Kinetics of the Hydroperoxide-dependent Dealkylation Reactions

T H E JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 255, No.20, Issue of October 25, pp. 9685-9692. 1980 Printed in U.S.A.

Kinetics of the Hydroperoxide-dependent Dealkylation Reactions Catalyzed by Rabbit Liver Microsomal CytochromeP-450* (Received for publication, April 2, 1979, and in revised form, May 21, 1980)

Dennis R. KoopS and Paul F. Hollenbergg From the Departments of Biochemistry and Pathology, Northwestern University Medical and Dental Schools, Chicago, Illinois 60611

Rabbit liver microsomal cytochrome P-450 catalyzes the dealkylation of a variety of substrates when organic hydroperoxides, peracids, or peroxyesters are substituted for NADPH and 02.The peroxide-supported demethylation of p-nitroanisole by rabbit liver microsomes exhibited normal Michaelis-Menten kinetics with respect to both substrates. The V,, values for the hydroperoxide-dependent reactions were dependent on the identity of the hydroperoxide and with several of the peroxides they were significantly greaterthan those obtained with NADPH and 0 2 . The pH profiles for the hydroperoxide-supported demethylations were similar to the profile for the NADPH-supported reaction and exhibited optima from 7.0 to 7.6, depending on the identity of the organic oxidant. The kinetic mechanism of the t-butyl hydroperoxidesupported demethylation of p-nitroanisole was determined. Plots of reciprocal velocity versus the reciprocal concentration of either substrate at several different fixed concentrations of the other substrate converged to common points of intersection on the negative side of the ordinate and above the abscissa, suggesting a sequential mechanism involving the formation of a ternary complex between cytochrome P-450, p-nitroanisole, and t-butyl hydroperoxide followedbyone or more reactions and the subsequent release of the products. Potassium cyanide was a competitive inhibitor with respect to t-butyl hydroperoxide and uncompetitive with respect to p-nitroanisole. t-Butyl alcohol, a product of the t-butyl hydroperoxide-supported reaction, was a noncompetitive inhibitor with respect to both substrates. These results, which indicate that the reaction proceeds via an Ordered Bi Bi mechanism in which p-nitroanisole binds to the enzyme prior to the binding of t-butyl hydroperoxide, are discussed in relationship to the peroxidase-type mechanism whichhas been suggested for the action of cytochrome P-450. * This research was supported in part by Grant R01 CA-16954 from the National Institutes of Health, United States Public Health Service, Department of Health, Education and Welfare, and by Grant A 74-30 from the Chicago and Illinois Heart Association. 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. $ Predoctoral Trainee of the United States Public Health Service, Grant GM-07263. The data in this paper are taken from a thesis submitted by D. R. K. in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biochemistry in the Graduate School of Northwestern University. Present address, Department of Biological Chemistry, The University of Michigan, Ann Arbor, Mich. 48109. 8 Recipient of a research fellowship from the Schweppe Foundation.

The liver microsomal mixed function oxidase system catalyzes the oxidation of a wide variety of substrates including drugs, steroids, fattyacids, alkanes,environmental pollutants, and other xenobiotics (1-3). This enzyme system has been resolved intothreecomponents,cytochrome P-450, cytochrome P-450 reductase, and phosphatidylcholinewith all three components being required for maximal activity of the reconstituted system (4,5). The separation purification and of the three components hasbeen reported by several laboratories (6-10). The reaction mechanism involves the binding of the substrate to the ferricenzyme followed by thetransfer of 1 electron from NADPH-cytochrome P-450 reductase to cytochrome P-450 resulting in the formationof a ferrous enzyme. substrate complex which then binds oxygen to give a ternary oxyferrous enzyme substrate complex (11-13). The transfer P-450 results in the formation of a 2nd electron to cytochrome of an “activated oxygen intermediate” which then hydroxylates theenzyme-bound substrate molecule. The natureof the activated oxygenating intermediate has not been determined and various activated forms of oxygen including superoxide (14) or peroxides (15-18) has been suggested. The isolation of the hydroperoxides of tetralin (15),fluorene (16), steroids (17), (18) from microsomal and 9,10-dimethyl-l,Z-benzanthracene reaction mixtures has been cited as evidence for the role of hydroperoxides in somebiological hydroxylations. Liver microsomal cytochrome P-450 can use organic hydroperoxides and peracids in place of NADPH and molecular oxygen to support a variety of reactions including N - and 0dealkylations (19-21) and the hydroxylation of aromatic substrates (22-25). In addition, other organic and inorganic oxidants can support certain cytochrome P-450-catalyzed reactions. For example,m-chloroperbenzoic acid supportsthe demethylation of benzphetamine (23), iodosobenzenesupports ethoxycoumarin o-deethylation (20), and NaC102 and NaI04 are active in the hydroxylation of steroids (25). Nordblom et al. (23) have demonstrated that highly purified rabbit liver cytochrome P-450 catalyzes the hydroxylation of cyclohexane and the dealkylation of benzphetamine, N,N-dimethylaniline, and several other N-methyl compoundswith cumene hydroperoxide as the oxidant. Thecumene hydroperoxide-supported demethylation of N,N-dimethylaniline results in the formation of equimolar quantities of formaldehyde, N-methylaniline, anda,a-dimethylbenzyl alcohol, and the oxygen atom incorporated into the product isderived from the peroxide (23).Therefore, Nordblom et al. (23) suggested that the interaction of the peroxide with cytochrome P-450 results in the formation of an activated oxygen-enzyme intermediate which is essentially the sameas the intermediateformed when NADPH and molecular oxygen are involved in the reaction and that cytochrome P-450 functions via a peroxidase-type mechanism in which the peroxide is formed at theactive site

9685

9686

Kinetics of Cytochrome P-450-catalyzed Dealkylations

of the hemeprotein by a 2-electron reduction of the molecular oxygen bound to the hemeprotein (23). The cytochrome P-450-catalyzed reactions supported by NADPH and O2 involve three substrates (NADPH, 0 2 , and the substrate tobe hydroxylated) and two proteins (NADPHcytochrome P-450 reductase and cytochrome P-450). The ability to substitute organic peroxides for NADPH (and thus eliminate the requirement for the NADPH-cytochrome P-450 reductase) and O2 provides a valuable model system for investigating the nature of the activated oxygen intermediate involved in cytochrome P-450-catalyzed oxidations. Since the organic peroxide-supported demethylation involves a single enzyme, cytochrome P-450, and 2 substrate molecules, the oxidant and p-nitroanisole, the system is amenable to steady state kinetic analysis using the techniques established for two substrate enzymes. In the present paper, we have characterized the peroxide-supported o-demethylation ofp-nitroanisole catalyzed by rabbit liver microsomal cytochrome P-450. The results of the steady state kinetic analysis of the peroxidesupported demethylation reaction indicate that the reaction proceeds by a sequential mechanism involving the formation of a ternary complex. EXPERIMENTAL PROCEDURES

Materials-The p-nitroanisole was obtained from the Eastman Chemical Co. (Rochester, NY) and recrystallized twice from petroleum ether (b.p. 30-60°C) prior to use. The organic peroxides were obtained from the following sources: t-butyl hydroperoxide from ICN Pharmaceuticals (Cleveland, OH); diacetyl peroxide, t-butyl peracetate, and peracetic acid from Pfaltz and Bauer, Inc. (Stamford, CT); t-butyl peroxybenzoate and cumene hydroperoxide from Matheson, Coleman and Bell (Norwood, OH), and t-amyl hydroperoxide from Polysciences Inc. (Warrington, PA). All were of the highest purity commercially available and the concentrations were determined iodometrically as described by Silbert and Swern (26). NADPH and bovine serum albumin were obtained from Sigma Chemical Co. (St. Louis, MO), t-butanol from Fisher Scientific Co. (Pittsburgh, PA), and potassium cyanide from J. T.Baker Chemical Co. (Phillipsburg, NJ). All other reagents were analytical reagent grade from commercial suppliers. Isolation of Microsomes-Liver microsomes were isolated from New Zealand white male rabbits weighing between 1.5 and 2.0kg using the methods of Matsubara etal. (27) and Powis and Boobis (28) as previously described (29). The washed pellet was resuspended in 100 mM potassium phosphate buffer, pH 7.70, containing 250 mM sucrose, at a protein concentration of 15 to 30 mg/ml. The suspension was divided into 1- and 2-ml aliquots, purged with Nz, frozen in dry ice-acetone, and stored at -90°C. Microsomes were used within 4 weeks of isolation. Under these storage conditions, there wasno significant loss of cytochrome P-450 or demethylase activity. The protein concentrations were determined by the method of Lowry et al. (30) using bovine serum albumin as the standard. The concentration of P-450 was determined from the reduced-CO difference spectrum of the microsomal preparations as described by Omura and Sato (31) using an extinction coefficient of 91 mM" cm" for the difference between the maximum absorbance in the 450 nmregion and the absorbance at 490 nm. The specific contents of the preparations used in these studieswere approximately 1.2 nmol of P-450/mg of protein. p-Nitroanisole Demethylase Assay-The steady state kinetic measurements were made using an Aminco DW-2 spectrophotometer operated in the dual wavelength mode with a sample wavelength of 403 nm, the absorbance maximum of thep-nitrophenolate anion, and a reference wavelength of490 nm. The rates obtained with this wavelength pair were identical with those obtained when the instrument wasused in the split-beam mode at 403 nm with identical reaction mixtures in the sample and reference cell and with the oxidant being added to the sample cell. The temperature of the cuvette was maintained at 30°C with a Haake circulating water bath and the actual temperature of the reaction mixture was periodically checked using a YSI model 42 SC Tele-Thermometer (YellowSprings Instruments, Inc.,Yellow Springs, OH). When peracetic acid or diacetyl peroxide were used to support the reaction, the DW-2 Anaerobic Cell Accessory was used to obtain the initial rates of the reactions. For these studies, the oxidant was placed on the plunger

assembly and the reaction was initiated without opening the sample compartment. As a result, initial rate data could be obtained within 2 s after the initiation of the reaction. Although the demethylation reactions supported by peracetic acid or diacetyl peroxide were often linear with time for less than 10 s, good initial rates could be obtained from the linear portion of the spectrophotometer tracings. The kinetic experiments were conducted in a final volume of 1.0 ml, except when the anaerobic cell accessory was used, in which case the final volume was 3.0 ml. In both cases, the reaction mixture consisted of 100 mM potassium phosphate buffer, pH 7.70, 1 to 2 mg of microsomal protein/ml, and various concentrations of p-nitroanisole, oxidant, and inhibitor. The p-nitroansiole and some of the oxidants were added in 5 pl of acetonitrile. The final concentration of acetonitrile in the reaction mixture was always limited to 1%or less. At this concentration, the acetonitrile had no effect on the rate of demethylation. The reaction mixture minus NADPH or the oxidant was placed in the spectrophotometer and when thermal equilibrium was attained (approximately 3 m i n ) the reaction was initiated by the addition of NADPH or the oxidant. The absorbance change at 403 nm relative to 490 nm was monitored with time. For the studies on the inhibition by cyanide, a stock solution of KCN wastitrated to pH 7.70 with monobasic potassium phosphate and kept cold (0-4°C) in a glass-stoppered volumetric flask. Just prior to the initiation of the reaction, an aliquot of the stock cyanide solution was added to the cuvette to give the final desired concentration and the cuvette was covered with parafilm to prevent the loss of cyanide. All experimental points were determined in duplicate, with the duplicates agreeing within 5% and the data points shown are the means of the determinations. Each experiment was performed a minimum of three times and the results presented here are representative data from a single experiment. The specific activity, expressed as nanomoles of p-nitrophenol formed per min per mgof protein, was calculated using an extinction coefficient of14.9cm" a t 403 nM for p-nitrophenol in 100 ~ B potassium I phosphate, pH 7.70. RESULTS

Peroxide-supported Demethylation of p-NitroanisolePreliminary experiments demonstrated that several different peroxides supported p-nitroanisole demethylation by rabbit liver microsomes in the absence of NADPH and 02.Since pnitrophenol, one of the products of the reaction, could be measured directly and continuously using a spectrophotometric assay, this reaction was used to investigate the kinetic mechanism of the peroxide-supported reactions. The increase in the absorbance of the reaction mixture at 403 nm relative to 490 nm as a function of time for several different concentrations of microsomal protein with t-butyl hydroperoxide as the oxidant is shown inFig. 1. Although the reactions exhibited good linearity with timeinitially, they were not linear indefinitely (note thecurvature observed after 30 s at a protein concentration of 1.72 mg). The length of time during which the reaction exhibited linearity with time was dependent on the level of the microsomal protein, the identity of the oxidant, the initial concentration of the oxidant, and theinitial concentration of p-nitroanisole. The reaction was linear for 3 to 5 min with t-butyl peroxybenzoate or t-butyl peroxyacetate, whereas it was linear for only 5 to 10 s with peracetic acid or diacetyl peroxide. The length of time during which the reaction was linear with time decreased with increasing concentrations of protein or oxidant. For studies reported here, the conditions were such that initial rates were obtained with good reproducibility (+5%). The rateof p-nitrophenol formation with t-butyl hydroperoxide present was linear with microsomal protein up toa conc mtration of approximately 2 mg of microsomal protein/ml (Fig. 1, inset). For the oxidants used in this study, the reaction was linear with protein up to approximately 2.0 mg/ml and an absolute requirement for microsomal protein was observed for all oxidants. The decreases in the ratesof the peroxide-supported demethylations observed with higher peroxide concentrations or longer incubationtimes were accompanied by decreases in the total cytochrome P-450 content of the microsomes, as measured by


9687

rr

ooeI -

0

8

>-

10

OoE

W

2

a

004

05

% 0.02

01

CI - 4 CI

0 2

(p-NITROANISOLE), mM 5

IO

15

20

25

30

35

40

TIME (SECONDS)

FIG. 1. Demethylationofp-nitroanisolesupported by t-butyl hydroperoxide. The reaction mixtures contained 100 mM potassium phosphate, pH 7.7, 3.0 m~ p-nitroanisole, 3.0 m~ t-butyl hydroper-

150

f

I

P-1 p w z

oxide, and the indicated levels of microsomal protein from control rabbit liver in a final volume of 1.0 ml. The reaction was initiated by addition of hydroperoxide at zero time and the increase in absorbance at 403 nm relative to 490 nm was monitored with time in an Aminco DW-2 spectrophotometer. The inset is a plot of the initial rate versus the level of microsomal protein.

I

100

W

k!3 u n

f 50 a the reduced CO-difference spectrum (31), as well as decreases J 0 in the total heme content, as measured by the formation of the pyridine hemochrome (32). pH Optima for the Peroxide-supported Demethylation0.2 04 06 08 The pH profiles for p-nitroanisole demethylation supported (p-NITROANISOLE), mM by t-butyl hydroperoxide, peracetic acid, or t-butyl peroxyFIG .2. Effect of the pnitroanisoleconcentration on the rate benzoate, exhibited broad optima centered from pH 7.4 to 7.7, while diacetyl peroxide had a broad optimum centered at pH of demethylation. A, the reaction mixture contained 100 mM potas6.8. These pH optima correspond well with the optimum of sium phosphate, pH 7.7,1 mM NADPH, 1.0 mg of microsomal protein, and various concentrations of p-nitroanisole in a final volume of 1.0 7.5 obtained by Nordblom et al. (23) for the cumene hydro- ml. The inset is the double reciprocal plot of the initial rate data. B , peroxide-supported demethylation of N-methylaniline by the the reaction mixture contained 100 mM potassium phosphate, pH 7.7, reconstituted system from rabbit liver. 0.50 mM peracetic acid, 1.5mgof microsomal protein, and various Effects of Substrate Concentrations on the Initial Velocity concentrations of p-nitroanisole in a final volume of 3.0 ml. The rate Patterns-The results obtained when the concentration of p- data were obtained using the anaerobic cell accessory for the Aminco nitroanisole was varied over a 20-fold range in the presence of DW-2 as described under “Experimental Procedures.” The inset is the double reciprocal plot of the data. 1 m~ NADPH or 0.50 mM peracetic acid (approximately 3 times the apparent K,,, for peracetic acid) are shown in Fig. 2, A and B, respectively. The microsomalenzymes exhibited TABLE I typical Michaelis-Menten saturation kinetics with respect to Apparent kinetic constants for p-nitroanisole with various p-nitroanisole when the demethylation reaction was supoxidants ported by either NADPH and O2or peracetic acid. The double The reaction mixtures contained 100 mM potassium phosphate, pH reciprocal plots of the initial rates uersus the p-nitroanisole concentrations (Fig. 2, A and B, insets) exhibited good linear- 7.7, 1.0 mg of microsomal protein, the indicated concentrations of the various oxidants and various concentrations of p-nitroanisole. The ity in both cases. For the NADPH-supported reaction, the kinetic parameters were determined from double reciprocal plots of apparent K,,,for p-nitroanisole was 0.02 mM and the apparent the initial rate data. V,,, was 1.60 nmol of p-nitrophenol formed/min/mg of proSubstrate Oxidant K,,, concentein. When peracetic acid was substituted for NADPH, the V”,,, tration apparent K,,, forp-nitroanisole was 0.17 m~ and the apparent nmol/mrn/ V,, was 167 nmol of p-nitrophenol formed/min/mg of promM mM mg prolern tein. 1.6 NADPH and Oz 1.00 0.02 Similar saturation kinetics with respect to p-nitroanisole 137.0 Peracetic acid“ 0.50 0.17 were observedwith a number of different oxidants. The kinetic Cumene hydroperoxide 12.0 1.50 0.97 constants obtained from these studies are shown inTable I. It 0.50 0.17 10.6 Diacetyl peroxide” t-Butyl hydroperoxide 0.13 2.10 3.7 should be pointed out that they are apparentkinetic constants 5.0 0.11 2.3 t-Butyl perbenzoic acid rather than true constants since they were not obtained under 1.0 0.12 0.7 t-Butyl peracetic acid conditions of saturation with respect to thesecond substrate. The anaerobic cell accessory of the Aminco DW-2 was used to Therefore, they are only estimates of the values which would be obtained at a saturating concentration of the second sub- initiate the reaction as described under “Experimental Procedures.” The reaction mixture contained 100 m M potassium phosphate, pH strate. For all of the peroxide-supported reactions, the appar- 7.7, 1.5 mg of microsomal protein, 0.50 mM peracetic acid, or 0.50 mM ent K,,, values for p-nitroanisole were greater than that ob- diacetyl peroxide, and varying concentrations of p-nitroanisole in a served with NADPH (Table I). The larger values of K,,, forp- final volume of 3.0 ml.

z

I.


9688

nitroanisole in the peroxide-supported reactions may be due to a competition between the peroxide and p-nitroanisole for the same binding site on the enzyme, as suggested by Nordblom et al. (23) for cumene hydroperoxide-supported benzphetamine demethylation or may reflect changes in the rate constants that constitute the K,,, in the two systems. The identity of the oxidant also had a very dramatic effect on the apparent V,,, for the demethylation (Table I). Theperacetic acid-supported reaction had a V, approximately 85 times that of the NADPH-supported reaction, while the V,, with cumege hydroperoxide or diacetyl peroxide was only 6 to 7 times greater. The Michaelis constants for several oxidants were similarly determined by measuring the initial rates of demethylation in the presence of varying concentrations of the oxidant at a fixed concentration of p-nitroanisole. The kinetic constants obtained from the slopes and intercepts of the double reciprocal plots of the initial rate data are given in Table 11. The K , values for the different oxidants vary from 0.07 mM for peracetic acid to 1.54 mM for t-butyl hydroperoxide. The apparent maximal velocities showa dependence on the identity of the oxidant similar to thatin Table I. The difference in the values of the maximal velocities between the two tables reflects the fact that these are apparent kinetic constants determined at a single concentration of the second substrate. It should be noted that theapparent K , values determined for p-nitroanisole with various oxidants (Table I) and for the oxidants (Table 11) were calculated using the concentration of the substrate added to the assay mixture. Since the assay mixture is at least a two-phase system (buffer solution and microsomal suspension) and since the substrates arenonpolar and therefore would tend to partition into the microsomal lipid, the values for the substrate concentrations used to calculate the K , values maydiffersignificantly from the actual substrate concentrations in the vicinity of the enzyme active site. Parry et al. (33) have suggested that the effect of ligand partitioning into membranes on the K , values for cytochrome P-450 and other membrane-bound enzymes is a function of the partition coefficient forthe substrate, K,, and the membrane concentration. For an enzyme active site which interacts with substrates in the aqueous phase, the G a P P w l l i approach K , true as the enzyme concentration approaches zero. If the substrate approaches the enzyme from the lipid will approach K m / K pas the enzyme concenphase, the KmaPP tration approaches zero. Thus, the true K , for a given substrate can be calculated by calculating KmaPP over a range of membrane concentrations and extrapolating to zeromembrane concentration. In addition, the value of K, for the substrate must be known. The K , values reported here were not corrected in this manner and therefore are apparent Km TABLEI1 Apparent kinetic constants for various oxidants in the dealkylation ofp-nitroanisole The conditions were as in Table I except that the concentration of p-nitroanisole was 1.0 mM and the oxidant concentrations were varied overa20-fold concentration range. The kinetic parameterswere determined from double reciprocal plots of the initial rate data. Oxidant

K, mM

0.07 Peracetic acid 0.39 Diacetyl peroxide 0.50 Cumene hydroperoxide" 1.54 t-Butyl hydroperoxide 0.29 t-Butyl perbenzoic acid 0.53 t-Butyl peracetic acid The concentration of p-nitroanisole was 3 mM.

V,X

nmol/min/mg protein

71.0 21.4 14.9 4.35 3.15 1.02

values. However, the K,,, values for p-nitroanisole in Table I can be compared since they were all determined at the same concentration of microsomal membrane and the K, for pnitroanisole is the same in all cases. Initial Velocity Experiments-The kinetic mechanism of the t-butyl hydroperoxide-supported demethylation reaction was investigated using initial velocity studies in which the concentrations of the two substrates were systematically varied and the results analyzed usingthe steady state assumption. When the concentration ofp-nitroanisole was varied at several different fiied concentrations of t-butyl hydroperoxide,double reciprocal plots of velocity against the concentration of p nitroanisole resulted in a family of straight lines which converged to acommon intersection point on the negative side of the ordinate and above the abscissa (Fig. 3). The double reciprocal plots of velocity against concentration of t-butyl hydroperoxide at several fixed concentrations ofp-nitroanisole also gave a family of straight lines having a common point of intersection on the negative side of the ordinate and above the abscissa (Fig. 4). These results are indicative of a sequential mechanism (34, 35), either ordered or random, involving the formation of a ternary complex between enzyme, p-nitroanisole, and t-butyl hydroperoxide,followed by one or more reactions and the subsequent release of the two products. They conclusively rule out a ping-pong mechanism in which one substrate reacts with the enzyme to form a substituted enzyme intermediate with the concomitant release of one product followed bythe binding of the second substrate to the substituted enzyme intermediate followed by a second reaction and subsequent release of the second product. The secondary plots obtained when the slopes and ordinate intercepts from the primary plots were replotted against the reciprocals of the corresponding concentrations of t-butyl hydroperoxide (inset, Fig. 3) or p-nitroanisole (inset, Fig. 4), were linear in all cases. The true kinetic constants for both substrates were calculated from the slopes and intercepts of these secondary plots as described by Segel (36). The K , for p-nitroanisole calculated in this manner was 0.071 mM while the K,,, for the t-butyl hydroperoxide was 1.0 mM. These values are slightly lower than the apparent K , values given in Tables I and 11,

(I-BUTYL HYDROPEROXIDE);'

-40

-30

-10 10 Ip-NITROANISOLEI"

-20

20

30

40

mM"

FIG. 3. Initial velocity patterns for pnitroanisole demethylation with p-nitroanisole as the variable substrate. The 1-ml reaction mixtures contained 100 mM potassium phosphate, pH 7.7, 1.35 mg of microsomal protein, and the concentrations of p-nitroanisole indicated. The concentrations of t-butyl hydroperoxide used 2.10 mM ( A ) , and 4.20 were: 0.26 mM ( O ) ,0.53 mM ( A ) , 1.05 m~ (n), m~ ( 0 ) The . insets are replots of the intercepts and slopes uersus the reciprocal of the corresponding t-butyl hydroperoxide concentration.

Kinetics of Cytochrome P-450-catalyzed Dealkylations I

I

'

I

V-1

I

L.

-4

-3

-2

-1

1

[t-BUTYL HYDROPEROXIDEl-l

2

3

4

9689

uersus [SI plot defies the same parameter, K,,,/Vmax).The secondary plot of the ordinate intercepts of the double reciprocal plots uersus the concentration of potassium cyanide (inset, Fig. 6) was linear, demonstrating thatthis is pure uncompetitive inhibition. The K , for cyanide, calculated from the slope of the secondary plot, was approximately 5.5 mM, which was in good agreement with the K , value obtained when t-butyl hydroperoxide was the variable substrate. These results clearly demonstrated that the reaction proceeds via an Ordered Bi Bimechanism in whichp-nitroanisole binds to the enzyme prior to t-butyl hydroperoxide. Inhibition by t-ButyZAlcohoZ-The effect of t-butyl alcohol, one of the products of the t-butyl hydroperoxide-supported demethylation ofp-nitroanisole, on the kinetics of the reaction was investigated. For these studies, the concentration of one

I

rnM"

Frc. 4. Initial velocity patterns for p-nitroanisole demethylation with t-butyl hydroperoxide as the variable substrate. Assay conditions wereas described in Fig.3. The concentrationsof p nitroanisole were: 0.025 m~ ( O ) ,0.05 m~ ( A ) , 0.10 m~ (O), 0.20 mM (A),and 0.40 mM ( 0 ) .Theinsetsarereplots of the slopesand intercepts from thereciprocalplots versus thereciprocals of the corresponding p-nitroanisole concentrations. respectively, where a single concentration of the second substrate was used. Inhibition by Cyanide-The formation of a ternarycomplex between enzyme, p-nitroanisole, and t-butyl hydroperoxide can occur by the random binding of the two substrates to the enzyme or can involve the preferential binding of one substrate to the enzyme prior to the addition of the second substrate. Kinetic experiments using dead-end inhibitors can differentiate between these two alternatives and can be used to determine the order of substrate binding forordered mechanisms (35, 37). For these studies, the concentration of one substrate was varied systematically in the presence of a fmed subsaturating level of the second substrate and various fixed concentrations of potassium cyanide, a dead-end inhibitor of the demethylation reaction. The double reciprocal plots obtained when the velocity of the demethylation reaction was determined as afunction of the t-butyl hydroperoxide concentration in the presence of 0, 2.5,5.0, 7.5, and 10 mM potassium cyanide and a f i e d concentration ofp-nitroanisole are shown in Fig. 5. The common intersection point on the ordinate indicated that cyanide was a competitive inhibitor with respect to t-butyl hydroperoxide. The replot of the slopes uersus the inhibitor concentration (inset, Fig. 5) gave a straight line demonstrating the pure competitive nature of the inhibition. The Kifor cyanide, calculated from the slope of the replot, was approximately 5.0 mM. The competitive nature of the inhibition indicates that cyanide, which is thought to bind as an axial ligand to the heme iron in cytochrome P-450 (38), and t-butyl hydroperoxide compete for the same binding site on the enzyme. When the concentration of t-butyl hydroperoxide was held constant at a subsaturating level (2.5 mM) and the concentration of p-nitroanisole varied in the presence of various Tied concentrations of potassium cyanide (0, 2.5, 5.0, 7.5, and 10 mM), the reciprocal plots of the initial rate data (Fig. 6) gave a series of parallel lines indicative of uncompetitive inhibition (35-37). The uncompetitive nature of the inhibition was conf m e d by plotting the initial rate da$a as [ S ] / u uersus [SI (Fig. 7). The intersection of the lines at a common point on the ordinateindicates that theslopes of the double reciprocal plots are identical (the slope term in a double reciprocal plot while the ordinate intercept in the [ S ] / u is equal to K,/V,,,

J It-BUTYL HYDROPEROXIDE I-'

rnM-l

FIG. 5. Double reciprocal plots of potassium cyanide inhibition data with t-butyl hydroperoxide as the variable substrate. The reactions contained 100 mr.1 potassium phosphate, pH 7.7, 0.96 mg of microsomal protein, 0.08 mM p-nitroanisole, various concentrations of potassium cyanide, and the indicated concentrations of t-butylhydroperoxidein a final volume of 1.0 ml. The concentrations of KCN were 0 mM ( O ) ,2.5 mM (A),5.0 mM ( O ) ,7.5 n m (A), and 10.0 mM ( 0 ) The . inset is a replot of the slopes uersus the corresponding concentrationsof KCN.

[p-NITROANISOLE 1'.

rnM-l

FIG. 6. Double reciprocal plots of potassium cyanide inhibition data with p-nitroanisole as the variable substrate. The reactionswerethe same as describedinFig. 5 except the t-butyl hydroperoxide was 2.5 m~ in all reactions and the concentrationofpnitroanisolewasvariedasindicated.Theconcentrations of KCN were: 0 mM ( O ) ,2.5 mM (A),5.0 mM (U),7.5 mM ( A ) , and 10.0 mM (@). The inset is a replot of the intercepts versus the corresponding concentration of KCN.


9690

of the substrates was varied in the region of its Michaelis constant in the presence of a fixed subsaturating concentration of the second substrate and different fixed concentrations of t-butyl alcohol and the initial rate of the reaction was determined. The highest level of t-butyl alcohol used for these studies (80 mM) had no effect on the cytochrome P-450 content of the microsomes as determined by measurement of the reduced-CO difference spectrum. The double reciprocal plots of the initial rate dataobtained when the t-butyl hydroperoxide concentration was varied in the presence of various levels of t-butyl alcohol (0, 20, 40, 50, and 80 mM) and a subsaturating level ofp-nitroanisole (0.06 mM, K,,, = 0.07 mM) are shown inFig. 8. The intersection of the lines on the abscissa indicates that t-butyl alcohol was a noncompetitive inhibitor. The secondary replots of the slopes and intercepts versus the corresponding concentrations of t-butyl alcohol - 4 0 - 3 0 - 2 0 -10 10 2 0 3 0 40 50 were linear (Fig. 8, zrrset), indicative of pure noncompetitive [p-NITROANISOLEI” mM” inhibition. When the concentration of t-butyl hydroperoxide was held constant a t a subsaturating level (1.8 mM, K , = 1.0 FIG. 9. Double reciprocal plots of t-butyl alcohol inhibition mM) and the p-nitroanisole concentration varied in the pres- data with p-nitroanisole as the variable substrate. Reactions ence of different fixed concentrations of t-butyl alcohol (0,20, were the same as described in Fig. 8 except the concentration of p 40, 60, and 80 mM), the reciprocal plots (Fig. 9) indicated that nitroanisole was varied as indicated and the concentration of t-butyl hydroperoxide was 1.80 mM in all reactions. The concentrations oftbutyl alcohol were: 0 mM (O), 20 mM ( A ) , 40 mM (n), 60 mM ( A ) , and 80 mM (0).The insets are replots of the slopes and intercepts uersus the corresponding concentrations of t-butyl alcohol.

t-butyl alcohol was also a noncompetitive inhibitor with respect to p-nitroanisole. The pure noncompetitive nature of this inhibition was demonstrated by the fact that thesecondary plots of the slopes and interceptsuersus the corresponding concentrations of t-butyl alcohol were linear (Fig. 9, inset). DISCUSSION

0.1 [p-N~troanisolel rnM

0

FIG. 7. Hanes plots for potassium cyanide inhibition withpnitroanisole as the variable substrate. Data fromFig. 6 were plotted in the form [p-nitroanisole]/velocityuersus [p-nitroanisole]. The concentrations of KCN were: 0 mM (O), 2.5 mM ( A ) , 5.0 mM ( O ) ,7.5 mM ( A ) , and 10 mM ( 0 ) .

- 0.5

0.5

[t-BUTYLHYDROPEROXIDEI-’

1.o

1.5

mM”

FIG. 8. Double reciprocal plots of t-butyl alcohol inhibition data with t-butyl hydroperoxide as thevariable substrate.The reaction mixtures contained 100 mM potassium phosphate, pH 7.7, 0.96 mg of microsomal protein, 0.05 mM p-nitroanisole, and various concentrations of t-butyl hydroperoxide andt-butyl alcohol in a final volume of 1.0 ml. The concentrations of t-butyl alcohol were: 0 mM (O), 20 mM ( A ) , 40 mM (a), 60 mM ( A ) , and 80 mM ( 0 ) The . insets are replots of the slopes and intercepts uersus the corresponding concentrations of t-butyl alcohol.

The ability of liver microsomal cytochrome P-450 to use a variety of oxidants in place of molecular oxygen and NADPH to support hydroxylation and dealkylation reactions has led to thehypothesis that thehydroperoxide-supported reactions andthe NADPH and On-supported reactions proceed by similar mechanisms involving the formation of the same activated oxygen intermediate (19-21, 23, 39). Furthermore, several laboratories have suggested that the peroxide-supported reactionsproceed via a “peroxidase-type’’mechanism, implicating a Compound I-type intermediate in the hydroxylation reactions (19, 20, 22-25). Since the organic peroxidesupported reactions involve a single enzyme, cytochrome P450, and 2 substrate molecules, the oxidant and the hydroxylatable substrate, this system is amenable to a steady state kinetic analysis using the techniques established for two substrate enzymes (34-37). It should be noted that rabbit liver microsomes contain multiple forms of cytochrome P-450. As a result,the microsome-catalyzed reactions may represent the composite of the activities of all forms of cytochrome P-450 present in the membrane which are able to use the organic oxidants to support thedemethylation of p-nitroanisole. Our observations that the double reciprocal plots are linear over the concentration ranges tested for the oxidants listed in Tables I and 11, indicate that the observed values of K , and V,,, represent the activity of a single enzyme which is able to use the organic oxidant to support the reaction or a number of enzymes which do not have significantly different K,,, values for the oxidants or for p-nitroanisole. The intersecting Lineweaver-Burk patterns obtained when the concentration of one substrate was varied systematically in the presence of different fixed concentrations of the second substrate (Figs. 3 and 4) indicate that the overall mechanism is sequential, either ordered or random, involving the forma-

9691

Kinetics of Cytochrome P-450-catalyzed Dealkylations tion of a ternary complex of enzyme, hydroperoxide, and pnitroanisole. These results are not consistent with a ping-pong mechanism involving the formation of a substituted enzyme intermediate andwhich is characterized by a seriesof parallel lines in the Lineweaver-Burk plots. Since cyanide is a deadend inhibitor for the reaction, it could be used to differentiate between an Ordered or Random mechanism ( 3 4 , 35, 37). Cyanide was a pure competitive inhibitor with respect to tbutyl hydroperoxide (Fig. 5) and a pure uncompetitive inhibitor with respect to p-nitroanisole (Fig. 6). These results are consistent withan Ordered mechanism in whichp-nitroanisole binds to the enzyme prior to thebinding of t-butyl hydroperoxide to form a kinetically significant ternary complex. Although cyanide binds to theferric form of cytochrome P-450 (38), the inhibition patterns observed with cyanide indicate that the cyanide binding is competitive with the second substrate tobind, t-butyl hydroperoxide, but occurs after thefirst substrate, p-nitroanisole, is bound (34, 35). This apparent anomaly may be due to a significant difference in the second order rate constants for the binding of p-nitroanisole and cyanide to theferric enzyme. Stopped flow spectrophotometric studies by Blake and co-workers' on the binding of cyanide and p-nitroanisole to purified rabbit liver LM2 as well as to microsomes indicate that the second order rate constant for the binding of p-nitroanisole to cytochrome P-450 is at least 2 orders of magnitude greater than the rateconstant for the binding of cyanide. A difference in the rate constants for binding of that magnitude would result in the apparentbinding order indicated in the dead-end inhibition experiments. Ohnishi and co-workers (40) havedemonstrated thatthe superoxide-supported dioxygenation of tryptophan by indoleamine 2, 3-dioxygenase proceeds by an Ordered Bi Bi mechanism. Although cyanide binds to the ferric form of the dioxygenase (40, 41), the inhibition pattern with respect to tryptophan is uncompetitive, suggesting that cyanide binds prior to theoxidant, but after the fist substrate,tryptophan, is bound. Since cyanide is thought to bind to cytochrome P450 as an axial ligand (38), the competitive nature of the inhibition with respect to t-butylhydroperoxide suggests that the hydroperoxide binds directly to the heme iron, an important featureif the peroxide interaction with cytochrome P-450 yields a Compound I-type intermediate as hasbeen proposed (22,23,39). Thereplots of the cyanide inhibition data (insets, Figs. 5 and 6) give an apparent K, for cyanide of about 5 mM which agrees fairly well with the spectral binding constant for cyanide of2.0 mM reported by Schenkman et al. (38). In addition, for an Ordered mechanism, the dissociation constant for the first substrate tobind to theenzyme can be calculated from the secondary plots of the initial rate data (insets, Figs. 3 and 4) (36). The dissociation constant for p-nitroanisole calculated in this manner is 0.25mM. This value is in good agreement with dissociation a constant of 0.21mM determined by spectral binding experiments with p-nitroanisole,' which gives a typical type I spectrum with a maximum at 390 nm and a trough at 423 nm. The agreement of these two independent methods for determining the dissociation constant for p-nitroanisole supports theOrdered mechanism indicated by the dead-end inhibition experiments. The inhibition experimentswith t-butyl alcohol, one of the products of the reaction, demonstrated that t-butyl alcohol was a pure noncompetitive inhibitor with respect to both substrates. Since cytochrome P-450-catalyzed reactions are essentially irreversible, the inhibition by &butyl alcohol can not be due to true product inhibition in which the reverse

' R. C. Blake, 11, D. R. Koop, and M. J. Coon, personal communication. D. R. Koop and P. F. Hollenberg, unpublished observations.

reaction competes with the forward reaction so that themeasured rate of product formation is decreased. However, the tbutyl alcohol could bind to the enzyme at or near the active site to form a catalytically inactive enzyme -&butyl alcohol complex. The addition of aliphatic alcohols such as t-butyl alcohol to liver microsomal preparations results in the formation of a reverse type I difference spectrum (42) which has been attributed to the formation of a modified ferrihemochrome resulting from the direct coordination of the hydroxyl group to the heme iron (42). The coordination of the alcohol to the heme iron would result in t-butyl alcohol acting as a dead-end inhibitor. The noncompetitive nature of the inhibition indicates that thet-butyl alcohol is acting as a nonexclusive inhibitor, the inhibition not being overcome by saturating concentrations of either substrate. The inhibition may result from the formation of an inactive enzyme .p-nitroanisole. tbutyl hydroperoxide. t-butyl alcohol complex, effectively reducing the level of enzyme available for catalysis while not affecting the binding of either substrate. Thepossibility that the alcohol is exerting a nonspecific effect on the microsomal membrane which reduces the enzyme available for catalysis can notbe excluded even though t-butyl alcohol at concentrations up to 80 m~ had no effect on the concentration of cytochrome P-450 as measured by the reduced-CO difference spectrum. The results of the initial velocity and dead-end inhibition studies of the t-butyl hydroperoxide-supported demethylation ofp-nitroanisole reportedhere areconsistent with the scheme shown in Fig. 10.These steady state results are also consistent with the stopped flow studies of Blake and Coon (43-45) on the interaction of a variety of hydroperoxides with highly purified rabbit liver cytochrome P-450LMz. Their results indicate that the hydroperoxide substrate binds in the dead time of the instrument followed by the apparently reversible interaction of the hydroperoxide with the enzyme. substrate complex forming two spectrally distinguishable intermediates (43, 44). The fist of these intermediates is thought to represent the catalytically important complex (45) and may correspond to the ternary complex suggested by the steady state analysis. The formation of a ternary complex as an intermediate in the reaction sequence is in contrast to the ping-pong mechanisms which have been demonstrated for peroxidases such as horseradish peroxidase (46,47) and chloroperoxidase (48). The catalytic mechanism for these enzymes involves the initial formation of a substituted enzyme intermediate, Compound I, which then interactswith the second substrate followed by release of the oxidized product. In contrast, the oxidation of ferrocytochrome c by cytochrome c peroxidase isolated from yeast (49) or Pseudomonas aeruginosa (50), appears to involve the compulsory addition of the substratesto the enzyme to form a ternary complexfollowed by a rapid oxidationreduction reaction resulting in the formation of Compound I prior to oxidation of the second substrate. Thus, while the results presented in this report indicate that the peroxidet-BUTANOL pNA-OH

t

tBHP

pTA

I E

+

-

E-pNA E-pNA-tBHP E-pNA-OH-tBUTANOL

t

E-1-BUTANOL

4 pNA-OH 1-BUTANOL

t

FIG. 10. Schematic diagram of the t-butyl hydroperoxidesupported o-demethylation ofp-nitroanisole catalyzed by liver microsomal cytochrome P-450.Abbreviations used are: pNA, pnitroanisole; tBHP, t-butyl hydroper0xide;pNA-OH, the 0-methyl01 of p-nitroanisole; and E , liver microsomal cytochrome P-450.


9692

supporteddemethylation of p-nitroanisole proceeds by a mechanism distinct from that of horseradish peroxidase or chloroperoxidase ( i e . initial formation of Compound I) the formation of such an intermediate after the initial formation of the ternary complex cannot be excluded. The extension of these mechanistic ideas to the NADPH-supported reaction would predict that thereduction of the ferrous cytochrome P450. oxygen substrate complex by a 2nd electron would result in the formation of an intermediate in which O2 was reduced to the level of peroxide ( i e . the ternary complex described above). In the absence of substrate the peroxide could then be released, as has been reported by Werringloer (51), while in the presence of substrate, the peroxide bond wouldbe cleaved releasing water and leaving a transient intermediate which could then react with the substrate. It remains to be determined whether such a transient intermediate is formed in the peroxide-supported reactions catalyzed by cytochrome P-450, and its relationship to the reactions supported by NADPH.

18. 19. 20. 21.

e

Acknowledgments-We thank Dr. Minor J. Coon and Dr. Robert C. Blake, 11, for their helpful discussions and for providing preprints of several manuscripts which are currently in press. We thank Miss Hisaye Kawaguchi for her assistance in the preparation of this manuscript. REFERENCES 1. Gillette, J. R., Davis, D. C., and Sasame, H.A. (1972) Annu. Reu. Pharmacol. 12, 57-84 2. Mannering, G. J. (1971) in Fundamentals of Drug Metabolism and Disposition (LaDu, B. N., Mandel, H. G., and Way, E. L., eds) pp. 206-252, Williams and Wilkins, Baltimore 3. Orrenius, S., and Ernster, L. (1974) in Molecular Mechanisms of Oxygen Activation (Hayaishi, O., ed) pp. 215-244, Academic Press, New York 4. Lu, A. Y. H., and Coon, M. J. (1968) J. Biol. Chem. 243, 13311332 5. Strobel, H. W., Lu, A. Y. H., Heidema, J., and Coon, M. J. (1970) J. Biol. Chem. 245,4851-4854 6. Van der Hoeven, T. A,, and Coon, M. J . (1974) J. Biol. Chem. 249,6302-6310 7. Ryan, D., Lu, A. Y. H., West, S., and Levin, W. (1975) J. Biol. Chem. 250, 2157-2163 8. Vermilion, J . L., and Coon, M. J. (1974) Biochem. Biophys. Res. Commun. 60, 1315-1322 9. Dignam, J. D., and Strobel, H. W. (1975) Biochem. Biophys. Res. Commun. 63,845-852 10. Yasukochi, Y., and Masters, B. S. S. (1976) J. Biol. Chem. 251, 5337-5344 11. Estabrook, R. W., Hildebrandt, A. G., Baron, J., Netter, K. J., and Leibman, K. (1971) Biochem. Biophys. Res. Commun. 42, 132-139 12. Gunsalus, I. C., Tyson, C. A., Tsai, R., and Lipscomb, J . D. (1971) Chem.-Biol. Interact.4,15-78 13. Ishimura, Y., Ullrich, V., and Peterson, J. A. (1971) Biochem. Biophys. Res. Commun. 42, 140-146 14. Strobel, H. W., and Coon, M. J . (1971) J. Biol. Chem. 246, 78267829 15. Chen, C., and Lin, C. C.(1968) Biochim. Biophys. Acta 170,366374 16. Chen, C., and Lin, C. C. (1969) Biochim. Biophys. Acta 184,634640 17. Van Lier, J . E., Kan, G., Langlois, R., and Smith, L. L. (1972) in

22. 23. 24. 25. 26. 27. 28.

Biological Hydroxylation Mechanisms(Boyd, G. S., and Smellie, R. M. S., eds) pp. 21-43, Academic Press, New York Chen, C., and Tu, M.-H. (1976) Biochem. J. 160,805-808 Rahimtula, A. D., and O’Brien, P. J. (1975) Biochem. Biophys. Res. Commun. 62,268-275 Lichtenberger, F., Nastainczyk, W., and Ullrich, V. (1976) Biochem. Biophys. Res. Commun. 70,939-946 Kadlubar, F. F., Morton, K. C.,and Ziegler, D.M. (1973) Biochem. Biophys. Res. Commun. 54, 1255-1261 Rahimtula, A. D., and OBrien, P. J. (1974) Biochem. Biophys. Res. Commun. 60,440-447 Nordblom, G. D., White, R. E., and Coon, M. J . (1976) Arch. Biochem. Biophys. 175, 524-533 Akhrem, A. A., Usanov, S. A., Dvornikov, S. S., and Metelitsa, D. I. (1977) React. Kinet. Catal. Lett. 6, 187-194 Hrycay, E.G., Gustafsson, J-.A,, Ingelman-Sundberg, M., and Ernster, L. (1976) Eur. J. Biochem. 61,43-52 Silbert, L. S., and Swern, D. (1958) Anal. Chem. 30,385-387 Matsubara, T., Prough, R. A., Burke, M. D., and Estabrook, R. W. (1974) Cancer Res. 34,2196-2203 Powis,G., and Boobis, A. R. (1975) Biochem. Pharmacol. 24,

1771-1776 29. Novak, R. F.,Koop, D. R., and Hollenberg, P. F. (1980) Mol. Pharrnacol. 17, 128-136 30. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193,265-275 31. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239,2370-2378 32. Fuhrhop, J-J., and Smith, K. M. (1975) Laboratory Methods in pp. 48-51, Elsevier Porphyrin and Metalloporphyrin Research, Scientific Publishing, New York 33. Parry, G., Palmer, D. N., and Williams, D. J. (1976) FEBS Lett. 67, 123-129 34. Cleland, W. W. (1963) Biochim. Biophys. Acta 67, 104-137 35. Cleland, W. W. (1977) Adu. Enzymol. 45, 273-387 36. Segel, I. H. (1975) Enzyme Kinetics: Behavior and Analysis of Rapid Equilibrium and Steady-State EnzymeSystems, pp. 560-591, John Wiley & Sons, New York 37. Cleland, W. W. (1963) Biochim. Biophys. Acta 67, 173-187 38. Schenkman, J. B., Remmer, H., and Estabrook, R. W. (1967) Mol. Pharmacol. 3, 113-123 39. Rahimtula, A. D., OBrien, P. J., Seifried, H. E., and Jerina, D. M.(1978) Eur. J. Biochem. 89,133-141 40. Ohnishi, T.,Hirata, F., and Hayaishi, 0.(1977) J. Biol. Chem. 252,4643-4647 41. Hayaishi, O., Hirata, F., Fujiwara, M., Ohnishi, T., and Nukiwa, T. (1975) FEBS Proc. Meet. 40, 131-144 42. Yoshida, Y., and Kumaoka, H. (1975) J. Biochem. (Tokyo) 78, 455-468 43. Blake, R. C., 11, and Coon, M. J . (1979) Fed. Proc. 38, 319 44. Blake, R. C., 11, and Coon, M. J. (1980) J. Biol. Chem. 255,41004111 45. Blake, R. C., 11, and Coon,M. J. (1980) in Microsomes, Drug Oxidations, and Chemical Carcinogenesis (Coon, M. J., Conney, A. H., Estabrook, R. W., Gelboin, H. V., Gillette, J . R., and OBrien, P. J., eds) pp. 303-306, Academic Press, New York 46. Hosoya, T.(1960) J. Biochem. (Tokyo) 47,794-803 47. Santimone, M. (1975) Biochimie 57,91-96 48. Hager, L. P., Thomas, J. A., and Morris, D. R. (1970) in Biochemistry of the PhagocyticProcess (Schultz, J., ed) pp. 67-87, North-Holland Publishing Co., Amsterdam 49. Yonetani, T., and Ray, G. S. (1966) J. Biol. Chem. 241,700-706 50. Ronnberg, M., and Ellfolk, N. (1975) A d a Chem. Scand. 3 29, 719-727 51. Werringloer, J. (1977) in Microsomes and Drug Oxidations (Ullrich, V., Roots, I., Hildebrandt, A., Estabrook, R. W., and Conney, A. H., eds) pp. 261-268, Pergamon Press, New York