity and a 57K protein which is distinct from P450XXI disappear from rat adrenal microsomes. ACTH restores both 57K protein and DMBA metabolic activity in hypophysectomized rats almost to the levels in intact untreated rats, but not to levels in ACTH-induced intact rats. The 57K protein has been partially purified from solubilized

W

HILE the rat adrenal cortex produces both glucocorticoids and mineralcorticoids through a series of hydroxylation reactions (1), it also has the capacity to metabolize hydrocarbons, such as benzo(a)pyrene and 7,12-dimethylbenz(a)anthracene (DMBA) (2). Steroid hydroxylation and aromatic hydrocarbon hydroxylation both require cytochrome P450 as their terminal oxidase. The cytochrome P450 superfamily of enzymes comprises both very specific hydroxylases required for hormone biosynthesis and enzymes involved in xenobiotic metabolism that exhibit characteristic, but often overlapping, substrate specificities (3).

In the rat, DMBA-induced adrenal necrosis, which destroys both the zona fasiculata and the zona reticularis, is dependent on ACTH (4) and estradiol (5), and it is prevented by certain P450 inhibitors and inducers (2). Adrenals are unaffected if DMBA is administered to mature rats that have been hypophysectomized for 14 days, but necrosis is restored if ACTH is coadministered. This toxicity parallels both changes in the amount of an ACTH-inducible (57K) microsomal protein and changes in microsomal polycyclic aromatic hydrocarbon (PAH) metabolism (6). DMBA is an extremely potent carcinogen in many tissues, including rat skin, mammary glands, and immune system (2, 7, 8). The carcinogenic, and potentially the necrotic, properties of DMBA are dependent on prior metabolic activation to a reactive electrophile that can covalently bind both DNA and protein (9,10). The most reactive electrophile is thought to be the Bay region diolepoxide, DMBA-3,4-diol-l,2-oxide (11, 12), or possibly the 7-hydroxy derivative of this oxide (13,14). The initial P450 product, DMBA-3,4-oxide, is hydrated to the cor-

Received November 19,1990. Address all correspondence and requests for reprints to: Dr. Colin R. Jefcoate, Department of Pharmacology, University of Wisconsin Medical School, 3780 Medical Science Center, 1300 University Avenue, Madison, Wisconsin 53706. * This work was supported in part by NIH Grants CA-16265 and DK-18585 (to S.O. and C.J.) and in part by NSF Grant 8908450-CTS (to C.P. and CM.). Portions of this work have been presented in abstract form at the Eighth International Symposium on Microsomes and Drug Oxidations, Stockholm, Sweden, June 25-29, 1990, and at the 74th Annual Meeting of the Federation of American Societies for Experimental Biology (FASEB), Washington, D.C., April 1-5,1990.

970

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 18:33 For personal use only. No other uses without permission. . All rights reserved.

CHARACTERIZATION OF A NOVEL RAT ADRENAL CYTOCHROME P450 responding trans-dihydrodiol by microsomal epoxide hydrolase before a second P450-mediated monooxygenation to the reactive electrophile. The adrenal cortex expresses several specific P450 enzymes responsible for steroidogenesis which are located in the mitochondria and the endoplasmic reticulum. Adrenal mitochondria contain both P450.3CC1 (XIA), which catalyzes side-chain cleavage of cholesterol, and P450n», (XIB), which catalyzed steroid 11/3- and 18hydroxylation of C-21 steroids. In most species, adrenal microsomes contain P450XXI, which catalyzes steroid 21-hydroxylation, and P450XVII, which catalyzes both steroid 17«-hydroxylation and C-17-C-20 bond cleavage (1). Although the rat adrenal lacks P450XVII, this enzyme is present in the rat testis (15). The selective effects of age (16, 17), inhibitors (18), and ACTH (6, 18, 19) suggest that the steroidogenic cytochromes P450 do not catalyze PAH metabolism in adrenal microsomes. PAH metabolic activity has been partially purified and separated from both steroid C-21 hydroxylase and 17«-hydroxylase-lyase activity in guinea pig adrenal cortex microsomes (21). The guinea pig isozyme presumed responsible for this activity is immunochemically related to rat P450IA1. It has a mol wt of approximately 52K, is male specific, is suppressed by ACTH, and is not affected by 3-methylcholanthrene (22, 23). This paper describes the purification to near homogeneity from rat adrenal microsomes of a 57K cytochrome P450 that effectively catalyzes DMBA metabolism. This purification has been achieved in one chromatographic step using a novel immobilized artificial membrane HPLC column composed of phosphatidylcholine attached to a silica matrix (IAM.PC) (24). We show that a rabbit antibody raised against this preparation recognizes a single ACTH-sensitive 57K protein and inhibits DMBA metabolism in rat adrenal microsomes, but it does not recognize other characterized rat microsomal P450s. This rat adrenal P450 exhibits very different characteristics from the guinea pig isozyme that is active in PAH metabolism.

Materials and Methods Animab Male Sprague-Dawley rats, 80-100 g, 6-8 weeks old, were used in all experiments, except where indicated. All rats, including hypophysectomized animals and age-matched controls, were purchased from Harlan Sprague-Dawley, Inc. (Madison, WI). Sprague-Dawley rat adrenals used for purifications, 1 The following abbreviations are used: P450IA1, P450c; P450IA2, P450d; P450IIA1, P450B; P450IIB1, P450b; P450IIB2, P450e; P450IIC6, P450k; P450IIC7, P450f; P450IIC11, P45oh; P450IIC12, P450i; P450IIC13, P450g; P450IIE, P450j| P450IIIA1, P450PCNi; P450XIA, P4508CC; P450XIB, P450,1; P450XXI, P4502I.

971

male ICR mouse adrenals, and male white-haired Hartley guinea pig adrenals were obtained frozen on dry ice from Bioproducts for Science, Inc. (Indianapolis, IN). Bovine adrenals were obtained from a local slaughterhouse and transported on ice.

Chemicals Corticotropin-(l-24) (Cortrophin Zinc) and dexamethasone (Hexadrol) were purchased from Organon, Inc. (West Orange, NJ); nitrocellulose and materials for gel electrophoresis from Bio-Rad Laboratories (Richmond, CA); and acetonitrile, diaminobenzidine, DMBA, and ethylene dichloride from Aidrich Chemical Co., Inc. (Milwaukee, WI). [3H]DMBA was purchased from Amersham Radiochemicals (Arlington Heights, IL) and purified by reverse phase HPLC before use. Unlabeled 7- and 12-hydroxy derivatives of DMBA (7-OHMMBA and 12-OHMMBA) were obtained from NCI Chemical Depository; 5,6-8,9-3,4-dihydrodiols of DMBA were prepared by nonenzymatic procedures previously described (26). NADP, glucose-6-phosphate, glucose-6-phosphate dehydrogenase, Trizma base, progesterone, cholic acid, dithiothreitol, ammonium thiocyanate, phenylmethylsulfonylfluoride (PMSF), Tween-20, Lubrol PX, and hydrogen peroxide were purchased from Sigma Chemical Co. (St. Louis, MO). Dilaurylphosphatidylcholine was purchased from Serdary Research Laboratories (London, Ontario, Canada). Methanol for HPLC, KH2PO4, K2HPO4, MgC12, cobalt nitrate, NaCl, and ethylacetate were purchased from Baker/Mallinckrodt (Phillipsburg, NJ). Glycerol, acetone, and sodium hydrosulfite were obtained from Fisher Scientific, Inc. (Pittsburgh, PA). IAM.PC column packing was generously supplied by Regis Chemical Co. (Morton Grove, IL). Purfied proteins and antibodies Epoxide hydrolase, cytochrome P450 reductase, and cytochrome P450 IA2 were purified as previously described (27). P450IA1 was purified using HPLC chromatography by previously described procedures (28). Rat cytochromes P450 of the IIC subfamily, P450IIA1, and rabbit polyclonal antibodies to rat IIA1 and IIC7 were generous gifts of Dr. Wayne Levin. Rat cytochrome P450 IIIAl and mouse monoclonal anti-IIIAl were generous gifts of Dr. Steven Wrighton. Rabbit antibovine P450XXI immunoglobulin G (IgG) was raised from bovine P450XXI generously provided by Dr. S. Narasimhulu. Rabbit polyclonal antibody to equine P450XVII was a generous gift of Dr. Ian Mason. Rabbit polyclonal antibody to rat epoxide hydrolase was a generous gift of Dr. Charles Kasper. We also used, as reference standards, rat cytochrome P450IIA1, IIA2, IIB1, and IIB2 individually expressed in HepG2 cells (29, 30), which were kindly donated by Dr. Frank Gonzales. Animal treatments To determine the effects of hypophysectomy and ACTH, groups of five rats were sc administered either 4 IU ACTH or 2 mg/kg dexamethasone diluted in 9% saline. This was performed twice daily for 14 days. Animals were killed 12 h after the final treatment. Hypophysectomized animals were given 5% sucrose in drinking water for three days postsurgery and


972

CHARACTERIZATION OF A NOVEL RAT ADRENAL CYTOCHROME P450

treated with ACTH beginning 2 weeks after surgery. Rats were killed by decapitation, exsanginated, and had their adrenals removed, trimmed of fat, and combined for microsome preparation. Microsome preparation Microsomes were prepared as previously described (31, 32) with minor modifications. Bovine adrenal cortex was scraped from the capsule and medulla. Otherwise, whole organs were placed in homogenization buffer [0.25 M potassium phosphate (pH 7.25), 0.15 M potassium chloride, 0.01 M EDTA (pH 7.5), 0.25 mM PMSF, and 0.1 mM dithiothreitol (DTT)] and homogenized (60% power, Tekmar tissuemizer, Cincinnati, OH). After a 20-min centrifugation at 15,000 X g, the supernatants were recentrifuged at 105,000 x g for 90 min. Pellets were resuspended in buffer [0.1 M sodium pyrophosphate (pH 7.25), 10 mM EDTA (pH 7.5), 0.25 mM PMSF, and 0.1 mM DTT] and homogenized using Potter-Evehjem homogenizers and recentrifuged at 105,000 X g for 60 min. Microsomal pellets were resuspended (10-40 mg protein/ml) in storage buffer [0.1 M potassium phosphate (pH 7.25), 10 mM EDTA, 20% glycerol, 0.1 mM DTT, and 0.25 mM PMSF] and frozen in liquid N2. Procedures were performed at 4 C. Microsomes from HepG2 cells were prepared as described, with the following modifications. Cells were first lysed in 5 ml hypotonic buffer [10 mM Tris (pH 7.5), 10 mM KC1, and 0.5 mM EDTA] on ice for 10 min, followed by the addition of 15 ml homogenization buffer and sonication (four 15-sec bursts at 50% power; Sonicator Cell Disrupter, Heat Systems-Ultrasonics, Inc., Plainview, NY). Microsomal pellets were suspended directly in a storage buffer containing 30% glycerol, eliminating the wash step. Purification of rat adrenal cytochrome P450 Microsomes were solubilized using sodium cholate recrystallized three times (32). Microsomes were added dropwise to storage buffer containing 2.5 mg cholate/mg microsomal protein to a final protein concentration of 1.5-3 mg/ml. Solubilization was carried out at 4 C with continual mixing. After mixing for an additional 30 min, microsomes were centrifuged at 150,000 X g for 60 min; solubilization efficiency was typically greater than 90%. Sodium cholate concentration in the supernatants was reduced before metabolism assays by treatment with Biobeads SM2, prepared as previously described (33), and dialysis against 10 vol storage buffer. Solubilized microsomes were incubated at 4 C for 2-3 h to denature P450XXI and then passed over a small 0.5 ml column of Biobeads SM2 (to remove lipid droplets present with adrenal microsomes) and recentrifuged at 105,000 X g to remove particulate material. Supernatants were filtered through a 0.2-^m filter and applied to a preparative (10 cm X 21.1 mm) IAM.PC HPLC column packed with 12-jum particles and fitted with a guard column of the same material. This preparative IAM.PC column was synthesized, as previously described (24), using silica propylamine obtained from Regis Chemical Co. (Morton Grove, IL). Backpressure was maintained at 400-600 psi, and flow rates were 5 ml/min. The column effluent was monitored at 405 nm. The column was washed with buffer A [0.1 M potassium phosphate buffer (pH 7.25), 20% glycerol, 1 mM

Endo«1991 Vol 129 • No 2

EDTA, 0.5% Na cholate, and 0.1 mM DTT] until absorbance at 405 nm approached baseline; then, a 10-min linear gradient to 0.8% (wt/vol) Lubrol PX in buffer A was applied, eluting some of the bound protein. After washing with 0.8% (wt/vol) Lubrol PX for an additional 15 min, a 20-min linear gradient to 1.3% (wt/vol) Lubrol in buffer A was applied, eluting a sharp peak with a pronounced tail of 405 nm absorbance. The column was then washed for an additional 30 min at 1.3% (wt/vol) Lubrol PX before returning to starting conditions. Elution profiles were obtained by measuring absorbance at 417 and 280 nm on individual column fractions. To test reproducibility during the course of these studies column runs were repeated at both universities using several different HPLC setups. Fractions absorbing at 417 nm were pooled, based on their banding patterns on sodium dodecyl sulfate (SDS)-polyacrylamide gels, and P450 content was measured spectrally by CO binding. Pools were dialyzed in more than 10 vol storage buffer for 12-24 h at 4 C and treated batchwise with Biobeads SM2 to remove detergents. Residual Lubrol PX was monitored, as previously described (34, 35). Pools containing P450 were concentrated to 1-2 mg protein/ml, using 2-ml microconcentrators (Centricon-30, Amicon Corp., Danvers, MA), assayed for Lubrol PX content, and treated again with Biobeads SM2 as necessary. Pools were aliquoted and stored in liquid nitrogen for subsequent reconstitution, immunoblotting, and animal immunizations. The detergent concentration should be below 0.1 mg Lubrol PX/ml for optimal reconstitution. Protein contents were determined using bicinchoninic acid (Micro-BCA) reagents from Pierce Chemical Co. (Rockford, IL). Antibody preparation In preparation for immunization, anesthetized male New Zealand White rabbits (ketamine, 30 mg/kg; xylazine, 5 mg/ kg; azopromazine, 3 mg/kg) were injected interdigitally with sterile Evans blue dye (0.1 mg/ml; 0.5 ml/hind foot). Rabbits were immunized by directly injecting 20-50 ng filter-sterilized antigen, emulsified in sterile complete Freund's adjuvant, into each popliteal lymph node, surgically exposed under anesthesia. Serum was collected 14 and 21 days after immunization, and rabbits were boosted with 20-30 ng antigen in incomplete Freund's adjuvant using intradermal injections to the back of the anesthetized rabbits. Additional boosting was carried out using several im or sc injections. Rabbit IgG was isolated, exactly as previously described (27), or small quantities were isolated using a SeletiSpher-10 protein-A HPLC column (Pierce Chemical Co.) with citrate buffer, as described by the manufacturer. Chicken IgY was isolated from eggs of immune chickens exactly as described previously (36). DMBA metabolism studies Microsomal assays for DMBA metabolism were performed as follows. A NADPH-generating system was prepared by combining 0.25 ml 1 M potassium phosphate buffer (pH 7.4), 0.15 ml NADP (15 mg/ml), 0.15 ml MgCl2 (0.1 M), 0.4 ml G6P (25 mg/ml), and 0.05 ml glucose-6-phosphate dehydrogenase (30 U/ml), and it was kept on ice until use; microsomes (0.03 mg) and 0.02 ml NADPH-generating mix were combined; 0.1 nmol epoxide hydrolase was added when indicated; and final volumes were adjusted to 0.1 ml with double distilled H2O. Reactions


CHARACTERIZATION OF A NOVEL RAT ADRENAL CYTOCHROME P450 were preincubated at 37 C for 5 min before initiation, with substrate (15 nM DMBA) dissolved in 3 n\ acetonitrile. Reactions were stopped after 15 min with 2 ml of a stopping reagent (ethyl acetate-acetone-0.1 M DTT, 2:1:0.003). Reconstitutions were carried out using purified epoxide hydrolase and cytochrome P450 reductase, essentially as described previously (27). For all reconstitutions, dilaurylphosphatidylcholine was prepared by sonication as a 0.05% (wt/vol) solution in 10 mM potassium phosphate (pH 7.5). All assay mixtures contained 0.02 ml NADPH-generating mix, 0.004 mg dilaurylphosphatidylcholine, 0.01 nmol cytochrome P450, 0.02 nmol cytochrome P450 reductase, and 0.05-0.1 nmol epoxide hydrolase in a final volume of 0.1 ml. For metabolism studies involving antibody inhibition, incubation mixtures were preincubated with specific antibodies (15-40 mg IgG/nmol P450) at 25 C for 30 min. Reaction mixtures were extracted by vortexing, and organic phases were evaporated under nitrogen, resuspended in 0.1 ml methanol, and analyzed by reverse phase HPLC. Separation of DMBA metabolites was accomplished exactly as previously described (25). In all cases reactions were initiated by the addition of DMBA (1.5 nmol), but in selected samples, purified [:'H]DMBA was added to a specific activity of 0.2-0.5 ftCi/ nmol. Samples with high turnover numbers were used to generate conversion factors (picomoles of DMBA metabolite per cm fluorescence peak ht) based on the specific activity of the DMBA substrate. These conversion factors were used to quantitate metabolites in samples with unlabeled substrate. Gel electrophoresis

Protein banding patterns were examined after electrophoresis on 7.5% nongradient SDS-polyacrylamide gels. Slab gels (0.75 mm thick) were prepared according to the method of Laemmli (37), using a Bio-Rad (Richmond, CA) dual slab gel apparatus. Aliquots of microsomes, solubilized microsomes, column fractions, and purified P450s were prepared in a sample buffer containing glycerol, SDS, mercaptoethanol, and bromophenol blue. Mol wt standards (BSA, 66K; catalase, 58K; glutamate dehydrogenase, 53K; ovalbumin, 45K; and carbonic anhydrase, 29K) were prepared in the same manner. Gels were either transferred to nitrocellulose for imunoblotting or stained with Coomassie brilliant blue to visualize proteins. Immunoblotting Proteins were transferred from acrylamide gels to nitrocellulose, as described by Towbin et ai (38), using a TE51 Transphor apparatus (Hoefer Scientific, San Francisco, CA) at 200 mamp for 2 h. Immunoreactive proteins were visualized using either the peroxidase-antiperoxidase and diaminobenzidine/ hydrogen peroxide substrate (39) or alkaline phosphatase-conjugated antibody and nitrobluetetrazolium/5-bromo-4-chloro3-indoylphosphate (NBT/BCIP) substrate, using immunoscreening reagents from Promega (Madison, WI). Goat antirabbit IgG and horseradish peroxidase-antiperoxidase were obtained from Miles Laboratories (Naperville, IL), and goat antirabbit alkaline phosphatase and NBT/BCIP were purchased from Promega.

973

Analytical methods The reduced carbon monoxide-bound cytochrome P450 content was determined by the methods of Omura and Sato (40, 41) using an extinction coefficient of 91 mM/cm for 490-450 nm with an Amicon DW-2 scanning spectrophotometer. Absolute oxidized spectra were used to measure total P450 and P420 using an extinction coefficient of 105 mM/cm for the absorbance peak at 417 nm (42, 43). Cytochrome P450 reductase activity and cytochrome-65 in chromatography fractions were identified spectrophotometrically by previously reported procedures (44, 45). Epoxide hydrolase and P450XXI were identified in chromatography fractions using SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting. Densitometry of Coomassie-stained gels and immunoblots was performed using a Zeineh Soft Laser Scanning Densitometer (model SL-504-XL), using a tungsten lamp. Densitometer scans were integrated using a Videophoresis II program (Biomed Instruments, Inc., Fullerton, CA). Three correlation coefficients were obtained using a Cricket Graph program (Cricket Software, Malvern, PA): the first by plotting the area of the 57K Coomassie-stained band in microsomes, measured by densitometry and normalized to 10 /xg protein, us. the area, measured by densitometry, of this 57K protein in microsomes on immunoblots for each of five treatment groups; the second by plotting the area of the 57K Coomassie-stained band in microsomes, normalized to 10 /tg protein vs. total DMBA metabolism (picomoles per mg/min) in microsomes for each of the five treatment groups; and the third by plotting the area of the 57K immunoblot in microsomes us. total DMBA metabolism (picomoles per mg/min) in microsomes for each of the five treatment groups.

Results To examine the relationship between a hormonally sensitive cytochrome P450 and DMBA metabolism, rat adrenal microsomes were prepared from five different treatment groups. Groups of age-matched rats were treated with ACTH daily for 14 days, subjected to 14 days of ACTH suppression by daily administration of dexamethasone, or left untreated. Two other agematched groups were either hypophysectomized for 30 days or hypophysectomized for 16 days, followed by 14 days of daily ACTH treatment. The metabolism of DMBA by microsomes isolated from each group of animals is shown in Table 1. It can be seen that 30 days of hypophysectomy completely removes DMBA metabolism, which ACTH treatment effectively restores to 80% of the activity in age-matched intact untreated animals. Furthermore, ACTH treatment of intact animals elevates activities to twice those in ACTH-treated hypophysectomized rats. Fourteen days of ACTH suppression by dexamethasone only decreases activities to 40% the activity in age-matched intact untreated animals. Both results suggest that a factor other than ACTH is lost through hypophysectomy, which contributes to the reg-


974


Endo-i99i Voll29«No2

TABLE 1. Hormonal regulation of DMBA metabolic activity from male Sprague-Dawley rat adrenal microsomes DMBA metabolite (pmol/mg protein-min) Treatment"

Hydroxymethyl

Dihydrodiols

Phenols TVv+al

Untreated ACTH DEX HX HX + ACTH

5,6

8,9

3,4

7

12

A"

B»

ND ND ND ND ND

8.7 14.2 4.9 ND 6.4

4.1

ND ND ND ND ND

ND ND ND ND ND

14.0 19.0 7.3 ND 10.1

2.3 2.7 2.8 1.3 2.9

5.8 2.1 ND 3.0

29.0 ± 41.7 ± 17.1 ± 1.3 ± 22.3 ±

5.3 0.1 2.3 0.1 4.2

Data are reported as the means of duplicate rat adrenal microsomal incubations. Variation was less than 20% of the mean. Totals are reported as the mean ± SD. All incubations were carried out in the presence of exogenously added EH, as described in Materials and Methods. ND, Not detectable. 0 Twenty-five litter-matched 80-g Sprague-Dawley rats were divided into five treatment groups of five animals each; two groups were hypophysectomized. On day 16, two groups received 4 IU corticotropin-(l-24) twice daily for 14 days; one group received 2 mg/kg dexamethasone twice daily for 14 days. Animals were killed 12 h after their last treatment. ACTH, corticotropin-(l-24) treated; DEX, dexamethasone treated; HX, hypophysectomized; HX + ACTH, hypophysectomized, ACTH treated. 6 Total identifiable DMBA phenols: phenol A includes 2-OH and 3-OH DMBA; phenol B includes 4-OH DMBA.

ulation of DMBA metabolism in rat adrenal. The distribution of microsomal DMBA metabolites characterizes the regioselectivity of the P450 cytochromes catalyzing the reaction. Adrenal microsomes (in vitro) are deficient in epoxide hydrolase (46), resulting in partial diversion of epoxides to phenols, rather than to dihydrodiols. This diversion reduces formation of DMBA-3,4-diol, the precursor of the most reactive electrophile formed, namely DMBA-3,4-diol-l,2-oxide. To prevent this loss, sufficient purified rat liver epoxide hydrolase was added to maximize dihydrodiol formation, thus providing an accurate picture of regioselectivity. In the presence of exogenously added epoxide hydrolase, the major rat adrenal microsomal metabolites are phenol A (2-hydroxy- and 3-hydroxy-DMBA), DMBA-8,9-diol, and DMBA-3,4-diol in approximately constant ratios. Unlike liver microsomal metabolism of DMBA, the more chemically reactive 7- and 12-methyl groups and the Kregion (5,6-bond) are resistant to monooxygenation. These consistent yet distinctive product ratios suggest that a single, hormonally regulated cytochrome P450 enzyme is responsible for this pattern of DMBA metabolism. We have previously shown that rabbit antibodies raised to P450IIB1 and P450IA1 are ineffective at inhibiting this metabolism (47). Previous work suggests a relationship between DMBA metabolism and a 57K adrenal microsomal protein (6, 20). A SDS-polyacrylamide gel of microsomes isolated from the same five treatment groups is shown in Fig. 1. Two proteins in the expected mol wt range of P450s (apparent mol wt, 52K and 57K) are highly sensitive to hormonal treatment. The more rapidly migrating protein corresponds in apparent mol wt to rat P450XXI (20) and cross-reacts with rabbit IgG raised against bovine P450XXI. This protein differs from the 57K protein in exhibiting comparable suppression by hypophysectomy

and dexamethasone treatment. The various treatments described above affect the intensity of the 57K band in close parallel to the effects on DMBA metabolism (r = 0.87). We attempted to purify this protein in order to establish whether it is indeed the cytochrome P450 responsible for DMBA metabolism. Five hundred rat adrenal glands yield only 100-200 mg microsomal protein. Although the total P450 content of rat adrenal microsomes ranges from approximately 0.3-0.7 nmol/mg protein, the relative staining intensities of the 52K and 57K bands on SDS-PAGE (4:1) suggest that approximately 20% of the total adrenal P450 corresponds to the 57K protein. Another purification protocol for microsomal cytochrome P450, previously employed by Montelius and Rydstrom (20) to partially purify this protein, provided a lower turnover number for DMBA, low yields, and incomplete separation from P450XXI. We now report substantially more success through HPLC of solubilized rat adrenal microsomes using a novel IAM.PC column (24). The IAM.PC column uses a solid phase membranelike matrix, formed by covalently binding a phosphatidylcholine analog to a silica support through amidation of the carboxyl-terminal group of one fatty acid chain. Thus, an immobile surface is presented to the solubilized proteins which mimics the surface of a phospholipid membrane. Proteins associate with this surface and are selectively eluted by a nonionic detergent gradient, presumably in inverse order of their affinity for the immobile phase. The column is particularity effective for purification of many forms of cytochrome P450 that typically bind with high affinity and require relatively high detergent concentrations for elution. Many microsomal proteins, including cytochrome P420 and cytochrome P450 reductase, do not bind to this column under the conditions employed or are eluted by



975

CoomassieBlue

mi

o

a: -66 K :-58K -53 K -45 K

-29 K H

H A

DMWIA1

FlG. 2. Elution profiles from preparative IAM.PC chromatography of rat adrenal microsomes. Filtered solubilized microsomes were applied to the preparative column (10 cm x 21.1 mm) fitted with a precolumn of the same material, and the column was washed with buffer A (see Materials and Methods). The column was then eluted using the gradient shown of the same buffer containing 2% Lubrol PX, designated buffer B. Column fractions were collected and combined into pools, designated PT, A, B, and C, with elution C being the most pure in the 57K P450.

•f

A FIG. 1. Hormonal effects on rat adrenal microsomes shown by SDSPAGE analysis. Analyses were carried out on the same microsomal preparations used to generate DMBA metabolism data presented in Table 1. Lanes are labeled in the following manner: U, untreated (30 ^g protein); H, hypophysectomized (30 days, 30 ng protein); H+A, hypophysectomized (16 days), followed by ACTH (14 days, 15 fig protein); A, ACTH (14 days, 15 ng protein); D, dexamethasone (14 days, 30 figprotein); MW, mol wt standards BSA (66K), catalase (58K), glutamate dehydrogenase (53K), ovalbumin (45K), and carbonic anhydrase (29K); IA1, P450IA1 (5 ^g protein). Arrows indicate the positions of the ACTH-sensitive proteins in the P450 region.

low concentrations of the nonionic detergent Lubrol PX (48). We can, therefore, achieve substantial separation of the more stable 57K P450 from the relatively unstable P450XXI, which converts predominantly to P420 upon solubilization with sodium cholate and incubation for several hours at 4 C. The more stable 57K P450 elutes with considerable purification at higher detergent concentrations. A typical elution profile (Figs. 2 and 3) shows three distinct pools: an initial pool containing epoxide hydolase and P450XXI (fraction A); a pool containing P450XXI, cytochrome-65, the 57K P450, and a 40K protein (fraction B); followed by a steady broad elution pool consisting of about 70% 57K P450 and 20% 40K protein. This 40K protein is not recognized by rabbit anti-P450XXI,

is not increased by ACTH, and is not antigenic in rabbits. Fraction C exhibits a specific P450 content of 2 nmol/ mg and contains very little P420. The low specific content for such a relatively pure preparation may reflect a loss of heme during the purification or the presence of apoprotein in the solubilized microsomes. Based on the estimated proportion of 57K P450 in the original microsomes, from 30-45% was recovered in the purest fraction C. Reconstitutions of these fractions with phospholipid, cytochrome P450 reductase, and epoxide hydrolase (Tables 2 and 3) show that DMBA metabolism is essentially absent in the pass-through, but it is present in each of the fractions (A, B, and C), increasing in parallel with increasing content of the 57K protein. Despite different specific activities, the product distribution is similar, suggesting catalysis by the same form of P450. Two preparations exhibiting a 6-fold difference in starting material (Exp 1, 45 mg; Exp 2, 268 mg) provided essentially the same recoveries of DMBA metabolism and P450 in fraction C. The specific activity of DMBA metabolism in fraction C (1.5 nmol/nmol P450-min) is at least 8 times higher than the activity in adrenal microsomes, consistent with the estimate that the rat 57K P450 (P450RAP) comprises 15-20% of the active microsomal P450. It is also noteworthy that increases in specific activities


976


Coomassie Blue

FIG. 3. Electrophoretic analysis of rat adrenal microsomal fractions eluting from the preparative IAM.PC column. A SDS-polyacrylamide gel of fractions eluting from the IAM.PC column. Lanes contain 10 jig protein from pooled column fractions shown in Fig. 2. PT, A, B, C, and rat adrenal microsomes are compared. Bl, B2, and B3 represent three sequential fractions across the peak of elution. Additional lanes contain: IA1, 5 ng P450IA1; and MW, mol wt standards BSA (66K), catalase (58K), glutamate dehydrogenase (53K), and ovalbumin (45K). Arrows indicate the positions of the ACTH-sensitive proteins in the P450 region.

(picomoles per mg) of fractions A, B, and C occur with progressive decreases in the 40K protein. Polyclonal rabbit antibodies have been raised to this preparation. Low doses of antigen were used in an attempt to generate antibody specific for the major protein in the preparation, IgG isolated from this antiserum recognized only a single 57K protein band in immunoblots of the partailly purified P450RAP and rat adrenal microsomes. Apparently the 40K ACTH-insensitive protein is neither antigenic nor a degradation product of P450RAP, since it is not recognized by the antibody on immunoblots. We have used this rabbit antibody (antiP450RAP) to verify that hormonal regulation of the 57K P450 parallels microsomal DMBA metabolism and is distinct from P450XXI (Fig. 4). Immunoblots with IgG isolated from this antiserum show that the 57K protein

Endo'1991 Vol 129 • No 2

is induced by ACTH, removed by hypophysectomy, and only partially suppressed by dexamethasone, similar to the changes in DMBA metabolism seen in the same microsomes. These hormonal effects on the 57K immunoblot correlate reasonably well with the effects on both the 57K Coomassie-stained protein and microsomal DMBA metabolism, (r = 0.87 and r = 0.91, respectively). The correlation between the densities of the 57K immunoblot with microsomal DMBA metabolism and the density of the Coomassie-stained band indicates that most of the band is attributable to P450RAP. Both the apparent mol wt and the hormonal regulation of P450RAP are distinct from those of P450XXI. Figures 5 and 6 show that rabbit anti-RAP failed to recognize purified rat liver P450s of the IA, IIA, IIB, IIC, HE, and IIIA families, even when present on blots at 100 times higher concentrations. Similarly, antibodies raised against members of these families did not recognize P450RAP. It is interesting to note that anti-P450RAP does specifically recognize a 55K benz(a)anthracene-inducible P450 in mouse embryo fibroblast 10T1/2 microsomes (25, 49) as well as a 57K protein that is induced by betanaphthoflavone in rat liver but is not detectable in untreated or phenobarbital-induced rat liver microsomes. This liver protein appears to be distinct from P450IA1, which is not recognized by anti-RAP. AntiRAP faintly blots protein in lane N in Figs. 5 (I) and 6 (III) and lanes D-I in Fig. 6 (I) with a slightly lower mol wt than IIBl and IIB2. It also faintly blots protein in lanes G-K in Fig. 5 (II) with a higher mol wt than P450 IIA1 and IIA2. This is interpreted as nonspecific binding to HepG2 microsomal proteins. P450RAP is immunochemically distinct from the major microsomal steroidogenic cytochromes P450XXI and P450XVII, and it could not be detected in mouse, guinea pig, or bovine adrenal microsomes (Fig. 7). Figure 8 shows that IgG isolated from a hyperimmune rabbit antiserum inhibits DMBA metabolism by a solubilized preparation of rat adrenal microsomes in a dosedependent fashion. Inhibition is nearly complete, although high levels of control IgG also decreased product formation, in part by interfering with extractions. Comparable incubations with anti-IAl IgG do not inhibit DMBA metabolism in rat adrenal microsomes (47). Discussion This paper describes the partial purification and characterization of a rat adrenal microsomal P450 (P450RAP), which is responsible for the high level of hormonally sensitive rat adrenal microsomal DMBA metabolism. This purification uses primarily a single step, high efficiency HPLC protocol to partially purify a 57K P450 in active form and separate it from P450XXI.



977

TABLE 2. DMBA metabolic activity from reconstituted column fractions DMBA metabolite (pmol/nmol P450- min)° Hydroxymethyl

Dihydrodiols

Sample Expl Microsomes Solubilized microsomes Elution A Elution B Elution C Exp2 Solubilized microsomes Elution C

5,6

8,9

3,4

7.55 17.5 11.5 40.5 61.5

7.5 109

21.5 32.5 21.7

31.5 343 647

108 245

57.3

4.4

72.0

14.6 172

605

Phenols 0

A

Ba

70.0 67.5 47.0

13.5

7

12

6.0

ND ND ND ND ND

279 475

ND ND

436

12.5 8.0

17.5 36.0 1.8

75.2

60.0

171 252 124 852

8.5 4.5

60.0 43.0

1507

18.8 67.0

1427

157

After detergent removal with Biobeads SM2, fractions were reconstituted with exogenous EH and NADPH-dependent cytochrome P450 reductase in dilaurylphosphatidylcholine vesicles. Data are reported as the means of duplicate incubations. Variation was less than 20% of the mean values shown. ND, Not determined. " Total identifiable DMBA phenols: phenol A includes 2-OH and 3-OH DMBA; phenol B includes 4-OH DMBA. TABLE 3. Immobilized artificial membrane chromatography of rat adrenal microsomes

Sample

Anti P450XXI I

Anti P450RAP II

SA Total Total activity (pmol (pmol protein DMBA/mg c 6 DMBA/min) (mg) P450 P420 protein-min) Total (nmol)0

Expl 45 Microsomes 17.8 Solubilized 17.6 45 microsomes Pass-through 8.1 36.4 35 Elution A 0.12 0.2 0.1 Elution B 0.24 0.3 0.1 0.44 Elution C 0.8 0.1 Exp2 Solubilized 268 209 microsomes Elution C 5.3 11.6 2.0

3,044 4,435 ND 25 256

68 99

ND 207

1,206

1,067 3,015

32,813

122

16,553

3,123

ND, Not determined. Cytochrome P450 was measured spectrally by reduced CO binding, and cytochrome P420 was calculated by subtracting nanomoles P450 from total nanomoles of enzyme measured by absorbance at 417, as described in Materials and Methods. h Protein concentrations were measured using a Micro BCA assay kit from Pierce Chemical Co. C DMBA metabolism was assayed in duplicate, and data for total metabolism and specific activity were calculated based on mean values. Variation was less than 20% of the mean values shown. 0

Although complete purification has not been achieved, several pieces of evidence establish the identity of the 57K ACTH-sensitive protein as the P450 responsible for rat adrenal microsomal DMBA metabolism. The analysis of adrenal microsomal preparations from rats in various endocrine states shows a close correlation between the density of the 57K anti-P450RAP immunoblot and the density of the cooresponding 57K Coomassie-stained protein band (r = 0.87). Taken together with the fact that this antibody inhibits rat adrenal

— - — U H H A D

U H H A D

FIG. 4. Hormonal effects on rat adrenal microsomes analyzed by Western blotting using rabbit polyclonal antibodies to P450XXI and P450RAP. Lanes are labeled in the following manner: U, untreated; H, hypophysectomized (30 days); H + A, hypophysectomized (16 days), followed by ACTH (14 days); A, ACTH (14 days); D, dexamethasone (14 days). Blot I, stained with antibody to P450XXI using peroxidase antiperoxidase and diaminobenzine/hydrogen peroxide, contains 10 ng protein/lane. Blot II, stained with anti-P450RAP using alkaline phosphatase and BCIP/NBT, contains 3 Mg protein/lane. Arrows indicate the positions of the ACTH-sensitive proteins in the P450 region. These proteins are recognized by anti-P450RAP (57K) and anti-P450XXI (52K), respectively.

microsomal DMBA metabolism, this correlation suggests that the 57K protein band consists of predominatly P450RAP. These data do not allow distinction between apoprotein and intact hemoprotein. Thus, a significant proportion of apo-P450RAP in adrenal microsomes may contribute to the low specific activity (2 nmol/mg) of preparations which, nevertheless, exhibit 50-70% of the 57K protein. The level of this 57K protein was correlated with DMBA metabolism, both through reconstitution of various fractions generated in purification and during hormonal manipulations of the adrenal that change activities more than 10-fold. Reasonable correlations were


978

CHARACTERIZATION OF A NOVEL RAT ADRENAL CYTOCHROME P450 I

A BC

D E F G H I

J

K

L

E n d o • 1991 Vol 129 • No 2

M N O P Q R S

A m i P4SOIA1 —45 ~58 —53

AntiP450RA°

-45

A B C D E

F G H I J

K

L M N O P

0

Anti-P45OIIA1

Ami P450RAP

FlG. 5. Western blot analysis of P450RAP compared to P450IA1, P450IA2 (5-1), P450IIA1, and IIA2 (5-II) 3-methylcholanthrene-inducible rat liver P450s indicates that P450RAP is immunochemically distinct from these xenobiotic microsomal P450s. Specific contents of purified P450s were 2 (RAP) to 12 (P450IA1) nmol P450/mg protein. Liver microsomes had a specific content of 1-1.5 nmol P450/mg protein, and specific contents of BA-induced 10T1/2 and P450s expressed in HepG2 cells were between 0-1 nmol P450/mg protein. Amounts loaded per lane are expressed in picomoles of P450, except for liver microsomes, 10T1/2 microsomes and recombinant P450IIA1 and IIA2 expressed in HepG2 cells, which are expressed as micrograms of total microsomal protein *(see below). BNF-RLM, Liver microsomes isolated from j8-napthoflavone-treated rats (40 mg/kg for 3 days). I. Anti-P45OIA1 and anti-P450RAP: A, B, and C, RAP (0.2, 0.5, and 1.0 pmol); D, E, and F, IA1 (0.1, 0.5, and 1.0 pmol); G, H, and I, IA2 (0.4, 0.1, and 2.0 pmol); J, K, and L, BNF-RLM (0.5, 1, and 1.5 fig); M, IIC7 (2.0 pmol); N, IIBl (2.0 pmol); O and P, BA 10T1/2 microsomes (5 and 10 ft); Q, R, and S, rat adrenal microsomes (5, 10, and 15 fig). II. Anti-P450IIAl and anti-P450RAP: A and B, RAP (0.2 and 2.0 pmol); C, D, and E, IIA1 (Levin; 0.2, 0.4, and 2.0 pmol); F, G, and H, IIAl* (5, 10, and 20 tig); I, J, and K, IIA2* (5, 10, and 20 n); L and M, BA-10T1/2 microsomes (5 and 10 fig); N and O, BNF-RLM (0.5 and 1 fig); P and Q, 10T1/2 microsomes (15 and 30 fig).

apparent between DMBA metabolism and the 57K protein by either Coomassie staining or anti-P450RAP immunoblot (r = 0.87 and r = 0.91, respectively). We also show that anti-P450RAP can completely inhibit metabolism of DMBA in rat adrenal microsomes. Although this antibody only recognizes the 57K protein on immunoblots of rat adrenal microsomes, an additional potentially active protein may be recognized under nondenaturing conditions. The correlations cited above provide strong, but not conclusive, evidence that inhibition arises from anti-RAP interacting with the same 57K protein that predominates in the partially purified P450RAP preparation. Although the mol wt of this P450RAP isozyme is similar to that of the major 3-methylcholanthrene-inducible P450IA1 (56K), we show that, contrary to previous suggestions (6, 20), these two forms are immunochemically distinct. Consequently, P450RAP is not immunologically related to a recently described xenobiotic guinea pig adrenal cytochrome P450, which is male specific, suppressed by ACTH, and immunochemically related to rat P450IA1 (22, 23). The apparent mol wt of P450RAP, however, confirms earlier suggestions that a 57K ACTH-sensitive adrenal microsomal protein is in-

deed responsible for adrenal DMBA metabolism. Stimulation by ACTH and suppression by hypophysectomy of activity and protein levels are consistent with previous reports concerning the hormonal sensitivity of rat adrenal PAH metabolism (6) and possibly DMBA-induced necrosis (2, 50). The purification procedure adopted for this low abundance cytochrome P450 is new, employing a novel HPLC matrix mimicking the physiochemical environment of cell membranes (24). Studies of the purification of P450 cytochromes from other tissues and other membranebound enzymes indicate that this method is generally applicable and offers significant advantages in speed, efficiency, and efficacy over more conventional hydrophobic interaction chromatography as an initial step in the purification of these enzymes (48). Lack of immunocross-reactivity on immunoblots suggests that P450RAP has little homology with P450s of the rat IA, IIA, IIB, IIC, and IIIA families, confirming previous studies that have used antibodies raised against cytochromes from these subfamilies to probe rat adrenal microsomes (50, 51). Surprisingly, P450RAP appears more closely related to a benz(a)anthracene-inducible P450 cytochrome (P450EF), which we have recently


CHARACTERIZATION OF A NOVEL RAT ADRENAL CYTOCHROME P450 |

A B G D E F

An,i-P450..B,

G H

I

979

J K L M N O P

.... , aaJJfrMfe^-.fr.

^ —45 —58 -53

Anli RAP

—45

A B C D E

F G H I

J K L M N O

P 0 K

Ami P450IIC7 $ & l

Anti P4SORAP

A B C D E F G H I J K L M N O

P 0

R S

Ami P45O1HA1

Anli-P450RAP — 45

FIG. 6. Western blot analysis of P450RAP compared to phenobarbital-inducible (6-1), constitutive (6-II), and dexamethasone-inducible (6-III) rat liver P450s indicates that P450RAP is immunochemically distinct from these xenobiotic microsomal P450s. Specific contents of purified P450s were 2 (RAP) to 8 (P450IIIA1) nmol P450/mg protein. Liver microsomes had a specific of 1-1.5 nmol P450/mg protein, and specific contents of benz(a)anthracene-induced 10T1/2 cell microsomes and P450s expressed in HepG2 cells were 0-1 nmol P450/mg protein. Amounts loaded per lane are expressed as picomoles of P450, except liver microsomes, 10T1/2 microsomes which are expressed as micrograms of total microsomal protein. * (see below), Recombinant P450IIB1 and IIB2 are expressed in HepG2 cells. PB-RLM, Liver microsomes isolated from phenobarbitaltreated rats (100 mg/kg for 3 days). DEX-RLM, Liver microsomes isolated from dexamethasone-treated rats (2 mg/kg for 6 days). I. Anti-P450IIBl and anti-RAP, A, B, and C, RAP (0.2, 0.5, and 2.0 pmol); D, E, and F, IIBl* (0.2, 0.5, and 1.0 pmol); G, H, and I, IIB2* (0.2, 0.5, and 1.0 pmol); J, K, and L, PB-RLM (0.5, 1.0, and 2.5 ng); M, IIC7 (2.0 pmol); N, IAl (2.0 pmol); O and P, BA 10T1/2 microsomes (25 and 10 /ig); Q and R, rat liver microsomes (0.5 and 1.5 ng). II. Anti-P450IIC7 and anti-P450RAP: A, B, and C, RAP (0.02, 0.1, and 1.0 pmol); D, E, and F, IIC7 (0.1, 1.0, and 2.0 pmol); G, IIC13 (1.0 pmol); H, IICll (1.0 pmol); I, IIC12 (1.0 pmol); J, IIEl (1.0 pmol); K, IIC6 (1.0 pmol); L, IIAl (1.0 pmol); M, N, and O, BA 10T1/2 microsomes (30, 20, and 10 fig); P, 10T1/2 microsomes (30 ng); Q and R, untreated-RLM (0.5, 1.0, and 1.5 ng). III. Anti-P450IIIAl and anti-P450RAP: A and B, rat adrenal microsomes (5 and 10 ng); C, D, and E, RAP (0.02, 0.2, and 2.0 pmol); F, G, and H, IIIAl (0.2, 0.5, and 2.0 pmol); I, J, and K, DEX-RLM (0.5, 1.0, and 2.5 ng); L, M, and N, IIBl* (0.2, 0.5, and 2.0 pmol); O, P, and Q, PB-RLM (0.5, 1.0, and 2.5 /xg); R and S, BA 10T1/2 microsomes (25 and 10 ng).

isolated from the mouse embryo fibroblast 10T1/2 cell line. The respective rabbit IgGs and chicken IgY crossreact with the heterologous P450s on immunoblots and metabolic inhibition assays, respectively, and the product distribution of DMBA metabolites catalyzed by the two P450 cytochromes are similar (25, 49). The immunochemical relatedness of P450RAP and P450EF indicates structural similarity, and comparison of the respective isolations is of interest. In both cases,

initial specific activities for DMBA metabolism were similar (~40 pmol P450EF/mg microsomal protein). In both cases, final preparations appeared to be approximately 70% homogeneous on Coomassie-stained 7.5% SDS-PAGE, yet a low specific content was exhibited (12 nmol P450/mg protein). P450RAP was partially purified in a single HPLC chromatographic step, and P450EF was partally purified by a conventional method using hydrophobic interaction chromatography, followed by



980 ANTIP450RAP

R M G B

ANTIP450XVII

ANTIP450XXI

R M G B

R M G B

FIG. 7. Western blot analysis of rat, mouse, guinea pig, and bovine adrenal microsomes, using anti-P450RAP, anti-P450XXI, and antiP450XVII. Eight micrograms of protein were loaded per lane. Lanes are labeled as follows: R, rat adrenal microsomes; M, mouse adrenal microsomes; G, guinea pig adrenal microsomes; and B, bovine adrenal cortical microsomes. Each blot was reacted with anti-P450RAP, antiP450XXI, or anti-P450XVII, as indicated. Mol wt standards, BSA (66K), catalase (58K), glutamate dehydrogenase (53K), and ovalbumin (45K), were transferred and visualized using Ponceau S; positions are labeled. The arrow indicates the position of P450RAP.

300

Endo«1991 Vol 129 • No 2

results suggest that if the 55K mouse embryo fibroblast P450 is a homolog of P450RAP, its induction and organ distribution are different in the two species. It is interesting to note that a hormonally regulated P450-mediated PAH metabolism has been reported in both rat testes and ovary (52, 53) and that an ACTHinducible PAH metabolism has been reported in human fetal adrenal tissue (50, 54, 55). Using rabbit anti-RAP IgG, we have recently blotted a 57K protein in rat testicular microsomes, which disappears after 30 days of hypophysectomy. This protein is present at 20-fold lower concentrations than P450RAP. Preliminary studies of DMBA metabolism by rat testicular microsomes indicate a product distribution very similar to that seen in rat adrenal microsomes, again suggesting that a P450RAPrelated protein is present in the rat testis. Future studies will use the distinctive product distribution of DMBA metabolites and anti-P450RAP to elucidate the distribution and possible functional significance of P450RAP. One crucial question concerning P450RAP is whether there is a natural substrate. The induction of this protein by pituitary hormones might suggest some type of physiological function. Parallel studies will be conducted to obtain further purification, sequence information, and cDNAs so that the identification, distribution, and regulation of the P450RAP can be further elucidated.

Acknowledgments

0

10 20 30 antl-P450RAPmg/nmol P450

FlG. 8. Inhibition of solubilized reconstituted rat adrenal microsomal DMBA metabolism by rabbit anti-P450RAP IgG. Reconstitutions of solubilized rat adrenal microsomes were carried out as described in Materials and Methods. DMBA metabolism was accomplished in duplicate reactions at several concentrations of Ig, ranging from 0-50 mg/nmol total P450. The variation between duplicate assays was within 10% of the mean values shown.

anion and cation exchange media. Although the rat adrenal and mouse embryo fibroblast P450s may be homologs in the two species, clear differences in absorbance maximums for reduced CO complex formation suggest otherwise [446 (EF) vs. 450 (RAP); data not shown]. Rabbit anti-RAP failed to detect a homologous protein in untreated mouse adrenal microsomes. Mouse adrenal microsomes catalyze DMBA metabolism with a product distribution distinct from that of P450RAP (our unpublished observation) and do not exhibit an immunoblot with anti-RAP. These combined

The authors wish to thank Drs. Brian McNamara and Maro Christou for helpful discussions and critique. The technical assistance of Ying Chai and Jeff Van Stelle, and the photographic assistance of Janelle Gerhardt were most helpful. We also thank Gonzalez, Charles Kasper, Wayne Levin, Ian Mason, Lynn Pottenger, Chris Turner, Neil Wilson, and Steven Wrighton for generous gifts of cytochromes P450 and antibodies.

References 1. Waterman M, Simpson E 1985 Regulation of the biosynthesis of cytochromes P-450 involved in steroid hormone synthesis. Mol Cell Endocrinol 39:81-89 2. Huggins C 1979 Experimental Leukemia and Mammary Cancer. University of Chicago Press, Chicago 3. Gonzalez F 1989 The molecular biology of cytochrome P-450s. Pharmacol Rev 40:243-288 4. Huggins C, Morii S 1964 Induced protection of adrenal cortex against 7,12-dimethylbenz(a)anthracene\ Influence of ethionine. Induction of menadione reductase. Incorporation of thymidine-H3. J Exp Med 119:923-942 5. Horvath E, Somlgui A, Korvacs K 1971 Effect of estradiol on 7,12dimethylbenz(a)anthracene-induced adrenocortical necrosis. Arch Geschwulstforsh 37/3:203-209 6. Guenther T, Nebert D, Menard R 1979 Microsomal aryl hydrocarbon hydroxylase in rat adrenal: regulation by ACTH but not by polycyclic hydrocarbons. Mol Pharmacol 15:719-728 7. Huggins CB, Grand LC, Brilliantes FP 1961 Mammary cancer induced by a single feeding of polynuclear hydrocarbons and its suppression. Nature 189:204-207


CHARACTERIZATION OF A NOVEL RAT ADRENAL CYTOCHROME P450 8. Huggins CB, Sugiyama T 1966 Induction of leukemia in the rat by pulse doses of 7,12-dimethylbenz(a)anthracene. Proc Natl Acad Sci USA 55:74-81 9. Dandekar S, Saraswaki S, Zarble H, Young LJT, Cardiff RD 1986 Specific activation of the cellular Harvey-ras oncogene in dimethylbenzanthracene-induced mouse mammary tumors. Mol Cell Biol 6:4104-4108 10. Montelius J, Papadopoulous D, Bengtsson M, Rydstrom J 1982 Metabolism of polycyclic aromatic hydrocarbons and covalent binding of metabolites to protein in the rat adrenal gland. Cancer Res 42:1479-1486 11. Sawicki JT, Moschel RC, Dipple A 1983 Involvement of both synand anti-dihydrodiol-epoxides in the binding of 7,12-dimethylbenz(a)anthracene to DNA in mouse embryo cell cultures. Cancer Res 43:3212-3218 12. Vigny P, Brunissen A, Phillips D, Cooper C, Hewer A, Grover P, Sims S 1985 Metabolic activation of 7,12-dimethylbenz(a)anthracene in rat mammary tissue. Fluorescence spectral characteristics of hydrocarbon-DNA adducts. Cancer Lett 26:5159 13. DiGiovanni J, Nebzydoskii A, Deccinia P 1983 Formation of 7hydroxymethyl-12-methylbenz(a)anthracene in mouse epidermis. Cancer Res 43:4221-4226 14. Christou M, Marcus C, Jefcoate C 1986 Selective interactions of cytochrome P-450 with the hydroxymethyl derivatives of 7,12dimethylbenz(a)anthracene. Carcinogenesis 7:871-877 15. Nishjihara M, Winters C, Buzko E, Waterman M, Dufau M 1989 Hormonal regulation of rat Leydig cell cytochrome P45017a mRNA levels and characterization of a partial length rat P45017a. Biochem Biophys Res Commun 154:151-158 16. Hallberg E, Montelius J, Rydstrom J 1982 Effect of ACTH on cytochrome P-450 content and DMBA metabolism in immature rat adrenal. Biochem Pharmacol 32:709-710 17. Colby H, Rumbaugh R, Stitzel R 1980 Changes in adrenal microsomal cytochrome(s) P-450 with aging in the guinea pig. Endocrinology 107:1359-1363 18. Dibartolomeis M, Jefcoate C 1984 The interrelationship of polycyclic hydrocarbon metabolism and steroidogenesis in primary cultures of bovine adrenal cortical cells. Mol Pharmacol 25:475486

19. Dibartolomeis M, Christou M, Jefcoate C 1986 Regulation of rat and bovine adrenal metabolism of polyaromatic hydrocarbons by adrenocorticotropin and 2,3,7,8-tetrachlorodibenzo-p-dioxin. Arch Biochem Biophys 246:428-438 20. Montelius J, Rydstrom J 1982 Partial separation of two different cytochrome P450 from rat adrenal cortex microsomes by affinity chromatography with immobilized NADPH-cytochrome C reductase as a ligand. In: Laitinen M, Hannien O (eds) Cytochrome P450 Biochemistry, Biophysics and Environmental Implications. Elsevier, Amsterdam, pp 349-351 21. Mochizuki H, Kominami S, Takemori S 1988 Examination of differences between benzo(a)pyrene and steroid hydroxylases in guinea pig adrenal microsomes. Biochim Biophys Acta 964:83-89 22. Black V, Barilla J, Russo J, Martin K 1989 A cytochrome P450 immunochemically related to P450c,d (P450 I) localized to the smooth microsomes and inner zone of the guinea pig adrenal. Endocrinology 124:2480-2493 23. Black V, Barilla J, Martin K 1989 Effects of age, adrenocorticotropin, and dexamethasone on a male specific cytochrome P450 localized in the inner zone of the guinea pig adrenal. Endocrinology 124:2494-2498 24. Pidgeon C, Venkataram U 1989 Immobilized artificial membrane chromatography: supports composed of membrane lipids. Anal Biochem 176:36-47 25. Pottenger L, Jefcoate C 1990 Characterization of a novel cytochrome P450 from the transformable cell line, C3H/10T1/2. Carcinogenesis 11:321-327 26. Tierney B, Hewer A, MacNicoll AP, Giovanni PG, Rattle HH, Walsh C, Grover PL, Sims P 1978 The formation of dihydrodiols by the chemical or enzymatic oxidation of benz(a)anthracene and 7,12-dimethylbenz(a)anthracene. Chem-Biol Interact 23:243-257 27. Wilson N, Christou M, Turner CR, Wrighton SA, Jefcoate C 1984

981

Binding and metabolism of benzo(a)pyrene and 7,12-hylbenz(a)anthracene by seven purified forms of rome P450. Carcinogenesis 5:1475-1483 28. Funae Y, Imaoka S 1985 Simultaneous purification of multiple forms of rat liver microsomal cytochrome P450 by high-performance liquid chromatography. Biochim Biophys Acta 842:119-132 29. Aoyama T, Korzekwa K, Matsunaga T, Nagata K, Gillette J, Gelboin HV, Gonzalez FJ 1990 cDNA-directed expression of rat IIAl and IIA2: catalytic activities toward steroids and otics and comparison with the enzymes purified from liver. Drug Metab Dispos 18:378-382 30. Aoyama T, Korzekwa K, Nagata K, Adesnik M, Reiss A, Lapenson DP, Gillette J, Gelboin HV, Waxman DJ, Gonzalez FJ 1989 Sequence requirements for cytochrome P-450 IIBl catalytic activity; alteration of the stereospecificity and regioselectivity of steroid hydroxylation by a simultaneous change of two hydrophobic amino acid residues to phenylalanine. J Biol Chem 264:21327-21332 31. Fahl W, Jefcoate C, Kasper C 1978 Characteristics of benzo(a)pyrene metabolism and cytochrome P450 heterogeneity in rat liver nuclear envelope and comparison to microsomal membrane. J Biol Chem 249:3106-3113 32. van der Hoeven TA, Coon MJ 1974 Preparation and properties of partially purified cytochrome P450 and reduced nicotinamide adenine dinucleotide phosphate cytochrome P450 reductase from rabbit liver microsomes. J Biol Chem 249:6302-6310 33. Holloway P 1973 A simple procedure for removal of Triton X-100 from protein samples. Anal Biochem 53:304-308 34. Garewal H 1973 A procedure for the estimation of microgram quantities of Triton X-100. Anal Biochem 54:319-324 35. Goldstein S, Blecher M 1975 The spectrophotometric assay for polyethoxy nonionic detergents in membrane extracts: a critique. Anal Biochem 64:130-135 36. Poison A, Coetzer T, Kruger J, van Maltzahn E, van der Merve KJ 1985 Improvements in the isolation of IgY from the yolks of eggs laid by immunized hens. Immunol Invest 14:323-327 37. Laemmli UK 1970 Cleavage of structural proteins during the assembly of the head group of bacteriophage T4. Nature 227:680685 38. Towbin H, Staehelin T, Gordon J 1979 Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76:43504354 39. Guengerich FP, Wang P, Davidson NK 1982 Estimation of isozymes of microsomal cytochrome P450 in rats, rabbits, and humans using immunochemical staining coupled with sodium dodecyl sulfate polyacrylamide gel electrophoresis. Biochemistry 21:16981706 40. Omura T, Sato R 1964 The carbon monoxide binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J Biol Chem 239:2370-2378 41. Omura T, Sato R 1964 The carbon monoxide binding pigment of liver microsomes. II. Solubilization, purification and properties. J Biol Chem 239:2379-2385 42. Yu CA, Gunsalus IC, Kataguri M, Suhara K, Takemori S 1974 Cytochrome P-450Cam I- Crystalization and properties. J Biol Chem 249:94-101 43. Fujita T, Shoeman DW, Mannering GJ 1973 Differences in P-450 cytochromes from livers of rats treated with phenobarbital and with 3-methylcholanthrene. J Biol Chem 248:2192-2201 44. Strobel HW, Dingnam JD 1978 Purification and properties of NADH-cytochrome P450 reductase. Methods Enzymol 52C:89-97 45. Estabrook RW, Werringloer J 1978 The measurement of difference spectra: application to the cytochromes of microsomes. Methods Enzymol 52C:212-221 46. Christou M, Jovanovich MC, Jefcoate C 1989 Epoxide hydratase: sex specific expression and rate limiting role in DMBA metabolism. Carcinogenesis 10:1883-1888 47. Christou M, Wilson NM, Jefcoate C 1987 Expression and function of three cytochrome P-450 isozymes in rat extrahepatic tissue. Arch Biochem Biophys 258:519-534 48. Pidgeon C, Stevens J, Otto S, Jefcoate C, Marcus C 1991 Immobilized artificial membrane chromatography: rapid purification of


982


functional membrane proteins. Anal Biochem 194:163-173 49. Pottenger L, Christou M, Jefcoate C 1991 Purification and immunological characterization of a novel P450 from C3H/10T1/2 cells. Arch Biochem Biophys 286:488-497 50. Hallberg E 1990 Metabolism and toxicity of xenobiotics in the adrenal cortex, with particular reference to 7,12-dimethylbenz(a)anthracene. J Biochem Toxicol 5:71 51. Bengtsson M, Hallberg E, Georgellis A, Rydstrom J 1990 On the identity of the xenobiotic-metabolizing form(s) of cytochrome P450 in endocrine organs. Cancer Lett 52:235-241 52. Lee IP, Suzuki K, Mukhtar H, Bend JR 1980 Hormonal regulation

Endo-1991 Vol 129 • No 2

of cytochrome P-450-dependent monooxygenase activity and epoxide-metabolizing enzyme activities in testis of hypophysectomized rats. Cancer Res 40:2486-2492 53. Bengtsson M, Rydstrom J 1973 Regulation of carcinogen metabolism in the rat ovary by the estrous cycle and gonadotrophin. Science 219:1437-1438 54. Hornsby PJ, Harris SE, Aldern KA 1985 Hormonal regulation of benzo[a]pyrene metabolism in human adrenocortical cell cultures. Biochem Biophys Res Commun 131:167-173 55. Juchau MR, Pedersen MG 1973 Drug biotransformation reactions in the human fetal adrenal gland. Life Sci 12:193-204


Polycyclic aromatic hydrocarbon metabolism in rat adrenal, ovary, and testis microsomes is catalyzed by the same novel cytochrome P450 (P450RAP).

Mitochondrial targeting of cytochrome P450 (CYP) 1B1 and its role in polycyclic aromatic hydrocarbon-induced mitochondrial dysfunction.

Cellular localization of cytochrome(s) P-450 metabolizing polycyclic aromatic hydrocarbons in the rat adrenal cortex.

Cytochrome P450 associated with insecticide resistance catalyzes cuticular hydrocarbon production in Anopheles gambiae.

Characterization of novel polycyclic aromatic hydrocarbon dioxygenases from the bacterial metagenomic DNA of a contaminated soil.

Cytochrome P450IA1 in hepatic lesions of a teleost fish (Fundulus heteroclitus) collected from a polycyclic aromatic hydrocarbon-contaminated site.

Human cytochrome P450 27C1 catalyzes 3,4-desaturation of retinoids.

Cyclopenta[c,d]pyrene: a highly mutagenic polycyclic aromatic hydrocarbon.

A CASE-SAR analysis of polycyclic aromatic hydrocarbon carcinogenicity.

Metabolism of an Alkylated Polycyclic Aromatic Hydrocarbon 5-Methylchrysene in Human Hepatoma (HepG2) Cells.

Cytochrome P450 maintenance and diazepam metabolism in cultured rat hepatocytes.

Metabolism of (-)-cis- and (-)-trans-rose oxide by cytochrome P450 enzymes in human liver microsomes.

Polycyclic aromatic hydrocarbon generation in heat-processed sundried salt.

Urinary polycyclic aromatic hydrocarbon biomarkers and diabetes mellitus.

Mechanism of action of food-associated polycyclic aromatic hydrocarbon carcinogens.

Stochastic atomistic simulation of polycyclic aromatic hydrocarbon growth in combustion.

New insights into thermal decomposition of polycyclic aromatic hydrocarbon oxyradicals.

Polycyclic aromatic hydrocarbons: from metabolism to lung cancer.

Erratum to: online determination of polycyclic aromatic hydrocarbon formation from a flame soot generator.

Urinary polycyclic aromatic hydrocarbon metabolites as biomarkers to woodsmoke exposure - results from a controlled exposure study.

Characterizing particulate polycyclic aromatic hydrocarbon emissions from diesel vehicles using a portable emissions measurement system.

Metabolism of nitro-polycyclic aromatic hydrocarbons.

Inhibition of Cytochrome P450 by Propolis in Human Liver Microsomes.

Online determination of polycyclic aromatic hydrocarbon formation from a flame soot generator.