Differential expression and function of three closely related phenobarbital-inducible cytochrome P-450 isozymes in untreated rat liver

ARCHIVES

Vol.

OF BIOCHEMISTRY

256, No. 2, August

AND

BIOPHYSICS

1, pp. 407-420,

198’7

Differential Expression and Function of Three Closely Related Phenobarbital-Inducible Cytochrome P-450 lsozymes in Untreated Rat Liver NEIL

M. WILSON,’

Environmental

MAR0

Toxicology

Center,

CHRISTOU, University

Received

AND

of Wisconsin,

January

COLIN Madison,

R. JEFCOATE2 Wiscmin

53706

2,198’7

The levels of expression of cytochromes P-450b and P-450e (both inducible by phenobarbital (PB) and differing by only 14 of 491 amino acids) in liver microsomes from untreated male rats were separately quantitated by Western blotting with a polyclonal antibody raised against P-450b that is equally effective against P-450e (anti P-450b/e). A protein with mobility identical to P-450e was detected in all microsomal samples. hlicrosomes from uninduced livers of individual male rats from five different strains exhibited only minor interstrain and interindividual variability in the expression of P450e (1’7 + 5 pmol P-450e/mg microsomal protein) with the exception of the Brown Norway strain (8.5 f 0.5 pmol P-450e/mg). Expression of P-450b varied widely from undetectable levels (~2 pmol/mg) in most Sprague-Dawley rats to about 50% of P-450e levels in Fischer and Brown Norway strains. Anti P-450b/e inhibited total metabolism of 7,12-dimethylbenz[a]anthracene (DMBA) by uninduced microsomes, to an extent dependent on rat strain (15-30%),predominantly through inhibition of formation of 12hydroxymethyl-‘7-methyl BA (12HOMMBA) (65-85%),the major metabolite of purified P-450e. A specific activity for P-450e-dependent DMBA metabolism was calculated from four sets of microsomes where the P-450b content was either undetectable or very low (0.7-1.0 nmol/nmol P-450e/min-l). Comparable calculated activities were, however, obtained from other untreated rat liver microsomes where P-450b levels were significant. Polymorphism in P-450b was detected but did not affect total P-450b expression or the sensitivity of DMBA metabolism to anti P-450b/e. A fourth band of greater mobility than P-450b (apparent M, < 50,000), was also recognized by anti P-450b/e. The intensity of this band did not vary among individual rats or among the different strains and therefore did not correlate with the sensitivity of microsomal DMBA metabolism to anti P-450b/e. A monoclonal antibody (MAb) against P-450b (2-66-3) recognized P-450’s b, b2, and e on Western blots but did not react with this higher mobility band. MAb 266-3 and two other MAbs produced against P-450b inhibited 12-methylhydroxylation of DMBA by untreated rat liver microsomes to the same extent as anti P-450b/e. Following PB induction, P-450b was induced to about double the level of P-450e in most rat strains examined. This study demonstrates that cytochrome P-450e is selectively expressed in livers of untreated rats and plays a substantial role in the metabolism of DMBA, despite its relatively low level of expression (~3% of the total spectrally detectable P-450). 0 1987 Academic Press, Inc. The microsomal cytochrome P-450 syskern is responsible for the metabolism of a wide variety of xenobiotics as well as en1 Present address: Department Health, School of Public Health

Medicine, University of Washington, Seattle, WA 98195. * To whom correspondence should be addressed at Department of Pharmacology, University of Wisconsin Medical School, 1300 University Avenue, Madison, WI 53706.

of Environmental and Community 407

0003-9861/87 $3.00 Copyright 0 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.

408

WILSON,

CHRISTOU,

dogenous compounds (1). Much of this diversity is due to the existence of multiple forms of this enzyme, with each isozyme exhibiting broad but overlapping substrate specificities (2). At least 14 distinct isozymes have been purified from rat liver microsomes (3-14), and it has been shown that several of these enzymes exhibit substantial sequence homology (1516). Studies utilizing clones representing specific P450 isozymes have demonstrated that there are several subgroups within the cytochrome P-450 gene family. P-450’s c and d represent one subfamily, which is inducible by a number of polycyclic aromatic hydrocarbons (PAH)3 (17, 18). P-450’s b and e are members of a second subfamily consisting of at least six distinct genes (19,20, 21), several of which are inducible by phenobarbital (PB) and a variety of other xenobiotics. Cytochrome P-450 PCN represents a third distinct subfamily of genes (22). In addition, several groups have reported the purification of distinct P-450 isozymes from untreated rat livers (4, 911). Recent work utilizing both immunological (23) and molecular biological (24) methods suggests that several of these P450 forms isolated from uninduced rats (i.e., P-4503 f, g, h, i, j, and k) may represent yet another distinct subfamily of P450 genes. Amino acid and cDNA analyses show that P-450’s b and e are extremely closely related, differing by only 14 out of 491 amino acids. Microheterogeneity in these enzymes has been demonstrated in different strains of rats (25,26). In addition, the close similarities in gene structure particularly the 5’-flanking region (19) lend added a Abbreviations used: DMBA, 7,12-dimethylbenz[a]anthracene; 12HOMMBA, 12-hydroxymethyl7-metbylbenz[a]anthracene; ‘IHOMMBA, 7-hydroxymethyl-12-methylben$o]anthracene; 3,4-diol, DMBA3,4-didhydrodiol; 5,6-diol, DMBA-5,6-dihydrodiol; 8,9diol, DMBA-8,9-dihydrodiol; P-450, cytochrome P-450; PAH, polycyclic aromatic hydrocarbons; SD, SpragueDawley; LE, Long-Evans; WI, Wistar; BN, Brown Norway; FI, Fischer F344; PB, phenobarbital; RID, radial immunodiffusion; PAP, peroxidase/anti-peroxidase; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; MAb, monoclonal antibody; NBS, nonspecific ascites fluid.

AND

JEFCOATE

interest to the control of the synthesis of these proteins. Although these two isozymes appear to be coordinately induced by PB with at least twice as much P-450b as P-450e expressed, preliminary studies suggest that the latter isozyme may be constitutively favored (27). Immunoquantitation of the low levels of P-450 cytochromes found in uninduced animals presents two major problems; detection of extraneous protein and an inability to distinguish active cytochrome from apoprotein. It is, therefore, important to establish a functional contribution for any immunologically detected P-450 cytochrome. Although most substrates exhibit indistinguishable product distributions when metabolized by P-450’s b and e (12, 28), we have found that they exhibit distinct regioselectivities in the metabolism of the PAH 7,12-dimethylbenz[alanthracene (DMBA) (29). P-450e showed a preference for hydroxylation at the 12methyl position, whereas P-450b produced more of the 7-hydroxymethyl metabolite. In this study, we have used Western blot analysis in conjunction with measurement of DMBA metabolism to examine the expression and function of cytochromes P450b and e. We demonstrate that P-450e is preferentially expressed relative to P-450b in uninduced rat liver and that it is responsible for a greater proportion of the metabolism of DMBA in microsomes isolated from uninduced rat livers than would be predicted from its relatively low levels of expression. MATERIALS

AND

METHODS

Materials. rH]DMBA was purchased from Amersham Radiochemicals (Arlington Heights, IL) and was purified by reverse-phase HPLC as previously described (29). Unlabeled DMBA metabolite standards were obtained from the NC1 Chemical Repository. Hydrogen peroxide, Tween 20, Lubrol PX, and agarose (Type II) were purchased from Sigma Chemical Co. (St. Louis, MO). PB was obtained from Mallinckrodt (St. Louis, MO). Goat anti-rabbit IgG and horseradish peroxidase/anti-peroxidase (PAP) were from Miles Laboratories (Naperville, IL), and 3,3’-diaminobenzidine was from Aldrich Chemical Co. (Milwaukee, WI). Nitrocellulose paper and materials for electrophoresis were obtained from Bio-Rad Laboratories (Richmond, CA). All other materials for purification of enzymes and for reconstitution of enzymatic ac-

CYTOCHROME

P-450e

IN

tivity were obtained from sources described previously (29). Preparation of microsmes and pwijied enzymes. Male SD rats weighing 100 g were obtained from the Holtzman Company (Madison, WI) and housed on wire-bottom cages for 5 days prior to use. Rats were given unlimited access to food (Lab Blox, Wayne Rodent Blox) and water, except that they were starved for 24 h prior to sacrifice. Rats treated with PB (100 mg/kg injected ip in 0.9% NaCl, 3 days) were housed for 5 days prior to induction. For experiments examining strain variation, Long-Evans (LE), Wistar I WI), Brown Norway (BN), and Fischer F344 (FI) rats I male, 100 g) were also obtained from the Holtzman Company. Microsomes were prepared as described previously (29). Cytochromes P-450b and e were purified from PB-induced SD rats (29). Immunological studies. Antibodies to purified cytochrome P-450b were generated in New Zealand White rabbits as described by Thomas et al (30). Purified IgG fractions were prepared using protein ASepharose (29). Radial immunodiffusion (RID) assays were performed as described by Thomas et al. (31). Since antibodies to P-450’s b and e cross-react completely, RID could only be used to determine total immunoreactive protein. Monoclonal antibodies (MAbs) to rat hepatic P-450b (2-66-3, 4-29-5, and 47-1; 32), prepared by the hydridoma technique at the National Cancer Institute of Laboratory of Molecular (:arcinogenesis, were a kind gift from Drs. H. V. Gelboin and F. K. Friedman. Electrophoresis and Western blotting. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed as described by Laemmli (33) with the exception that electrophoresis was run at 30 mA per gel and was continued for 60 min after the dye front had eluted from the gel to optimize the separation of P-450’s b and e. We have found that the separation and sharpness of protein bands varies considerably with different lots and suppliers of SDS. Thus, it may be necessary to modify these conditions if the source of SDS is changed. Proteins were transferred to nitrocellulose as described by Towbin et al (34) using a Hoeffer (San Francisco, CA) TE51 Transphor apparatus at a setting of 300 mA for 2 h. Immunoreactive proteins were visualized using the PAPdiaminobenzidine reaction, as described by Guengerich et al. (35), with the exception that phosphate-buffered saline containing 0.3% Tween 20 (36) was substituted for the bovine serum albumin-fetal calf serum buffer. This resulted in a significant decrease in nonspecific staining. In addition, PAP was used at a di:ution of 112000 as suggested by Domin et al. (37). A dose-response curve was generated to determine the amount of anti P-450b/e IgG required to give optimal :staining. Immunoreactive proteins were quantitated by :scanning with a Zeineh model SL-504-XL Soft Laser

UNINDUCED

RAT

409

LIVER

densitometer. Peak areas of the resulting traces were then calculated using a Hewlett-Packard digitizer. Standard curves with known amounts of P-450b or e ranging from 5 to 100 ng were run on each gel. In addition, several lanes containing known amounts of both P-450’s were run and quantitated to ensure that the two forms could be resolved by the densitometer. For determination of immunoreactive bands in microsomal samples, duplicate lanes with each of three different amounts, ranging from 1 to 50 pg of microsomal protein, were electrophoresed and blotted. When greater than 50 pg of protein was run, it was found that the excess protein in the cytochrome P450 region (i.e., 48-56 kDa) interfered with staining, presumably by preventing optimal binding of the primary antibody to the antigen. For microsomal samples which contained a third immunoreactive band which migrated between P-450’s b and e (P-450b2), the three peak areas were calculated as shown in Fig. 1. The total area under the curve for all three peaks was first determined. Then the areas representing P-450’s b and e were each calculated by measuring the outer halves of their respective peaks and then doubling each value. The area attributable to P-450bz (i.e., the middle band) was then determined by subtracting the values for P-450’s b and e from the total peak area. Metabolism studies. Total oxidation of DMBA was determined by the method of Van Cantfort et al. (38). Regioselectivity of metabolism of DMBA was determined by HPLC as described previously (39). Conditions for metabolism by microsomes or by purified cytochromes P-450b or e in reconstituted vesicles were

e

bz=T-2(0.5b+OSe)

FIG. 1. Densitometric scan of untreated SD rat microsomes (30 pg), that contained detectable of P-450’s b, bz, and e, illustrating the method for calculation of the peak area attributable to bp. Details of this method are described under terials and Methods.

liver levels used P-450 Ma-

410

WILSON.

CHRISTOU,

as previously described (29), with the exception that 5 PM DMBA was used for studies with uninduced microsomes (to allow conversion of a greater percentage of DMBA to metabolic products), whereas 15 pM DMBA was used for PB-induced microsomes. Preliminary experiments demonstrated that there were no significant differences in the regioselectivity of DMBA metabolism by uninduced microsomes at either 5 or 15 PM DMBA. For studies involving inhibition of metabolism by the polyclonal antibody, anti P-450b/e, or the MAbs (2-66-3, 4-29-5, and 4-7-l), microsomes were preincubated with each IgG (2.5 mg/nmol total P-450) at 22°C for 25 min. Preliminary experiments showed that this concentration of each antibody provided maximal inhibition with microsomes isolated from uninduced rats. Control incubations were carried out in the presence of equivalent concentrations of preimmune IgG or nonspecific ascites fluid (NBS) for anti P-450b and MAbs, respectively. The contribution of cytoehromes P-450b, ba or e to microsomal DMBA metabolism was determined by comparing metabolism by microsomes in the presence of saturating levels of anti P-450b/e IgG to that observed in control incubations. In the absence of detectable levels of P-450’s b and ba, the difference between these two incubations was attributed to microsomal cytochrome P-45Oe. The rate of metabolism by P-450e was then determined by dividing this microsomal rate by the amount of P450e present, as determined by Western blots. Analytical methods. Proteih concentrations were determined according to the method of Lowry et al (40) using bovine serum albumin as the standard. Total spectrally detectable cytochrome P-450 content was determined by the method of Omura and Sato

AND

JEFCOATE

(41) using an extinction for 450-490 nm.

of 91 mM-’

Coefficient

cm-’

RESULTS

Quantitation of cytochromes P-450b and P-45Oe. A recent study from this laboratory (29) showed that P-450’s b and e exhibit unique patterns of metabolism of DMBA in a reconstituted system. To examine whether these isozymes function differently in the microsomal membrane, it was necessary to quantitate each protein separately. By modifying the conditions for SDS-gel electrophoresis (as described in detail under Materials and Methods), these two closely related PB-inducible isozymes were separated sufficiently to allow their individual densitometric quantitation on Western blots (Fig. 2). The antibody used was raised in rabbits against pure P-450b and recognized both P-450b and P-450e but not six other purified rat liver P-450 cytochromes. For both P-450b and P-450e, the response to anti P-450b/e was linear from 5 to 100 ng, as shown in the densitometric traces. Liver microsomal samples from untreated and PB-treated SD rats were analyzed by Western blotting and compared to authentic P-450b and e standards (Fig. 3, lane 1). As expected, bands comigrating

b

FIG. 2. Western blot of purified P-450b (lanes 1, 2, and 3; 10, 40, and 80 ng respectively), P-450e (lanes 4, 5, and 6; 10, 40, and 80 ng respectively), and mixtures of the two isozymes (lane 7, 40 ng each; and lane 8,80 ng each). SDS-PAGE, electrophoretic transfer, and immunochemical staining were performed as described in detail under Materials and Methods. Densitometric scans of each sample are shown directly below each lane. Note that the amounts in lane 8 represent the point at which the standard curve for each isozyme deviates from linearity.

CYTOCHROME

2

1

FIG. 3. Western blot I, 40 ng each) and liver ther PB-induced (lane 30 pg) SD rats. Each the pooled livers from

P-450e

IN

UNINDUCED

of purified P-450’s b and e (lane microsomes isolated from ei2, 1 pg) or untreated (lane 3, microsomal sample represents six rats.

with P-450’s b and e were detected in microsomes from PB-treated animals (Fig. 3, lane 2). Quantitation of the two bands showed that P-450b was present at nearly two times the level of P-450e (Table I). Microsomal samples from individual uninduced SD rat livers contained predominantly one band which comigrated with cytochrome P-450e (Fig. 3, lane 3). Comparison with authentic P-450e standards indicated a P-450e content of 18 k 6 pmol/ mg (n = 12). A band comigrating with cytochrome P-450b was detectable in only 2 of 12 microsomal samples examined (data not shown). Immunoquantitation of this band indicated very low levels of expression (2-3 pmol/mg). A third band with mobility

TABLE

I

CONTENT OF PB-INDUCIBLE CYTOCHROME P-450 SPRAGUE-DAWLEY RAT LIVER MICROSOMES Cytochrome

P-450

Western Treatment None PB

Spectral (total)

411

LIVER

intermediate to that of forms b and e and referred to in this report as P-450bz, was also seen in some microsomal preparations isolated from both untreated and PBtreated rats. This is consistent with previous reports of polymorphic variation of P-450b in rats (25). P-450b2 was detectable in only 3 of 12 microsomal preparations of uninduced SD rat liver microsomes (data not shown). Similar to P-450b, the levels of expression of P-450bz were significantly lower than those of P-450e (Table I). With the relatively large sample loads required for detection of P-450’s by anti P450b/e in uninduced rat liver microsomes (30-50 pg), an additional prominent band of greater mobility than either P-450b or P-450e was observed on blots (Fig. 3, lane 3). The intensity of this band was fairly constant in all the microsomal samples examined. A MAb produced against P-450b (2-66-3; (32)) recognized cytochromes P450b, be, and e on Western blots but did not detect this band of lower molecular weight (Fig. 4, compare lanes 2 and 3 with lanes 7 and 8). It did, however, stain a different band of still greater mobility in uninduced microsomal samples (Fig. 4, lanes 2 and 3). Since the inhibitory effects of MAb 2-66-3 on the 12-methyl hydroxylation of DMBA by untreated rat liver microsomes was comparable to that exhibited by the polyclonal anti P-450b/e (see later sections), it was concluded that these addi-

3

SPECIFIC

RAT

ISOZYMES

IN

(pmol/mg)’

blot”

b

640 + 200 1910 + 250

P-450b 3f 3 648 f 128

P-450bz 3.5* 5 326 +19

P-450e

(P-450

18f 6 473 + 170

a Data are reported as the means f SD of duplicate analyses from 12 individual rats. b Total P-450 was determined by the reduced-CO spectral method (41). ’ Individual isozymes were quantitated by Western blot analyses as described in detail Methods. d Total anti P-450b/e reactive protein was determined by RID (31).

RIDd b + b + e)

20+ 10 1240 + 250

under

Materials

and

412

FIG. 4. Western blot of microsomal samples stained with either monoelonal antibody 2-66-3 (lanes l-5) or polyclonal anti P-450b/e (lanes 6-10). Lanes and protein loads were as follows: P-450b and P-450e (lanes 1,5,6, and 10; 50 ng each), uninduced LE rat liver (lanes 2 and 7; 30 pg each), uninduced SD rat liver (lanes 3 and 8; 30 gg each), and PB-induced SD rat liver (lanes 4 and 9; 1 pg each).

tional bands of apparent M, < 50,000 do not represent proteins that contribute significantly to DMBA metabolism particularly via 12-methyl hydroxylation. In addition, since the bands detected by each antibody are clearly separated from P450’s b and e, their presence did not interfere with the quantitation of the latter proteins on Western blots. The total levels of proteins immunoreactive with anti P-450b were also determined using the RID technique (Table I), which has been used previously by a number of investigators (3, 11, 31). The levels of total protein calculated by this method are in good agreement with the sum of P450’s b, bz, and e obtained with Western blots. This suggests that the higher mobility band that is a major contributor to the Western blot is not detected by the RID method. Variation in expression of P-450’s b, b2, and e in uninduced liver microsomes of different rat strains. To examine whether the observed variability in expression of these closely related PB-inducible isozymes occurred in other rat strains, microsomes were prepared from six individual untreated rats from each of four additional strains, including two outbred, LE and WI, and two inbred, BN and FI. Blots of microsomes from a representative individual from each strain are shown in Fig. 5. Cytochrome P-450e was detected in all of the livers tested (Table II). Expression of this isozyme was relatively constant within each strain and ranged between 8 pmol/ mg (BN) and 18 pmol/mg (all other strains examined). In contrast, expression of P-

450’s b and bz exhibited a different pattern in each strain. With this selection of SD rats, three of the six mircosomal samples contained low levels of P-450bz, (3-5 pmol/ mg) and none of these individuals showed any detectable P-450b. None of the LE rats contained any P-450b2, whereas five of them contained levels of P-450b near the detection limit (2-3 pmol/mg). All of the WI rats contained consistent levels of P450bz (6 f 1 pmol/mg), while no P-450b was observed. All of the individuals from the two inbred strains contained exclusively P450b (BN = 4 f 1 pmol/mg and FI = 9 f 3 pmol/mg). In all rats of all strains examined, the band of lower molecular weight detected by anti P-450b/e in SD rats was observed at essentially constant levels (Fig. 4). Expression of P-450’s b, b,, and e in PBinduced rat livers. Western blots of liver microsomes from individual rats of each strain pretreated with PB were also examined (Table III). In outbred SD and LE rats, there was considerable variability

FIG. 5. Western blot of purified P-450’s b and e (lanes 1 and 6; 100 ng each) and 30 pg each of liver microsomal protein isolated from untreated rats from four different strains: lane 2, Wistar (WI); lane 3, Brown Norway (BN); lane 4, Long-Evans (LE); lane 5, Sprague-Dawley (SD).

CYTOCHROME

P-450e

IN

UNINDUCED

TABLE EXPRESSION

RAT

II

OF PB-INDUCIBLE CYTOCHROME P-450 ISOZYMES IN UNTREATED MICROSOMES FROM FIVE RAT STRAINS Cytochrome Western

Strain

P-450b

SD 1 6 Averaged

<2 <2 <2

413

LIVER

P-450

RAT LIVER

(pmol/mg)’

blot*

P-450bZ

P-450e

<2 4.5 1.5 f 2.5

Spectral” (total)

20.5 12.0 16.7 f 6.0

790 970 930 f 110

22.2 17.0 18.3 + 3.3

850 590 720 f 210

17.1 11.4 15.2 3~ 2.4

480 550 610 + 130

LE 3 6 Average

7.0 2.8 3.0 f 2.3

<2 <2 <2

WI 2 5 Average

12 <2 <2

4.7 5.8 5.6 f 0.7

BN 3 6 Average

4.2 3.7 3.9 + 0.4

<2 <2 <2

7.9 8.7 8.5 f 0.5

530 450 590 f 160

1 2 Average

12.3 8.3 9.0 + 3.0

12 <2 <2

23.2 12.5 17.2 2 5.5

1000 560 780 + 220

FI

’ Data are reported for two individual animals from each rat strain that exhibited high variability in the expression of cytochromes P-450b, ba, or e. *Individual isozymes were quantitated by Western blot analyses as described in detail under Materials and Methods. ‘Total P-450 was determined by the reduced-CO spectral method (41). d Average values represent the means f SD for six individual animals from each strain.

between individuals in the relative induction of P-450’s b and b2 by PB. While most of the individual rats examined expressed both of these isozymes, in these particular groups two of the six animals from each of these strains contained only P-450bz (data not shown). The selective expression of these two allozymes in PB-induced WI, BN, and FI rats was consistent with our previous observations in uninduced animals. Thus, only P-450bz was detected in WI rats and only P-450b was seen in BN and FI livers. P-450e was induced to about half the extent of the sum of P-450’s b and b2 in SD,

LE, WI, and FI rats but to approximately the same level in BN rats (Table III). Eflect of anti P-.&Ob/e on microscvmal metabolism of DMBA. The role of cyto-

chromes P-450b and e in the microsomal metabolism of DMBA was determined by investigating the inhibitory effect of anti P-450b/e and the MAbs 2-66-3,4-29-5, and 4-7-l. Previous work (29) showed that purified P-450e preferentially hydroxylates DMBA at the 12- versus the 7-methyl position, while P-450b forms more of the 7hydroxymethyl metabolite (‘7HOMMBA). The regioselectivity of metabolism by un-

414

WILSON,

CHRISTOU,

AND

TABLE EXPRESSION

OF PB-INDUCIBLE

P-450

ISOZYMES

III

IN PB-TREATED

Cytochrome Western P-450b

Strain SD LE WI BN FI

JEFCOATE

P-450

RATS FROM FIVE DIFFERENT (pmol/mg)a

blotb

P-450b2

648 f 128 240+ 40 <20 575 c 150 581 f 56

Spectral’ (total)

P-450e

326+ 19 364 f 63 712 + 241 <20 <20

473 + 249 + 346 + 518k 3252

170 24 202 117 81

1910 1700 1720 1910 2223

a Data are reported as the means + SD of duplicate analyses from six individual rats b Individual isozymes were quantitated by Western blot analysis as described in detail Methods. ‘Total P-450 was determined by the reduced-CO spectral method (41).

treated liver microsomes from five different rat strains was determined using selected microsomal preparations that were also used for quantitation of P-450's b and e (Table IV). In the selected groups of SD, WI, and BN rats, formation of the 12HOMMBA metabolite predominated over that of 7HOMMBA. In contrast, microsomes isolated from LE rats generated equal amounts of 12HOMMBA and

METABOLISM

OF DMBA

Diols

150 330 120 140 233

from each strain. under Materials and

IV

BY MICROSOMES

Percentage

f + + f f

7HOMMBA, whereas FI microsomes produced more ‘IHOMMBA. Differing patterns were also observed in the production of the three DMBA dihydrodiols. While microsomes from SD and LE rats produced more DMBA-3,4-dihydrodiol(3,4-diol) than DMBA-5,6-dihydrodiol(5,6-diol) or DMBA (8,9-dihydrodiol (8,9-diol), metabolism by BN and FI liver microsomes resulted in equal amounts of 8,9- and 3,4-diol and less

TABLE REGIOSELECTIVE

STRAINS

distribution

ISOLATED of DMBA

FROM LIVERS OF UNTREATED

RATS

metabolitesa

Hydroxymethyl

Strain

5,6-

8,9-

3,4-

7-OH

12-OH

Phenols

SD LE WI BN FI

13 14 5 10 9

11 11 14 17 12

17 18 6 15 13

20 25 23 17 35

35 26 41 30 27

4 6 11 11 3

‘Numbers are expressed as percentage distribution analysis as described under Materials and Methods. Each performed with microsomes isolated from two individual incubations was within 10% in all cases. b Total identifiable phenols which include 2-, 3-, 4-, 8-, ’ Rate of total DMBA metabolism was determined by

b

Total metabolism” (pmol/mg/min) 58+10 73klO 48f 9 40-t 6 53* 3

of total primary metabolites determined value represents the means of duplicate rats of each strain (n = 2). Variability and 9-OH DMBA. the method of Van Cantfort

et al. (38).

by HPLC incubations of duplicate

CYTOCHROME

P-450e

IN TABLE

EFFECT

OF ANTI P-450b/e

IJNINDIJCED

metabolite

V

(pmol/mg/min)”

Dials itrain

IgG

additiond

5,6-

8,9-

SD 1

Preimmune Anti P-450b/e Preimmune Anti P-450b/e Preimmune Anti P-450b/e Preimmune Anti P-450b/e Preimmune Anti P-450b/e Preimmune Anti P-450b/e Preimmune Anti P-450b/e Preimmune Anti P-450b/e Preimmune Anti P-450b/e Preimmune Anti P-450b/e

4.6” (1V 6.8 (14) 6.2 (10) 10.9 (14) 1.4 (5) 2.2 (14)

4.8 (+33)e’* 5.7 (+12)* 6.3 (+14)* 6.6

LE 3 LE 6 WI2 WI5 BN3 BN6 FI 1 FI 2

(f$ 4.7 (9) 4.8

(21) 4.1 (15)

(+W 5.8 (+26)* 4.3 (+25)* 5.1 (+12) * 6.4 (+22)* (:; 5.3 (+13)*

415

LIVER

ON THE REGIOSELECTIVE METABOLISM OF DMBA MICROSOMES FROM FIVE RAT STRAINS DMBA

SD 6

RAT

BY UNTREATED

or percentage

LIVER

inhibition*

Hydroxymethyl 3,4-

7-OH

12-OH

6.1

6.9 (13) 11.0 (4) 13.7

17.7 (79) 14.4

(+a 9.1 (+4) 8.8

(+16) 13.6

(2) 1.6 (0) 3.5 (+23) 3.3

(+Q 8.0

(2) 6.0 (7) 6.8 (7)

(2) 16.1 (4) 6.7 (11) 10.8 (11) 5.5 (13) 6.4 (14) 15.4 (18) 15.2 (+I)

(72) 13.4 (65) 14.3 (64) 13.3 (77) 17.2

(80) 9.3 (65) 11.1 (70) 12.6 (87) 12.2 (88)

Phenols” 2.2 (0) 2.1 (+lo) 4.0 (+15)* 2.8 (+7) 3.8 (+21)* 3.8 (+13) * 4.3 (+9) 3.6 (t14)* 2.0 (+lo) 1.6 (+19)*

Total

(ii) 58

(22) 62 (11) (YE) (E) ($ (;t) 47 (15) 58

(22) (ii)

a Data are reported as the means of duplicate incubations for each microsomal sample (n = 2). Variability was within 10% in all cases. *Percentage inhibition of formation of each metabolite by anti P-450b/e when compared to control incubations with equivalent amounts of preimmune IgG. ‘Total identifiable phenols which include 2-, 3-, 4-, 8-, and g-OH DMBA. d Each IgG was added at a final concentration of 2.5 mg/nmol total P-450. “Plus indicates increase over the rate observed in control incubations. * Significantly different (P < 0.05) from control values.

5,6-diol. In contrast, WI microsomes generated predominantly the 8,9-diol, with less than half as much 5,6- and 3,4-diol. Anti P-450b/e inhibited total DMBA metabolism to an extent that was dependent on the rat strain (10 to 32%;Table V). The greatest inhibitory effects were observed in microsomes prepared from SD, WI, and FI rats (22-32%), while metabolism by LE and BN rat livers was inhibited only half as much. However, with microsomes from all rat strains, anti P-450b/e inhibited predominantly 12HOMMBA formation (65~85%) (Table V). Formation of 5,6-diol and ‘7HOMMBA were inhibited to

a much smaller extent (2-18%), whereas that of formation of 8,9-diol and DMBA phenols was stimulated to a small extent. The stimulation of 8,9-diol formation in the presence of anti P-450b/e was statistically significant (P < 0.05) in four of five groups (Table V) and resulted in a 20-40% increase in the ratio of 8,9- to 5,6-diol in all groups. Saturating concentrations of the MAb 266-3 (2.5 mg IgG/nmol total P-450) and two other MAbs produced against P-450b (32) strongly inhibited metabolism of DMBA by untreated SD rat liver microsomes (Table VI). Similar to the polyclonal antibody (anti P-450b/e), the most potent inhibitory

416

WILSON,

CHRISTOU, TABLE

EFFECTS

JEFCOATE

VI

OF THREE MONOCLONAL ANTIBODIES PRODUCED AGAINST P-450b ON THE REGIOSELECTIVE METABOLISM OF DMBA BY SPRAGUE-DAWLEY RAT LIVER MICROSOMES DMBA

metabolite

Dials

NBS 2-66-3 4-29-5 4-7-l 2-66-3

AND

(pmol/mg/min)”

or percentage

inhibition*

Hydroxymethyl

MAb additionsd

5,6-

8,9-

3,4-

7-HO

12-HO

Phenols”

(ii) (49) (45) (25)

9.8 (50) (15) (+lo)e (35)

10.5

(22)

18.3 (75)

(10) (24)

(82) (88)

2.6 (23) (5) (+13)

+ anti P-450b/e

7.3” (35) b (23) (15) (35)

(20)

(78)

(20)

Total

(ii) (45) (46) (35)

‘Each value represents the average of duplicate determinations, using microsomes isolated from two individual SD rats (n = 2). Variability of duplicate incubations was within 10% in all cases. b Percentage inhibition of formation of each metabolite by each MAb when compared to control incubations with equivalent amounts of nonspecific ascites fluid (NBS). A control incubation with preimmune IgG plus NBS (not shown) was not significantly different from that of NBS alone. Variability in percentage inhibition was within 15% in all cases. ‘Total identifiable phenols which include 2-, 3-, 4-, 8-, and 9-OH DMBA. d Each MAb was added at a final concentration of 2.5 mg/nmol total P-450. ‘Plus indicates increase over the rate observed in control incubations.

effect of each of these MAbs was toward 12-methyl hydroxylation (‘75~85%). However, these MAbs also inhibited other metabolic pathways of DMBA to varying degrees. Addition of both MAb 2-66-3 and anti P-450b/e in incubation mixtures did not increase the extent of inhibition of DMBA 12-methyl hydroxylation, in comparison to that observed with separate additions of each antibody (Table VI). The similarity in the immunoreactivity of MAb 2-66-3 and anti P-450b/e with P-450’s, b, bB, and e but not the lower molecular weight proteins, on Western blots, provide further evidence that the latter protein does not have a significant role in DMBA metabolism. Comparison of the inhibitory effects of anti P-450b/e (Table V) with the levels of expression of P-450’s b and bz in individual rats (Table II) indicated that the extent of inhibition of either 12-methyl hydroxylation or total DMBA metabolism was independent of the variability in expression of these allozymes. More specifically, in the outbred strains SD and LE where P-450’s b and bz were either undetectable (SD No. 1 and No. 3) or expressed at levels sever-

alfold lower than those of P-450e (LE No. 6), the extent of inhibition of 12HOMMBA formation by anti P-450b/e (9-16 pmol/ mg/min) was comparable to that observed in the two inbred strains (BN and FI) and also in WI rats (7-13 pmol/mg/min) where P-450b or bz were consistently expressed at levels 30-60% of those of P-450e. The specific activities of P-450e toward 12methyl hydroxylation and total DMBA metabolism by untreated liver microsomes were calculated in two outbred rat strains (SD and LE) where the allozymes P-450b and bz were expressed at either undetectable or very low levels (Table VII). The calculated total activities (0.7-1.0 nmol/nmol P-450e/min-‘) are only two- to threefold lower compared to those previously determined for purified P-450e in a reconstituted system (29). However, even when P-450b was present at 30-60% of P-450e levels, the specific activity for total DMBA metabolism and lZmethy1 hydroxylation based on P-450e content remained in the same range (data not shown). The specific activities of P-450e toward 7HOMMBA and 5,6-diol formation were similarly calculated based upon antibody

CYTOCHROME

P-450e

IN

UNINDUCED

TABLE

RAT

417

LIVER

VII

CALCULATEDMETABOLISMOFDMBABYCYTOCHROMEP-~~O~INUNTREATEDSPRAGUE-DAWLEY AND LONG-EVANS RAT LIVER MICROSOMES Metabolite Strain SD 1 SD 6 LE3 LE 6 Pure

P-450e

*

(nmol/nmol

P-450e/min)”

5,6-Diol

‘IHOMMBA

12HOMMBA

Total

0.04 0.08 0.03 0.04

0.04 0.04 0.04 0.06

0.67 0.84 0.52 0.50

0.78 1.0 0.73 0.78

0.6

0.7

1.5

3.3

a Specific activities were calculated based on the percentage inhibition of formation of each metabolite by anti P-450b/e (Table V) and the amount of P-450e determined by Western blot analysis (Table II). The microsomal samples used for these calculations contained either undetectable or very low amounts of P-450b or P-450ba (~3 pmol/mg), with the exception of LE 3 which contained a higher level of P-450b (7 pmol/mg). *Rates of formation of individual metabolites by pure P-450e in a reconstituted system were determined as previously described (29).

inhibition data (Table VII). These activities were at least lo-fold lower compared to those previously observed in a reconstituted system (29). This may be due to increased formation of these metabolites by other isozymes when metabolism by P-450e is eliminated. This hypothesis is supported by the fact that anti P-450b/e caused increased formation of DMBA-8,9-diol and DMBA phenols (Table V). Regioselective metabolism of DMBA was also examined in microsomes prepared from five different strains of rats that were pretreated with PB (data not shown). With all rats examined, the predominant metabolites were the 3,4-diol(8-12%),5,6-diol (lo-18%), 7HOMMBA (12-19%), and 12HOMMBA (21-28%). The selective formation of 12HOMMBA versus ‘7HOMMBA, regardless of the relative levels of P-450b versus P-450bz, suggests a strong contribution of P-450e to metabolism despite its lower levels relative to P-450b (Table III). DISCUSSION

Previous studies have emphasized the coordinate expression of cytochromes P450b and P-450e (42, 43). In the present study, we have used Western blot analysis and the regioselectivity of DMBA metabolism to show that cytochrome P-450e is

consistently expressed in uninduced livers from five strains of rats. The average contents range from 1.7% (BN) up to 2.6% (WI) of total P-450. The fact that a stress as great as 24 h starvation did not affect this constitutive expression of P-450e (unpublished data) suggests that this constitutive expression is not particularly susceptible to the physiological status of the liver. Cytochrome P-450b is expressed constitutively at a much lower level and in a more complex manner. First, the polymorphism described previously for P-450b (20,44) was observed in Western blot analyses as an additional band (designated P-450bz) migrating between P-450b and P-450e. Only P-450b was expressed in the inbred FI and BN strains while only P-450bz was expressed in the outbred group of WI rats. The expression of P-450b or P-450bz in these strains was 30-60% of that of P-450e. With outbred SD and LE only occasional animals expressed P-450b or P-450bz and even then at levels near the detection limit (2 pmol/mg). Thus, there appears to be polymorphism in the constitutive expression of these low levels of P-450b. This variability is not maintained following induction by PB where P-450b is expressed at twice the level of P-450e in four of five rat strains. BN rats exhibited equal levels

418

WILSON,

CHRISTOU.

AND

JEFCOATE

of P-450b and P-450e apparently due to the selectivity (unpublished data). Although Prelatively high induction of P-450e. The 450g is present in uninduced microsomes selectivity between P-450b and P-450ba in (44,48), its DMBA monooxygenase activity the various strains examined was mainin a reconstituted system is not sensitive tained following induction, confirming that to anti P-450b/e (unpublished data). The the differences are due to variation in the observation that both the polyclonal anti P-450b locus. It therefore appears that FI P-450b/e and MAb 2-66-3 inhibit 12-methyl and BN rats are homozygous for the P-450b hydroxylation to a comparable extent even allele while this selection of WI rats are though the latter does not react with the homozygous for the P-450b2 allele. SD and same lower molecular weight protein furLE rats may be homozygous for either, ther emphasizes that this pathway is atwhile a proportion of the animals were tributable to cytochromes P-450e or Pheterozygous and expressed both proteins, 450b. Finally, quantitative comparison of confirming the codominant expression of the rates of anti P-450b/e-sensitive DMBA these alleles. Our finding of three immumetabolism with the P-450e content indinochemically reactive bands detectable on cates a fairly constant specific activity (0.7Western blots is consistent with two recent 1.0 nmol/nmol P-450e/min) for the two studies showing that there are at least outbred strains of rats where P-450’s b and three PB-inducible mRNAs and that each ba are expressed at either undetectable or message codes for a protein with unique very low levels. This compares with a maxmobility on SDS-PAGE (45, 46). imum turnover number for pure P-450e of Inhibition of metabolism of DMBA in 3.3 in a reconstituted system containing uninduced microsomes by a polyclonal an- stoichiometric amounts of P-450 reductase. tibody developed against P-450b indicated The finding that this DMBA metabolizing that P-450 cytochromes cross-reacting activity calculated in terms of P-450e conwith this antibody account for 15-30% of tent, remains fairly constant in other this activity. The highest inhibition was strains where P-450’s b and bz are exseen in SD rats that expressed detectable pressed at appreciable levels, raises queslevels of P-450e but not P-450b, thus tions with respect to the contribution of strongly implicating P-450e in this activity. the latter isozymes in the metabolism of However, there are multiple genes in the DMBA by uninduced rat liver microsomes. P-450b family that could contribute to this Another point of concern is that foractivity (19, 20, 44-47). Of major concern mation of two other metabolites of DMBA is a protein of lower apparent molecular that are also catalyzed by P-450e are only weight that was detected more strongly marginally sensitive to anti P-450b/e. A than P-450e on Western blots of microsource of this discrepancy is suggested by the fact that formation of 8,9-diol is insomes from uninduced rats. This protein, however, did not interact with MAb 2-66- creased significantly over control incuba3 which recognizes P-450’s b and e on tions in four of five strains (P < 0.05) by Western blots. The selectivity of the inhibaddition of anti P-450b/e. A probable exitory effects of the polyclonal anti P-450b/ planation for this increase in 8,9-diol formation and the insensitivity of formation e also supports the involvement of P-450e in the metabolism of DMBA by untreated of 5,6-diol and 7HOMMBA is that selective rat liver microsomes. Thus, in all rats ex- inhibition of cytochrome P-450b and Pamined, this antibody predominantly low- 450e causes metabolism to be redirected to ers DMBA metabolizing activity through other P-450 cytochromes that can actively inhibition of 12-methyl hydroxylation (65- metabolize DMBA to a variety of metab78%). This matches the regioselectivity of olites. This redirection could arise from purified P-450e which preferentially hy- competition among P-450 cytochromes for droxylates DMBA at the 12- versus the ‘7- a common limiting component, possibly P450 reductase. Recent studies on Warfarin methyl position (29). Of 10 characterized (49) suggest that P-450e may rat liver cytochromes, P-450g is the only metabolism other isozyme that shows this positional be particularly effective in such competi-

CYTOCHROME

P-450e

IN

tion for P-450 reductase. It is, therefore, possible that the observed inhibition of DMBA metabolism by anti P-450b/e underestimates the contribution of P-450e. This works establishes that P-450e is expressed constitutively in five strains of rats in preference to the closely related P-450b. By contrast we confirm that in four of these tive strains, P-450b is expressed preferentially after PB induction. Quantitation of the respective mRNA for P-450e and P450b suggest that these differences arise from selective transcription of the respective genes (50). This very small proportion of total P-450 nevertheless contributes up LO 30% of metabolism of DMBA by uninduced rat liver microsomes. ACKNOWLEDGMENTS We thank Karen Wipperfurth and Mary Kay Acker for their assistance in the preparation of this manuscript. This work was supported by the National Institutes of Health Grant CA 16265. N.M.W. and M.C. were supported by National Institutes of Environmental Health Sciences Training Grant ES 07015. REFERENCES 1. CONNEY, A. H. (1982) Cancer Res. 42,4875-4917. 2. Lu, A. Y. H., AND WEST, S. B. (1980) PharmucoL Rev. 31,277-295. D. E., THOMAS, P. E., REIK, L. M., AND LEVIN, (1982) Xenobiotica 12,727-744.

3. RYAN, W.

4. RYAN, D. E., IIDA, S., WOOD, A. P. E., LIEBER, C. S., AND LEVIN,

BioL &em. 259,1239-1250. D. E., RAMANATHAN,

5. RYAN,

L., IIDA,

W., W.

THOMAS, (1984) J.

S., THOMAS,

P. E., HANIU, M., SHIVELY, J. E., LIEBER, C. S., AND LEVIN, W. (1985) J. BioL Chem. 260,6385-

6393. 6. GUENGERICH,

F.

P.,

DANNAN,

S. T., MARTIN,

M. V., AND

Biochemistry

21,6019-6030.

7. WRIGHTON, S. A., MAUREL, WATKINS, P. B., YOUNG, P. S. (1985) Biochemistry 8. TAMBURINI, MAKOWSKI,

G.

KAMINSKY,

A.,

WRIGHT, L. S. (1982)

P., SCHEUTZ, E. G., B., AND GUZELIAN,

24,2171-2178.

P. P., MASSON, H. R. J., MORRIS,

A., B.,

BAINS, S. K., AND GIBSON,

G. G. (1984) Eur. J. Biochem. 139,235-246. 9. CHENG, K., AND SCHENKMAN, J. B. (1982) J. BioL

Chem 10.

JANNSON,

257,2378-2385. I., MOLE,

J. BioL Chem.

J., AND

SCHENKMAN,

J. B. (1985)

260,7084-7093. 11. KAMATAKI, T., MAEDA, K., YAMAZOE, Y., NAGAI, T., AND KATO, R. (1983) Arch Biochem Biophys. 225,758-770.

UNINDUCED

RAT

419

LIVER

12. WAXMAN, D. J., AND WALSH, C. (1982) J. Biol. Chem. 257.10446-10457. 13. WAXMAN, D. J., AND WALSH, C. (1983) Biochemistry 22,4846-4855. 14. WAXMAN, D. J., Ko, A., AND WALSH, C. (1983) J. BioL Chem. 258,11937-11947. 15. YUAN, P. M., RYAN, D. E., LEVIN, W., AND SHIVELY, J. E. (1983) Proc. NatL Acad. Sci. USA 80,11691173. 16. SOGAWA, K., GOTOH, O., KAWAJIRI, K., AND FUJIIKURIYAMA, Y. (1984) Proc. Natl. Acad Sci. USA 81,5066-5070. 17. SOGAWA, K., GOTOH, O., KAWAJIRI, K., HARADA, T., AND FUJII-KURIYAMA, Y. (1985) J. Biol Ckm. 260,5026-5032. 18. KAWAJIRI, K., GOTOH, O., TAGASHIRA, Y., SOGAWA, K., AND FUJII-KURIYAMA, Y. (1984) J. BioL Chem. 259,10145-10149. 19. SUWA, Y., MIZUKAMI, Y., SOGAWA, K., AND FUJIIKURIYAMA, Y. (1985) J. BioL Chem. 260, 79807984. 20. RAMPERSAUD, A., AND WALZ, F. G. (1983) Proc. Natl. Acd Sci. USA 80, 6542-6546. 21. ATCHISON, M., AND ADESNIK, M. (1986) Proc. Natl. Acad. Sci. USA 83,2300-2304. 22. GONZALEZ, F. J., NEBERT, D. W., HARDWICK, J. P., AND KASPER, C. B. (1985) J. Biol. Chem. 260, 7435-7441. 23. BANDIERA, S., RYAN, D. E., LEVIN, W., AND THOMAS, P. E. (1985) Arch. Biochem. Biophys. 240,478482. 24. GONZALEZ, F. G., KIMURA, S., SONG, B.-J., PASTEWKA, J., GELBOIN, H. J. P. (1986) J. BioL Chem.

25. VLASUK,

G. P., GRAYHEB,

V.,

AND

HARDWICK,

261, 10667-10672. J., RYAN,

L., THOMAS, P. E., LEVIN, W., AND (1982) Biochemistry 21,789-798.

D. E.,

REIK,

WALZ,

F. G.

26. SAKAI, H., HINO, Y., AND MINAKAMI, S. (1983) Biochem. J. 215,83-89. 27. REIK, L. M., LEVIN, W., RYAN D. E., MAINES, S. L., AND THOMAS, P. E. (1985) Arch. Biochem. Bie phys. 242,365-382. 28. RYAN, D. E., THOMAS, P. E., AND LEVIN, W. (1982) Arch. B&hem. Biophys. 216,272-288. 29. WILSON, N. M., CHRISTOU, M., TURNER, C. R., WRIGHTON,

S. A.,

Carcinogenesis 30. THOMAS,

P. E.,Lu, J., AND LEVIN,

AND

JEFCOATE,

C. R. (1984)

5,1475-1483. A. Y. H., RYAN, D. E., KAWALEK, W. (1976) J. BioL Chem 251,1385-

1391. 31. THOMAS, LEVIN,

P. E., KORZENIOWSKI, D., RYAN, D., AND W. (1979) Arch Biochem. Biophys. 192,

524-532. 32. PARK, S. S., FUJINO, F.

P.,

AND

T., MILLER,

GELBOIN,

H.

(1984)

B&hem.

PharmacoL 33. LAEMMLI, 685.

33.2071-2081. U. K. (1970) Nature

H., GUENGERICH, V.

(London)

227,680-

420

WILSON,

CHRISTOU,

AND

34. TOWBIN, H., STAEHELIN, T., AND GORDON, J. (1979) Proc. NatL Acad Sci. USA 76,4350-4354. 43.

35. GUENGERICH, F. P., WANG, P., AND DAVIDSON, N. K. (1982) Biochemistry 21,1698-1706. 36. TSANG, V. C. W., PERALTA, J. M., AND SIMONS, A. R. (1983) in Methods in Enzymology (Langone, J. J., and Van Vunakis, H. H., Eds.), Vol. 92, pp. 377-391, Academic Press, Orlando. 37. DOMIN, B. A., SERABJIT-SINGH, C. J., AND PHILPOT, R. M. (1984) Anal Biochem 136,390-396. 38. VAN CANTFORT, J., DEGRAEVE, J., AND GIELEN, J. E. (1977) Biochem. Biophys. Res. Commun. 79,505-512. 39. CHRISTOU, M., WILSON, N. M., AND JEFCOATE, C. R. (1984) Carcinogenesis 5,1239-1247. 40. LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L., AND RANDALL, R. J. (1951) J. BioL Chem. 193, 265-275.

49.

41. OMURA, T., AND SATO, R. (1964) 2370-2378.

50.

42. VLASUK,

J. BioL Chem.

G. P., RYAN, D. E., THOMAS,

44.

45. 46.

47. 48.

239,

P. E., LEVIN,

JEFCOATE

W., AND WALZ, F. G. (1982) Biochemistry, 21, 6288-6292. PARKINSON, A., THOMAS, P. E., RYAN, D. E., REIK, L. M., SAFE, S. H., ROBERTSON, L. W., AND LEVIN, W. (1983) Arch. Bicchem Biophys. 225,263-215. WALZ, F. G., VLASUK, G. P., OMIECINSKI, C. J., BRESNICK, E., THOMAS, P. E., RYAN, D. E., AND LEVIN, W. (1982) J. BioL Chem 257,4023-4026. KUMAR, A., RAPHAEL, C., AND ADESNIK, M. (1983) J. BioL Chem. 258,11280-11284. PICKETT, C. B., JETER, J. L., WANG, R., AND Lu, A. Y. H. (1983) Arch. B&hem. Biophys. 225, 854-860. ATCHISON, M., AND ADESNIK, M. (1986) Proc. NatL Acad Sci. USA 83,2300-2304. BANDIERA, S., RYAN, D. E., LEVIN, W., AND THOMAS, P. E. (1984) Arch. Biochem. Biophys. 240,478482. KAMINSKY, L. S., AND GUENGERICH, F. P. (1985) Eur. J. Biochem. 149,479-489. OMIECINSKI, C. J. (1986) Nucleic Acid Res. 14,15251539.