THE JOURNAL OF BIOLOGICAL CHEMISTRY Val. 266, No. 2, Issue of January 15, pp. 735-739,1991 Printed in U. S. A.

Interactions Among Cytochromes P-450 in the Endoplasmic Reticulum DETECTION OF CHEMICALLY CROSS-LINKED COMPLEXES WITH MONOCLONAL ANTIBODIES*

(Received for publication, July 26, 1990)

Kenneth AlstonS, Richard C. Robinson, Sang S. Park, Harry V. Gelboin, and Fred K. Friedmans From the Laboratory of Molecular Carcinogenesis, National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892 and the $Department of Chemistry, Benedict College, Columbia, South Carolina 29204

The quaternary structure of rat liver cytochrome P- 450 within microsomal membranes from 3-methyl- cholanthrene-treated rats was examined by a novel chemical cross-linking-monoclonal antibody approach. Complex formation among the different forms of P- 450 was probed by cross-linking of membrane proteins followed by immunopurification with a monoclonal an- tibody (mAb) to P-450c, the major 3-methylcholan- threne-inducible form. Subsequent immunoblot analy- sis of the immunopurified proteins with this mAb in- dicated that P-45Oc formed complexes with other microsomal proteins. Immunoblots with mAbs to dif- ferent P-450s were carried out to identify the P-450s that were cross-linked to P-45Oc. This approach de- tected specific cross-linking of P-45Oc to P-450 2a. Immunoinhibition experiments suggest that P-450 2a further metabolizes the primary phenols produced by P-45Oc-catalyzed hydroxylation of benzo[a]pyrene. Complex formation among membrane-bound enzymes has implications for their catalytic efficiency and an approach combining cross-linking and monoclonal an- tibody-based characterization of cross-linked proteins will be useful for elucidating such membrane protein macrostructures.

The cytochromes P-450 are a family of enzymes that cata- lyze the oxidation of a wide variety of lipophilic compounds (1-5). These include xenobiotics such as drugs and carcino- gens, as well as endogenous compounds such as prostaglan- dins, fatty acids, and steroids. The various forms of P-4501 differ in their substrate and product specificities and reactiv- ities. The ultimate disposition of substrates is therefore influ- enced by the types and amounts of P-450s present. Regulation of the P-450 expression is governed by numerous factors, including age, sex, and hormonal status, and environmental and endogenous inducers.

The P-450s and other mixed function oxygenase proteins are membrane-bound (6). A substrate molecule that diffuses in the membrane and is transformed by a P-450 may be a

* This work was supported in part by National Institutes of Health Grant CRR08112 and the Summer Faculty Research Fellowship Program of the American Society of Biochemistry and Molecular Biology. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Institute, Bldg. 37, Rm. 3324, Bethesda, MD 20892. 5 To whom correspondence should be addressed National Cancer

’ The abbreviations used are: P-450, cytochrome P-450; BP, benzo[a]pyrene; mAb, monoclonal antibody; MC, 3-methylcholan- threne; MC-microsomes, liver microsomes from MC-treated rats; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electro- phoresis; sulfo-SADP, sulfosuccinimidyl(4-azidopheny1dithio)pro- pionate.

-

substrate for other P-450s it subsequently encounters. There are many examples of P-450 substrates undergoing parallel and/or consecutive reaction sequences at multiple sites, in- cluding benzopyrene (7) and steroids (8-10). The partitioning of metabolite flux among the many possible pathways yields a multiplicity of products whose distribution depends upon the type and amount of P-450s present as well as factors that govern metabolite transfer among the P-450s.

Formation of enzyme-enzyme complexes generally facili- tates metabolite transfer between these enzymes (11, 12) owing to the spatial proximity of the active sites. Such com- plexes in the P-450 system would enhance metabolite flux for consecutive reactions catalyzed by the complexed P-450s. Since the type and amount of P-450s in these complexes may regulate metabolic flux in the membrane, characterization of their organizational structure has implications for studies in drug metabolism and carcinogen activation.

Chemical cross-liaking is an approach for structural studies in the range of 5-20 A and is particularly useful for elucidation of the quaternary structure of membrane proteins (13). Nu- merous membrane proteins have been examined by this ap- proach (14). We have investigated the quaternary structure of P-450s via a cross-linking approach, using monoclonal antibodies as probes for cross-linked P-450s. Reaction of microsomal membrane proteins from MC-treated rats with the reversible heterobifunctional cross-linking reagent sulfo- SADP revealed that both P-45Oc and P-450 2a were cross- linked. This combined cross-linking-antibody approach will be applicable for elucidating such structural relationships for membrane protein systems.

MATERIALS AND METHODS

Preparation of Microsomes-Treatment of male Sprague-Dawley (8 week) rats with MC (15) was as previously described. Liver micro- somes were prepared by differential centrifugation (16). Protein con- centration was determined by the BCA method (Pierce Chemical CO.).

Monoclonal Antibodies-These were prepared and characterized as previously described for mAb 1-36-1 and 1-7-1 to the MC-induced P- 450c (15), mAb 1-98-1 to P-450j (17), and mAb 2-13-1 to P-450 2a (18). These mAbs were purified from mouse ascites fluid as described (19).

Cross-linking with Sulfo-SADP-Cross-linking with sulfo-SADP (Pierce Chemical Co.) was carried out in 0.1 M sodium phosphate, pH 7.4, containing 0.1 mM EDTA, 3 mM magnesium chloride, and 20% glycerol at 22 “C in a final volume of 0.65 ml. The sulfo-SADP (20 mM in DMSO) was added to 0.5 ml of microsomes (2.5 mg/ml) to yield 1 mM sulfo-SADP and incubated with mixing at 22 “C for 30 min. 10 mM N-ethylmaleimide (added from a freshly prepared 0.5 M stock solution in ethanol) was added to block unreacted sulfo-SADP and membrane protein sulfhydryls. The mixture was then irradiated at 270 nm and 22 “C for 10 min, using a SLM DW2000 spectropho- tometer. 0.1 M glycine was added to a final concentration of 10 mM and incubated for 15 min. The microsomes were then solubilized by addition of 10% Emulgen 911 (Kao-Atlas, Japan) to a final concen-

735

736 Endoplasmic Reticulum Cytochrome P-450 Interactions tration of 0.5% and subjected to immunopurification.

Immunopurification-The immunopurification procedure was as described previously (20) with some modifications. MAb 1-36-1 to P- 450c was immobilized on a Sepharose support and mixed with the solubilized, cross-linked microsomes at a ratio of 0.1 ml of resin/mg of protein. After mixing at 22 "C for 15 min, the suspension was centrifuged at 5000 X g and the supernatant removed. The resin was washed twice with 1 ml of 40 mM potassium phosphate, pH 7.0, 40 mM potassium phosphate, 1 M NaCI, pH 7.0, and 4 mM potassium phosphate, pH 7.0. All solutions contained 0.1% Emulgen 911 and 20% glycerol. Following the final wash, 0.1 M glycine, pH 3.0, was added in an amount equal to that of the resin. After mixing and a 5- min incubation, the mixture was centrifuged and the supernatant adjusted to pH 7.4 by addition of 0.08 volumes of 1 M Tris-HC1, pH 8.4.

Immunodetection of Cross-linked Proteins-The immunopurified samples were subjected to SDS-PAGE followed by immunoblotting (21). All samples were analyzed in the presence and absence of 2% B- mercaptoethanol. The immunoblots were developed by sequential incubations with mAb in 0.25 mg/ml ascites fluid, alkaline phospha- tase-conjugated goat anti-mouse IgG (1:lOOO dilution) (K&P Labo- ratories), and 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetra- zolium substrate (K&P Laboratories). For immunodetection of NADPH-dependent cytochrome P-450 reductase, a 1:lOOO dilution of rabbit antiserum (provided by Dr. J. Hardwick, Argonne National Laboratory) was used in conjunction with alkaline phosphatase- labeled goat anti-rabbit antibody. Immunostained areas were quan- titated by densitometry with a Beckman DU-8 spectrophotometer equipped with scanning accessory.

Benzo[a]pyrene Hydroxylation-The fluorometric assay of Nebert and Gelboin (22) was employed. This assay detects phenol metabolites of BP, primarily the highly fluorescent 3-hydroxy-BP. Immunoinhi- bition of microsomal metabolism of these phenols to nonfluorescent secondary metabolites was assessed using a 1:lOO dilution of anti- serum containing antibody to P-450 2a (23) (provided by Dr. F. J. Gonzalez, National Institutes of Health) or P-450d (24); phosphate- buffered saline was used as a control.

RESULTS

The strategy of the cross-linking-mAb procedure is as fol- lows: 1) covalently cross-link membrane proteins; 2) use a mAb to immunopurify the primary protein of interest; and 3) use immunodetection methods to identify other proteins that may be cross-linked to the primary immunopurified protein.

The heterobifunctional, cleavable, cross-linking reagent used in this study was sulfo-SADP. This reagent has a succi- nimidyl moiety that reacts with primary amines and an aryl azide that photochemically generates a nitrene that nonspe- cifically inserts into many types of covalent bonds on neigh- boring proteins (25). After cross-linking proteins in the intact microsomal membrane, these are solubilized and the target P-450 antigen is extracted from the resulting preparation by immunopurification with Sepharose-bound mAb. The immu- nopurified product contains mAb-specific P-450 along with any proteins cross-linked to this P-450. These are subjected to SDS-PAGE in the absence and presence of P-mercaptoeth- anol, which cleaves this disulfide cross-linker to yield the individual components of the cross-linked complex. Cross- linked P-450s are then identified by immunoblotting with a panel of mAbs to different P-450s. P-450s that are not cross- linked migrate as monomers in the region of 46,000-58,000 daltons, regardless of the presence of 8-mercaptoethanol. Cross-linked P-450s migrate as polymers that may not enter the gel, depending on their size, in the absence of prior cleavage. Upon cleavage, these P-450s migrate as monomers.

The cross-linking mAb procedure was carried out with microsomes from MC-treated rats and mAb 1-36-1 to P-450c, the major MC-inducible form of P-450. The proteins are shown in the SDS-PAGE analysis in Fig. 1. MC-microsomes ( l a n e 3) have more inducible proteins in the P-450 region than control microsomes (lane'2). As a control immunopuri- fication in the absence of cross-linking, a single band (lane 4 )

1 2 3 4 5 6 7 0 1

MK C MC U XL U XL MK MlCS +PME -PME

FIG. 1. SDS-PAGE analysis of proteins immunopurified from cross-linked MC-microsomes with mAb 1-36-1. Lanes I and 8 contain molecular weight markers (MK). Lanes 2 and 3 contain 10 pg of microsomes from control (C) and MC-treated rats. Lanes 4 and 5 contain proteins immunopurified from MC-microsomes that were untreated (U) or cross-linked (XL) with sulfo-SADP, respec- tively. Lanes 6 and 7 are the same as 4 and 5 except that 8- mercaptoethanol (BME) was omitted from the sample buffer. Proteins were stained with Coomassie Blue.

1 2 3 4 5 6 . . - " -_i"rr

C MC U XL U XL

MlCS +PME -PME

FIG. 2. Immunoblot analysis of mAb 1-36-1 immunopuri- fied proteins using mAb 1-36-1 for immunodetection. Lanes 1 and 2 contain 10 pg of microsomes from control (C) and MC-treated rats. Lanes 3 and 4 contain proteins immunopurified from MC- microsomes that were untreated (U) or cross-linked (XI,) with sulfo- SADP, respectively. Lanes 5 and 6 are the same as 3 and 4 except that P-mercaptoethanol was omitted from the sample buffer.

corresponding to P-45Oc is obtained from total microsomal proteins. Upon cross-linking and subsequent cleavage with 0- mercaptoethanol, P-45Oc as well as additional faster migrating proteins are also present ( l a n e 5). The intensities of these bands are markedly reduced in the absence of prior cleavage ( l a n e 7), presumably owing to their presence as high molecular weight complexes that are visible near the origin of the gel. Omission of P-mercaptoethanol from the control immunopu- rification does not affect P-45Oc band intensity, but results in reduced band sharpness ( l a n e 6 ) .

The proteins immunopurified with mAb 1-36-1 were ana- lyzed on immunoblots using this mAb (Fig. 2). The induction of P-45Oc by MC is exemplified by the immunostained band in MC-microsomes (lane 2) and its absence in microsomes from untreated rats ( l a n e I). In the absence of prior cross- linking, immunodetectable P-45Oc migrated a t its monomeric position both in the presence and absence of P-mercaptoeth- anol (lanes 3 and 5). Upon cross-linking and immunopurifi- cation, monomeric P-45Oc was visible upon cleavage (lane 4 )

Endoplasmic Reticulum Cytochrome P-450 Interactions 737

but virtually disappeared in the absence of P-mercaptoethanol (lane 6) with the concomitant appearance of higher molecular weight P-45Oc near the origin. These results indicate that P- 450c molecules in the microsomal membrane are cross-linked either to other P-45Oc molecules, other forms of P-450, and/ or other proteins.

Since several of the immunopurified and cross-linked mi- crosomal proteins observed on SDS-PAGE (Fig. l , lane 5) migrated slightly faster than P-45Oc and appeared in the P- 450 region (46,000-58,000 daltons), we examined the possi- bility that some of these may be P-450s. Several mAbs to different P-450s were thus used as probes in immunoblots of the proteins immunopurified from MC-microsomes with mAb

An immunoblot using mAb 1-7-1, which recognizes both P- 450c and P-450d (26), detects these P-450s as the upper and lower bands, respectively, in MC-microsomes (Fig. 3, lane I). In the cross-linked and immunopurified mixture (lune 2), mAb 1-7-1 immunodetected only the P-45Oc band seen with mAb 1-36-1 (lanes 3 and 4) but not a lower P-450d band. One might expect that since P-45Oc and P-450d are both induced by MC they may migrate to the same membrane regions when they are synthesized and thus have a tendency to associate with each other. Our results do not support this expectation since the P-45Od/P-45Oc ratio of staining intensity of MC-micro- somes (lane I) is 0.3 and would be expected to be the same for the cross-linked product if these P-450s exhibited identical cross-linking behavior. An upper limit for the degree of P- 450c to P-450d cross-linking can be determined from the sensitivity limits of the immunoblot procedure: we estimate that the P-450d band would be visible for a ratio of staining intensities greater than 0.05. Comparing this value to the 0.3 value of MC-microsomes indicates that less than 17% of P- 450d is cross-linked to P-45Oc.

Immunodetection using mAb 2-13-1 to P-450 2a reveals the same single band in MC-microsomes (lane 5) as in the cross- linked-immunopurified product (lane 6). To ascertain whether this band is distinct from that seen using mAb 1-36-1, a mixing experiment was performed. When immunostained with a mixture of mAbs 1-36-1 and 2-13-1, two distinct bands corresponding to the P-45Oc and P-450 2a forms recognized by these two mAbs, respectively, were observed in both MC- microsomes (lane 7) and cross-linked-immunopurified prod- uct (lane 8). These P-450s are thus cross-linked. Furthermore, the relative intensity of the two bands in MC-microsomes is similar to that of the immunopurified material, suggesting a

1-36-1.

1 2 3 4 6 6 7 8

~

““I”” “ 1

1-7-1 1-36-1 2-13-1 1-36-1

2-1 3-1 and

FIG. 3. Immunoblot analysis of mAb 1-36-1 immunopuri- fied proteins using various mAbs for immunodetection. All samples contained 6-mercaptoethanol. Lanes I, 3, 5, and 7 contain MC-microsomes. Lanes 2, 4, 6, and 8 contain proteins from these microsomes cross-linked with sulfo-SADP and immunopurified with mAb 1-36-1. mAb 1-7-1 was used for immunodetection for lanes I and 2, mAb 1-36-1 for lanes 3 and 4, mAb 2-13-1 for lanes 5 and 6, a mixture of mAbs 1-36-1 and 2-13-1 for lanes 7 and 8.

U XL MlCS RED

FIG. 4. Immunoblot analysis of mAb 1-36-1 immunopuri- fied proteins using anti-reductase antibody for detection. All samples contained P-mercaptoethanol. Lanes I and 2 contain proteins immunopurified from untreated (U) and cross-linked ( X L ) MC- microsomes, respectively. Lane 3 contains 10 pg of MC-microsomes, in which the major band was identified as reductase by reference to purified reductase (0.2 pg) in lane 4 .

strong association between these P-450s. Immunoblotting of mAb 1-36-1 immunopurified material

with mAb 1-98-1 to P-450j detected this form in microsomes but not in cross-linked product (data not shown). This result along with the nondetection of any cross-linked P-450d thus demonstrates that the observed cross-linking of P-45Oc to P- 450 2a is form-specific and is not a universal characteristic of

As P-450 is known to associate with reductase (6), we examined the cross-linked, immunopurified P-45Oc for reduc- tase content. An immunoblot with anti-reductase antibody (Fig. 4) revealed that reductase was indeed present in cross- linked, immunopurified microsomes (lane 2) and absent in immunopurified, untreated microsomes (lane I). This positive control experiment thus validated our cross-linking approach for detection of P-450 complexes.

Since the cross-linking data indicated a structural interac- tion between P-45Oc and P-450 2a, we next explored the metabolic consequences of this relationship. Using limiting levels of substrates, the disappearance of the primary metab- olites produced by each one of these P-450s was monitored in the presence and absence of inhibitory antibody to the other P-450. We thus measured the phenol products of BP hydrox- ylation, a reaction predominantly catalyzed in MC-micro- somes by P-45Oc (15, 27), and which we could barely detect using microsomes from untreated rats, which lack P-45Oc. As shown in Fig. 5 (control curve), a rapid increase in the level of B P phenols was followed by a decline as they were con- verted by microsomal P-450s to secondary products, which may include diphenols (28-30). An antibody to P-450 2a inhibited the further metabolism of primary BP phenols while antibody to P-450d, used as a control for nonspecific antibody effects, had no effect. In addition, the anti-P-450 2a antibody did not affect the rate of BP hydroxylation, as measured under reaction conditions with excess substrate. These results demonstrate that P-450 2a plays a role in secondary metabo- lism of BP phenols produced by P-45Oc.

We performed similar measurements of the GB-hydroxy product of P-450 2a-catalyzed testosterone hydroxylation, since liver microsomes catalyze the conversion of mono- to dihydroxysteroids (9). The inhibitory mAb 1-7-1 to P-450~ had no effect on the rate of disappearance of this primary metabolite (data not shown), suggesting that P-45Oc has no role in its further metabolism.

P-450s.

DISCUSSION

The electron-transport chain of the mixed function oxygen- ase system is composed of the P-450s and other proteins such

738 Endoplasmic Reticulum Cytochrome P-450 Interactions

0 anti-P-450 2a anti-P-450 d i

time (min)

FIG. 5. Effect of antibody to P-450 2a on microsomal BP hydroxylation products. The standard assay was carried out using 5 pg of MC-microsomes and 25 pmol of BP substrate at 37 "C. One- ml aliquots from the reaction mixture were withdrawn at the indicated times and the hydroxylated BP (BP-OH) metabolites were measured. Assays were carried out with the addition of anti-P-450 2a and anti- P-450d antibodies and phosphate-buffered saline (control).

as NADPH-cytochrome P-450 reductase and cytochrome b5. Several aspects of the organization of this membrane system require delineation: 1) the structural relationship among mol- ecules of the same P-450 form, 2) the relationship among different P-450 forms, 3) the relationship between the P-450s and other mixed function oxygenase proteins, and 4) the topographical distribution of the monooxygenase proteins relative to other membrane components such as lipids.

Some of these questions have been addressed with both microsomal and reconstituted systems (reviewed in Ref. 6). The interaction of P-450 with reductase and lipid has been especially well studied. The long rotational relaxation times of microsomal P-450 suggests immobilization in the mem- brane by either self-association or association with other membrane components to form large complexes. Solution studies also demonstrate a tendency of P-450 to aggregate. These studies do not distinguish between whether microsomal P-450 polymerization is among homologous or heterologous forms of P-450, although the functional implications are different for each situation.

Complex formation among P-450s may influence the met- abolic flux in two ways. First, a monomeric P-450 may be catalytically altered through an alloplex interaction mecha- nism (31) in which the binding interaction with another P - 450 modifies its active site conformation. Second, the spatial proximity of the reactive centers on enzyme-enzyme com- plexes facilitates metabolite transfer for consecutive reactions and enhances metabolite flux through the complexed enzymes (11,12). The topographical distribution and quaternary struc- ture as well as the type and amount of P-450s in the membrane regulates metabolic flux and influences substrate disposition. There are many examples of P-450 substrates undergoing oxygenation reactions a t multiple sites, with benzopyrene as a well-characterized example of a molecule which undergoes parallel consecutive reactions at multiple sites to yield dozens of products (7). Another example of multiple P-450-catalyzed reactions is that of the sequential hydroxylation of steroid molecules. One report on the formation of mono- and dihy- droxysteroid metabolites raised the question of whether me- tabolite transfer between the catalytic P-450s involves trans- membrane transport, or transport between P-450s that are juxtaposed in the membrane (9). The cross-linking-immuno- purification method in this report presents one approach to addressing this problem.

P-45Oc efficiently hydroxylates polycyclic aromatic hydro- carbons while P-450 2a hydroxylates steroids at the 66-posi- tion. The functional implications of a complex between these P-450s that metabolize different classes of substrates is thus not readily apparent. The substrate and reactivity profiles of P-450s are broad and overlapping, however, and one of these P-450s may act on the product of the other P-450. One must also consider that the substrate specificity profiles of P-450s are generally determined for the isolated enzymes with a large excess of substrate and thus do not necessarily reflect the activities in the microsomal membrane under physiological conditions. Clusters of P-450s in the milieu of the membrane may act as a "sink" to concentrate the specific substrate of one P-450 form in the vicinity of a second P-450 form in the cluster and thus enhance metabolism through the second P- 450. However, when the second P-450 is in the isolated state it may have a low activity toward the same substrate.

In order to investigate the metabolic consequences of the P-45Oc interaction with P-450 2a, we carried out experiments using specific antibodies to these P-450s to determine whether the primary metabolite of one P-450 is subsequently acted on by the other P-450. We used limiting substrate in order to rapidly consume substrate, an approach previously reported for following secondary metabolism of BP (32) and andro- stenedione (9). We found no evidence that the 6P-hydroxy- testosterone metabolite produced by P-450 2a is further me- tabolized by P-45Oc. However, the data indicate a role for P- 450 2a in secondary metabolism of the BP phenols produced by P-45Oc. We have thus demonstrated a functional as well as a structural interaction between two P-450s. P-450 2a may exert its effect in several ways, most directly via oxygenating a BP phenol; although multiply hydroxylated BP metabolites have been reported (28-30, 33), the specific P-450 forms responsible for their formation are unknown. Alternatively, P-450 2a may interact with P-45Oc in such a manner as to inhibit the dissociation rate for release of newly hydroxylated BP-phenols from P-45Oc.

Previous chemical cross-linking and electrophoresis studies of liver microsomes showed that proteins in the P-450 region were cross-linked (34-37), but their identities were not estab- lished since nonspecific probes were used for protein detec- tion. The approach in this report relies on the specificity of mAbs as well as immunopurification to eliminate non-cross- linked proteins and is potentially more useful since mAbs can be prepared to any desired membrane protein of interest.

Consideration of the kinetics of chemical cross-linking (36) leads to the conclusion that cross-links are formed between proteins in a complex (which may include proteins that are closely associated, i.e. spatially proximate) and not proteins that randomly collide during lateral diffusion in the mem- brane. In this report one of the cross-linker functional groups is photoreactive and generates a nitrene whose lifetime is on the order of one ms (25), which further decreases the likeli- hood of cross-linking of other than protein complexes. The observed cross-linking of P-45Oc and P-450 2a thus most likely derives from an actual complex of P-450s.

The absence of cross-linking between P-45Oc and other mAb-specific P-450s (P-450d and P-45Oj) suggests that P- 450c does not form complexes with these forms. It is possible, however, that a complex will not be cross-linked owing to improper distance and/or orientation of the reactive groups on the proteins. This is unlikely, however, since both func- tionalities of sulfo-SADP are quite reactive with the protein surface. The target for the succinimidyl functionality is an amino group. Each P-450 molecule thus contains a number (30-50) of potentially reactive lysines that are quite evenly

Endoplasmic Reticulum Cytochrome P-450 Interactions 739

distributed along the primary sequences. Furthermore these charged groups are more likely to be exposed to aqueous medium and thus exposed on the protein surface rather than enveloped in the membrane. The nitrene generated from sulfo-SADP is likewise a highly reactive moiety that inserts into many types of covalent bonds (25) and thus will react with many otherwise inert amino acids.

In summary, the utility of chemical cross-linking for eluci- dation of organization of membrane proteins is established. When coupled with specific mAb-based immunopurification- immunodetection methods in the approach described in this report, further details of the interactions among membrane protein systems may be revealed, and complex systems such as the P-450s may be examined. Although interactions be- tween P-450 and other microsomal proteins have been exten- sively studied, these previous studies have dealt with either 1) a single defined P-450 reconstituted in artificial membrane systems, or 2) the state of P-450s within microsomes, but without defining which P-450 form(s) contribute to the ob- served parameters. In contrast, the current approach probes the state of defined P-450s in a "natural" microsomal milieu which includes not only natural membrane components but the natural mixture of different P-450 forms.

REFERENCES

1. Lu, A. Y. H., and West, S. B. (1980) Pharmacol. Reu. 31, 277-

2. Adesnik, M., and Atchison, M. (1985) CRC Crit. Reu. Biochem.

3. Ortiz de Montellano, P. R. (ed) (1986) Cytochrome P-450: Struc- ture, Mechanism and Biochemistry, Plenum Publishing Co., New York

4. Nebert, D. W., and Gonzalez, F. J. (1987) Annu. Reu. Biochem.

5. Gonzalez, F. J. (1989) Pharmacol. Reu. 40, 243-288 6. Ingleman-Sundberg, M. (1986) in Cytochrome P-450, Structure,

Mechanism and Biochemistry, (Ortiz de Montellano, P. R., ed) pp. 119-160, Plenum Publishing Co., New York

7. Gelboin, H. V. (1980) Physiol. Reu. 60, 1107-1165 8. Gustafsson, J.-A., Mode, A., Norstedt, G., and Skett, P. (1983)

9. Sheets, J . J., and Estabrook, R. W. (1985) Biochemistry 24,

295

19,247-305

56,945-993

Annu. Reu. Physiol. 45, 51-60

6591-6597 10. Waxman, D. J. (1988) Biochem. Pharmacol. 37, 71-84 11. Srivastava, D. K., and Bernhard, S. A. (1986) Curr. Top. Cell.

12. Srivastava, D. K., and Bernhard, S. A. (1987) Science 234,1081-

13. Peters, K., and Richards, F. M. (1977) Annu. Reu. Biochem. 46,

Regul. 28, l -68

1086

523-551

14. Das, M., and Fox, C. F. (1979) Annu. Reu. Biophys. Bioeng. 8,

15. Park, S. S., Fujino, T., West, D., Guengerich, F. P., and Gelboin,

16. Lu, A. Y. H., and Levin, W. (1972) Biochem. Biophys. Res.

17. KO, I. Y., Park, S. S., Song, B. J., Patten, C., Tu, Y., Hah, Y. C., Yang, C. S., and Gelboin, H. V. (1987) Cancer Res. 47, 3101- 3109

18. Park, S. S., Waxman, D. J., Miller, H., Robinson, R. C., Attisano, C., Guengerich, F. P., and Gelboin, H. V. (1986) Biochem. Pharmacol. 35, 2859-2867

19. Stanker, L. H., Vanderlaan, M., and Juarez-Salinas, H. (1985) J. Immunol. Methods 76, 157-169

20. Cheng, K. C., Gelboin, H. V., Song, B.-J., Park, S. S., and Friedman, F. K. (1984) J. Biol. Chem. 259, 12279-12284

21. Towbin. H., Staehelin. T.. and Gordon. J. (1979) Proc. Natl. Acad.

165-193

H. V. (1982) Cancer Res. 42, 1798-1808

Commun. 46, 1334-1339

Sci. U. S.'A. 76, 4350-4354 22. Nebert. D. W.. and Gelboin. H. V. (1968) J. Biol. Chem. 243.

624216249 '

, .

23. Gonzalez, F. J., Song, B. J., and Hardwick, J. P. (1986) Mol. Cell.

24. Myers, M., Liu, G., Miller, H., Gelboin, H. V., Robinson, R. C., and Friedman. F. K. (1990) Biochem. Bioohvs. Res. Commun.

Bid. 6, 2969-2976

169,171-176' . "

25. Bavlev. H. (1983) Photogenerated ReaPents in Biochemistrv and MoLculur'Biology, pp. Tl-14, ElsevieiScientific Publishing Co., Amsterdam

26. Cheng, K. C., Park, S. S., Krutzsch, H. C., Grantham, P. H., Gelhoin, H. V., and Friedman, F. K. (1986) Biochemistry 25,

27. Thomas, P. E., Reik, L. M., Ryan, D. E., and Levin, W. (1981) J.

28. Wiebel, F. J. (1975) Arch. Biochem. Biophys. 168,609-621 29. Capdevila, J., Estabrook, R. W., and Prough, R. A. (1978)

30. Prough, R. A., Saeki, Y., and Capdevila, J. (1981) Arch. Biochem.

31. Yip, Y. K., Waks, M., and Beychok, S. (1972) J. Biol. Chem. 247,

32. Capdevila, J., Renneberg, R., Prough, R. A., and Estabrook, R. W. (1980) in Genetic and Environmental Factors in Experimen- tal and Human Cancer (Gelboin, H. V., Macmahon, B., Mat- sushima, T., Sugimura, T., Takayama, s., and Takehe, H., eds) pp. 45-58, Japan Scientific Societies Press, Tokyo

33. Ribiero, O., Kirkby, C. A,, Hirom, P. C., and Millburn, P. (1985) Carcinogenesis 10, 1507-1511

34. McIntosh, P. R., and Freedman, R. B. (1979) FEBS Lett. 105, 217-223

35. McIntosh, P. R., and Freedman, R. B. (1980) Biochem. J. 187, 227-237

36. Baskin, L. S., and Yang, C. S. (1982) Biochim. Biophys. Acta

37. Baskin, L. S., and Yang, C. S. (1982) Biochem. Biophys. Res.

2397-2402

Biol. Chem. 256, 1044-1052

Biochem. Biophys. Res. Commun. 82,518-525

Biophys. 212, 136-146

7237-7244

684,263-271

Commun. 108. 700-707