ARCHIVES

OF BIOCHEMISTRY

Vol. 291, No. 1, November

AND

BIOPHYSICS

15, pp. 24-30,199l

Isoelectric Focusing of Cytochrome P450: Isolation of Six Phenobarbital-Inducible Rat Liver Microsomal lsoenzymes Maja

Oertle,’

Laboratorium *Department

Received

Djordje

fuer Biochemie of Biochemistry

February

Filipovic,*

Christoph

Richter,

Kaspar.

H. Winterhalter,

and Ernest0

I, ETH Zentrum, Universitaetstrasse 16, CH-8092 Zurich, Switzerland; and Beckman Institute, University of Illinois, Urbana, Illinois 61801

15, 1991, and in revised

form

July

E. Di Iorio

and

25, 1991

previously considered to be homogeneous, into three peaks using anion exchange HPLC. Rampersaud and Walz (5) demonstrated by two-dimensional gel electrophoresis the existence of at least six phenobarbital-inducible homologous P45Os in rat liver microsomes. Using the same technique Rampersaud (6) showed microheterogeneity of P450h (7) and P450-2c (8). Similar results were reported by Pickett et al. (9), who detected the presence of three polypeptides of similar size, but different charges, in their preparation of the major phenobarbital-inducible rat liver P450 fraction. At the nucleic acid level, microheterogeneity of rabbit liver P450IIB4 (nomenclature according to Nebert et al. (10)) was demonstrated by sequencing cDNA (11). More recently Oesch et al. (12) prepared antipeptide antibodies against the regions where the rat liver microsomal P450IIBl and P450IIB2 differ in their primary structure. With these antibodies they were able to detect a new member of the P450IIBl family as well as Cytochromes P450 (P450) play an important role in an additional P450IIB2-related enzyme. In view of a possible physiological significance of this the metabolism of endogenous and xenobiotic compounds. Multiple forms of P450 occur in liver microsomes, some multiplicity of P45Os, their characterization is of great of which can be induced specifically by certain drugs. This importance. This requires the purification of the various multiplicity was first recognized in the course of the pu- P45Osto homogeneity and in a native form. In this report rification of distinct P450 isozymes from liver microsomes we describe an IEF’ procedure which can be used for both of untreated and treated animals, and, more recently, by analytical and preparative purposes, and yields, contrary isolation and structural analysis of cDNA and genomic two analogous approaches previously reported (l), enzyclones encoding structurally distinct P45Os (for reviews matically active cytochrome P450. Using the high resosee Refs. (1, 2)). lution power of this technique we were able to resolve the At the protein level, high resolution purification methP450IIB subfamily into six fractions, which were partially sequenced. ods such as HPLC and two-dimensional gel electrophoresis have been applied to separate P45Os. Bansal et al. (3, 4) were able to resolve a rat liver P450 preparation, * Abbreviations used: IEF, isoelectric focusing; Chaps, 3-[(3-cholA procedure for the isolation of native proteins from membranes by isoelectric focusing is described. It was used to resolve into six components the major fraction of cytochrome P460, obtained from liver microsomes of phenobarbital-treated rats, after chromatography on DE52 cellulose. When eluted from the gel, these proteins are in a native form as shown by (a) the light absorption spectra of the Soret region of their reduced carbonyl derivatives, all characterized by maxima around 450 nm, and (b) their enzymatic activities toward three different substrates. Characterization by a monoclonal antibody and partial sequence analysis of tryptic peptides reveal that three of the IEF-purified proteins have PISOIIBl character, whereas the other three are related to P450IIB2. o 1991 Academic press, IUC.

i To whom correspondence should be addressed at present address: Steroid Laboratory, Department of Medicine, University Hospital, CH8091 Zurich, Switzerland. Fax: (01) 255 44 47.

amidopropyl)dimethylamino]propanesulfona~, ELISA, enzyme-linked immunoabsorbant assay; PBS, phosphate-buffered saline: 2.2 mM KCl, 1.1 mM KH2P0,, 138 mM NaCl, 8.1 mM Na,HPO,; SDS-PAGE, sodium dodecyl Sulfate-polyacrylamide gel electrophoresis; Tris, tris(hydroxymethyl)aminomethane

24 All

Copyright 0 1991 rights of reproduction

0003-9&X61/91 $3.00 by Academic Press, Inc. in any form reserved.

ISOELECTRIC

MATERIALS

AND

FOCUSING

METHODS

Protein isolation. Phenobarbital-inducible forms of P450 (P45OIIBl and P450IIB2) were isolated from liver microsomes of phenobarbitaltreated male Sprague-Dawley rate by anion exchange chromatography on a Whatman DE-52 column as described elsewhere (13). The peaks corresponding to P450IIBl and P450IIB2 were pooled, dialyzed against 15 mM potassium phosphate buffer (pH 7.4), containing 20% glycerol, and subsequently applied to a hydroxylapatite column equilibrated with 15 mM potassium phosphate buffer (pH 7.4), containing 20% glycerol and 0.2% Emulgen 911. The resin was washed with 3 column vol of 45 mM potassium phosphate buffer (pH 7.4), containing 20% glycerol and 0.2% Emulgen 911. Subsequently, detergents were removed by extensive washing with 15 mM potassium phosphate buffer (pH 7.4), containing 20% glycerol, until no absorbance at 280 nm could be detected in the effluent. P450 was eluted with 150 mM potassium phosphate buffer (pH 7.4), containing 20% glycerol. NADPH cytochrome P450 reductase was isolated from phenobarbitalinduced rat liver microsomes by affinity chromatography on agarosehexane-adenosine-2’5’diphosphate (PL Biochemicals) as described by Shephard et al. (14). Total, smooth and rough microsomes were prepared according to the method of De Pierre and Dallner (15). Isoelectric focusing. The standard procedure described by Pharmacia (16) applies to water soluble biopolymers. In order to be able to use IEF with membrane-derived proteins we introduced the following modifications: An IEF-stock solution was prepared by mixing 360 mg agarose IEF (Pharmacia, Uppsala, Sweden), 3.6 g sorbitol (Fluka, Buchs, Switzerland), 6 ml of 87% glycerol (Merck, Darmstadt, Germany), and 26.4 ml Hz0 and heating to approximately 75°C. For an analytical gel (90 X 120 X 0.5 mm), 10.5 ml of the stock solution was added to a mixture of 110 mg Chaps (Sigma, St. Louis MO), 40-220 mg Triton X-100 (Serva, Braunschweig, Germany) or Nonidet P-40 (Serva, Braunschweig, Germany) or Emulgen 911 (Kao Atlas, Tokyo, Japan) and 800 /.d Ampholines pH 3.5-10 (LKB, Bromma, Sweden). The mixture was stirred to homogeneity with a glass rod and care was taken to avoid the formation of air bubbles. The gel was cast on a preheated glass plate covered with a gel bond film (LKB Bromma, Sweden) and allowed to harden at 4’C for at least 1 h, preferably overnight. After it was removed from the casting unit the gel was carefully wiped with a Whatman filter paper and then transferred to the electrophoresis chamber (LKB Multiphor II, LKB, Bromma, Sweden) thermostated at 12°C. The electrode strips were soaked in 50 mM HsSO, for the anode and 1 M NaOH for the cathode and placed onto the gel after having removed the excess solution by pressing them between three layers of filter paper. The samples were applied using an LKB Mylar application mask or filter papers. Focusing was obtained with about 3000 V X h using an LKB Macrodrive power supply set to U = 2000 V, Z = 20 mA, and P = 5 W. Prefocusing was not required. After 1 h the application mask was removed and when necessary, the electrode strips were wiped. Erythrocyte lysate was used as a colored marker to follow the focusing process. Immediately after focusing the gels were stained with benzidine for heme and/or with silver or Coomassie blue for protein. For preparative purposes thicker gels (up to 5 mm) of the same composition as the analytical ones were prepared, except for the use of a 3:l mixture of ampholines pH 6-8 and pH 3.5-10 in order to improve the resolution. On a 105 X 240 X 5-mm gel up to 2 mg protein (in a maximum volume of 600 ~1) could be applied with a Whatman filter paper near the anodic side. Also in this case prefocusing was not required and about 4000 V X h were needed to get well-focused bands. Heme proteins were visible without any staining procedure; however, when necessary, a narrow strip was cut on both sides of the gel and stained. The areas with the bands were cut out and proteins eluted from the gel by freezing and thawing in 50 mM potassium phosphate buffer (pH 7.4) containing 20% glycerol. Detergents were removed from the individual bands on small hydroxylapatite columns (1 X 3 cm) as described

OF

P450

above. Complete detergent at 280 nm and by HPLC.

25 removal

was checked

both by the absorbance

Benzidine staining. Heme proteins were visualized according to a modified method of Crosby and Forth (17). About 50 mg of benzidine (Merck, Darmstadt, Germany) or tetramethyl benzidine (Fluka, Buchs, Switzerland) was mixed with 9 ml of ethanol. Nine milliliters of 50% acidic acid was added and the suspension mixed and poured onto the gel. After 5 min the color was developed by the addition of 5% HsOs. Silver staining. Most of the methods reported in the literature for staining of gels containing ampholytes failed when applied tc the agarose IEF gels described above. The only procedure giving satisfactory results was the one described by Black (18). The essential feature of thii method is the inclusion of glycerol in the solutions used in the handling of the gel before silver staining. The resulting stained gels could be dried and kept in the dark for a long time. Coomussie blue staining. The procedure given by Pharmacia (16) was used with slight modifications. After completion of the focusing experiment the gel was fixed in 10% TCA for 30 min, and subsequently the ampholines were washed out by a 2 X 30-min immersion in destain solution (methanohacetic acidwater = 3:1:6). The gel was then put on a glass plate, covered with three sheets of Whatman filter paper and a second glass plate, and pressed, applying a weight of about 1 kg. After 15 min the gel was dried with a hair dryer until it became clear, stained for 15 min in destain solution containing 0.2% Coomassie brilliant blue R-250 (Serva, Braunschweig, Germany), and extensively washed in destain solution. Solid phuse ELZSA. The murine monoclonal antibody ae4 against P450IIB was prepared and characterized according to Schweizer et al. (19). Microtiter plates were coated with 50 ~1 of a 1.2 pg/ml solution of the antigen in PBS. After saturation of nonspecific adsorption sites with PBS containing 2% bovine serum albumin, plates were incubated with serially diluted monoclonal antibody, 1:5 dilutions were made starting from 1 mg/ml. The wells were treated with a 1:500 dilution of phosphatase-conjugated goat antibodies specific for mouse IgG (Sigma). The antigen-antibody complexes were visualized with 4-nitrophenyl phosphate (Fluka, Buchs, Switzerland), and the absorbance at 405 nm measured on an EIA reader (Bio-Rad, Richmond CA). Between each step, unbound protein was removed with ice cold PBS. Determination of the enzymatic o&i&y. P450 activity was measured in a system reconstituted as follows: Equimolar amounts (between 0.2 and 1 nmol) of the P45Os and NADPH cytochrome P450 reductase were incubated for 10 min at 15’C with a clear, freshly sonicated solution of 1-a-dilauroyl-3-phosphatidylcholine (300 pg/ml) to give a protein to lipid ratio of 1:lO in 50 mM potassium phosphate buffer (pH 7.4) containing 20% glycerol. Activity toward ‘I-ethoxycoumarin was measured with a fluorimetric assay (20), p-nitroanisole 0-demethylation by following thep-nitrophenol formation as described elsewhere (21) and Ndemethylation of benzphetamin from the generation of formaldehyde using the method of Nash (22). Peptide mapping. For digestion with trypsin typically 200 H of P450 was used. The peptides were separated on a nucleosile 300-5-Cs column (Macherey-Nagel, Dueren, Germany) with a Hewlett-Packard 1090 liquid chromatograph. The experimental conditions were the same as described by Yuan et al. (23). Sequence analysis. All chemicals used for the sequencing work were purchased from Applied Biosystems (Foster City, CA). Sequencing was done on an Applied Biosystems 470A sequencer and 120 phenylthiohydantoin derivative analyzer. Other naethds. Protein concentrations were determined according to Lowry (24) using bovine serum albumin as a standard. P450 concentrations were calculated from the reduced carbonyl difference spectra as reported by Omura and Sato (25). SDS-PAGE was performed according to Laemmli (26) except that the concentrations of Tris and glycine in the running buffer were doubled.

26

OERTLE

ET

AL. a

RESULTS Protein purification. After hydroxylapatite chromatography the heme content of our P450 preparations was in the average 14.5 nmol per milligram protein. This material was subjected to isoelectric focusing. It resulted in at least six distinct bands, all detectable by benzidine, Coomassie blue, and/or silver staining. They are designated b1 to bV1 and have isoelectric points between 7.4 and 7.9 (Fig. 1). Validation of the IEF approach. A number of controls were made to consolidate the validity of our purification method. The results can be summarized as follows: (a) The overall P450 IEF pattern was not influenced by the use of acrylamide or Sephadex instead of agarose as gel matrix (data not shown). (b) In addition to Chaps, other detergents such as Triton X-100, Emulgen 911, and Nonidet P-40 could be added without influencing the result of the separation procedure (data not shown). (c) The isoelectric pattern of the individual IEF-purified proteins did not change with time (Fig. l), as clearly shown by the refocusing experiment. It is worth mentioning that b1 and bI1 cannot be completely separated from each other due to their very close isoelectric points, and that bIV is not homogeneous, but consists of two bands. (d) The same six bands were observed when partially purified P450 was applied to the gel or when solubilized microsomes were subjected to IEF analysis followed by staining for heme by benzidine. No difference was detectable between microsomal preparations of a single rat or from a pool of 10 livers (data not shown). Furthermore, separation of the microsomes into the rough and smooth fractions showed an uneven distribution of the six bands, i.e., bI1, bII1, and bV1 were predominantly found in the smooth fraction, whereas bIV was mainly detectable in the rough fraction. Characterization and identification of the IEF-purified bands. The material eluted from the IEF gel contained, after detergent removal, between 9 and 14 nmol heme per milligram protein. On SDS-PAGE the individual IEFpurified subfractions exhibited slightly different molecular weights, ranging between 51,800 and 53,200 (Fig. 2). The reduced-CO Soret difference spectra of the purified bands are shown in Fig. 3. It is important to notice that no conversion to P420 could be detected. As summarized in Table I, small, but significant differences in the absorption maxima of the ferrous, carbonyl, and ferric derivatives between the various subfractions were found, the largest being in the reduced unliganded form. Catalytic activities of the six proteins toward p-nitroanisole, 7-ethoxycoumarin, and benzphetamine were determined in a reconstituted system. As shown in Fig. 4, they exhibit characteristic patterns for each isolated band. Thus, p-nitroanisole is best metabolized by bII1, benzphetamine by bV, and ethoxycoumarin by bV1. It should be noted that the low activities reported in Fig. 4

b

c

d

e

f

A -

a

9 -

bw

-

bv

C

b

B

e

b PH 8

cf -

c bvl bv blv bill iI’

-.““cir;s-,

PH 7

FIG. 1. Isoelectric focusing of P450IIB. (A) Refocusing of P450IIB subfractions. Immediately after focusing, the bands were cut out and the protein was eluted by freezing and thawing. Lanes a through f show refocusing of the individual IEF-purified P450IIB subfractions b1 through bV1, lane g shows the pattern obtained when P450IIB fraction obtained after hydroxylappatite chromatography and removal of detergents is applied on the gel. The bands were evidenced by Coomassie blue staining. (B) Comparative IEF analysis of purified P450IIBl and P450IIB2 from the laboratory of Dr. D. J. Waxman and our preparations prior to isoelectric focusing purification. Lane a, our P450IIB preparation; lane b, P450IIB2; and lane c, P450IIBl. Staining was done with Coomassie blue. (C) IEF analysis of microsomes. Lane a, our P450IIB fraction as a reference; lane b, solubilized smooth microsomes of phenobarbitaltreated rata; and c, solubilized rough microsomes of phenobarbital-treated rats. Staining was done with benzidine.

are related to the presence of 20% glycerol in our buffers. In a control experiment carried out on microsomes the enzymatic activities in the absence of glycerol were found to be at least three times higher than in its presence, most likely due to a viscosity effect. The results of an ELISA with monoclonal antibody ae4 are shown in Fig. 5. The antibody did not cross-react with P450 isozymes other than rat P450IIB (data not shown). In particular, no reaction was observed upon incubation with rabbit P450IIB4 which exhibits 77% sequence homology with P450IIBl. The ae4 monoclonal antibody recognized subfractions b1, bI1, and bIV, but did not react

ISOELECTRIC

e

f

analysis. Lanes a and e are molecular weight from the eluate of the DE-52 cellulose column.

standards,

a

FIG. 2. SDS-PAGE h is P450IIB pooled

FOCUSING

b

c

d

with subfractions bII1, bV, and bV1 (see Fig. 5). In a Western blot none of these proteins could be evidenced with this monoclonal antibody (data not shown). This result is consistent with a conformational epitope (19). We have compared the products of our purification procedure with P450IIBl and P45011B2.3 In Fig. 1B we show the results of this comparison, which identifies one of our bIV components as P450IIB2 and bV1 as P450IIBl. N-terminal sequence analysis has proven to be useful for characterizing the various P450 isozymes (27). The sequence of the first 10 residues was found to be the same for all six bands and corresponds to that previously reported for P450IIBl and P450IIB2. This renders unlikely the possibility that the multiplicity of bands observed in IEF is due to contamination with any P450 isozymes other than P450IIB. Peptide mapping after tryptic digestion of all bands is consistent with the results reported by Yuan et al. (23), i.e., about 50 peptide peaks were observed after HPLC separation. The tryptic maps of all subfractions were highly similar, indicating extensive homology. Therefore, we first focused our sequencing efforts on the peptides that correspond to the region carrying the 13 differences between P450IIBl and P450IIB2 (23). The results of this analysis (Table II) unequivocally demonstrate that bI1, bV, and bV1 belong to the P450IIBl family, whereas b1, bI1, and bIV have P450IIB2 structural features. Suva et al. (28) reported that position 392 is occupied by an Arg in P450IIBl and a Leu in P4501IB2, whereas Yuan et al. (23) and Mizukami et al. (29) found Leu in both isozymes. 3 P450IIBl

and P450IIB2

were generous

gifts

of Dr. D. J. Waxman.

OF

27

P450

9

h

lanes b through

i

g show purified

bands

b1 through

bV1, and lane

We found no evidence for Arg at position 392 in P450IIBltype subfractions. Ongoing sequencing work focuses now on differences between the individual bands. Up to now we have found two sequencedifferences, both in the sameprotein region. In b1 Tyr 268 is replaced by a Ser, and in bII1 Thr 255 by a Val and Pro 258 by a Gly. DISCUSSION

Validation of the IEF approach. A serious warning on possible artifacts generating from the application of isoelectric focusing to membrane-derived P45Os was given as early as 1979 by Guengerich (30). We have taken artifacts into very serious consideration and therefore performed our IEF fractionation under a variety of experimental conditions. The results of these controls can be summarized as follows: The electrophoretic pattern obtained when microsomal P450 was applied on the IEF gel was independent of the gel composition, both in terms of gel matrix and detergents used, and was also observed in solubilized microsomes, even though distinct quantitative differences were found between rough and smooth microsomes (Fig. 1). Furthermore, refocusing of each of the individual bands, at different times after purification, did not show any trace of reappearance of the other components (Fig. 1). These findings not only excluded that artifacts related to the use of IEF were responsible for the multiplicity of bands observed, but also eliminated the possibility that other phenomena, such as lipids binding to the enzyme or partial oxidation, were responsible for the observed IEF pattern. The possibility that electrophoretic heterogeneity was related to aggregation equi-

28

OERTLE

ET

AL. TABLE

I

Soret Maxima of the Individual IEF-Purified P450IIB Subfractions IEF band

Ferric

Ferrous

I II III IV V VI

416.2 416.0 416.0 416.3 417.0 416.5

418.3 418.1 417.0 418.0 419.2 421.0

Note. Spectra 7.4), containing

bVI

bV 4 g 0 _-___-

.-_--.---

--_--

__----~. /’

/ J -1 ;‘g -3

A nm

:Y

A nm

:

FIG. 3. Reduced CO difference spectra of the IEF-purified P450IIB subfractions. Spectra were recorded in 50 mM potassium phosphate buffer (pH 7.4) containing 20% glycerol and plotted with relative absorbance units.

libria or other types of modification of the enzyme, occurring either during the first chromatographic steps or during IEF, was also ruled out by the above controls. What could be ascribed to aggregation is the incapability to resolve the various P450 subfractions on the DE-52 column. Characterization and isolation. The molecular masses of the individual IEF purified bands were all around 52 kDa (Fig. 2). The reduced carbonyl difference spectra showed maxima in the 450-nm region and displayed no detectable amount of P420 (Fig. 3). The proteins exhibited catalytic activities toward three different substrates in a reconstituted system. Therefore, during the focusing of the structure of the protein around the heme pocket, the substrate access region and the area responsible for the interaction with the NADPH cytochrome P450 ,reductase remained preserved. These results clearly show that contrary to what previously was reported in the literature (1, 30), membrane-derived P45Os can be purified by IEF without denaturation.

were taken in 50 mM potassium 20% glycerol.

Ferrous-CO 449.8 450.3 444.9 449.7 450.0 450.0 phosphate

buffer

(pH

The N-terminal sequences of the six P45Os as well as the sequence data of the tryptic fragments (Table II) excluded that the multiplicity observed in IEF was due to contamination with isozymes other than P450IIB and showed that the resolving power of our procedure is such that in only one step we could resolve the P45OIIB family into a number of components never reached before. Upon very careful and restrictive selection of the DE-52 fractions in the region corresponding to the P450IIBl peak we were able to obtain pure bV1. In addition, with reference standards of P450IIBl and P450IIB2 we could identify bV1 as P450IIBl and bIV as P450IIB2 (Fig. 2). These results were corroborated by the sequencing of selected tryptic fragments from the region where the 13 differences between P450IIBl and P450IIB2 are located. As shown in Table II, in addition to bVI, bII1 and bV also display the structural features of P450IIBl. Furthermore,

150

100

50

0 I mB

II enzphetamine

III m

7-Ethoxycoumarin

IV

V m

VI p-Nitroonisole

FIG. 4. Catalytic activities of the individual IEF-purified P450IIB subfractions. The results are given as percentage of the activity measured for the P450IIB after chromatography on hydroxylapatite and removal of detergents. 100% activity corresponds to 6.1+ 0.35,0.32 f 0.013, and 0.26 + 0.009 nmol product/nmol P450/min, respectively, for bensphetamine, 7-ethoxycoumarin, and p-nitroanisole.

ISOELECTRIC

FOCUSING

OF

29

P450

by at least three members. That microheterogeneity is responsible for the multiplicity of bands detected by our IEF approach is in keeping with the results of Friedberg et al. (31) who detected in rat liver the expression of a P450IIB gene, which, according to their RNAse A protection investigation, does not correspond to either P450IIBl or P450IIB2. Furthermore, Komori et al. (11) described microheterogeneity of the major phenobarbitalinducible form of rabbit microsomal P450 on the cDNA level. In view of all these investigations, biophysical and biochemical properties of membrane-derived P45Os should be reexamined with completely homogeneous preparations. Moreover, microheterogeneity of P450 might account, at least to a certain extent, for the broad substrate specificity of these proteins. FIG. 6. Reactivity of monoclonal antibody ae4 with the individual IEF bands. ELISA plates were coated with the different P450IIB subfractions and incubated with serially diluted ae4 monoclonal antibody, 1:5 dilutions starting from 1 mg/ml (see Materials and Methods).

all residues from this variable region sequenced up to now classify b1, bI1, and bIV as P450IIB2. The same picture also emerges from our immunological investigation. The monoclonal antibody ae4 recognizes only b1, bI1, and bIV, thus suggesting that these three bands result from a microheterogeneity of P450IIB2. Each of the individual subfractions displays unique spectroscopic properties, electrophoretic mobilities, and catalytic activities, even though the sequence data underline their kinship to the same protein family. Post-translational modifications or different gene products with very high sequence homology are both potential explanations for these findings. The first possibility cannot be completely ruled out, but on the basis of the preliminary sequence data for b1 and bII1 described at the end of the Results section we favor the second. Our sequence data point to the conclusion that P450IIBl and P450IIB2 are not members of the same subfamily, but each belongs to two distinct groups formed TABLE

II

P450

303

IIBl IIB2

S G

bI bII bIII bIV bV bVI

Note.

Data

321

322

A T

E V

A

E

A A

E E

were

337 L P

L L

obtained

339 T S

T T

344 S T

acid position 360

363

367

S A

V A

V L

A A S A s

A A V A v

L

by sequencing

tryptic

V L v

407

419

S T

A T

T S T

A T

S

A

peptides.

473 K M

We are indebted to Prof. I. C. Gunsaius for valuable suggestions, to Dr. D. J. Waxman for providing P450IIBl and P450IIB2 purified in his laboratory, and to Peter James for running the sequence analyzer. Benzphetamine was kindly provided by Mrs. Feithknecht from the Upjohn Co. This work was supported by the Swiss National Foundation Grants 3.973-084, 3.199-085, and 3.189-088. REFERENCES 1. Guengerich, F. P. (1987) press, Boca Raton, FL. 2. Gonzalez,

F. J. (1989)

Mammalian

Pharmacol.

Cytochromes

P-450,

CRC-

Reu. 40,243-288.

3. Bansal, S. K., Love, J. H., and Gurtoo, H. L. (1983) Biachem. Biophys. Res. Commun. 117,268-274. 4. Ban&, S. K., Love, J. H., and Gurtoo, H. L. (1985) Eur. J. Biochem. 146,23-33. 5. Rampersaud, A., and Walz, F. G., (1983) Proc. Natl. Acad. Sci. USA 80.6542-6546. 6. Rampersaud, A., Waxman, D. J., Ryan, D. E., Levin, W., and Walz, F. G. (1983)

Arch.

Biochem.

Biophys.

243,

174-183.

7. Ryan, D. E., Iida, S., Wood, A. W., Thomas, P. E., Lieber, and Levin, W. (1984) J. Biol. Chem. 259,1239-1250. 8. Waxman,

D. J. (1984)

J. Biol.

Chem.

259,

C. S.,

15,481-15,491.

9. Pickett, C. B., Jeter, R. L., Wang, R., and Lu, A. Y. H. (1983) Arch. Biochem. Biophys. 255,854-863. 10. Nebert, D. W., Nelson, D. R., Coon, M. J., Estabrook, R. W., Feyereisen, R., Fuji-Kuryiama, Y., Gonzalez, F. J., Guengerich, F. P., Gun&us, I. C., Johnson, E. F., Loper, J. C., Sato, R., Waterman, M. R., and Waterman, D. J. (1991) DNA Cell Bid. lO, l-14. 11. Komori, M., Imai, Y., Tsunasawa, S., and Sato, R. (1988) Biochem-

Amino Acids Varying between P4501IBl and P450IIB2 Amino

ACKNOWLEDGMENTS

478 G A

istry27,73-80. 12. Oesch, F., Waxman, D. J., Morrissey, J. J., Honscha, W., Kissel, W., and Friedberg, T. (1989) Arch. Biochem. Biqhys. 270,23-32. 13. Ryan, D., Thomas, P. E., and Levin, W. (1982) Arch. Biochem. Biophys. 216,272-288. 14. Shephard, E. A., Pike, S. F., Rabin, B. R., and Phillips, I. P. (1983) And. Biachem. 129,430-436. 15. De Pierre, J., and Dallner, G. (1976) in Biochemical Analysis of Membranes (Mady, E. H., Ed.), pp. 79-131, Wiley, New York. 16. Pharmacia Fine Chemicals (1982) in Isoelectric Focusing, Principles and Methods, Ljungforetagen AG, Otebro, Sweden.

30 17. Crosby, 18. Black,

OERTLE W. H., and Forth, J. A. (1985)

F. W. (1956)

Electrophoresis

Blood

4,380.

6, 27.

19. Schweiser, M., Peter, M. A., Filipovic, D., Tinner, R., Bosshard, H. R., and Oertle, M. (1991) Arch. Biochem. Biophys. 288,64-70. 20. Ullrich, V., and Weber, 353,1171-1177.

P. (1972)

21. Richter, C., Azzi, A., Weser, 252,5061-5066. 22. Nash,

T. (1953)

Biochem.

Hoppe-Seykr’s

U., and Wendel,

2. Physiol. A. (1977)

Chem.

J. BzbZ. Chem.

J. 55,416-421.

23. Yuan, P. M., Ryan, D. E., Levin, W., and Shively, Natl. Acad. Sci. USA SO, 1169-1172.

J. E. (1983)

Proc.

ET

AL.

24. Lowry, 0. H., Rosenbourgh, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275. 25. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239, 2370-2378. 26. Laemmli, U. K. (1970) Nature 227,680-685. 27. Black, S. D., and Coon, M. J. (1986) in Cytochrome P-450 (Ortiz de Montellano, P. R., Ed.), p. 177, Plenum, New York/London. 28. Suva, Y., Mizumkami, Y., Sogawa, K., and Fuji-Kuriyama, Y. (1985) J. Biol. Chem. 260, 7980-7985. 29. Mizukami, Y., Sogawa, K., Suva, Y., Muramatsu, M., and Fuji-Kuriyama, Y. (1983) Proc. Natl. Acad. Sci. USA 80,3958-3962. 30. Guengerich, F. P. (1979) Biochim. Biophys. Acta 577, 132-141. 31. Friedberg, T., Grassow, M. A., and Oesch, F. (1990) Arch. Biochem. Biophys. 279, 161-173.