RtHo

JW

R

tsnccpthfPfavwccypt(coassldstiA

Nd

7

Biochemical and Biophysical Research Communications 270, 777–781 (2000)

doi:10.1006/bbrc.2000.2495, available online at http://www.idealibrary.com on

esidue 285 in Cytochrome P450 2B4 Lackinghe NH2-Terminal Hydrophobic Sequenceas a Role in the Functional Associationf NADPH–Cytochrome P450 Reductase

ohannes Schulze, Katharina Tschop, Michael Lehnerer, and Peter Hlavica1

alther-Straub-Institut fur Pharmakologie und Toxikologie der LMU, Nussbaumstrasse 26, D-80336 Munich, Germany

eceived February 23, 2000

Cytochrome P450 (P450 or CYP; EC 1.14.14.1) en-zmop(FtctmPn

bahflCmttcodmPtoiCf

M

dlG

Cytochrome P450 2B4 (CYP2B4) lacking the NH2-erminal signal anchor sequence (2–27) was used totudy the impact of replacement of histidine with ala-ine at position 285 on electron transfer from NADPH-ytochrome P450 reductase (P450R). Absorption andircular dichroism spectra of the recombinant hemo-roteins indicated that amino acid substitution nei-her grossly perturbed the geometry of the immediateeme vicinity nor the global polypeptide backbone

olding. Fitting of the initial-velocity patterns of450R-directed reduction of the ferric CYP2B4 (2–27)

orms to the Michaelis–Menten kinetics revealed anpproximately 3.5-fold increase in the apparent Km

alue for the electron donor of the H285A mutant,hile its reductive capacity (Vmax) remained un-

hanged; this caused a strong drop in reductive effi-iency of the engineered enzyme. Circumstantial anal-sis suggested that impaired association of the redoxartners accounted for this phenomenon. Thus, dele-ion of the positive charge at position 285 of CYP2B42–27) might have disrupted contacts with oppositelyharged entities on the P450R surface. Measurementsf the stoichiometry of aerobic NADPH consumptionnd H2O2 production disclosed the oxyferrous H285Apecies to autoxidize more readily compared with thehortened wild type. This was assumed to arise fromess efficient coupling of the system due to defectiveonation of the second electron by P450R. These re-ults are consistent with the view that His-285 in theruncated CYP2B4 is of importance in the functionalnteraction with the flavoprotein reductase. © 2000

cademic Press

Abbreviations used: P450 or CYP, cytochrome P450; P450R,ADPH-cytochrome P450 oxidoreductase; SDS–PAGE, sodium do-ecyl sulfate–polyacrylamide gel electrophoresis.

1 To whom correspondence should be addressed. Fax: 49-89-5160-207. E-mail: Hlavica@lrz.uni-muenchen.de.

777

ymes represent a group of structurally related he-oproteins catalyzing the oxidative transformation

f a diversity of endogenous and exogenous com-ounds (1). Although NADPH–P450 oxidoreductaseP450R; EC 1.6.2.4), containing one molecule each ofAD and FMN, has long been recognized to consti-ute a regular component of the P450-dependent mi-rosomal electron-transport chain (2), the role of cy-ochrome b 5 as an intermediate carrier seems to beore complex, as it can either enhance or inhibit450-mediated activities in an isoform-specific man-er (3).The problem of donor/acceptor recognition has

een the most important and intriguing one in therea of P450 research. Thus, electrostatic as well asydrophobic mechanisms appear to foster electronow from P450R to the rabbit liver microsomalYP2B4 isozyme (4, 5). Moreover, repulsive forcesight be operative in this process, providing a cer-

ain looseness required for optimal spatial orienta-ion of the redox partners (6). In accord with thisoncept, previous chemical-modification studies byur laboratory suggested positively charged histi-ine(s) in CYP2B4 to participate in salt bridge for-ation with carboxylate groups on the surface of the450R molecule (7, 8). Indeed, we have recently iden-ified via site-directed mutagenesis His-226 as beingne of the critical residues (9). Here, we report on themportance of a histidine at position 285 of theYP2B4 polypeptide in controlling electron transfer

rom the flavoprotein reductase.

ATERIALS AND METHODS

Materials. NADPH, glucose 6-phosphate, glucose-6-phosphateehydrogenase (EC 1.1.1.49), glucose oxidase (EC 1.1.3.4) and cata-ase (EC 1.11.1.6) were obtained from Roche Diagnostics (Mannheim,ermany). Dilauroyl L-a-phosphatidylcholine and hexobarbital were

0006-291X/00 $35.00Copyright © 2000 by Academic PressAll rights of reproduction in any form reserved.

purchased from Sigma (Deisenhofen, Germany). All other reagentsw

(S(sp5swcsGJfc(b9bf

Wota

tU

tldPrps

dmml

wv(m1TatRfibww

wdtppsbad

adduct at 450 nm upon rapid mixing of the contents of the twos

cdsDiMwgdq

R

tdiEttdpnnSAotfC

tpetspnbagvtsssaitat(2cci

Vol. 270, No. 3, 2000 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

ere of the highest purity commercially available.

Protein expression and purification. The vector pGEX-KT-2B42–27), serving to express truncated CYP2B4 fused to glutathione-transferase (10), was generously provided by Dr. Minor J. Coon

University of Michigan, Ann Arbor, MI). The set of sense and anti-ense oligonucleotide primers designed to change histidine atosition 285 (59-AGCAGCGAGTTCGCCCACCAGAACCTC-39 and9-GATGAGGTTCTGGGCGTGGAACTCGCT-39) was chemicallyynthesized by Eurogentec (Heidelberg, Germany). The mutationas introduced using the QuikChange protocol (11). Transformants

arrying cDNA encoding for the desired amino acid exchange wereelected using the T7Sequencing Kit (Pharmacia Biotech, Freiburg,ermany). The recombinant fusion proteins were expressed in E. coliM109 (Stratagene, Heidelberg, Germany), liberated from the trans-erase moiety by thrombin treatment, and purified to a specificontent of about 14 nmol/mg protein exactly as described previously10). The concentration of the pigments was determined as indicatedy Omura and Sato (12) using a molar absorption coefficient of1,000 M21 cm21. The final preparations were subjected to analysisy SDS–PAGE on slab gels containing 12% (m/v) acrylamide (13)ollowed by Western blotting (14).

P450R, isolated from hepatic microsomes of male New Zealandhite rabbits as reported elsewhere (15), exhibited a specific activity

f 40 mmol of cytochrome c reduced/min per mg protein. The flavopro-ein was quantified by its absorbance at 456 nm using a molarbsorption coefficient of 21,400 M21 cm21 (16).

Spectral measurements. Electronic absorption spectra of the pro-eins were recorded at room temperature with a ShimadzuV1601PC spectrophotometer; optical pathlength was 1.0 cm.CD spectra in the far ultraviolet and visible regions were moni-

ored at 20°C with a Jasco J-715 spectropolarometer; optical path-engths were 0.1 and 1.0 cm, respectively. Measurements were con-ucted in 50 mM sodium phosphate (pH 7.4) containing 1–5 mM450. The standard conditions were as follows: band width, 1nm;esponse, 1 s; step resolution, 0.1 nm. Blanks (buffer without hemo-rotein) were routinely recorded and subtracted from the originalpectra. On average, data from 10 scans were accumulated.Hexobarbital binding to P450 was measured at 20°C by optical

ifference spectroscopy in media containing 2 mM hemoprotein and 2M barbiturate in 100 mM sodium phosphate (pH 7.4) supple-ented with 20% (v/v) glycerol. Spectra were recorded in the wave-

ength region from 360 to 430 nm; optical pathlength was 1.0 cm.

Enzyme assays. The NADPH-supported reduction of ferric P450as measured at 20°C in reaction mixtures composed of 2 mM P450,arying amounts of P450R, 48 mM dilauroyl phosphatidylcholinesonicated until clarification was observed), 1 mM hexobarbital, 100M glucose, glucose oxidase (400 mg/ml), and catalase (75 mg/ml) in

00 mM sodium phosphate (pH 7.4) containing 20% (v/v) glycerol.he system was preincubated at room temperature for 15 min tollow efficient association of the redox proteins in the micellar ma-rix. Subsequently, the samples were gassed for 5 min with CO.eactions were initiated by the rapid addition of NADPH to yield anal concentration of 1 mM using a set of plunger cuvettes. Absor-ance changes at 450 nm relative to those at 500 nm were recordedith an Aminco DW-2 spectrophotometer operated in the dual-avelength mode.In some studies, rates of reduction were assessed in systems inhich ferric P450 and P450R did not exist in the preformed complexescribed above. In this case, 4 mM P450 and 0.2 mM P450R, respec-ively, were separately combined in two syringes with 48 mM phos-holipid dissolved in 100 mM phosphate buffer (pH 7.4) in theresence of 1 mM hexobarbital, 1 mM NADPH, and a deoxygenatingystem. The samples were incubated at 20°C for 15 min. After gentleubbling of the mixtures with CO, the syringes were transferred ton Aminco-Morrow stopped-flow apparatus. The extent of P450 re-uction was determined by monitoring the formation of the carbonyl

778

yringes.NADPH oxidase activity was measured at 20°C in aerobic media

omprising 2 mM P450, 0.5 mM P450R, 48 mM dilauroyl phosphati-ylcholine, 2 mM hexobarbital and 50 mM NADPH in 100 mModium phosphate (pH 7.4) supplemented with 20% (v/v) glycerol.isappearance of the reduced cofactor was followed by the decrease

n absorbance at 340 nm using a molar absorption coefficient of 6,22021 cm21 (17). For determining H2O2 production, the above mixturesere fortified with NADPH-regenerating system consisting of 10 mMlucose 6-phosphate, 6 mM MgCl2, and glucose-6-phosphate dehy-rogenase (5 mg/ml) and incubated at 20°C for 20 min. Peroxide wasuantified by the ferrithiocyanate method (18).

ESULTS AND DISCUSSION

Spectral characterization of the recombinant pro-eins. Truncated CYP2B4 lacking amino acid resi-ues 2–27 and its mutated derivative bearing alaninen place of histidine at position 285 were expressed in. coli fused to glutathione S-transferase and purified

o apparent homogeneity following cleavage of the pro-eins with thrombin. The pigments did not undergoegradation during the expression and purificationrocesses, as evidenced by immunoblot analysis (dataot shown). The two hemoprotein species generatedormal CO-reduced spectra, with the position of theoret bands ranging from 451.4 to 451.6 nm. The A418/390 ratios in the electronic absorption spectra of thexidized pigments varied from 2.1 to 2.3. These spec-ral characteristics are comparable with those reportedor highly purified rabbit liver microsomal low-spinYP2B4 (19).To recognize potential effects of amino acid substitu-

ion on the structure of the heme-substrate-bindingocket, the ability of the truncated wild type and itsngineered congener to interact with hexobarbital ashe model substrate was assessed in terms of type Ipectral complex formation. As shown in Figure 1, theeak-to-trough difference at the wavelength pair 380m/417 nm in the spectrum arising from barbiturateinding to the H285A variant was not significantlyberrant from that for the control enzyme, disproving arossly perturbed geometry of the immediate hemeicinity. Furthermore, the CD spectrum recorded withhe deletion mutant in the visible region was veryimilar to that for CYP2B4 (2–27) with respect to bandhape and peak position (Figure 2B), ruling out a sub-tantial rearrangement of the heme site. CD-spectralnalysis in the far ultraviolet region revealed almostdentical spectra for the two enzyme forms (Figure 2A),he optical tracings being corrected for the presence ofpohemoprotein on the basis of the A 418/A 278 ratios inhe absorption spectra of the individual preparations20). Taking into account mean residue ellipticities at22 nm of 221,400 and 223,300 deg cm2 dmol21 for theontrol and H285A species, respectively, histidine ex-hange did not appear to introduce a major alterationn the global polypeptide backbone folding (21).

tsptmttrtpsstrhcdocrfs3Hcv2Pttt

Since hexobarbital has been previously proposed toatPmiupniheswrerwpttdic

H(cp

Cotp

Vol. 270, No. 3, 2000 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

Reduction kinetics of the ferric CYP2B4 (2–27) pro-eins under anaerobic conditions. CYP2B4 (2–27) wasubjected to site-directed mutagenesis to study the im-act of a specific amino acid deletion at position 285 onhe productive interaction with P450R. Selection of theutant was based on both a tentative map of func-

ional domains located on the CYP2B4 surface (22) andhe well-known involvement of basic residues in P450Recognition (4). Measurements were conducted withhe redox proteins reconstituted into phospholipid. Thereincubation period chosen was adequate to compen-ate for the slow anchoring to the matrix of CYP2B4pecies lacking the hydrophobic tail portion (23), sohat limited availability of lipid-bound P450s could beuled out to diminish the statistical probability of theemoproteins of encountering P450R attached to mi-ellar phospholipid (24). Analyses were carried out un-er strictly anaerobic conditions to avoid autooxidationf the terminal acceptor proteins; the velocities of ac-umulation of ferrous pigment thus primarily reflectedates of electron transfer. The kinetic data derivedrom such assays are plotted in Figure 3A and areummarized in Table 1. As can be seen, there was a.5-fold increase in the apparent Km value for the285A/P450R complex in the presence of hexobarbital

ompared with the truncated wild type, while the Vmax

alue was indistinguishable from that for CYP2B4 (2–7), indicating that in the presence of elevated levels of450R the mutant P450 functioned normally. Fromhese parameters, the reductive efficiency (Vmax/Km) ofhe engineered system could be calculated to be 32%hat of the wild-type-dependent route.

FIG. 1. Spectral interaction of hexobarbital with recombinantYP2B4 (2–27) and its H285A congener. Difference spectra werebtained by reacting 2 mM hexobarbital with 2 mM truncated wildype (—) or histidine-mutated protein (–) in 100 mM sodium phos-hate (pH 7.4) containing 20% glycerol.

779

ccelerate CYP2B4 reduction in an indirect wayhrough favoring association of the pigment with450R (25), the observed fall in reductive efficiencyight have been the consequence of a mutation-

nduced defect in barbiturate binding. However, eval-ation of the hexobarbital-elicited difference spectraresented in Figure 1 did not lend much support to thisotion. Circumstantial analysis of the reduction kinet-

cs disclosed that replacement of histidine with alaninead directly affected the interaction of the modifiednzyme with its redox partner: when the ferric,ubstrate-bound H285A construct was rapidly mixedith NADPH-reduced P450R, so that association of the

edox components constituted the rate-limiting step inlectron transfer, the initial velocity of hemoproteineduction leveled down to a value 16% that for theild-type-catalyzed reaction (Figure 3B), suggestingerturbed reactivity toward each other of the proteinso represent the overriding mechanism responsible forhe observed rise in the apparent Km value for theonor/acceptor couple (Table 1). Based on the spectralnvestigations shown in Figures 1 and 2, substantialhanges in the critical architecture of the mutant’s

FIG. 2. CD spectra for recombinant CYP2B4 (2–27) and its285A derivative. CD spectra in the far ultraviolet (A) and visible

B) regions were recorded for solutions containing 1 to 5 mM trun-ated wild type (—) or mutated enzyme (–) in 50 mM sodium phos-hate (pH 7.4).

atit(firtida

pst(mapsar(pPu

As is evident from Table 1, the rate of hexobarbital-dwwwtmvalswecicgemtrrffmpptO

tC3l

tw

KVVNHH

hPddgNwm

HwtPfaPapm(

Vol. 270, No. 3, 2000 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

ctive site or gross structural anomalies due to defec-ive global protein folding could be dismissed as havingmposed constraints on the precollisional spatial orien-ation of the partners. Rather, modification of CYP2B42–27) was thought to have injured the electron trans-er interface of the pigment. Thus, deletion of the cat-onic charge at position 285 might have impeded short-ange contacts with complementary anionic entities onhe surface of the electron donor. Indeed, charge shield-ng by high salt concentrations has been previouslyemonstrated to block the transfer of reducing equiv-lents from P450R to CYP2B4 (4).

Stoichiometry of NADPH consumption and hydrogeneroxide formation with CYP2B4 (2–27) proteins recon-tituted with P450R under aerobic conditions. To fur-her characterize the effect of mutagenesis of CYP2B42–27) on electron transfer, NADPH oxidation waseasured along with hydrogen peroxide production in

erobic media containing P450/P450R, micellar phos-holipid, and hexobarbital as the substrate. In suchystems, NADPH oxidase activity has been shown to belmost entirely due to CYP2B4-dependent catalysisather than cofactor disposal by the P450R segment26). To assess turnover with a physiologically relevantrotein stoichiometry, assays were carried out at a450R-to-P450 molar ratio far below that routinelysed (26).

FIG. 3. NADPH-sustained reduction of CYP2B4 (2–27) and its285A variant by P450R. In a first set of experiments (A), reactionsere started by the addition of NADPH to anaerobic solutions con-

aining ferric P450 protein preincubated with specified amounts of450R in the presence of hexobarbital; initial rates of electron trans-

er were fitted to the Michaelis–Menten kinetics. In a second set ofssays (B), NADPH-reduced P450R was rapidly reacted with ferric450 in the presence of barbiturate to study the time course ofssociation of the redox partners; the molar ratio of flavo- to hemo-rotein in the mixtures was 1:20. The data points presented refer toeasurements with truncated wild type (E) and the mutant H285A

F), and are the means of three experiments.

780

ependent NADPH consumption by the H285A variantas consistently decreased by some 30% comparedith the truncated wild type, while H2O2 formationas marginally stimulated. On the whole, the propor-

ion of peroxide produced to NADPH utilized by theutated pigment was significantly (P , 0.005) ele-

ated by a factor of 1.6. That some conformationallteration in the H285A protein might have destabi-ized its iron-hydroperoxo intermediate to release H2O2

eemed unlikely. Therefore, the amino acid deletionas assumed to have caused the oxyferrous mutantnzyme to autoxidize more readily owing to less effi-ient introduction of the second electron usually serv-ng to tightly couple the system. Based on this concept,ritical evaluation of the data collected in Table 1 sug-ested the apparent efficiency of transfer of the secondlectron to the H285A species to be depressed to aarkedly lower extent compared with that of introduc-

ion of the first electron. This discrepancy could beeconciled by assuming that the small structural rear-angement related to the shift from substrate-bound,erric high-spin P450 to the oxyferrous low-spin con-ormer was beneficial to the functional coupling of the

odified hemoprotein with P450R. This might haveartly resulted from improved association of the redoxartners or/and the drastic rise in the midpoint poten-ial of phenobarbital-inducible ferrous P450 following

2 binding (27).Collectively, we have provided evidence of the impor-

ance in P450R recognition of His-285 in the truncatedYP2B4 isozyme. Judging from recent computer-aidedD models of CYP2B4 (28, 29), the basic residue isocated at the start of the putative I helix. In addition

TABLE 1

Initial-Velocity Kinetic Pattern of NADPH-Driven Elec-ron Transfer to CYP2B4 (D2–27) Proteins Reconstitutedith P450R

Parameter measured

CYP2B4 (D2–27) species tested

Wild type H285A mutant

m (mM) 0.57 6 0.03 2.00 6 0.10max (nmol P450 reduced/min) 0.80 6 0.05 0.89 6 0.04max/Kmax (ml/min) 1.40 6 0.08 0.45 6 0.02ADPH oxidation (nmol/min) 3.17 6 0.29 2.31 6 0.142O2 production (nmol/min) 1.04 6 0.02 1.19 6 0.052O2 produced/NADPH oxidized 0.32 6 0.02 0.52 6 0.02

Note. The apparent Km and Vmax values, reflecting theexobarbital-stimulated reactivities and reductive potencies of450R toward the recombinant hemoproteins under anaerobic con-itions, were taken from the double-reciprocal initial-velocity plotsepicted in Fig. 3A; calculations were carried out by nonlinear re-ression analysis using the Corel QuattroPro 7.0 software. AerobicADPH utilization and peroxide-linked ferrithiocyanate formationere followed at 340 and 480 nm, respectively. The data representeans 6 SEM of three experiments.

to being involved in redox partner binding, His-285 hasbtstadlpaTsesmpta

A

Fi

R

10. Pernecky, S. J., Olken, N. M., Bestervelt, L. L., and Coon, M. J.

1

111

11

1

1

1

2

2

2

2

2

2

2

22

23

3

3

3

Vol. 270, No. 3, 2000 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

een predicted to contribute to a salt link to Glu-250 inhe proposed G helix, serving to coordinate the inter-ection of the two helices (30). Using site-directed mu-agenesis, the G helix has been furthermore reported toccommodate a series of basic and aromatic residuesefining part of the P450R-binding domain (9). Theatter also comprises structural elements located in theutative C/C* helices (31) and a known point of inter-ction residing in the b2(2)–b1(3) turn of CYP2B4 (32).he sum of these findings substantiates previous ob-ervations by this and other laboratories hinting at thexistence of multiple P450R contact regions on theurface of CYP2B4 (7, 33). Work is underway to gain aore detailed molecular delineation of such specific

rotein-protein associations related to the step of elec-ron transfer with mammalian P450s for which therere no crystallographic data.

CKNOWLEDGMENTS

This work was supported by Grant Hl 1/15-1 from the Deutscheorschungsgemeinschaft. The authors are indebted to Cortina Keil-

ng for competent technical assistance.

EFERENCES

1. Porter, T. D., and Coon, M. J. (1991) J. Biol. Chem. 266, 13469–13472.

2. Lu, A. Y. H., and Coon, M. J. (1968) J. Biol. Chem. 243, 1331–1332.

3. Morgan, E. T., and Coon, M. J. (1984) Drug Metab. Dispos. 12,358–364.

4. Bernhardt, R., Kraft, R., Otto, A., and Ruckpaul, K. (1988)Biomed. Biochim. Acta 47, 581–592.

5. Black, S. D., French, J. S., Williams, C. H., and Coon, M. J.(1979) Biochem. Biophys. Res. Commun. 91, 1528–1535.

6. Voznesensky, A. I., and Schenkman, J. B. (1992) J. Biol. Chem.267, 14669–14676.

7. Golly, I., and Hlavica, P. (1993) Biochim. Biophys. Acta 1142,74–82.

8. Hlavica, P., Lehnerer, M., and Eulitz, M. (1996) Biochem. J. 318,857–862.

9. Lehnerer, M., Schulze, J., Achterhold, K., Lewis, D. F. V., andHlavica, P. (2000) J. Biochem. 127, 163–169.

781

(1995) Arch. Biochem. Biophys. 318, 446–456.1. Papworth, C., Bauer, J. C., Braman, J., and Wright, D. A. (1996)

Strategies 8, 3–4.2. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239, 2379–2385.3. Laemmli, U. K. (1970) Nature 227, 680–685.4. Towbin, H., Staehelin, T., and Gordon, J. (1979) Proc. Natl.

Acad. Sci. USA 76, 4350–4354.5. Hlavica, P., and Hulsmann, S. (1979) Biochem. J. 182, 109–116.6. French, J. S., Guengerich, F. P., and Coon, M. J. (1980) J. Biol.

Chem. 255, 4112–4119.7. Horecker, B. L., and Kornberg, A. (1948) J. Biol. Chem. 175,

385–390.8. Hildebrandt, A. G., Roots, I., Tjoe, M., and Heinemeyer, G.

(1978) Methods Enzymol. 52, 342–350.9. Haugen, D. A., and Coon, M. J. (1976) J. Biol. Chem. 251,

7929–7939.0. Janig, G. R., Friedrich, J., Smettan, G., Bernhardt, R., Ristau,

O., and Ruckpaul, K. (1985) Biomed. Biochim. Acta 44, 1071–1082.

1. Chen, Y. H., Yang, J. T., and Martinez, H. M. (1972) Biochem-istry 11, 4120–4131.

2. Hlavica, P. (1992) in Cytochrome P450: Biochemistry and Bio-physics (Archakov, A. I., and Bachmanova, G. I., Eds.), pp. 216–221, INCO-TNC, Joint Stock Company, Moscow.

3. Shank-Retzlaff, M. L., Raner, G. M., Coon, M. J., and Sligar,S. G. (1998) Arch. Biochem. Biophys. 359, 82–88.

4. Lehnerer, M., Schulze, J., Pernecky, S. J., Lewis, D. F. V., Eulitz,M., and Hlavica, P. (1998) J. Biochem. 124, 396–403.

5. Backes, W. L., and Eyer, C. S. (1989) J. Biol. Chem. 264, 6252–6259.

6. Nordblom, G. D., and Coon, M. J. (1977) Arch. Biochem. Biophys.180, 343–347.

7. Guengerich, F. P. (1983) Biochemistry 22, 2811–2820.8. Chang, Y. T., Stiffelman, O. B., Vasker, I. A., Loew, G. H.,

Bridges, A., and Waskell, L. (1997) Protein Eng. 10, 119–129.9. Lewis, D. F. V., and Lake, B. G. (1997) Xenobiotica 27, 443–478.0. Schulze, J., Lehnerer, M., Lewis, D. F. V., and Hlavica, P. (1998)

Biochem. Mol. Biol. Int. 44, 1147–1155.1. Bridges, A., Gruenke, L., Chang, Y. T., Vasker, I. A., Loew, G.,

and Waskell, L. (1998) J. Biol. Chem. 273, 17036–17049.2. Bernhardt, R., Makower, A., Janig, G. R., and Ruckpaul, K.,

(1984) Biochim. Biophys. Acta 785, 186–190.3. Davydov, D. R., Knyushko, T. V., Kanaeva, I. P., Koen, Y. M.,

Samenkova, N. F., Archakov, A. I., and Hui Bon Hoa, G. (1996)Biochimie 78, 734–743.