New perspectives on the conformational equilibrium regulating multi-phasic reduction of cytochrome P450 2B4 by cytochrome P450 reductase

https://doi.org/10.1016/S0162-0134(03)00310-6Get rights and content

Abstract

The pre-steady-state reduction of cytochrome P450 (P450) 2B4 by P450 reductase (reductase) was modeled by assuming that an equilibrium between three catalytic conformers of P450 regulates the multi-phasic reduction of the enzyme. This model was compared to a model of reduction involving a minimum number of phases. Based on several criteria, the former model seems to provide an improved fit to the reduction data. Substrates were divided into two groups based on their effects at different concentrations of reductase. Surprisingly, in the presence of some substrates (group 1) but not others (group 2), the rate of reduction was actually slower with an excess of reductase than with equimolar reductase and P450. Presumably, oxidized reductase binds differently to P450 than reduced reductase. A schematic model based on two sites of interaction between reductase and P450 2B4 is offered to explain the unusual reduction kinetics with the two different groups of substrates.

Introduction

The superfamily of cytochrome P450 (P450) monoxygenases is comprised of enzymes that are expressed in virtually all species and have a wide substrate specificity including both endogenous and xenobiotic compounds [1]. Metabolism by the mammalian enzymes of this superfamily involves a complex catalytic cycle in which two electrons that are used for the mixed function oxidation of substrates are supplied by the redox partners, cytochrome b5 and/or cytochrome P450 reductase (reductase) [2]. In most instances, the first electron in the cycle is delivered by P450 reductase, whereas the second can be transferred from either redox partner [3].

The first electron reduction of P450 enzymes by P450 reductase has been well studied [4], [5], [6], [7], [8] and characterized as a biphasic process under conditions in which there is an excess of reductase relative to P450. The source of the biphasic nature of P450 reduction is controversial and has not been completely elucidated after more than 30 years of debate. At least four explanations have been proposed to explain biphasic reduction of P450 enzymes [4], [6], [7], [8].

First, it was suggested that biphasic reduction was due to the limiting amounts of P450 reductase in microsomes [6]. However, this theory was disproved when studies with purified enzymes using excess reductase still showed biphasic reduction kinetics. It was later suggested that the multiple reduction states of the reductase might reduce P450 enzymes at different rates [7]. Although completely reduced reductase did catalyze reduction of P450 faster than half-reduced enzyme, the proportion of half-reduced reductase at given NADPH concentrations did not correspond to the proportion of P450 reduced in the slow phase of the process [9]. Later, it was proposed that the rate of reduction of P450 was controlled by the spin state of the enzyme and that only the high spin form of the enzyme could be reduced by P450 reductase [8]. Thus, it was believed that the slow phase of reduction was dependent on the relatively slow conversion of low spin P450 to the high spin form. With certain P450 enzymes [5], [10], this relationship seems to hold true. For instance, the reduction potential of high spin P450cam is significantly higher than that of the low spin enzyme [10]. However, this relationship cannot be generalized across the entire P450 superfamily because many isoforms clearly display no correlation between the rate of reduction and the proportion of high spin P450 [11]. A negative correlation between reduction rate and high spin enzyme has even been observed in some isoforms [12]. Furthermore, the slow phase of reduction of P450 has not been shown to correlate with the rate of the spin state transition because the latter is an extremely rapid process [4].

Ultimately, Backes and Eyer [4] concluded that the slow phase of reduction of P450 2B4 could be attributed to a fraction of the P450 being in an “incompetent” conformation that was not able to form a functional complex with the P450 reductase. Thus, the slow phase of reduction was dependent on a relatively slow transition to the competent conformation. Substrates were shown to perturb the equilibrium between these putative conformers in favor of the functional conformation. However, Backes and Eyer did not speculate on the nature of this conformational equilibrium.

The Backes and Eyer study [4] also showed that the rate of association of P450 2B4 and reductase is dependent on substrate and is proportional to the rate of P450 2B4 reduction. In the previous study in this journal issue [30], the pre-steady-state reduction of P450 2B4 required at least three phases. We used equimolar reductase and P450 concentrations in the experiment. Thus, there was a significant amount of unbound P450 and reductase at the start of the reduction. The extra phase needed for reduction was attributed to the binding and reduction of free P450 by reductase. If this assumption is correct, the phase should disappear when the reductase concentration is saturating. In this study, we compared P450 2B4 reduction with different concentrations of reductase in an attempt to identify a phase representing unbound P450 at sub-saturating reductase concentrations.

In the previous work [30], we also obtained data suggesting that progression through the P450 catalytic cycle may also be limited by a conformational equilibrium. We proposed that the rate of change from one conformation to another is inhibited dramatically when reduced P450 reductase is bound to the P450. Furthermore, we speculated that there was a mixture of three functionally significant conformations of P450 at each stage of the catalytic cycle. One conformation is required for reduction of the ferric intermediate of P450 by the P450 reductase. Another conformation allows the reductase to deliver the second electron in the catalytic cycle and reduce the oxyferrous intermediate of P450 2B4. The third conformation is compatible with the perferryl iron-oxo intermediate. This conformation facilitates effective oxidation of the substrate. This model is supported by recent work using monochromatic X-ray diffraction to analyze freeze-trapped intermediates of the P450 cam catalytic cycle which demonstrates discrete conformations of the enzyme at three different stages of the cycle [13]. In the present study, we test the possibility that the conformation changes proposed to be involved in the rate-limiting steps of catalysis also regulate the multi-phasic pre-steady-state reduction of P450 (described above).

More specifically, it is proposed that the equilibrium between three catalytically significant conformations is influenced by substrate and redox state of the heme group. We propose that the fast phase of reduction is attributed to P450 enzymes that exist in the conformation favored by the ferric species when the enzyme is bound to reduced reductase. According to the model used to explain our findings in the preceding study, the slow phase(s) of reduction of ferric P450 2B4 result from reduced reductase being bound to P450 forms that are compatible with other intermediates of the catalytic cycle (e.g. oxyferrous or perferryl intermediates). Thus, the conformation regulates the functioning of the enzyme at each respective stage of the cycle. In other words, ferric P450 cannot be reduced until the P450 adopts the form that is favored by the ferric species in the putative conformational equilibrium. Because we propose three functional conformations of P450 2B4, there could be three phases of reduction (not including that attributed to subsaturating reductase concentrations (discussed above)) unless the reduced reductase binds equally to the other conformers. It seems possible that two closely overlapping phases of reduction might be interpreted as one slow phase of reduction. One of the objectives of this paper is to model the pre-steady-state reduction of P450 2B4 by reductase using three phases that presumably correspond to the putative catalytic conformers. This type of model is then compared to the conventional biphasic reduction routinely used in the literature. We looked for substrate-related correlations in the data in order to test the validity of the two models. In addition, we measured reduction with both saturating and sub-saturating reductase concentrations, and these data from both models were compared to the substrate-dependent reductase and P450 dissociation constants calculated in the preceding study.

As shown from previous studies [4] and from the data herein, substrate alters P450 and subsequently affects reduction by reductase on a time scale that is too rapid to measure on a stopped flow spectrophotometer. This would seemingly indicate that the substrate-related changes are not involved in the regulation of P450 reduction. However, according to our postulate, these rapid substrate-induced changes may occur in the time preceding reduction of the reductase by NADPH. Once the reductase is reduced and bound to P450, the rate of conformational equilibrium is inhibited. At this point, reduction of the P450 in unfavorable conformations (those related to the oxyferrous and perferryl species, respectively) would require the relatively slow dissociation of P450 and reductase before these conformers could rapidly change to that favoring the reduction of the ferric P450.

Two constraints are required to accommodate the model from our previous manuscript. First, the rate of equilibrium for the catalytic conformers of P450 is dramatically inhibited by the binding of reduced P450 reductase. Second, cytochrome b5- and substrate-related changes in the dissociation constant for P450 reductase and P450 2B4 interaction are mediated through changes in the rate of association of the proteins (kon). Our model predicts that the duration of binding between the proteins (koff) is considered to be relatively constant with or without cytochrome b5 and substrate. Otherwise, the putative conformation changes needed for progression of catalysis would be delayed.

By comparing the pre-steady-state reduction of P450 2B4 at subsaturating and saturating concentrations of P450 reductase, evidence is presented that oxidized reductase does indeed bind and interact with P450 differently from reduced reductase. In addition, when reduction was modeled as a triphasic phenomenon that is consistent with an equilibrium between three distinct conformers, rate constants for the slow phases of reduction were unaltered in the presence of different substrates. This is to be expected if the dissociation of the reductase–P450 complex limits the rate of reduction in the slow phases and if the substrates do not alter the koff for the interaction of the proteins. Thus, this paper provides evidence for all of the major constraints of this model proposed in the preceding study in this journal.

Section snippets

Materials

Benzphetamine, aminopyrine, resorufin, l-α-dilauryl-sn-glycero-3-phosphocholine (DLPC), and NADPH were purchased from Sigma Chemical Co. (St. Louis, MO, USA). Benzyloxyresorufin (BRF), 7-ethoxy-4-trifluoromethylcoumarin (7EFC), and 7-hydroxy-4-trifluoromethylcoumarin (7HFC) were purchased from Molecular Probes (Eugene, OR, USA). Formaldehyde (37%), hydrogen peroxide (30%), and solvents of the highest purity available were purchased from Fisher Scientific (Pittsburgh, PA, USA).

Enzyme sources

P450 2B4 was

Reduction of P450 2B4 in the presence of saturating and sub-saturating concentrations of P450 reductase

As stated in the preceding study of this journal, the reduction of P450 2B4 is comprised of an additional phase when equimolar concentrations of reductase and P450 were used. We speculated that the binding of the reductase to the free P450 2B4 was rate-limiting and that this was responsible for the slower reduction relative to the P450 that was bound to reductase when the NADPH was added. In order to test this hypothesis, we measured the rate of reduction of P450 2B4 in the presence of

Comparison of kinetic modeling using the equilibrium hypothesis and a minimum number of phases

The hypothesis proposed in the preceding manuscript invokes three different catalytic conformers in the regulation of catalysis. Fig. 3 shows that the rates of reduction with equimolar amounts of reductase and P450, as measured by the sum of the products of the rate constants and the proportions of P450 reduced in the respective phases (Table 2), also seem to correlate with the rates of metabolism (taken from Table 1 of Ref. [21]). The fact that the putative conformations invoked in our model

Abbreviations

    7EFC

    7-ethoxy-4-trifluoromethyl-coumarin

    BZP

    benzphetamine

    Ap

    aminopyrine

    BRF

    7-benzyloxyresorufin

    PCA

    protocatechuic acid

    PCD

    protocatechuate 3,4-dioxygenase

    N.D.

    not detected

Acknowledgements

The authors would like to thank Dr. David P. Ballou for technical assistance in the experiments with stopped flow spectrometry. Dr. John Teiber is thanked for helpful discussion. We also thank Hsia-lien Lin for the purification of enzymes used in this study. This work was supported, in part, by NIH grant CA-16954.

References (30)

  • T.D. Porter et al.

    J. Biol. Chem.

    (1991)
  • W.L. Backes et al.

    Biol. Chem.

    (1989)
  • J.A. Peterson et al.

    J. Biol. Chem.

    (1976)
  • W.L. Backes et al.

    J. Biol. Chem.

    (1988)
  • S.G. Sligar et al.

    Biochem. Biophys. Res. Commun.

    (1979)
  • C.S. Eyer et al.

    Arch. Biochem. Biophys.

    (1992)
  • D.A. Haugen et al.

    J. Biol. Chem.

    (1976)
  • A.L. Shen et al.

    J. Biol. Chem.

    (1989)
  • T. Omura et al.

    J. Biol. Chem.

    (1964)
  • E.S. Roberts et al.

    Arch. Biochem. Biophys.

    (1995)
  • C. Bull et al.

    J. Biol. Chem.

    (1981)
  • J.R. Reed et al.

    J. Inorg. Biochem.

    (2003)
  • D.R. Davydov et al.

    Biochimie

    (1996)
  • C.A. Hasemann et al.

    Structure

    (1995)
  • B.C. Kunz et al.

    Biochim. Biophys. Acta

    (1991)
  • Cited by (0)

    View full text