- Email: [email protected]
Comparative kinetic studies of the dealkylation reaction and steroid hydroxylations in various oxygen and carbon monoxide:oxygen atmospheres
ARCHIVES OF BIOCHEMISTRY AND BIOPHYSICS Vol. 224, No. 2, July 15, pp. 614-624, 1983
Comparative Kinetic Studies of the Dealkylation Reaction and Steroid Hydroxylations in Various Oxygen and Carbon Monoxide:Oxygen Atmospheres’ STUART
J. HAMILL,
DAVID Y. COOPER,2 HEINZ OTTO ROSENTHAL3
SCHLEYER,
AND The Harr&m
Department of Surgical Research and The Department of Pharmacology, University of Pennsylvania, Philadelphia, Pennsylvania 19104 Received November
School of Medicine,
16, 1982, and in revised form March 28, 1983
The time-course kinetics of the cytochrome P-450-catalyzed dealkylations of the exogenous compounds benzphetamine, ethylmorphine, codeine, and 7-ethoxycoumarin were compared to the hydroxylation of the endogenous compound testosterone. Using liver microsomes from phenobarbital-induced rats, the time course of the demethylations of ethylmorphine, codeine, and especially benzphetamine was characterized by a fast initial phase of enzymatic activity and then a steady decline in the rate throughout the remainder of the reaction. In contrast, under the same experimental conditions, both the dealkylation of ‘7-ethoxycoumarin and the hydroxylation of testosterone showed no initial fast phase of activity and a constant rate of product formation for most of the remainder of the time course. The difference also held for the carbon monoxide inhibition studies in which the degree of inhibition of the demethylation reactions by a variety of CO:Oz mixtures was time dependent, in contrast to the constant, timeindependent degree of CO inhibition of the other two reactions. The kinetics of the demethylation reactions could not be explained by enzyme destruction, back reaction, or product adduct formation and were further confirmed by measurements of the rate of Oz utilization and NADPH oxidation. The complexity of the demethylation reaction should be taken into consideration in any detailed studies of the monooxygenation reaction system.
Enzymatic N-demethylation reactions catalyzed by hepatic microsomes, usually considered to require cytochrome P-450 for oxygen activation, are among the most commonly encountered biological oxidations of organic molecules. N-dealkylation of this class of compounds is an indispensible step in the metabolic processes of
eliminating lipid-soluble drugs, carcinogens, and other xenobiotics that may enter the living organism (1). Furthermore, one of these N-demethylations, the demethylation of benzphetamine, is frequently selected as a substrate to serve as standard for assaying activities of purified and microsomal P-450-containing systems (2). Despite the frequent occurrence and importance of dealkylation reactions in the metabolism of toxic and therapeutic agents, there have been few detailed studies on the basic kinetics of these microsomal systems. Most published studies estimate enzyme activity at a single time in-
i Aided by Grant AM-04434-20 from the NIH. ’ To whom correspondence should be addressed: 319 Medical Education Building, School of Medicine of the University of Pennsylvania, 36th and Hamilton Walk, Philadelphia, Pa. 19104. * Deceased July 14, 1980.
0003-9861/83 $3.00 Copyright All rights
0 1983 by Academic Press, Inc. of reproduction in any form reserved.
614
KINETICS
OF DEALKYLATION
AND
terval of 5,10, or 15 min, an approach which allows little kinetic evaluation of the demethylase activity. Earlier work of this laboratory on the cytochrome P-450-dependent oxidative demethylation of a number of drugs, including benzphetamine, found kinetic anomalies in terms of CO inhibition and light reversal of CO inhibition as compared with the hydroxylation of endogenous substrates such as testosterone (3). These studies were further complicated by our inability to obtain optimal results of CO inhibition and light reversal using the purified system. As a consequence, in these investigations we used the microsomal preparation, which has the necessary proportions of enzyme and structural components allowing reproducible ‘kinetic measurements. These experiments produced a more consistent set of experimental results and served as a basis for further experimental comparison, especially in our case, for the measurement of light reversal of CO inhibition. In order to compare these kinetic data in detail, we studied the time course of the monooxygenation of a variety of exogenous substrates (benzphetamine, ethylmorphine, codeine, and 7-ethoxycoumarin) in conjunction with the endogenous substrate testosterone by means of product formation, selective inhibitors (e.g., CO), and O2 consumption. These results outline the differences in the kinetics of the metabolism of the endogenous and exogenous substrates and underscore the importance of this type of detailed microsomal study in order to explain many of the variations seen in studies of different monooxygenation systems. MATERIALS
AND
METHODS
Materials The chemicals used in these studies were obtained from the following sources: benzphetamine hydrochloride was a kind gift of Dr. Paul O’Connell of the
’ Abbreviations used: G-6-P, glucose 6-phosphate; G-6-PD, glucose-6-phosphate dehydrogenase; PB, phenobarbital.
HYDROXYLATION
REACTIONS
615
Upjohn Company; codeine phosphate was from Mallinkrodt; ethylmorphine (Dionine) was from Merck Chemical Company; ‘I-ethoxycoumarin, testosterone, NADP, NADPH, Trizma base, and glucose 6-phosphate (G-6-P)’ were from Sigma Chemical Company; methanol and sodium azide were from Fisher Scientific Company; glucose-6-phosphate dehydrogenase (G-6-PD) was from Boehringer-Mannheim; and gas mixtures were from Union Carbide Corporation.
Methods Induction. of c&chrome P-450. Male Sprague-Dawley rats weighing 200-250 g were induced by pretreatment with phenobarbital (PB) (80 mg/kg in 0.9% NaCI) administered intraperitoneally once a day for 4 consecutive days. Noninduced rats (controls) were given the same volume of the vehicle (0.9% NaCI) without PB for 4 consecutive days as previously described (4). The rats were sacrificed Microsomd preparations. by decapitation, the livers perfused to remove hemoglobin, and the hepatic microsomes prepared by previously described methods (5). The microsomal pellet was suspended in 0.25 M sucrose to yield a concentration of 50 nmol of P-450lml sucrose and frozen in 2-ml aliquots in liquid nitrogen for use when needed. Cytochrwme P-450 akternination The amount of P450 in the microsomal preparations was determined with an Aminco DW-2 spectrophotometer by difference spectroscopy as previously reported (6), using the extinction coefficient of 91 mM-' cm-’ of Omura and Sato (7). Microsomal protein was determined by the method of Lowry (8). Assay of cytochrome P-450 activity. All reactions were run at 25 f O.l”C in a modified Erlenmeyer flask (25 ml) with a tightly fitted stopper pierced with two needles to allow continuous gas flow through the reaction chamber over the surface of the incubation mixture. The reaction mixture had a final volume of 10 ml, with final concentrations of the following ingredients: MgCIZ (5 mM), EDTA (1 mM), NaN3 (1 mM) with benzphetamine (1 mM), ethylmorphine (5 mM), ‘Iethoxycoumarin (0.3 mM), or testosterone (0.33 mM) in 100 mM Tris-phosphate buffer, pH 7.8. The substrates 7-ethoxycoumarin and testosterone were dissolved in ethanol, 0.5 ml of the solution was placed in the empty vessel, and the ethanol was removed by evaporation with a stream of nitrogen. Appropriate amounts of microsomes in 0.25 M sucrose were added to 8.5 to 9.0 ml of the 100 mM Tris-phosphate buffer containing the reaction components listed above, and the volume was adjusted to 9.8 ml with 0.25 M sucrose. The incubation vessel containing this mixture was placed in a constant-temperature shaking (120 rpm) water bath and allowed to equilibrate for 5 min at
616
HAMILL
25°C with the gas mixture to be used in the experiment. Final P-450 concentrations ranged from 0.2 to 2.0 FM, corresponding to a protein concentration of 0.1 to 1.0 mg/ml. The reaction was started, except where indicated, by addition of NADPH or a NADPHgenerating system preequilibrated with the desired gas mixture at 25°C. The generating system contained, as final concentrations: NADP (1 mM), G-6-P (5 mM), and G-6-PD (1 IU/ml). In certain experiments the reaction was started by addition of 0.2 ml of a NADPH solution which yielded a final reduced pyridine nucleotide concentration of 1 mM. The codeine, ethylmorphine, and benzphetamine demetbylation reactions were stopped by adding 1.0 ml of the reaction mixture to a deproteinating reagent consisting of either 1.5 ml 15% trichloroacetic acid or 1 ml of saturated BaOH followed by 0.5 ml of 0.35 rd ZnSO1. The suspensions were centrifuged at 10,OOOgfor 10 min to sediment the denatured protein, after which 2 ml of the deproteinated supernatant was removed for analysis of the product, formaldehyde, formed during the reaction. Deethylation of Il-ethoxycoumarin and hydroxylation of testosterone reactions were stopped by adding 5 ml of cold dichloromethane as previously described (3, 9). Product analysis. Formaldehyde was measured with Nash reagent (10). ‘I-Hydroxycoumarin (I-methylumbelliferone) formation was continuously monitored in an Eppendorf fluorometer, with an exciting wavelength of 363 nm and an emission wavelength of 456 nm as described by Ullrich and Weber (11). The ‘I-hydroxycoumarin formation was also measured in the kinetic gassing experiments after extraction of the product by a modification of the method of Greenlee and Poland (9). The formation of S@-,?a-, and 16cu-testosterone was measured using [4%]testosterone (52 pCi/mmol) obtained from New England Nuclear Corporation. The dichloromethane extract (10 ml) was evaporated to dryness and resuspended in 100 ~1 of a methanol:chloroform mixture (1:l by volume). Ten microliters was spotted per lane on a heat-activated (llO’C, 30 min) thin-layer chromatographic plate (prescored, Analtech silica gel GF) and developed using an acetone:isopropyl ether (1:4 by volume) solvent system. The metabolites and substrates were identified by observation under fluorescent light of cold standards spotted on adjacent lanes. The amounts of substrate and product were determined by scanning the thinlayer plate with a Bioscan Imaging Radiation Detector System (Bioscan BID System 100/800) which works as a proportional counter (12). CO inhibition studies. The CO inhibition of P-450dependent monooxygenases was studied for CO:Oz ratios that ranged from 0.0 to 8.0 with oxygen concen-
ET AL.
trations of 4, 10,20, and 50% by volume, with Nz as the inert carrier gas. The gases used were either commercially mixed and obtained from the source given above or mixed with Wustoff gas-mixing pumps. The appropriate gas mixture was flowed over the top of the shaking reaction mixture by means of an inlet needle piercing the stopper tightly fitted into the Erlenmeyer flask. After preincubation of the system for a sufficient time (5 min) to ensure adequate equilibration of the reaction system with the desired gas, the reaction was initiated by addition of a standard solution containing NADPH or a NADPH-generating system.
Other Methods Hydrogen per& (H@$ formation Hydrogen peroxide formation was measured with the catalase (1 mg/ml)-methanol (100 PM) trapping system described originally by Chance and Theorell(13). Azide was omitted from the reaction mixture in these studies. 0, cmumption rates. The concentration of Oz in the reaction system and the rate of Oz utilization were measured polarographically using an oxygen electrode especially designed by Dr. Rudolph H. Eisenhardt that allows Oz consumption to be studied in any desired controlled gas atmosphere. This cell has a sample volume of 3.0 ml and is fitted with two holes in the top which are covered with a carefully machined hood. The sample and reagents are added through one of the ports, while the other is open to allow equilibration with the gas used for the experiment. Following equilibration, the reaction was initiated with NADPH or the NADPH-generating system. The rate of Oz consumption was monitored continuously. P-450 destruction The destruction of P-450 as a function of the time course of the reaction was measured with and without substrate. The reaction system was the same as described above. A l.O-ml sample was taken at the appropriate reaction time and added to 2.0 ml of Tris-phosphate buffer that had been prebubbled with 100% CO for 5 min and that contained sufficient sodium dithionite to ensure complete reduction of P-450. The amount of P-450 destruction was determined with the Aminco DW-2 spectrophotometer using difference spectroscopy. The reference cell contained the P-450 suspension in Tris buffer with CO and dithionite which had been incubated with no NADPH-generating system present.
RESULTS
The time course of formaldehyde mation by liver microsomes from
forPB-
KINETICS
OF DEALKYLATION
AND HYDROXYLATION
Time
REACTIONS
61’7
(mitt)
FIG. 1. Formation of formaldehyde in the oxidative dealkylation of benzphetamine by rat liver microsomes. PB-induced microsomes (5 FM P-450) and benzphetamine (1.0 maa) are preincubated for 5 min in a standard Tris-phosphate buffer (106 mM, pH 7.8, at 25’C with EDTA (1.0 m?d),NaNa (1.0 mM), and MgClz (5.0 mM)) in a 10% O&O% Na gas atmosphere. The reaction is started by the addition of an NADPH-generating system, NADP+ (1.0 mM), G-6-P (5.0 mbf), and G-6-PD (1.0 IU/ ml), which has been preincubated for 3 min in the gas atmosphere used. The error bars give the experimental range of the measurements from three experiments.
treated rats with benzphetamine as substrate is shown in Fig. 1. The time course, for analytical purposes, is divided into three segments. The first, lasting approximately 1 min (i.e., up to the 1-min reading or slightly longer as determined by linear regression analysis), consists of a rapid initial phase of activity. The second, which lasts up to the 15-min sample, consists of a nearly linear rate of formaldehyde formation. The final portion, the interval between the 15- and 30-min samples, is characterized by a steady decline in the rates of product formation. The progress curve shows that the kinetics of benzphetamine demethylation deviate significantly from what is usually considered ideal behavior for a single enzyme-catalyzed reaction. The rate of formaldehyde formation is not only nonlinear with time but the line representing the rate of formation does not extrapolate to zero time. Because of the importance of the zero blank value, this control was analyzed in a number of ways. First, in each reaction a zero time blank in addition to the standard blank was determined by the addition of the NADPH-generating system after the deproteinating reagent. There was no detectable difference between the two
values. Second, in one set of experiments, samples were withdrawn as fast as possible (approximately 3 s) after activation of the reaction mixture with the NADPHgenerating system and produced no detectable amounts of product formation. Finally, the reaction was run under a CO:Oz gas mixture of 8:l which inhibited the first segment of the reaction up to 88% or practically to the baseline. We carried out further studies of the early rates by preincubating the microsomes with the NADPH-generating system (NADP+, G-6-P, and G-&PD) and by initiating the reaction by addition of the substrate benzphetamine. This procedure with a 1-min preincubation increased the product formation for the first minute by 10% over control values, while a 5-min preincubation yielded an even greater increase of 30%. pH Dependence of benzphetamine demeth&ztion The oxidative demethylation of benzphetamine was determined at a number of pHs (range 7.0 to 8.2 in Trisphosphate buffer) to assess the effect of hydrogen ion concentration on the reaction kinetics. Figure 2 illustrates that changes in pH have a much more significant effect on the rapid first phase, further distin-
618 25 5: Q Li
HAMILL
ET AL.
I
I
I
I
I ,
zo-
0) : c \ ; g0 t s? 0 E c
15-
lo-
5
sI, ______-
--
_-
-*-
-----*----
--
*-
--_ -4
I
I
01 7.0
I
I
I
J 82
7.0
74 PH
FIG. 2. The dependence of the specific rates of formaldehyde formation in the oxidative dealkylation of benzphetamine on the pH of the system. PB-induced microsomes (0.33 PM P-450) and benzphetamine (1.0 mM) are preincubated in the standard Tris-phosphate buffer reaction mixture (adjusted to a pH range from 7.0 to 8.2) in 10% Oz:90% Nz gas atmosphere and the reaction is begun as before with NADPH (1.0 mM NADP+)-generating system (see Fig. 1). Specific rates for the reaction are calculated from the nanomoles of formaldehyde formation in the selected time period per minute per nanomole of P-450 (nmol product min- 1 nmol P-450-i). Segment l(0); segment 2 (- - -); segment 3 (0).
guishing this part of the reaction from the remainder. These results were independent of the microsomal concentration as the re-
I
I
action carried out with higher microsomal concentrations produced essentially the same results.
I
I
I
300 -
3- Ethylmorphine
Time
(min)
FIG. 3. Formation of formaldehyde in the oxidative dealkylation of codeine and ethylmorphine by rat liver microsomes. PB-induced microsomes (1.0 PMP-450) and codeine (5 mM) or ethylmorphine (5 mM) are preincubated in the standard Tris-phosphate buffer reaction mixture in a 10% Oz:90% Nz gas atmosphere. The reaction is started by addition of an NADPH (NADP+ 1.0 mM)-generating system (see Fig. 1). Error analysis as in Fig. 1.
KINETICS
OF DEALKYLATION
AND
Kinetics qf the metabolism of ethylmorphine and codeine. When the time courses of the demethylation of ethylmorphine and codeine were studied, they were found to have similar but not identical kinetics to that observed with benzphetamine (Fig. 3). Both substrates had rapid initial rates of formaldehyde formation, but not as pronounced as that found with benzphetamine. Furt.hermore, there was little or no decrease with either substrate in rate of formaldehyde formation after the first 15 min. This was particularly true with ethylmorphine. The codeine demethylation reaction kinetics fell between those of ethylmorphine and benzphetamine. CO inhibition of benxphetamine and ethylmorphine demethylation. We studied the demethylation of benzphetamine and
TABLE
I
CO INHIBITION OF THE OXIDATIVE DEALKYLATION OF BENZPHETAMINE AND ETHYLMORPHINE Relative rate (01) Reaction time period (min) 0.0 1.0 2.5 5.0 10.0 15.0 20.0
--t
1.0 2.5 5.0 10.0 15.0 20.0 30.0
Benzphetamine 0.52 0.49 0.39 0.26 0.24 0.29 0.32
HYDROXYLATION
619
REACTIONS TABLE
II
CO INHIBITION OF THE DEALKYLATION OF BENZPHETAMINE (BP) AND ETHYLMORPHINE (EM) BY VARIOUS CO:Oz MIXTURES Relative rate (a)
Reaction time period (min)
4% 02 +4% co
4% o2 +8% CO
4% 02 + 32% CO
BP
EM
BP
EM
BP
EM
0.0 1.0 2.5 5.0 10.0 15.0
0.48 0.32 0.29 0.28 0.40 0.40
0.59 0.59 0.61 0.42 0.44 0.46
0.40 0.15 0.11 0.15 0.32 0.32
0.47 0.42 0.43 0.27 0.27 0.28
0.24 0.17 0.03 0.02 0.14 0.16
0.27 0.15 0.10 0.10 0.10 0.17
1.0 2.5 --t 5.0 -t 10.0 - 15.0 - 20.0
Note. Phenobarbital-induced microsomes (0.5 pM P-450) are preincubated in the appropriate gas mixtures for 5 min in the standard Tris-based buffer as described in Table I. In this case the 0a:CO mixture is 4% Oz + 4,8, or 32% CO in Nz as inert carrier gas, and the O2 gas mixture is 4% Oz in Nz. a is defined as the specific rate (nmol product min-’ nmol P-450-i) in the CO:O* mixture divided by the specific rate in 02 alone or (Y = V(C0 + O,)/V(O,) where V stands for the specific rate.
Ethylmorphine 0.56 0.57 0.45 0.39 0.41 0.40 0.43
Note. Phenobarbital-induced microsomes (0.5 &M P-450) are preincubated in the appropriate gas mixtures for 5 min in the standard Tris-based buffer (100 mM, pH 7.8, at 25°C with EDTA (1.0 mM), azide (1.0 mM), and MgC& (5.0 m&i) with either benzphetamine (1.0 mM) or ethylmorphine (5.0 mM)). The reaction is started by the addition of an NADPH-generating system (NADP+ (1.0 mM), G-6-P (5.0 mM), and G-6PD (1.0 III/ml)) preincubated for 3 min. In this case, the Oz + CO mixture is 10% Oa:lO% CO in Nz as inert carrier gas and the Oz gas is 10% O2 in NZ. o is defined as the specific rate (nmol product min-’ nmol P-450-i) in the CO:Oz mixture divided by the specific rate in Oz alone or 01 = V (CO + O,)/V(O,) where V stands for the specific rate.
ethylmorphine over a wide range of CO to Oz ratios. The most routinely used gas mixture of 10% CO2 and 10% CO in nitrogen produced widely varying degrees of inhibition of benzphetamine demethylation during the time course of the reaction (Table I). The inhibition of ethylmorphine demethylation varied less than that of benzphetamine. Both reactions, however, demonstrated the least CO inhibition during the first segment of the reaction which increased to the highest degree in the second segment and then decreased slightly in the third. With different CO to O2ratios (Table II) the extent of inhibition varied but the pattern of CO inhibition remained essentially the same. These results are derived from one set of experiments, but numerous repetitions of this experiment produced essentially the same progression of CO inhibition.
620
HAMILL
Time
ET AL
(min)
FIG. 4. Formation of G@hydroxytestosterone from testosterone by rat liver microsomes. PBinduced microsomes (1.0 PM P-450) and [4-“Cjtestosterone (0.33 mM) are preincubated in the standard Tris-phosphate buffer reaction mixture and the reaction is initiated as described in Fig. 1 with an NADPH-generating system. (0) In 10% O&lO% N2 gas mixture; (0) in 10% Or:101 CO:60% Nz gas mixture. The error bars are the experimental variation for the individual experiment.
Reaction kinetics of the 6@-,7~, and 16ah~droxylations of testosterone. The hydroxylations of testosterone by hepatic microsomes were investigated under identical reaction conditions, so that the hydroxylations of an endogenous steroid substrate could be directly compared with those of
0
the dealkylation of ethylmorphine and benzphetamine. Each of the two major hydroxylation products, 6@ and 16cu-hydroxytestosterone, was formed in a 10% O2 atmosphere at a constant rate that lasted for the first 5 min of the reaction (Figs. 4 and 5). The rates of product formation di-
IO
20 Time
30
(mid
FIG. 5. The formation of 16a-hydroxytestosterone from testosterone by rat liver microsomes. PBinduced microsomes (1.0 FM P-450) and [4-‘“Cjtestosterone (0.33 mM) are preincubated in the standard Tris-phosphate buffer reaction mixture and the reaction is initiated as described in Fig. 1 with a NADPH-generating system. (0) In 10% O&IO% Nx gas mixture; (0) in 10% 0x:10% CO:6095 N2 gas mixture. Error analysis as in Fig. 4.
KINETICS
OF DEALKYLATION
AND
HYDROXYLATION
IO
20 Time
REACTIONS
621
30
(mid
FIG. 6. The formation of 7-hydroxycoumarin from ‘I-ethoxycoumarin by rat liver mierosomes. PB-induced microsomes (0.45 PM P-450) and ‘I-ethoxycoumarin (0.33 mM) are preincubated in the standard Tris-phosphate buffer in the appropriate gas mixture and the reaction is begun as described in Fig. 1 with a NADPH-generating system. (0) In 10% O#O% N2; (0) in 10% 02:10% CO:30% Ne gas mixture. The reaction in all cases was monitored continuously. The error bars give the experimental variation of the measurements.
minished thereafter as a consequence of a high degree of enzyme destruction. Importantly there was no rapid initial burst of activity and the average specific rates over the 30-min period of monitoring were 2.0,0.56, and 0.19 nmol product min-’ nmol P-450p1 for the 6p-, 16a-, and 7cy-hydroxylations of testosterone, respectively. The degree of CO inhibition remained nearly constant (Figs. 4 and 5), +5% through the time course for the 6p and 16a reactions. The kinetics of the ‘7a reaction were more difficult to evaluate as this hydroxylation proceeds at a very low rate and the values were at the limit of the accuracy of the assay procedure. Kinetics of deethylatim of 7-ethoxycmmarin. We studied the rate of deethylation of ‘I-ethoxycoumarin because the formation of its fluorescent oxidation product can be monitored continuously in a fluorometer. In addition, these studies gave us another substrate for comparison with the kinetics of the dealkylations and hydroxylations of this study. This deethylation, unlike the demethylations, had a linear rate of product formation throughout most of the time course of the reaction (Fig. 6). The specific rate of the deethylation was 1.8 nmol prod-
uct min-’ nmol P-450-‘, a value well below these observed with the N-demethylation reactions studied here. The degree of CO inhibition by a 10% CO:lO% Oz mixture (CO:Oz = 1:l) was 43 + 5% and remained essentially constant through the time period the reaction was monitored. Microsd hydrogen peroxide fn-mution. Since peroxide formation by microsomes has been demonstrated by a number of laboratories (14,15) and since peroxide has been implicated in microsomal mixedfunction oxidations, we studied the time course of HzOz formation in the presence and absence of various substrates using the catalase-methanol trapping system of Chance and Theorell(13) to measure H202 formation. We found that H,Oz (50 PM maximum/PM P-450) was formed in a timedependent manner with a number of substrates. However, the kinetics for formation of HzOz in the first 15 min of the reaction are different from those of the corresponding substrate and argue against the direct involvement of H202 in dealkylation reactions (data not shown). Relationship of demethylutim to oxygen concentration. Since oxygen is one of the three substrates involved in monooxygen-
622
HAMILL TABLE
III
O2 CONSUMPTION
Substrate Benzphetamine a b Ethylmorphine ; Codeine a b No substrate a b
Initial rate (CM min-‘)
Total 02 consumed (ef)
Ratio NADPH/ l/2 o*
24.2 14.8
48.3 39.4
0.91 1.12
13.6 12.2
45.2 43.3
0.97 1.01
16.0 8.6
46.2 41.8
0.95 1.06
14.8 7.7
45.0 36.2
0.98 1.21
Note. Phenobarbital-induced microsomes (2.0 pM P-450) are preincubated in air for 5 min in the standard Tris-based buffer as described in Table I. The reaction is started by the addition of NADPH (100 @d final concentration) and followed continuously in a specially designed microelectrode system (see Materials and Methods). A subsequent addition of NADPH (100 CM) was made as soon as there was no further Op consumption from the previous addition. Initially, Op concentration (in air) equals 240 PM and measurements were based on full-scale deflection equaling this value. Initial rates were measured in the first 10 s of the reaction. The ratio is defined as the prd NADPH injected/prd l/2 Oe eonsumed. An ideal P-450 reaction would produce a ratio of 0.5. All ratio values were adjusted for the fact that under these experimental conditions, only 88% of the NADPH can be oxidized.
ating reactions, the kinetics of the demethylation reactions were studied in atmospheres containing various concentrations of Oz. Higher Oz pressures (50% 0,) increased demethylation rates in the early initial segment of the reaction with the three substrates (35% with benzphetamine), although the effect was much less with ethylmorphine (10%) and codeine (15%). With all three substrates, the last two-thirds of the reactions were essentially insensitive to increases in oxygen concentration, except in the case of benzphetamine where there was a slight increase (10%) in rates with 50% Oz in the 2.5- to 15-min segment of the reaction.
Oxygen consumption by hpatic
micro-
somes. Oxygen consumption was determined polarographically with an oxygen electrode. The results of experiments with
ET AL.
three different substrates are summarized in Table III. The initial rates of O2 consumption and the ratio of NADPH added for, each l/2 O2 consumed for two consecutive 100 PM NADPH additions are compared. These results are in general agreement with the kinetics of formaldehyde formation observed with these three substrates (Figs. 1 and 3). The reaction system containing benzphetamine had the highest initial rates of O2 consumption as well as the greatest Oa consumption with the first injection of NADPH. Thereafter, both the rate and the amount of 0, consumed declined significantly with further additions of NADPH (complete data not shown). With ethylmorphine, the rates of O2 consumption were more nearly constant during the NADPH additions (decreasing 37 vs 65% for benzphetamine). The amount of O2 consumption and variation in rates with codeine as substrate were nearly the same as those found with ethylmorphine. None of the substrates studied had ratio of Oa consumed to NADPH oxidation of 0.5, the value expected if there was a stoichiometric relationship between the oxygen consumption and NADPH oxidation. Possible causes of nonlinearity. Three possible causes of nonlinearity suggested by data in the literature were investigated: (i) the formation of an inactive complex of cytochrome P-450 with substrate or product; (ii) selective induction of a form of P450 by PB with different kinetics; and (iii) inactivation of P-450 during the reaction by enzyme destruction. (i) Product adducts formation was ruled out because under these experimental conditions (1.0 mM benzphetamine) there was no spectrophotometric evidence of substrate or product combining with the hemoprotein. Also, with chromatographic separation of the extract from the reaction mixture, we were able to detect only one product and the substrate. (ii) Since the kinetics of benzphetamine demethylation were the same with control microsomes as with those from PB-induced rats, there is no evidence that PB induces a different P-450 system. (iii) The gradual decline in specific dealkylation rates late in the reaction course
KINETICS
OF DEALKYLATION
AND
could be explained in part by P-450 destruction. We did not, however, detect any significant enzyme destruction associated with the initial stages of the reaction.
DISCUSSION
The experiments reported in this paper have demonstrated that demethylations of benzphetamine, ethylmorphine, and codeine by hepatic microsomes are catalyzed by a much more complex enzymatic process than the hydroxylation of testosterone and the deethylation of ‘I-ethoxycoumarin, and most likely represent a family of enzymatic systems that contribute to the overall rates. To a large extent, the complexity of the demethylation reaction stems from the existence, especially in the case of benzphetamine, of the initial rapid segment of activity. Using a variety of zero blank methods we consistently found an increased rate of product formation for the first 1 to 2.5 min of the reaction. The distinction of this first segment from the remainder of the reaction was further supported by its increased pH sensitivity and the ability to selectively affect this part of the reaction by various Oz pressures. Since microsomes are very heterogeneous mixtures, the ability of CO to inhibit the reaction is essential in determining the involvement of P-450, a hemeprotein, in these reactions. Other possible monooxygenation reaction systems such as flavin amine mixed-function peroxidase (19, 20) and catalase, which can under certain experimental conditions catalyze the demethylation of various substrates, are not CO inhibitable. Recent investigations (21, 22) have demonstrated that prostaglandin synthetase also metabolizes a number of N-alkyl compounds. However, this enzyme system is completely dependent upon the addition of arachidonic acid, is not functional with NADPH, and is not competitively inhibited by CO. In addition, it only dealkylates those compounds where the nitrogen atom is directly attached to an aromatic moiety. Our results demonstrate
HYDROXYLATION
REACTIONS
623
that not only are these reaction systems easily CO inhibitable, thus indicating that a CO-combining system is the predominant pathway, but that the degree of CO inhibition is dependent on the CO:Oz ratio. This is consistent with the concept that as an oxidase, the reaction rates will be dependent upon the competition between CO and O2 for the reaction site. Importantly, however, the degree of CO inhibition of the demethylation reactions (the end product of this CO:Oz competition) varied widely as a function of the reaction time. These results imply the involvement of at least one more CO-inhibitable reaction. In these systems then, the observed rates of CO inhibition would be a function of the degree of contribution and inhibition of each enzymatic process. In contrast to this is the time-independent degree of CO inhibition of the monooxygenation of testosterone and ‘I-ethoxycoumarin. The complexity of these systems is further emphasized by the nonstoichiometry of Oz consumption. One to one and onehalf nanomoles of Oz are consumed for each nanomole of NADPH oxidized. This is up to three times greater than the 0.5 nmol of O2 that should be consumed per nmol of NADPH oxidized, if the relationship maintained the strict stoichiometry of a monooxygenase-catalyzed reaction. It is apparent from these experiments that the demethylation reaction is much more complex than the hydroxylation or deethylation. These reactions appear to be composed of a number of catalytic systems which can contribute to the overall rate of product formation. The evidence for this is the variation in turnover rates (especially the fast early segment), the timedependent degree of CO inhibition, and the lack of stoichiometry between NADPH and O2 consumption. As a consequence, any further study of the microsomal monooxygenation reaction as well as the use of the purified system containing only P-450 must keep in mind the intricacies of the demethylation reaction. In the accompanying paper, the light-reversal properties of CO inhibition of these substrates are described and compared as a means of bet-
624
HAMILL
ter understanding these reactions by the determination of the P-450 involvement. REFERENCES 1. COOPER, D. Y., SCHLEYER, H., LEVIN, S., EISENHARDT, R. H., NOVACK, B. G., AND ROSENTHAL, 0. (1979) Drug M&u& Rev. 10.153-185. 2. PETERS, M. A., AND FOUTS, J. R. (1970) Biochem Ph4mwmml 19,533~544. 3. HAMILL, S., COOPER, D. Y., SCHLEYER, H., AND ROSENTHAL, 0. (1980) in Biechemistry, Biophysics and Regulation of Cytochrome P-450 (Gustzfasson, J. A., et al, eds.), pp. 337-341, Elsevier/North-Holland, Amsterdam/New York. 4. COOPER,D. Y., SCHLEYER, H., THOMAS, J. H., VARS, H. M., AND ROSENTHAL, 0. (1975) in Cytochromes P-456 and 4. Structure, Function and Interaction. (Cooper, D. Y., et al, eds.), pp. 81102, Plenum, New York. 5. DALLNER, G. (1974) in Methods in Enzymology (Fleisher, S., and Packer, L., eds.), Vol. 31, pp. 191201, Academic Press, New York. 6. COOPER, D. Y., SCHLEYER, H., ROSENTHAL, O., LEVlN, W., Lu, A. H. Y., KUNTZMAN, R., AND CONNEY, A. H. (1977) Eur. J. Bimhem 74,6875. 7. OMURA, T., AND SATO, R. (1962) J. Bid Chem 237, 1375-1381. 8. LOWRY, 0. H., ROSENBROUGH, N. J., FARR, A. L.,
ET AL.
9. 10. 11. 12. 13. 14. 15. 16. 17.
18. 19. 20. 21.
22.
AND RANDALL, R. J. (1951) J. Biol Chem 193, 265-275. GREENLEE, W., AND POLAND, A. (1978) J. Pharmix01 Exp. Thzr. 205, 595-605. NASH, T. (1953). Biochtm J. 55,416-421. ULLRICH, V., AND WEBER, P. (1972) HqnpeSeyler’s 2. Physid Chem 353,1171-1177. BAIRD, W. M., DIAMOND, L., AND SHULMAN, S. (1979) Ad Biochrem 103.64-69. CHANCE, B., AND THEORELL, H. (1951) Acta C&m Scana! 5,1127-1133. NORDBLOOM, G. D., AND COON, M. J. (1976) Arch Biochem. Biophps. 175,343-349. NORDBLOOM, G. D., AND COON, M. J. (1977) Arch. Btihem. Biophys. 130,524-533. FRANKLIN, M. R. (1971) Xaobiotica 1,581~588. WERRINGLOER, J., AND ESTABROOK, R. W. (1979) in The Induction of Drug Metabolism (Estabrook, R. W., and Lindenlaub, E., eds.), pp. 269307, F. K. Schattauer Verlag, Stuttgart. WERRMGLOER, J., AND ESTABROOK, R. W. (1973) Life Sci 13, 1319-1330. MACHINIST, J. M., ORME-JOHNSON, W. H., AND ZIEGLER, D. M. (1966) Biochemistry 5.2939-2943. UEHELEKE, H. (1973) Drug Metub. Dispos 1,299313. SIVARAJAH, K., LASKER, J. M., ELING, T. E., AND ABOU-DONIA, M. B. (1982) Mel Pharmacd 21, 133-141. LASKER, J. M., SIVARAJAH, K., MASON, R. P., KALYANARAMAN, B., ABOU-DONIA, M. B., AND ELING, T. E. (1981) J. Bid Chem 256.7764-7767.
Comparative Kinetic Studies of the Dealkylation Reaction and Steroid Hydroxylations in Various Oxygen and Carbon Monoxide:Oxygen Atmospheres’ STUART
J. HAMILL,
DAVID Y. COOPER,2 HEINZ OTTO ROSENTHAL3
SCHLEYER,
AND The Harr&m
Department of Surgical Research and The Department of Pharmacology, University of Pennsylvania, Philadelphia, Pennsylvania 19104 Received November
School of Medicine,
16, 1982, and in revised form March 28, 1983
The time-course kinetics of the cytochrome P-450-catalyzed dealkylations of the exogenous compounds benzphetamine, ethylmorphine, codeine, and 7-ethoxycoumarin were compared to the hydroxylation of the endogenous compound testosterone. Using liver microsomes from phenobarbital-induced rats, the time course of the demethylations of ethylmorphine, codeine, and especially benzphetamine was characterized by a fast initial phase of enzymatic activity and then a steady decline in the rate throughout the remainder of the reaction. In contrast, under the same experimental conditions, both the dealkylation of ‘7-ethoxycoumarin and the hydroxylation of testosterone showed no initial fast phase of activity and a constant rate of product formation for most of the remainder of the time course. The difference also held for the carbon monoxide inhibition studies in which the degree of inhibition of the demethylation reactions by a variety of CO:Oz mixtures was time dependent, in contrast to the constant, timeindependent degree of CO inhibition of the other two reactions. The kinetics of the demethylation reactions could not be explained by enzyme destruction, back reaction, or product adduct formation and were further confirmed by measurements of the rate of Oz utilization and NADPH oxidation. The complexity of the demethylation reaction should be taken into consideration in any detailed studies of the monooxygenation reaction system.
Enzymatic N-demethylation reactions catalyzed by hepatic microsomes, usually considered to require cytochrome P-450 for oxygen activation, are among the most commonly encountered biological oxidations of organic molecules. N-dealkylation of this class of compounds is an indispensible step in the metabolic processes of
eliminating lipid-soluble drugs, carcinogens, and other xenobiotics that may enter the living organism (1). Furthermore, one of these N-demethylations, the demethylation of benzphetamine, is frequently selected as a substrate to serve as standard for assaying activities of purified and microsomal P-450-containing systems (2). Despite the frequent occurrence and importance of dealkylation reactions in the metabolism of toxic and therapeutic agents, there have been few detailed studies on the basic kinetics of these microsomal systems. Most published studies estimate enzyme activity at a single time in-
i Aided by Grant AM-04434-20 from the NIH. ’ To whom correspondence should be addressed: 319 Medical Education Building, School of Medicine of the University of Pennsylvania, 36th and Hamilton Walk, Philadelphia, Pa. 19104. * Deceased July 14, 1980.
0003-9861/83 $3.00 Copyright All rights
0 1983 by Academic Press, Inc. of reproduction in any form reserved.
614
KINETICS
OF DEALKYLATION
AND
terval of 5,10, or 15 min, an approach which allows little kinetic evaluation of the demethylase activity. Earlier work of this laboratory on the cytochrome P-450-dependent oxidative demethylation of a number of drugs, including benzphetamine, found kinetic anomalies in terms of CO inhibition and light reversal of CO inhibition as compared with the hydroxylation of endogenous substrates such as testosterone (3). These studies were further complicated by our inability to obtain optimal results of CO inhibition and light reversal using the purified system. As a consequence, in these investigations we used the microsomal preparation, which has the necessary proportions of enzyme and structural components allowing reproducible ‘kinetic measurements. These experiments produced a more consistent set of experimental results and served as a basis for further experimental comparison, especially in our case, for the measurement of light reversal of CO inhibition. In order to compare these kinetic data in detail, we studied the time course of the monooxygenation of a variety of exogenous substrates (benzphetamine, ethylmorphine, codeine, and 7-ethoxycoumarin) in conjunction with the endogenous substrate testosterone by means of product formation, selective inhibitors (e.g., CO), and O2 consumption. These results outline the differences in the kinetics of the metabolism of the endogenous and exogenous substrates and underscore the importance of this type of detailed microsomal study in order to explain many of the variations seen in studies of different monooxygenation systems. MATERIALS
AND
METHODS
Materials The chemicals used in these studies were obtained from the following sources: benzphetamine hydrochloride was a kind gift of Dr. Paul O’Connell of the
’ Abbreviations used: G-6-P, glucose 6-phosphate; G-6-PD, glucose-6-phosphate dehydrogenase; PB, phenobarbital.
HYDROXYLATION
REACTIONS
615
Upjohn Company; codeine phosphate was from Mallinkrodt; ethylmorphine (Dionine) was from Merck Chemical Company; ‘I-ethoxycoumarin, testosterone, NADP, NADPH, Trizma base, and glucose 6-phosphate (G-6-P)’ were from Sigma Chemical Company; methanol and sodium azide were from Fisher Scientific Company; glucose-6-phosphate dehydrogenase (G-6-PD) was from Boehringer-Mannheim; and gas mixtures were from Union Carbide Corporation.
Methods Induction. of c&chrome P-450. Male Sprague-Dawley rats weighing 200-250 g were induced by pretreatment with phenobarbital (PB) (80 mg/kg in 0.9% NaCI) administered intraperitoneally once a day for 4 consecutive days. Noninduced rats (controls) were given the same volume of the vehicle (0.9% NaCI) without PB for 4 consecutive days as previously described (4). The rats were sacrificed Microsomd preparations. by decapitation, the livers perfused to remove hemoglobin, and the hepatic microsomes prepared by previously described methods (5). The microsomal pellet was suspended in 0.25 M sucrose to yield a concentration of 50 nmol of P-450lml sucrose and frozen in 2-ml aliquots in liquid nitrogen for use when needed. Cytochrwme P-450 akternination The amount of P450 in the microsomal preparations was determined with an Aminco DW-2 spectrophotometer by difference spectroscopy as previously reported (6), using the extinction coefficient of 91 mM-' cm-’ of Omura and Sato (7). Microsomal protein was determined by the method of Lowry (8). Assay of cytochrome P-450 activity. All reactions were run at 25 f O.l”C in a modified Erlenmeyer flask (25 ml) with a tightly fitted stopper pierced with two needles to allow continuous gas flow through the reaction chamber over the surface of the incubation mixture. The reaction mixture had a final volume of 10 ml, with final concentrations of the following ingredients: MgCIZ (5 mM), EDTA (1 mM), NaN3 (1 mM) with benzphetamine (1 mM), ethylmorphine (5 mM), ‘Iethoxycoumarin (0.3 mM), or testosterone (0.33 mM) in 100 mM Tris-phosphate buffer, pH 7.8. The substrates 7-ethoxycoumarin and testosterone were dissolved in ethanol, 0.5 ml of the solution was placed in the empty vessel, and the ethanol was removed by evaporation with a stream of nitrogen. Appropriate amounts of microsomes in 0.25 M sucrose were added to 8.5 to 9.0 ml of the 100 mM Tris-phosphate buffer containing the reaction components listed above, and the volume was adjusted to 9.8 ml with 0.25 M sucrose. The incubation vessel containing this mixture was placed in a constant-temperature shaking (120 rpm) water bath and allowed to equilibrate for 5 min at
616
HAMILL
25°C with the gas mixture to be used in the experiment. Final P-450 concentrations ranged from 0.2 to 2.0 FM, corresponding to a protein concentration of 0.1 to 1.0 mg/ml. The reaction was started, except where indicated, by addition of NADPH or a NADPHgenerating system preequilibrated with the desired gas mixture at 25°C. The generating system contained, as final concentrations: NADP (1 mM), G-6-P (5 mM), and G-6-PD (1 IU/ml). In certain experiments the reaction was started by addition of 0.2 ml of a NADPH solution which yielded a final reduced pyridine nucleotide concentration of 1 mM. The codeine, ethylmorphine, and benzphetamine demetbylation reactions were stopped by adding 1.0 ml of the reaction mixture to a deproteinating reagent consisting of either 1.5 ml 15% trichloroacetic acid or 1 ml of saturated BaOH followed by 0.5 ml of 0.35 rd ZnSO1. The suspensions were centrifuged at 10,OOOgfor 10 min to sediment the denatured protein, after which 2 ml of the deproteinated supernatant was removed for analysis of the product, formaldehyde, formed during the reaction. Deethylation of Il-ethoxycoumarin and hydroxylation of testosterone reactions were stopped by adding 5 ml of cold dichloromethane as previously described (3, 9). Product analysis. Formaldehyde was measured with Nash reagent (10). ‘I-Hydroxycoumarin (I-methylumbelliferone) formation was continuously monitored in an Eppendorf fluorometer, with an exciting wavelength of 363 nm and an emission wavelength of 456 nm as described by Ullrich and Weber (11). The ‘I-hydroxycoumarin formation was also measured in the kinetic gassing experiments after extraction of the product by a modification of the method of Greenlee and Poland (9). The formation of S@-,?a-, and 16cu-testosterone was measured using [4%]testosterone (52 pCi/mmol) obtained from New England Nuclear Corporation. The dichloromethane extract (10 ml) was evaporated to dryness and resuspended in 100 ~1 of a methanol:chloroform mixture (1:l by volume). Ten microliters was spotted per lane on a heat-activated (llO’C, 30 min) thin-layer chromatographic plate (prescored, Analtech silica gel GF) and developed using an acetone:isopropyl ether (1:4 by volume) solvent system. The metabolites and substrates were identified by observation under fluorescent light of cold standards spotted on adjacent lanes. The amounts of substrate and product were determined by scanning the thinlayer plate with a Bioscan Imaging Radiation Detector System (Bioscan BID System 100/800) which works as a proportional counter (12). CO inhibition studies. The CO inhibition of P-450dependent monooxygenases was studied for CO:Oz ratios that ranged from 0.0 to 8.0 with oxygen concen-
ET AL.
trations of 4, 10,20, and 50% by volume, with Nz as the inert carrier gas. The gases used were either commercially mixed and obtained from the source given above or mixed with Wustoff gas-mixing pumps. The appropriate gas mixture was flowed over the top of the shaking reaction mixture by means of an inlet needle piercing the stopper tightly fitted into the Erlenmeyer flask. After preincubation of the system for a sufficient time (5 min) to ensure adequate equilibration of the reaction system with the desired gas, the reaction was initiated by addition of a standard solution containing NADPH or a NADPH-generating system.
Other Methods Hydrogen per& (H@$ formation Hydrogen peroxide formation was measured with the catalase (1 mg/ml)-methanol (100 PM) trapping system described originally by Chance and Theorell(13). Azide was omitted from the reaction mixture in these studies. 0, cmumption rates. The concentration of Oz in the reaction system and the rate of Oz utilization were measured polarographically using an oxygen electrode especially designed by Dr. Rudolph H. Eisenhardt that allows Oz consumption to be studied in any desired controlled gas atmosphere. This cell has a sample volume of 3.0 ml and is fitted with two holes in the top which are covered with a carefully machined hood. The sample and reagents are added through one of the ports, while the other is open to allow equilibration with the gas used for the experiment. Following equilibration, the reaction was initiated with NADPH or the NADPH-generating system. The rate of Oz consumption was monitored continuously. P-450 destruction The destruction of P-450 as a function of the time course of the reaction was measured with and without substrate. The reaction system was the same as described above. A l.O-ml sample was taken at the appropriate reaction time and added to 2.0 ml of Tris-phosphate buffer that had been prebubbled with 100% CO for 5 min and that contained sufficient sodium dithionite to ensure complete reduction of P-450. The amount of P-450 destruction was determined with the Aminco DW-2 spectrophotometer using difference spectroscopy. The reference cell contained the P-450 suspension in Tris buffer with CO and dithionite which had been incubated with no NADPH-generating system present.
RESULTS
The time course of formaldehyde mation by liver microsomes from
forPB-
KINETICS
OF DEALKYLATION
AND HYDROXYLATION
Time
REACTIONS
61’7
(mitt)
FIG. 1. Formation of formaldehyde in the oxidative dealkylation of benzphetamine by rat liver microsomes. PB-induced microsomes (5 FM P-450) and benzphetamine (1.0 maa) are preincubated for 5 min in a standard Tris-phosphate buffer (106 mM, pH 7.8, at 25’C with EDTA (1.0 m?d),NaNa (1.0 mM), and MgClz (5.0 mM)) in a 10% O&O% Na gas atmosphere. The reaction is started by the addition of an NADPH-generating system, NADP+ (1.0 mM), G-6-P (5.0 mbf), and G-6-PD (1.0 IU/ ml), which has been preincubated for 3 min in the gas atmosphere used. The error bars give the experimental range of the measurements from three experiments.
treated rats with benzphetamine as substrate is shown in Fig. 1. The time course, for analytical purposes, is divided into three segments. The first, lasting approximately 1 min (i.e., up to the 1-min reading or slightly longer as determined by linear regression analysis), consists of a rapid initial phase of activity. The second, which lasts up to the 15-min sample, consists of a nearly linear rate of formaldehyde formation. The final portion, the interval between the 15- and 30-min samples, is characterized by a steady decline in the rates of product formation. The progress curve shows that the kinetics of benzphetamine demethylation deviate significantly from what is usually considered ideal behavior for a single enzyme-catalyzed reaction. The rate of formaldehyde formation is not only nonlinear with time but the line representing the rate of formation does not extrapolate to zero time. Because of the importance of the zero blank value, this control was analyzed in a number of ways. First, in each reaction a zero time blank in addition to the standard blank was determined by the addition of the NADPH-generating system after the deproteinating reagent. There was no detectable difference between the two
values. Second, in one set of experiments, samples were withdrawn as fast as possible (approximately 3 s) after activation of the reaction mixture with the NADPHgenerating system and produced no detectable amounts of product formation. Finally, the reaction was run under a CO:Oz gas mixture of 8:l which inhibited the first segment of the reaction up to 88% or practically to the baseline. We carried out further studies of the early rates by preincubating the microsomes with the NADPH-generating system (NADP+, G-6-P, and G-&PD) and by initiating the reaction by addition of the substrate benzphetamine. This procedure with a 1-min preincubation increased the product formation for the first minute by 10% over control values, while a 5-min preincubation yielded an even greater increase of 30%. pH Dependence of benzphetamine demeth&ztion The oxidative demethylation of benzphetamine was determined at a number of pHs (range 7.0 to 8.2 in Trisphosphate buffer) to assess the effect of hydrogen ion concentration on the reaction kinetics. Figure 2 illustrates that changes in pH have a much more significant effect on the rapid first phase, further distin-
618 25 5: Q Li
HAMILL
ET AL.
I
I
I
I
I ,
zo-
0) : c \ ; g0 t s? 0 E c
15-
lo-
5
sI, ______-
--
_-
-*-
-----*----
--
*-
--_ -4
I
I
01 7.0
I
I
I
J 82
7.0
74 PH
FIG. 2. The dependence of the specific rates of formaldehyde formation in the oxidative dealkylation of benzphetamine on the pH of the system. PB-induced microsomes (0.33 PM P-450) and benzphetamine (1.0 mM) are preincubated in the standard Tris-phosphate buffer reaction mixture (adjusted to a pH range from 7.0 to 8.2) in 10% Oz:90% Nz gas atmosphere and the reaction is begun as before with NADPH (1.0 mM NADP+)-generating system (see Fig. 1). Specific rates for the reaction are calculated from the nanomoles of formaldehyde formation in the selected time period per minute per nanomole of P-450 (nmol product min- 1 nmol P-450-i). Segment l(0); segment 2 (- - -); segment 3 (0).
guishing this part of the reaction from the remainder. These results were independent of the microsomal concentration as the re-
I
I
action carried out with higher microsomal concentrations produced essentially the same results.
I
I
I
300 -
3- Ethylmorphine
Time
(min)
FIG. 3. Formation of formaldehyde in the oxidative dealkylation of codeine and ethylmorphine by rat liver microsomes. PB-induced microsomes (1.0 PMP-450) and codeine (5 mM) or ethylmorphine (5 mM) are preincubated in the standard Tris-phosphate buffer reaction mixture in a 10% Oz:90% Nz gas atmosphere. The reaction is started by addition of an NADPH (NADP+ 1.0 mM)-generating system (see Fig. 1). Error analysis as in Fig. 1.
KINETICS
OF DEALKYLATION
AND
Kinetics qf the metabolism of ethylmorphine and codeine. When the time courses of the demethylation of ethylmorphine and codeine were studied, they were found to have similar but not identical kinetics to that observed with benzphetamine (Fig. 3). Both substrates had rapid initial rates of formaldehyde formation, but not as pronounced as that found with benzphetamine. Furt.hermore, there was little or no decrease with either substrate in rate of formaldehyde formation after the first 15 min. This was particularly true with ethylmorphine. The codeine demethylation reaction kinetics fell between those of ethylmorphine and benzphetamine. CO inhibition of benxphetamine and ethylmorphine demethylation. We studied the demethylation of benzphetamine and
TABLE
I
CO INHIBITION OF THE OXIDATIVE DEALKYLATION OF BENZPHETAMINE AND ETHYLMORPHINE Relative rate (01) Reaction time period (min) 0.0 1.0 2.5 5.0 10.0 15.0 20.0
--t
1.0 2.5 5.0 10.0 15.0 20.0 30.0
Benzphetamine 0.52 0.49 0.39 0.26 0.24 0.29 0.32
HYDROXYLATION
619
REACTIONS TABLE
II
CO INHIBITION OF THE DEALKYLATION OF BENZPHETAMINE (BP) AND ETHYLMORPHINE (EM) BY VARIOUS CO:Oz MIXTURES Relative rate (a)
Reaction time period (min)
4% 02 +4% co
4% o2 +8% CO
4% 02 + 32% CO
BP
EM
BP
EM
BP
EM
0.0 1.0 2.5 5.0 10.0 15.0
0.48 0.32 0.29 0.28 0.40 0.40
0.59 0.59 0.61 0.42 0.44 0.46
0.40 0.15 0.11 0.15 0.32 0.32
0.47 0.42 0.43 0.27 0.27 0.28
0.24 0.17 0.03 0.02 0.14 0.16
0.27 0.15 0.10 0.10 0.10 0.17
1.0 2.5 --t 5.0 -t 10.0 - 15.0 - 20.0
Note. Phenobarbital-induced microsomes (0.5 pM P-450) are preincubated in the appropriate gas mixtures for 5 min in the standard Tris-based buffer as described in Table I. In this case the 0a:CO mixture is 4% Oz + 4,8, or 32% CO in Nz as inert carrier gas, and the O2 gas mixture is 4% Oz in Nz. a is defined as the specific rate (nmol product min-’ nmol P-450-i) in the CO:O* mixture divided by the specific rate in 02 alone or (Y = V(C0 + O,)/V(O,) where V stands for the specific rate.
Ethylmorphine 0.56 0.57 0.45 0.39 0.41 0.40 0.43
Note. Phenobarbital-induced microsomes (0.5 &M P-450) are preincubated in the appropriate gas mixtures for 5 min in the standard Tris-based buffer (100 mM, pH 7.8, at 25°C with EDTA (1.0 mM), azide (1.0 mM), and MgC& (5.0 m&i) with either benzphetamine (1.0 mM) or ethylmorphine (5.0 mM)). The reaction is started by the addition of an NADPH-generating system (NADP+ (1.0 mM), G-6-P (5.0 mM), and G-6PD (1.0 III/ml)) preincubated for 3 min. In this case, the Oz + CO mixture is 10% Oa:lO% CO in Nz as inert carrier gas and the Oz gas is 10% O2 in NZ. o is defined as the specific rate (nmol product min-’ nmol P-450-i) in the CO:Oz mixture divided by the specific rate in Oz alone or 01 = V (CO + O,)/V(O,) where V stands for the specific rate.
ethylmorphine over a wide range of CO to Oz ratios. The most routinely used gas mixture of 10% CO2 and 10% CO in nitrogen produced widely varying degrees of inhibition of benzphetamine demethylation during the time course of the reaction (Table I). The inhibition of ethylmorphine demethylation varied less than that of benzphetamine. Both reactions, however, demonstrated the least CO inhibition during the first segment of the reaction which increased to the highest degree in the second segment and then decreased slightly in the third. With different CO to O2ratios (Table II) the extent of inhibition varied but the pattern of CO inhibition remained essentially the same. These results are derived from one set of experiments, but numerous repetitions of this experiment produced essentially the same progression of CO inhibition.
620
HAMILL
Time
ET AL
(min)
FIG. 4. Formation of G@hydroxytestosterone from testosterone by rat liver microsomes. PBinduced microsomes (1.0 PM P-450) and [4-“Cjtestosterone (0.33 mM) are preincubated in the standard Tris-phosphate buffer reaction mixture and the reaction is initiated as described in Fig. 1 with an NADPH-generating system. (0) In 10% O&lO% N2 gas mixture; (0) in 10% Or:101 CO:60% Nz gas mixture. The error bars are the experimental variation for the individual experiment.
Reaction kinetics of the 6@-,7~, and 16ah~droxylations of testosterone. The hydroxylations of testosterone by hepatic microsomes were investigated under identical reaction conditions, so that the hydroxylations of an endogenous steroid substrate could be directly compared with those of
0
the dealkylation of ethylmorphine and benzphetamine. Each of the two major hydroxylation products, 6@ and 16cu-hydroxytestosterone, was formed in a 10% O2 atmosphere at a constant rate that lasted for the first 5 min of the reaction (Figs. 4 and 5). The rates of product formation di-
IO
20 Time
30
(mid
FIG. 5. The formation of 16a-hydroxytestosterone from testosterone by rat liver microsomes. PBinduced microsomes (1.0 FM P-450) and [4-‘“Cjtestosterone (0.33 mM) are preincubated in the standard Tris-phosphate buffer reaction mixture and the reaction is initiated as described in Fig. 1 with a NADPH-generating system. (0) In 10% O&IO% Nx gas mixture; (0) in 10% 0x:10% CO:6095 N2 gas mixture. Error analysis as in Fig. 4.
KINETICS
OF DEALKYLATION
AND
HYDROXYLATION
IO
20 Time
REACTIONS
621
30
(mid
FIG. 6. The formation of 7-hydroxycoumarin from ‘I-ethoxycoumarin by rat liver mierosomes. PB-induced microsomes (0.45 PM P-450) and ‘I-ethoxycoumarin (0.33 mM) are preincubated in the standard Tris-phosphate buffer in the appropriate gas mixture and the reaction is begun as described in Fig. 1 with a NADPH-generating system. (0) In 10% O#O% N2; (0) in 10% 02:10% CO:30% Ne gas mixture. The reaction in all cases was monitored continuously. The error bars give the experimental variation of the measurements.
minished thereafter as a consequence of a high degree of enzyme destruction. Importantly there was no rapid initial burst of activity and the average specific rates over the 30-min period of monitoring were 2.0,0.56, and 0.19 nmol product min-’ nmol P-450p1 for the 6p-, 16a-, and 7cy-hydroxylations of testosterone, respectively. The degree of CO inhibition remained nearly constant (Figs. 4 and 5), +5% through the time course for the 6p and 16a reactions. The kinetics of the ‘7a reaction were more difficult to evaluate as this hydroxylation proceeds at a very low rate and the values were at the limit of the accuracy of the assay procedure. Kinetics of deethylatim of 7-ethoxycmmarin. We studied the rate of deethylation of ‘I-ethoxycoumarin because the formation of its fluorescent oxidation product can be monitored continuously in a fluorometer. In addition, these studies gave us another substrate for comparison with the kinetics of the dealkylations and hydroxylations of this study. This deethylation, unlike the demethylations, had a linear rate of product formation throughout most of the time course of the reaction (Fig. 6). The specific rate of the deethylation was 1.8 nmol prod-
uct min-’ nmol P-450-‘, a value well below these observed with the N-demethylation reactions studied here. The degree of CO inhibition by a 10% CO:lO% Oz mixture (CO:Oz = 1:l) was 43 + 5% and remained essentially constant through the time period the reaction was monitored. Microsd hydrogen peroxide fn-mution. Since peroxide formation by microsomes has been demonstrated by a number of laboratories (14,15) and since peroxide has been implicated in microsomal mixedfunction oxidations, we studied the time course of HzOz formation in the presence and absence of various substrates using the catalase-methanol trapping system of Chance and Theorell(13) to measure H202 formation. We found that H,Oz (50 PM maximum/PM P-450) was formed in a timedependent manner with a number of substrates. However, the kinetics for formation of HzOz in the first 15 min of the reaction are different from those of the corresponding substrate and argue against the direct involvement of H202 in dealkylation reactions (data not shown). Relationship of demethylutim to oxygen concentration. Since oxygen is one of the three substrates involved in monooxygen-
622
HAMILL TABLE
III
O2 CONSUMPTION
Substrate Benzphetamine a b Ethylmorphine ; Codeine a b No substrate a b
Initial rate (CM min-‘)
Total 02 consumed (ef)
Ratio NADPH/ l/2 o*
24.2 14.8
48.3 39.4
0.91 1.12
13.6 12.2
45.2 43.3
0.97 1.01
16.0 8.6
46.2 41.8
0.95 1.06
14.8 7.7
45.0 36.2
0.98 1.21
Note. Phenobarbital-induced microsomes (2.0 pM P-450) are preincubated in air for 5 min in the standard Tris-based buffer as described in Table I. The reaction is started by the addition of NADPH (100 @d final concentration) and followed continuously in a specially designed microelectrode system (see Materials and Methods). A subsequent addition of NADPH (100 CM) was made as soon as there was no further Op consumption from the previous addition. Initially, Op concentration (in air) equals 240 PM and measurements were based on full-scale deflection equaling this value. Initial rates were measured in the first 10 s of the reaction. The ratio is defined as the prd NADPH injected/prd l/2 Oe eonsumed. An ideal P-450 reaction would produce a ratio of 0.5. All ratio values were adjusted for the fact that under these experimental conditions, only 88% of the NADPH can be oxidized.
ating reactions, the kinetics of the demethylation reactions were studied in atmospheres containing various concentrations of Oz. Higher Oz pressures (50% 0,) increased demethylation rates in the early initial segment of the reaction with the three substrates (35% with benzphetamine), although the effect was much less with ethylmorphine (10%) and codeine (15%). With all three substrates, the last two-thirds of the reactions were essentially insensitive to increases in oxygen concentration, except in the case of benzphetamine where there was a slight increase (10%) in rates with 50% Oz in the 2.5- to 15-min segment of the reaction.
Oxygen consumption by hpatic
micro-
somes. Oxygen consumption was determined polarographically with an oxygen electrode. The results of experiments with
ET AL.
three different substrates are summarized in Table III. The initial rates of O2 consumption and the ratio of NADPH added for, each l/2 O2 consumed for two consecutive 100 PM NADPH additions are compared. These results are in general agreement with the kinetics of formaldehyde formation observed with these three substrates (Figs. 1 and 3). The reaction system containing benzphetamine had the highest initial rates of O2 consumption as well as the greatest Oa consumption with the first injection of NADPH. Thereafter, both the rate and the amount of 0, consumed declined significantly with further additions of NADPH (complete data not shown). With ethylmorphine, the rates of O2 consumption were more nearly constant during the NADPH additions (decreasing 37 vs 65% for benzphetamine). The amount of O2 consumption and variation in rates with codeine as substrate were nearly the same as those found with ethylmorphine. None of the substrates studied had ratio of Oa consumed to NADPH oxidation of 0.5, the value expected if there was a stoichiometric relationship between the oxygen consumption and NADPH oxidation. Possible causes of nonlinearity. Three possible causes of nonlinearity suggested by data in the literature were investigated: (i) the formation of an inactive complex of cytochrome P-450 with substrate or product; (ii) selective induction of a form of P450 by PB with different kinetics; and (iii) inactivation of P-450 during the reaction by enzyme destruction. (i) Product adducts formation was ruled out because under these experimental conditions (1.0 mM benzphetamine) there was no spectrophotometric evidence of substrate or product combining with the hemoprotein. Also, with chromatographic separation of the extract from the reaction mixture, we were able to detect only one product and the substrate. (ii) Since the kinetics of benzphetamine demethylation were the same with control microsomes as with those from PB-induced rats, there is no evidence that PB induces a different P-450 system. (iii) The gradual decline in specific dealkylation rates late in the reaction course
KINETICS
OF DEALKYLATION
AND
could be explained in part by P-450 destruction. We did not, however, detect any significant enzyme destruction associated with the initial stages of the reaction.
DISCUSSION
The experiments reported in this paper have demonstrated that demethylations of benzphetamine, ethylmorphine, and codeine by hepatic microsomes are catalyzed by a much more complex enzymatic process than the hydroxylation of testosterone and the deethylation of ‘I-ethoxycoumarin, and most likely represent a family of enzymatic systems that contribute to the overall rates. To a large extent, the complexity of the demethylation reaction stems from the existence, especially in the case of benzphetamine, of the initial rapid segment of activity. Using a variety of zero blank methods we consistently found an increased rate of product formation for the first 1 to 2.5 min of the reaction. The distinction of this first segment from the remainder of the reaction was further supported by its increased pH sensitivity and the ability to selectively affect this part of the reaction by various Oz pressures. Since microsomes are very heterogeneous mixtures, the ability of CO to inhibit the reaction is essential in determining the involvement of P-450, a hemeprotein, in these reactions. Other possible monooxygenation reaction systems such as flavin amine mixed-function peroxidase (19, 20) and catalase, which can under certain experimental conditions catalyze the demethylation of various substrates, are not CO inhibitable. Recent investigations (21, 22) have demonstrated that prostaglandin synthetase also metabolizes a number of N-alkyl compounds. However, this enzyme system is completely dependent upon the addition of arachidonic acid, is not functional with NADPH, and is not competitively inhibited by CO. In addition, it only dealkylates those compounds where the nitrogen atom is directly attached to an aromatic moiety. Our results demonstrate
HYDROXYLATION
REACTIONS
623
that not only are these reaction systems easily CO inhibitable, thus indicating that a CO-combining system is the predominant pathway, but that the degree of CO inhibition is dependent on the CO:Oz ratio. This is consistent with the concept that as an oxidase, the reaction rates will be dependent upon the competition between CO and O2 for the reaction site. Importantly, however, the degree of CO inhibition of the demethylation reactions (the end product of this CO:Oz competition) varied widely as a function of the reaction time. These results imply the involvement of at least one more CO-inhibitable reaction. In these systems then, the observed rates of CO inhibition would be a function of the degree of contribution and inhibition of each enzymatic process. In contrast to this is the time-independent degree of CO inhibition of the monooxygenation of testosterone and ‘I-ethoxycoumarin. The complexity of these systems is further emphasized by the nonstoichiometry of Oz consumption. One to one and onehalf nanomoles of Oz are consumed for each nanomole of NADPH oxidized. This is up to three times greater than the 0.5 nmol of O2 that should be consumed per nmol of NADPH oxidized, if the relationship maintained the strict stoichiometry of a monooxygenase-catalyzed reaction. It is apparent from these experiments that the demethylation reaction is much more complex than the hydroxylation or deethylation. These reactions appear to be composed of a number of catalytic systems which can contribute to the overall rate of product formation. The evidence for this is the variation in turnover rates (especially the fast early segment), the timedependent degree of CO inhibition, and the lack of stoichiometry between NADPH and O2 consumption. As a consequence, any further study of the microsomal monooxygenation reaction as well as the use of the purified system containing only P-450 must keep in mind the intricacies of the demethylation reaction. In the accompanying paper, the light-reversal properties of CO inhibition of these substrates are described and compared as a means of bet-
624
HAMILL
ter understanding these reactions by the determination of the P-450 involvement. REFERENCES 1. COOPER, D. Y., SCHLEYER, H., LEVIN, S., EISENHARDT, R. H., NOVACK, B. G., AND ROSENTHAL, 0. (1979) Drug M&u& Rev. 10.153-185. 2. PETERS, M. A., AND FOUTS, J. R. (1970) Biochem Ph4mwmml 19,533~544. 3. HAMILL, S., COOPER, D. Y., SCHLEYER, H., AND ROSENTHAL, 0. (1980) in Biechemistry, Biophysics and Regulation of Cytochrome P-450 (Gustzfasson, J. A., et al, eds.), pp. 337-341, Elsevier/North-Holland, Amsterdam/New York. 4. COOPER,D. Y., SCHLEYER, H., THOMAS, J. H., VARS, H. M., AND ROSENTHAL, 0. (1975) in Cytochromes P-456 and 4. Structure, Function and Interaction. (Cooper, D. Y., et al, eds.), pp. 81102, Plenum, New York. 5. DALLNER, G. (1974) in Methods in Enzymology (Fleisher, S., and Packer, L., eds.), Vol. 31, pp. 191201, Academic Press, New York. 6. COOPER, D. Y., SCHLEYER, H., ROSENTHAL, O., LEVlN, W., Lu, A. H. Y., KUNTZMAN, R., AND CONNEY, A. H. (1977) Eur. J. Bimhem 74,6875. 7. OMURA, T., AND SATO, R. (1962) J. Bid Chem 237, 1375-1381. 8. LOWRY, 0. H., ROSENBROUGH, N. J., FARR, A. L.,
ET AL.
9. 10. 11. 12. 13. 14. 15. 16. 17.
18. 19. 20. 21.
22.
AND RANDALL, R. J. (1951) J. Biol Chem 193, 265-275. GREENLEE, W., AND POLAND, A. (1978) J. Pharmix01 Exp. Thzr. 205, 595-605. NASH, T. (1953). Biochtm J. 55,416-421. ULLRICH, V., AND WEBER, P. (1972) HqnpeSeyler’s 2. Physid Chem 353,1171-1177. BAIRD, W. M., DIAMOND, L., AND SHULMAN, S. (1979) Ad Biochrem 103.64-69. CHANCE, B., AND THEORELL, H. (1951) Acta C&m Scana! 5,1127-1133. NORDBLOOM, G. D., AND COON, M. J. (1976) Arch Biochem. Biophps. 175,343-349. NORDBLOOM, G. D., AND COON, M. J. (1977) Arch. Btihem. Biophys. 130,524-533. FRANKLIN, M. R. (1971) Xaobiotica 1,581~588. WERRINGLOER, J., AND ESTABROOK, R. W. (1979) in The Induction of Drug Metabolism (Estabrook, R. W., and Lindenlaub, E., eds.), pp. 269307, F. K. Schattauer Verlag, Stuttgart. WERRMGLOER, J., AND ESTABROOK, R. W. (1973) Life Sci 13, 1319-1330. MACHINIST, J. M., ORME-JOHNSON, W. H., AND ZIEGLER, D. M. (1966) Biochemistry 5.2939-2943. UEHELEKE, H. (1973) Drug Metub. Dispos 1,299313. SIVARAJAH, K., LASKER, J. M., ELING, T. E., AND ABOU-DONIA, M. B. (1982) Mel Pharmacd 21, 133-141. LASKER, J. M., SIVARAJAH, K., MASON, R. P., KALYANARAMAN, B., ABOU-DONIA, M. B., AND ELING, T. E. (1981) J. Bid Chem 256.7764-7767.