Spin-labeling studies of rat liver NADPH-cytochrome P450 reductase: Conformation and function relationship

ARCHIVES

OF BIOCHEMISTRY

AND BIOPHYSICS

Vol. 296, No. 1, July, pp. 73-80, 1992

Spin-Labeling Studies of Rat Liver NADPH-Cytochrome P450 Reductase: Conformation and Function Relationship Shashi P. Singh and Lawrence H. Piette’ Department of Chemz’stry and Biochemistry, Utah State University, Logan, Utah 84322-0300

Received July 3, 1991, and in revised form February

7, 1992

ESR spin-labeling studios designed to yield information regarding the relationship between function and conformation of rat liver NADPH-cytochrome P450 reductase (EC 1.6.4.2) were carried out. The purified enzyme was spin labeled by a nitroxide derivative of p-chloromercuribenzoate. Two conditions for spin labeling were employed: (i) the presence of NADP+, yielding an active siteprotected spin-labeled reductase, and (ii) the absence of NADP+, yielding completely spin-labeled reductase. Reductase in which the active site was protected by binding NADP+ and then spin-labeled retains most of its enzymatic activity; on the other hand, completely spin-labeled reductase is devoid of any enzymatic activity. Completely spin-labeled reductase yields a two-component resolved ESR spectrum that reflects two classes of spin-labeled binding sites, a strongly immobilized (S) and a weakly immobilized (W) site. The ratio of W/S provides a valuable parameter for studying the relationship between function and conformation. Structural perturbants, such as urea, KCl, and pH, were employed to determine their effects on the activity of the enzyme and their relationship to changes in the conformational state of the reductase. It was further observed that the enzymatically active spinlabeled derivative generated superoxide radical in the presence of NADPH and cytochrome c, which in turn reQ ISSZ ~cndemi~ duced completely the attached spin-label. Press, Inc.

NADPH-cytochrome P450 reductase constitutes an important component of the microsomal monooxygenase system. The local microenvironment of the enzyme, NADPH-cytochrome P450 reductase, exerts a profound influence on its enzymatic properties. Strobe1 and Coon (1) showed that in the presence of high salt concentration, 1 To whom correspondence 0003-9661/92

should be addressed.

$5.00

Copyright 0 1992 by Academic Press, Inc. All rights of reproduction

in any form

reserved.

the reductase was more effective in the reduction of cytochrome. Protease-solubilized reductase, on the other hand, which lacks the hydrophobic membrane-anchoring segment, is devoid of the capability to reduce either cytochrome P450 or cytochrome b5. Bilimoria and Kamin (2), however, reported that high salt concentration restored the protease-solubilized reductase’s ability to reduce cytochrome bg. Thus, under certain conditions and in particular environments even a nonsubstrate becomes an excellent electron acceptor for the reductase. These results suggest that changes in the conformation of the reductase may occur with changes in the enzyme’s microenvironment. Very little is known about these conformational changes or how to measure them. Studies of those conditions that bring about conformational changes in the reductase and other components of the microsomal electron transport system are needed in order to understand the mechanism of electron transport in the complex. A beginning in this direction was made by Grover and Piette (3) in their early attempts to use spin-labeling of the reductase to study possible conformational changes occurring in the protein during electron transport. This report builds on that early work. In this report we are attempting to understand the conformation and function relationship of the reductase by using a spin-labeling technique designed to reveal changes specifically at the active site of the enzyme. All seven available cysteinyl sulfhydryl groups on the reductase isolated from rat liver microsomal fractions can be labeled with p-chloromercuribenzoate (3-5, 7). Several studies using sulfhydryl-modifying reagents indicate that two or three cysteinyl residues are located at or near the binding site of NADPH on the catalytic segment of the reductase (3, 6-9). Our spin-labeling technique has allowed us to follow motional perturbations of different sulfbydryl locations on the enzyme, reflected by their ESR spectra as a function of conformational change. The conformational per73

74

SINGH

AND

turbants that we have chosen to use are urea, ionic strength, and pH. Each of these is known to affect the enzyme’s activity, therefore we can correlate ESR spectral changes with activity and hopefully conformation. It is understood that the label itself is a perturbant and thus may not reflect the true conformation of the active site. MATERIALS

AND METHODS

Materials. Male Sprague-Dawley rats were used for the liver microsomal preparation. Phenobarbital was obtained from Fisher. A nitroxide ester derivative of p-chloromercuribenzoate (PCMB-SL)z was provided by Dr. Kalman Hides, University of Pets, Pets, Hungary. Bromelin (EC 3.4.22.4), flavin mononucleotide (FMN), flavin adenine dinucleotide (FAD), Sephadex G-25 and G-100, DEAE-cellulose, sodium azide, and bovine serum albumin (BSA) were obtained from Sigma Chemical Co. Ethylenedinaminetetraacetic acid (EDTA) were from MCB Manufacturing Chemists, Inc. Acetonitrile (HPLC grade) was a Baker analyzed reagent. All other chemicals used in the experiments were of analytical grade. Commercial enzymes were diluted in the appropriate buffer before use. PM-30 ultrafiltration membranes were obtained from Amicon. 2’,5’-ADP-Sepharose-4B affinity chromatographic material was obtained from Pharmacia. Spectroscopic measurements were performed using a Beckman DU-7 spectrophotometer. Flavin assays were performed using a Gilford Fluro IV spectrofluorometer. ESR spectra were taken on a Varian E-9 spectrometer equipped with an on line microcomputer for spectral analysis. NADPH-cyPurification of NADPH-cytochrome P450 reductase. tochrome P450 reductase was purified from rat liver by the protease solubilization method described by Pederson (10). Cytochrome c reduction assay. The activity of NADPH-cytochrome P450 reductase for cytochrome c reduction was assayed at 3O“C in the reaction mixture essentially using a method similar to that described by Phillips and Langdon (11). The rate of cytochrome c reduction was determined at 550 nm using an extinction coefficient of 21 mM-’ cm-’ (12) for the conversion of oxidized to reduced cytochrome c. The reductase concentration in the assay solution was 3 to 4 pM. Specific activities were expressed as micromoles of cytochrome c reduced per minute per milligram of protein. NADPH oxidation activity of NADPHNADPH oxidase activity. cytochrome P450 reductase was determined using an extinction coefficient of 6.2 mM-’ cm-’ at 340 mM (6). The NADPH oxidase activity of the reductase was measured in a reaction mixture containing 10 mM potassium phosphate buffer, pH 7.7, containing 10% glycerol and 60 to 65 fig/ml of the native or spin-labeled reductase. The reaction was initiated by the addition of NADPH at room temperature. Specific activities were expressed as nanomoles of NADPH oxidized per minute per milligram of protein. The determination of Estimation of flnvin content of the reductase. FMN and FAD was performed according to the method of Faeder and Siegel (13). All solutions were prepared using 0.1 M potassium phosphate buffer, pH 7.7, containing 0.1 mM EDTA (flavin assay buffer). The enzyme concentration for flavin assay was 20 to 25 PM. Flavin concentrations in stock solutions were determined by measuring absorbance at 450 nm using an extinction coefficient of 12.2 mM-’ cm-’ for FMN and 11.3 mM-’ cm-’ for FAD (14). Preparation of acetykzted cytochrome c. The procedure used in the preparation of acetylated cytochrome c was that described by Azzi et al. (15).

’ Abbreviations used: PCMB-SL,p-chloromercuribenzoate; BSA, bovine serum albumin; ASP-SL, active site-protected spin-labeled reductase; camp-SL, completely spin-labeled reductase; s, strongly immobilized label; W, weakly immobilized label; DTT, dithiothreitol; cyt. c, cytochrome c.

PIETTE FMN depleted NADPHPreparation of FMN-depleted reductase. cytochrome P450 reductase was prepared by the method of Vermillion and Coon (5). Spin labeling of the reductase was Spin labeling of the reductase. performed at room temperature. The NADPH binding site on the reductase was protected by the addition of 1 mM NADP+ to a 3 to 4 pM enzyme solution and incubating for 10 min on ice. A stock solution of 2 mu PCMB-SL in acetonitrile was prepared and a suitable aliquot of the stock solution was evaporated under nitrogen so as to yield a lofold excess of label over enzyme. The enzyme solution prepared in the presence or in the absence of NADP+ was then added. The reaction was allowed to occur for 1.5 h at room temperature. Spin-labeled reductase was separated from free PCMB-SL by filtering the reaction mixture through a Sephadex G-25 column. Fractions of 1 ml were collected by eluting the column with 10 mM potassium phosphate buffer, pH 7.7, containing 10% glycerol. Peak protein fractions were pooled together and concentrated using cold solid polyethylene glycol. Protein was determined Protein determinution. Bradford (16) using BSA as standard.

by the method of

RESULTS AND DISCUSSION

NADPH-cytochrome P450 reductase was purified to homogeneity from rat liver microsomes. Spin labeling with PCMB-SL was performed under two different conditions so as to modify the sulfhydryl groups on the reductase at two locations. In one case, the -SH groups at the NADPH binding site of the enzyme were labeled along with other -SH groups on the protein (camp-SL). In the other case, the NADPH binding site was protected by NADP+ so that the - SH groups at this site were not modified with PCMB-SL(ASP-SL). We had previously shown that this labeling technique, based upon double integration of the ESR spectra showed that five sulfhydryls are labeled in the ASP-SL prep and seven in the camp-SL preparation (3). Table I presents a tabulation of various reductase preparations and their activities toward cytochrome c reduction, NADPH oxidation, and flavin content. The FMN/FAD ratio of 0.86 for native enzyme was found to be very close to that of 0.91 obtained by Vermilion and Coon (4). Cytochrome c reduction and NADPH oxidation activities of native reductase were found to be 43 pmol mine1 mg-l and 20 nmol min-’ mg’, respectively. This activity compared reasonably well with the activity of 37 to 57 pmol-l mg-l reported by others (6,17-19). Variation in the activity could be due to modification in the methods of preparation or the storage conditions. There is a loss in FMN content of the reductase during its purification as its levels were found to be less than those obtained for FAD (Table I). Completely spin-labeled reductase lost its ability to reduce cytochrome c almost completely. Less than 1% of its original activity toward cytochrome c was recovered. NADPH oxidation activity was also drastically reduced (only 3% of the original activity remained). Of the original FAD content, 94% was depleted when the reductase was completely labeled by PCMB-SL. FAD depletion in the completely labeled preparation exceeded that of FMN

SPIN

LABELING

OF RAT

LIVER

NADPH-CYTOCHROME

(30% of the original FMN content was retained by the reductase). This was considered to be significant since under many manipulations of the reductase, retention of FAD is favored over that of FMN (6). The differential depletion due to complete labeling of the protein’s sulfhydryl groups was consistent with the result obtained by Grover and Piette (3,6). Since FAD appears to be located near the NADPH binding site of the reductase (17, 20), its loss could be due to PCMB-SL modification of a sulfhydryl group at the active site in the vicinity of the FAD binding site. It may also suggest that the integrity of the sulfhydryl group at the NADPH site is essential for the enzyme to maintain its ability to bind FAD. When the -SH group at the NADPH binding site was protected by the addition of NADP+, the resultant reductase retained most of its activity toward cytochrome c reduction, as well as NADPH oxidation (88 and 85% of the activity was retained, respectively). Loss of flavin content was also minimal (90% of FAD and 92% of FMN content of the original reductase was maintained). This indicates that the -SH group at the NADPH binding site may have an important role in maintaining structural stability of the reductase (8). This also implies that - SH groups at the other regions on the reductase play a relatively minor role as far as enzyme activity toward cytochrome c reduction and NADPH oxidation are concerned. ESR spectra of the spin-labeled reductase are shown in Fig. 1. Free PCMB-SL in 10 mM potassium phosphate buffer (pH 7.7) containing 10% glycerol yields an ESR spectrum with three characteristic sharp lines (Fig. 1A). All spin-labeling studies were performed in 10 mM potassium phosphate buffer to avoid any effects of high ionic strength on the enzyme conformation. Figure 1B represents spectra obtained when the NADPH active site of the reductase was protected by NADP+ during PCMBSL labeling. The spectrum is characteristic of labels attached at multiple sites with hindered mobility. Repeated elution through Sephadex G-25 or dialysis against buffer did not change the relative intensity of the peaks. Thus

PCMB-SL

ASP-Reductase

Comp-SLReductase

20 G FIG. 1. Comparative ESR spectra. (A) Free PCMB-SL, (B) ASP-SL reductase prepared in the presence of 1 mM NADP+, and (C) Comp-SL reductase prepared without the addition of NADP+. ESR spectral parameters: field set, 3370 G; scan range, 160 G; receiver gain, different in all spectra; modulation amplitude, varied from 1.25 to 3.2 G.; modulation frequency, 100 kHz; microwave power, 50 mW; frequency, 9.515 GHz; time constant, varied from 0.5 to 1.0 s; scan time, 4 min.

we can confidently say the spectrum represents covalently bound label in which the labels are weakly immobilized. When the reductase was spin-labeled without the protection of the NADPH binding site, completely spin-la-

TABLE Properties

of Various

75

P450 REDUCTASE

I

Reductase

Preparations Activity

Flavin content Reductase preparation

FMN

FAD

1. Native reductase 2. Comp-SL reductase 3. ASP-SL reductase

8.4 2.4 7.7

10.2 0.6 9.2

FMN/FAD 0.86 4.00 0.78

cyt. c reduction 43.0 0.3 37.7

NADPH oxidation 20.5 0.6 17.4

Note. Flavin content is expressed in nmol rng-i. Cytochrome c reduction and NADPH oxidation activities are expressed in pmol mini mg-’ and nmol min-’ mg-‘, respectively. (Acetylated cytochrome c reductase activity was found to be 1.4 pmole mini mg-i). The FMN content of the reductase should equal its FAD content. However, during purification, the enzyme is partially depleted of FMN. Procedures for the assay of cytochrome c reductase and NADPH oxidase activity are described in the Materials and Methods section. Comp-SL: completely spin-labeled reductase, ASP-SL, active-site-protected spin-labeled reductase.

76

SINGH

AND

beled reductase was the product. The resultant ESR spectra of this product is shown in Fig. 1C. The ESR spectrum of completely spin-labeled reductase showed two major components reflecting different degreesof spin label immobilization, namely a strongly immobilized (S) label and a weakly immobilized (W) label. There may be more than one population of weakly immobilized labels present. The strongly immobilized label is characterized by the additional peaks observed at low and high fields in addition to the broadened three-line spectrum. These strongly immobilized peaks are most likely due to labels attached to the - SH groups at the NADPH binding site on the reductase. These peaks were not observed when labeling was carried out in the presence of NADP+; i.e., the NADPH site was protected. The weakly immobilized labels are most likely associated with binding to cysteines at the surface of the protein. The amplitude ratio W/S is a crude measure of the relative freedom of the labels to move in their different environments. Since the W component (weakly immobilized) has a line width much narrower than that of the S component (strongly immobilized), the W/S ratio is very sensitive to any intensity change in the W component (20). Small changes in the conformational state of the protein can cause small changes in the environment of the label which are reflected as a change in the mobility of the label. These changes in label mobility can produce significant changes in the W/S ratio (20-22). Thus the W/S ratio can provide a valuable parameter for studying the relationship between function and conformation of a protein. Labeling Specifiity

The specificity of labeling to the cysteine sulfhydryls and not to any other amino acid residue was determined by treating the labeled protein with DTT, which is known to reduce the cysteinyl group back to its native -SH. ESR spectra obtained after the treatment of spin-labeled reductase with 1.5 mM DTT were identical to that shown in Fig. 1A. Addition of 1.5 mM DTT to the completely labeled reductase preparation resulted in the release of PCMB-SL from the protein within 10 min at room temperature. This result suggests that PCMB-SL is covalently labeled only to the various -SH groups on the reductase and not to any other amino acid. Effects of Perturbants

on ESR Spectra

The effects of the perturbants, urea, ionic strength, and pH, which can alter protein conformation were monitored as a function of the W/S ratio and correlated with activity. The results are shown in Figs. 2-5. Urea. The effect of urea denaturation on the active site region of the reductase is shown in Fig. 2. As the concentration of urea was increased the S component of the ESR spectrum of the completely spin-labeled reduc-

PIETTE

A

4M Urea

FIG. 2. Effect of urea on the NADPH active site of the reductase. The reaction mixture contained 162 pg of the completely spin-labeled reductase with the indicated concentration of urea in 10 mM potassium phosphate buffer (pH 7.7) containing 10% glycerol (volume, 250 ~1) at room temperature (A) without urea, (B) 1 M urea, (C) 2 M urea, (D) 4 M urea, and (E) 6 M urea. ESR spectral parameters are as described in the legend to Fig. 1.

tase started to diminish, and at 6 M urea the S component of the spectrum was almost completely gone. The W component of the spectrum, on the other hand, increased in intensity. The effect of urea on the changes in the spectra was reversible. The complete reversibility of urea denaturation was demonstrated after dialyzing the urea out of the reductase solution (Fig. 2F). The strongly immobilized component reappeared and the spectrum was identical with that obtained without the addition of urea. The correlation between structural change due to urea and activity of the reductase is presented in Fig. 3. The W/S ratio of the spin-labeled reductase and the cytochrome c reduction activity of the native reductase was plotted against urea concentration. As the concentration of urea was increased in the assay medium, the activity of the reductase decreased. A significant effect on the activity was observed after the addition of 3 M urea. This was accompanied by a change in the conformational state of the protein, which was observed in the changes in the W/S ratio at high concentrations of urea. Changes in the

SPIN

LABELING

OF RAT

NADPH-CYTOCHROME

18

80 A

LIVER

-I

i $

-10 2

-8

0 -6 -4 -2

0-o 0

12

3

4

5

8

7

Urea 04) FIG. 3.

Cytochrome c reduction activity of the native reductase and W/S ratio of the Comp-SL reductase as a function of urea. (A) W/S ratio, (B) cytochrome c reduction activity. Details of the cytochrome c reductase assay of the enzyme are described in the Materials and Methods section.

S component of the ESR spectra of the reductase most likely represent conformational changes occurring at the NADPH binding site. It seems likely that the integrity of the active site region of the reductase is affected by urea and this in turn is related to activity. It may also be possible that the overall three-dimensional conformation of the reductase is disrupted by urea, which may in turn cause structural collapse around the active site. The effect of a monovalent salt (i.e., Ionic strength. KCl) on the active site of the reductase was studied by the addition of increasing amounts (0.3 to 1.4 M) of KC1 to the completely spin-labeled reductase. The effect of the increase in salt concentration is apparent in Fig. 4, in which the W/S ratio was plotted as a function of KC1 concentration. The W/S ratio increased as the salt concentration was increased up to 0.6 M KCl, and then a decrease in the ratio was observed. The overall change in W/S ratio, however, is within a narrow range of 2 to 3. There is no drastic change in the ratio as there was with urea (Fig. 3). Potassium chloride affects the reductase through a different mechanism. It might modulate the enzymatic activity due to change in ionic strength of the medium by altering the structure of the solvent at the active site (2). The small increase in the ratio with the increase in the KC1 concentration up to 0.6 M might be due to a “salting in” effect. The decrease in the ratio at higher concentration of the salt (0.6 to 1.4 M) probably indicates a “salting out” phenomenon. This change in W/S ratio correlates with the change in its enzymatic activity, as measured by cytochrome c reduction. Maximum enzymatic activity was observed at 0.6 M KCl. The W/S ratio was found to be highest at this salt concentra-

77

P450 REDUCTASE

tion. The result is consistent with that found by Bilimoria and Kamin (2) although they used a different salt, namely ammonium sulfate. The KC1 effect was also found to be reversible. ESR spectra obtained after dialyzing the KC1 out of the medium yielded the same W/S ratio (approx. 2) as that obtained without the addition of KC1 (result not shown). It should be pointed out that this change in W/S with KC1 and activity is very small. The actual value of W/S for maximum activity is only slightly different from that with no KC1 2.0-2.9, whereas complete loss of activity due to denaturants such as urea yield a W/S of 17. The absolute value of the ratio of W/S has no real significance except when it is compared to a perturbing influence such as a denaturant, etc. The ratio is dependent both upon the integrity of the active site and the extent of labeling of the entire protein. Samples used to study the effects of urea, KCl, and pH were all different. Thus the W/S ratio at optimum activity varied because of the different extent of labeling, particularly the SH groups not at the active site which contribute principally to the W signal. The mechanism by which salt interacts with proteins is not well understood (24). Two general mechanisms of interaction have been considered, a direct ionic interaction with the protein, i.e., the polar groups of the polypeptide backbone that are internalized in the native structure, and an indirect interaction through changes in solvent (H,O) structure. The latter includes effects of the ions on the interaction between water molecules and the polar side chains of the protein. The KC1 effect on the conformational state of the reductase may have been mediated by both of these factors. pH effect. The effect of pH on the activity of the reductase is shown in Fig. 5, where the W/S ratio and the

,b E T .E E

60

3.0

50

2.8

40

2.6

$

30

2.4

ii? m

5 Y >, .Gj 'E

20 2.2

2 5

2

10

g

2.0

0

1.8 0

0.5

1.0

1:5

210

KCI (M) FIG. 4. Effect of KC1 on the cytochrome c reductase activity and the W/S ratio of camp-SL reductase. (A) W/S ratio, (B) cytochrome c reductase activity of the native reductase. Details of the cytochrome c reductase assay of the enzyme are described in the Materials and Methods section. W/S ratio was obtained from the effect of Kl on the ESR spectrum of completely spin-labeled reductase (spectra not shown). ESR spectral parameters are same as described in the legend to Fig. 1.

78

SINGH

AND

the NADPH site was achieved when labeling was carried out in the presence of NADP+. It is safe to assume that the cysteinyl residue at the NADPH binding site plays a vital role in maintaining the functional integrity of the enzyme.

-7 = *E

PIETTE

40

330

A

v5

20

E r

‘0

Effect of Electron Transport on ESR Spectra B li.$l 0

3

6

9

12

PH FIG. 5. Cytochrome c reductase activity of the native reductase and the W/S ratio of the camp-SL reductase as a function of pH. (A) W/S ratio and (B) cytochrome c reductase activity of the native enzyme. Details of the cytochrome c reductase assay are described in the Materials and Methods section. The W/S ratio was obtained from the effect of pH on the ESR spectrum of completely labeled reductase. The solution contained 165 pg of the completely spin-labeled reductase in 10 mM potassium phosphate buffer and 10% glycerol at the indicated pH (volume, 250 ~1). ESR spectra were recorded at room temperature. (A) pH 2.9, (B) pH 5.6, (C) pH 7.7, (D) pH 9.7, (E) pH 12.2.

cytochrome c reductase activity are plotted as a function of pH. Activity of the reductase was found to be maximum at pH 7.7. A change in the conformational state of the reductase at different pH was indicated by changes in the W/S ratio. The shape of the graph in Fig. 5 also indicates that the structural integrity of the reductase is maintained between pH 6 and 10. Extensive disorganization in the structure was observed outside of this pH range, reflected by a dramatic change in the W/S ratio. An increase in W/S ratio was observed on both sides of pH 7.7, indicating an “opening up” of the active site of the reductase. A similar increase in W/S ratio was also observed in the case of urea denaturation (Fig. 3). The effect of pH on the structural integrity of the reductase was reversible, as was the case with the other two environmental perturbants (spectra not shown). Effects of pH and urea denaturation are similar, as indicated by the high W/S ratio obtained at extreme values of these perturbants while salt modulates the activity by changing the ionic strength of the media and altering activity by a mechanism other than dramatic conformational change at the active site, as reflected by the change in the W/S ratio. The effect of NADPH or NADP+ addition on the active site of the completely spin-labeled reductase is presented in Fig. 6. When the W/S ratio was plotted against either NADPH or NADP+ concentration, no change in the ratio was observed. Even at a very high concentration of NADPH or NADP+, the W/S ratio remained unchanged, as did the lineshape of the spectra. This probably indicates that when the -SH group at the NADPH binding site was occupied by a spin label, neither NADPH nor NADP+ could be accommodated. This is a reasonable assumption since effective blocking of the labeling of the cysteines at

The effect of electron transport on the ESR spectra of active spin-labeled reductase was studied by following the ESR signal as a function of the complete enzyme activity measurement, i.e., addition of NADPH and cytochrome c to the active site-protected enzyme. The ESR signal for the nitroxide was completely reduced in 30 min after the addition of NADPH to the reaction mixture containing 0.176 mM cytochrome c (Fig. 7). The question was raised: what causes the reduction of the ESR signal? To answer this question a set of experiments which would indicate which components of the electron transport system might be involved in the signal reduction was designed. Possible candidates causing the reduction of the label are NADPH itself or 0, radicals, which are known to be produced in this system during enzymatic oxidation of NADPH (3). When active site-protected spin-labeled reductase was used in the presence of acetylated cytochrome c and NADPH, the ESR signal was not reduced within the time limit of the experiment (Fig. 7A). It is well known that acetylated cytochrome c cannot be reduced directly by the reductase (15, 25). It can, however, be reduced by 0,. When the completely spin-labeled reductase was used instead of active site-protected reductase in the presence of cytochrome c and NADPH, the ESR’ spectral signal was also not reduced (Fig. 7B). The absence of signal reduction in this case is due to the fact that completely spin-labeled reductase does not have enzymatic activity. Using active site-protected spin-labeled reductase and NADPH alone, without cytochrome c, the ESR signal is

6 0 ‘S m a

6 NADPH

84 3 2

0

2

4

6

6

10

-I

12

NADPH or NADP+ (mM) FIG. 6. Effect of NADPH and NADP+ on the NADPH active site of the reductase. Reaction mixture contained 91 pg of completely spinlabeled reductase in 10 mM potassium phosphate buffer (pH 7.7) containing 10% glycerol (volume, 250 ~1) with the indicated amount of either NADPH or NADP+. ESR spectra were recorded at room temperature. (A) W/S ratio in the presence of NADPH, and (B) W/S ratio in the presence of NADP+.

SPIN

LABELING

OF RAT

A

n

0

10

I 20

T 30

1 40

I 50

I 60

I 70

LIVER

NADPH-CYTOCHROME

0

I 60

90

Time (min) FIG. 7. Effects of NADPH and cytochrome c on ESR signals of spinlabeled reductases. The reaction mixtures contained 10 mM potassium phosphate buffer (pH 7.7) containing 10% glycerol (volume, 250 ~1). ESR spectra were recorded at room temperature. ESR spectral parameters are as described in the legend to Fig. 1. (A) 71 kg of ASP-SL reductase, 1.4 mM NADPH, and 0.147 mM acetylated cytochrome c; (B) 71 pg of camp-SL reductase, 1.4 mM NADPH, and 0.176 mu cytochrome c; (C) 70 pg of ASP-SL reductase, and 1.4 mM NADPH; (D) 61 pg FMNdep ASP-SL reductase, 1.4 mM NADPH, and 0.176 mM cytochrome c; (E) 81 kg of ASP-SL reductase, and 0.176 mM cytochrome c; (F) 56 kg of ASP-SL reductase, 1.4 mM NADPH, 0.176 mM cyt. c, and 240 units of SOD; (G) 40 pg of ASP-SL reductase, 1.4 mM NADPH, and 0.176 mM cytochrome c.

partially reduced slowly (Fig. 7C). The ESR spectral signal, however, was reduced very rapidly and completely when both cytochrome c and NADPH were added (Fig. 7G). When the reaction was carried out in the presence of FMN-depleted spin-labeled reductase, cytochrome c, and NADPH, the ESR signal was not reduced (Fig. 7D). This is explained by the fact that FMN-depleted reductase has virtually no activity (0.9 pmol mine1 mg-l) as compared to the active site-protected spin-labeled reductase (Table I). The ESR spectral signal reduction was also not observed in the presence of active site-protected spinlabeled reductase and cytochrome c alone (Fig. 7E). Even after 90 min of incubation, there was very little change in the signal intensity. The complete system of enzyme plus NADPH and cytochrome c was essential for the reduction. When superoxide dismutase (SOD) was added to the reaction mixture containing active site-protected spin-labeled reductase, cytochrome c, and NADPH, the nitroxide ESR signal was not reduced (Fig. 7F). This suggests the possibility that superoxide anions are responsible for the reduction of the nitroxide radical, with a concomitant loss of the ESR signal. The fact that acetylated cytochrome C and NADPH had little or no effect on reducing the spin-label signal suggests that if 0; is responsible for the reduction it is only produced in the presence of the complete system, i.e., with NADPH and cytochrome C. If superoxide generated by the reductase is responsible for the reduction of the nitroxide signal, a superoxide-

P450 REDUCTASE

79

generating system such as xanthine/xanthine oxidase (26), may also be capable of reducing the ESR signal of the nitroxide-labeled reductase. To examine this hypothesis, active site-protected spin-labeled reductase was added to the reaction mixture containing xanthine oxidase, and the reaction was initiated by the addition of xanthine. ESR spectra were recorded at various time intervals. The ESR signal was reduced completely within 40 min. This result was similar to what was found with the reductase in the presence of cytochrome c and NADPH (Fig. 7G). This result strongly suggests the involvement of superoxide in the reduction of the ESR signal. Since the ESR signal reduction occurred only in the presence of cytochrome c with the reductase and NADPH, it seems probable that cytochrome c plays a significant role in the generation of the superoxide radicals. In the presence of acetylated cytochrome c, activity of the reductase was reduced to an insignificant level (i.e., 1.4 pmol min-’ mg-l) (Table I). This also suggests that superoxide generation requires reduced cytochrome c. The effect of cytochrome c concentration on the ESR spectra of the active site-protected spin-labeled reductase is represented in Fig. 8. As the concentration of cytochrome c in the reaction mixture was increased from 0.06 to 0.32 mM, the time to reduce the ESR signal was drastically reduced. At 0.06 mM cytochrome c the ESR signal was slowly reduced, and complete reduction of the signal was not achieved even after 90 min. When the cytochrome c con-

Time (min) FIG. 8. Effect of cytochrome c on the ESR signal of ASP-SL reductase. Reaction mixture contained 76 pg of ASP-SL reductase, 1.4 mM NADPH, and the indicated amount of cytochrome c in 10 mM potassium phosphate buffer (pH 7.7) containing 10% glycerol (volume, 250 ~1). (A) 0.06 mM cyt. c, (B) 0.192 mM cyt. c, (C) 0.32 mM cyt. c. ESR spectral parameters are as described in the legend to Fig. 1.

80

SINGH

AND

centration was increased to 0.19 mM, the time for complete reduction of the signal was less than 40 min. At 0.32 mM cytochrome c the signal was reduced within 20 min. A similar experiment was designed to investigate the effect of increasing the amount of NADPH at a fixed concentration of cytochrome c. As the concentration of NADPH was raised from 0.8 to 4.0 InM, the time taken for the ESR signal reduction was reduced from 43 to 23 min (result not shown). Although the time taken for nitroxide ESR signal reduction was decreased as the concentration of NADPH was increased, this effect became more efficient when the amount of cytochrome c was increased. The effect of an increase in cytochrome c on the rate of the ESR signal reduction is much greater than that of NADPH. It took less than one-tenth of the amount of cytochrome c to achieve the same rate of ESR signal reduction as with NADPH. Paraquat is known to enhance the production of superoxide radical in the presence of cytochrome P450 reductase and NADPH (27). The redox cycling property of paraquat depends on its ability to undergo one-electron reduction by a physiological reductant, (i.e., reductase: NADPH) followed by reoxidation of the paraquat radical by molecular oxygen. This results in catalytic production of the superoxide radical. The effect of paraquat on the ESR signal reduction of the active site-protected spinlabeled reductase was pronounced (results not shown). The ESR signal was reduced within 3 min in the presence of NADPH and paraquat. This faster rate of signal reduction is considered to be due to the presence of paraquat and its ability to catalytically produce 0; in the presence of the reductase and NADPH. Cytochrome c in the reductase system must be performing a function similar to that of paraquat, namely, the catalytic production of superoxide radical. Cytochrome c is reduced by the enzyme followed by subsequent reoxidation by oxygen to yield the superoxide radical, which converts the nitroxide radical of the spin-labeled reductase to a nonspin compound. This reoxidation must be mediated by the enzyme in some way since simple autoxidation of cytochrome c to produce 0, is much too slow. It would appear that this spin-labeling technique cannot only yield structural functional information on conformation at the active site relative to activity but can also give some insight into the nature of the electron transport taking place.

PIETTE

REFERENCES 1. Strobel, H. W., and Coon, M. J. (1971) J. Bid. Chem. 246,78267829. 2. Bilimoria, M. M., and Kamin, H. (1975) Ann. N. Y. Acad. Sci. 212, 428-448. 3. Grover, T. A., and Piette, L. H. (1983) in Frontiers in Biochemical and Biophysical Studies of Proteins and Membranes, (Liu et al., Eds.), pp. 321-330, Elsevier Science, New York. 4. Vermilion, J. L., and Coon, M. J. (1978a) J. Biol. Chem. 253, 2694-2704. 5. Vermilion, J. L., and Coon, M. J. (1978b) J. Bid. Chem. 253, 8812-8819. 6. Grover, T. A., and Piette, L. H. (1981) Arch. Biochem. Biophys. 212,105-114. 7. Masters, B. S. S., Bilimoria, M. H., Kamin, H., and Gibson, Q. H. (1965) J. Biol. Chem. 240,4081-4088. 8. Lazar, T., Ehrig, H., and Lumber, L. (1977) Eur. J. Biochem. 76, 365-371. 9. Nishimoto, Y., and Shibata, Y. (1981) B&him. Biophys. Acta 662, 291-299. 10. Pederson, T. C., Burge, J. A., and Aust, S. D. (1973) J. Biol. Chem. 248,7143-7151. 11. Phillips, A. H., and Langdon, R. G. (1962) J. Bid. Chem. 237, 2652-2660. 12. Massey, V. (1959) B&hem. Biophys. Acta 34, 255-256. 13. Faeder, E. J., and Siegel, L. M. (1973) Ann. Biochem. 53, 332336. 14. Whitby, L. B. (1953) Biochem. J. 53,437-42. 15. Azzi, A., Montecucco, C., and Richter, C. (1975) Biochem. Biophys. Res. Commun. 65,597-603. 16. Bradford, M. (1976) Anal. Biochem. 72,248-254. 17. Vermilion, J. L., Ballow, D. P., Massey, V., and Coon, M. J. (1981) J. Biol. Chem. 256,266-277. 18. Yasukochi, Y., and Masters, B. S. S. (1976) J. Bid. Chem. 251, 5337-5344. 19. Yasukochi, Y., Peterson, J. A., and Masters, B. S. S. (1979) J. Biol. Chem. 254,7097-7104. 20. Iyanagi, T., Makino, R., and Anan, F. K. (1981) Biochemistry 13, 1722-1730. 21. Fung, L. W. M., Litvin, S. D., and Reid, T. (1983) Biochemistry 22,864-869. 22. Fung, L. W. M., and Johnson, M. E. (1984) in Current Topics in Bioenergetics (Lee, C. P., Ed.), pp. 107-157, Academic Press, New York. 23. Fung, L. W. M. (1981) Biochemistry 20,7162-7166. 24. Haschemeyer, R. H., and Haschemeyer, A. E. V. (1973) in Proteins, pp. 368-386, Wiley, New York. 25. Kuthan, H., Ullrich, V., and Estabrook, R. H. (1982) B&hem. J. 203,551-558. 26. Thornalley, P. J., and Bannister, J. V. (1985) in CRC Handbook of Methods for Oxygen Radical Research (Greenwald, R. A., Ed.), pp. 133-236, CRC Press, Boca Raton, FL. 27. Frank, D. M., Arora, P. K., Blumer, J. L., and Sayre, L. M. (1987) Biochem. Biophys. Res. Commun. 147,1095-1104.