Mutation effects of a conserved threonine (Thr243) of cytochrome P450nor on its structure and function

Journal of Inorganic Biochemistry 82 (2000) 103–111 www.elsevier.nl / locate / jinorgbio

Mutation effects of a conserved threonine (Thr243) of cytochrome P450nor on its structure and function a a a b a, Eiji Obayashi , Hideaki Shimizu , Sam-Yong Park , Hirofumi Shoun , Yoshitsugu Shiro * a

b

RIKEN Harima Institute /SPring-8, 1 -1 -1 Kouto, Mikazuki-cho, Sayo, Hyogo 679 -5148, Japan The Institute of Applied Biochemistry, University of Tsukuba, Tsukuba, Ibaraki 305 -8572, Japan Received 3 March 2000; received in revised form 1 August 2000; accepted 4 August 2000

Abstract Threonine 243 of cytochrome P450nor (fungal nitric oxide reductase) corresponds to the ‘conserved’ Thr in the long I helix of monooxygenase cytochrome P450s. In P450nor, the replacement of Thr243 with Asn, Ala or Val makes the enzymatic activity dramatically reduce. In order to understand the roles of Thr243 in the reduction reaction of NO by P450nor, the crystal structures of three ˚ resolution and at cryogenic Thr243 mutants (Thr243→Asn, Thr243→Val, Thr243→Ala) of P450nor were determined at a 1.4-A temperature. However, the hydrogen-bonding pattern in the heme pocket of these mutants is essentially similar for that of the WT enzyme. This suggests that the determination of the structure of the NADH complex of P450nor is required, in order to evaluate the role of Thr243 in its enzymatic reaction. We attempted to crystallize the NADH complex under several conditions, but have not yet been successful.  2000 Elsevier Science B.V. All rights reserved. Keywords: Mutation effects; Conserved threonine; Cytochrome P450nor; Structure; Function

1. Introduction Cytochrome P450nor (CYP 55A1), which has been isolated from denitrifying fungi [1–4], is a thiolate-bound hemoprotein with a molecular weight of 46 000. This enzyme is classified as a member of the P450 superfamily, since its primary structure has an approximately 25% identity to a number of other bacterial P450s [2]. The crystal structure of the protein is essentially the same as those of other P450s, which are available thus far [5]. However, P450nor is not a monooxygenase, but, rather, a nitric oxide (NO) reductase [1]: 2NO1NAD(P)H1 1 1 H →N 2 O1H 2 O1NAD(P) . Based on spectroscopic and kinetic data on this reaction, we have proposed that the overall enzymatic reaction (scheme 1) consists of three chemical reactions [6]; Fe 31 1 NO → Fe 31 NO

(1)

Fe 31 NO 1 NADH →I] 1 NAD 1

(2)

I] 1 NO 1 H 1 → Fe 31 1 N 2 O 1 H 2 O

(3)

*Corresponding author. Fax: 181-791-58-2818. E-mail address: [email protected] (Y. Shiro).

where I] is designated as the reaction transient intermediate which is generated in the catalytic reaction. The formation 31 of I, ] i.e,. the NADH-dependent reduction of the Fe NO form, is the rate-determining step in the overall reaction. Recent time-resolved resonance Raman investigations suggest that the intermediate is possibly in a Fe 21 (NO)2 H 1 state, which is suitable for the subsequent N–O bond cleavage and N–N bond formation in the third reaction step [7]. In order to understand the molecular mechanism of the NO reduction in structural terms, we have also evaluated the crystal structures of P450nor in several states [5,8]. In the structure, it was found that two characteristic amino acid residues, Ser286 and Thr243, are located in the heme active site. The hydroxyl group of Ser286 is hydrogenbonded to a water molecule (Wat74) just adjacent to the iron-bound NO. Through the specific hydrogen-bonding network of Ser286 and Wat74 with Asp393 and other water molecules, the active site is connected to the solvent region. The importance of the Wat74 and the Ser286 / Asp393 hydrogen-bonding network in the P450nor function has been discussed on the basis of a number of different sets of experimental data [5,8]. For example, in the case of the Ser286Val mutant, which did not show any of the NO reduction activity, Wat74 was removed from its

0162-0134 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0162-0134( 00 )00161-6

104

E. Obayashi et al. / Journal of Inorganic Biochemistry 82 (2000) 103 – 111

original position and the Ser286 /Asp393 hydrogen-bonding network was disrupted, resulting in the disconnection of the active site to the solvent region. As a result, we proposed that the Wat74 and the Ser286 /Asp393 hydrogenbonding network act as a proton donor and a proton delivery pathway, respectively, during the NO reduction reaction by P450nor. Thr243 is present on the long I helix. The hydroxyl group of Thr243 is located in the direction of the sixth iron site, and is hydrogen-bonded to several water molecules [5]. Since the hydroxyl group of Thr243 interacts with the main-chain carbonyl oxygen of Ala239 through an abnormal hydrogen-bonding pattern, the I helix is distorted at this position. In the crystal structure, which was determined at cryogenic temperature, the hydrogen-bonding network, including Thr243, is connected to the solvent water [8]. Our recent mutagenesis work showed that the replacement of Thr243 with another amino acid residues significantly affects the enzymatic activity of P450nor [9]. This finding suggests an important role for Thr243 in the NO reduction reaction. The position of Thr243 of P450nor corresponds to that of the ‘conserved’ Thr residues in monooxygenase P450s [2,5]. The conserved Thr of P450 is thought to play a crucial role in the proton delivery in monooxygenase reactions [10–19]. When the conserved Thr of P450s, e.g., Thr252 of d-camphor hydroxylase cytochrome P450cam, was modified by the mutagenesis, the resultant mutants do not exhibit the monooxygenation reactivity, but produce hydroperoxide (H 2 O 2 ) as a reduced O 2 product (uncoupling reaction) [13–16]. In the crystal structure of Thr252Ala mutant of P450cam, location of water molecules and hydrogen-bonding networks in the heme active site were disturbed, and consequently the proton transfer to the iron-bound O 2 cannot be systematically controlled in the catalytic cycle [15]. In an effort to understand the roles of Thr243 of P450nor in the NO reduction, we determined the crystal structures of the Thr243 mutants (Thr243→Asn, Thr243→Val, Thr243→Ala) of P450nor in the ferric–NO state at a high resolution and at cryogenic temperature.

state were prepared by previously reported methods, and were converted to the corresponding NO adducts [8]. All crystals were flash-frozen in liquid nitrogen for both storage and data collection at cryogenic temperatures (100 K).

2.2. X-ray crystallography Analysis of the diffraction images showed that the crystals of the recombinant enzymes were isomorphous to that of the fungal enzyme. High-resolution data were obtained at the synchrotron radiation source of the BL44B2 (Riken Beam Line 2) station in the SPring-8 (Harima, Japan). Data collection, processing, and refinement were performed according to methods reported previously [5,8,21], and statistics are given in Table 1. The structures of the native fungal enzyme in the P2 1 2 1 2 1 space group (PDB ID: 1CL6), without the coordinates of the 243 residue and water molecules, were used as an initial model for refinement of the structures of the mutants. After a first round of rigid-body, simulated annealing and B-factor refinements, a u2Fo 2Fc u map was calculated. A high electron density was found at the 243 position, and manual rebuilding of the 243 residue and water molecules were then carried out. Structural evaluations of the final models of the WT and the mutant enzymes using PROCHECK indicated that more than 92% of the residues are in the most favorable regions of the Ramachandran plot, and that no residues are in disallowed regions. Coordinates of the ferric–NO complexes of T243N, T243V and T243A mutants have been deposited at the Brookhaven Databank: ID codes 1F25, 1F26, and 1F24, respectively.

2.3. Spectral measurements Absorption spectra were measured at 208C using a Hitachi U-3000 spectrophotometer. The resonance Raman spectra were measured using a Jasco NR-1800 spectrometer; details of the procedure have been described previously [20].

2. Materials and methods

3. Results and discussion

2.1. Enzyme and crystal preparations

3.1. Preparation and measurement of activities of Thr243 mutants of cytochrome P450 nor

The WT and mutant (T243A, T243V, T243N) enzymes of P450nor were expressed using the expression system of the recombinant enzyme in E. coli JM109, which had been constructed previously, and were purified according to procedures described elsewhere [5,20]. The catalytic activities of the enzymes were assayed by spectrophotometrically monitoring the rate of the NADH consumption [6]. The concentration of NADH was determined from the absorbance at 340 nm using a molar absorption coefficient of 6.22 mM 21 cm 21 . Single crystals in the ferric resting

The Thr243 mutants of P450nor were prepared using our expression system [5,8], which is different from that constructed by Okamoto et al. [9]. The recombinant proteins prepared by Okamoto et al. were not suitable for the preparation of single crystals, because their N-termini were modified [9]. The enzymatic activities of our Thr243 mutants were measured by spectrophotometrically following the rate of NADH consumption in the NO reduction reaction (Fig. 1). We previously reported that the activity

E. Obayashi et al. / Journal of Inorganic Biochemistry 82 (2000) 103 – 111

105

Table 1 Crystal parameters, data collection and structure refinement of the Thr243 mutants of cytochrome P450nor Data set

T243N

T243A

T243V

˚ Resolution range (A) ˚ Unit cell dimensions (A)

100–1.4 a554.56 b581.92 c585.70

100–1.4 a554.58 b581.76 c585.64

100–1.4 a554.57 b581.94 c585.53

685 047 / 76 306

696 775 / 76 100

586 895 / 72 706

100 / 100

99.9 / 100

99.6 / 97.5

3.8 / 12.9

3.8 / 13.0

4.4 / 13.0

Refinement statistics ˚ Resolution range (A) s cut-off R factor (%)b Free R factor (%) Solvent

10–1.4 0.0 14.2 20.4 586

10–1.4 0.0 13.9 20.7 626

10–1.4 0.0 12.8 18.5 638

Rms deviations from ideals ˚ bond lengths (A) ˚ bond angles (A)

0.010 2.074

0.010 2.135

0.010 2.141

Ramachandran plot residues in most favorable regions (%) residues in additional allowed regions (%) residues in disallowed regions (%)

91.7 8.3 0.0

92.0 8.0 0.0

91.3 8.7 0.0

Reflections, measured / unique Completeness (%), overall / outer shell (1.48–1.40) R merge (%)a overall / outer shell (1.48–1.40)

a

R merge 5SuIi 2kIlu / SuIi u, where Ii is the intensity of an observation and kIl the mean value for that reflection and the summations are overall reflections. The free R factor was calculated with 5% of the data. b R factor5S h uuFo (h)u2uFc (h)uu / S h Fo (h), where Fo and Fc are the observed and calculated structure factor amplitudes, respectively.

thus measured is nearly comparable to that obtained by measurement of the rate of formation of N 2 O [6]. The mutants showed very low levels of NO reduction activity, compared with the wild-type (WT) enzyme: WT (100%), T243N (2%), T243V (0.01%) and T243A (3%),1 indicating the functional importance of Thr243 in the NO reduction. 1

Fig. 1. Time courses for the absorption changes of NADH during the reduction of NO catalyzed by the WT and some Thr243 mutant enzymes of P450nor: (a) WT, (b) T243N, (c) T243A and (d) T243V. The catalytic conditions are: [NADH]50.16 mM, [NO]51 mM and [Enzyme]510 nM.

In the previous study, Okamoto et al. prepared the Thr243Asn mutant of P450nor using an expression system that they constructed by themselves [9]. In this system, the isolated recombinant protein contains a modified amino acid sequence at N-terminal region, MALLLAVFP instead of MASGAPSFP, in which the hydrophobic tail, LLLAV, was present instead of SGAPS. When the NO reduction activity of the T243N mutant was measured by following the NADH consumption, it contained 60% of the activity of the WT enzyme. The mutant gives the Fe–CO and the Fe–NO stretching frequencies at 487 and 529 cm 21 in the resonance Raman spectra, respectively. These results are surprisingly different from those of the T243N that we prepared in this study; the present T243N mutant is only 2% active, and gives the Fe–CO and the Fe–NO stretching frequencies at 490 and 530 cm 21 , respectively. At present, it is unclear why such differences were observed in the T243N mutants upon the N-terminal modification. However, since the N-terminal region of the native enzyme is exposed to the solvent region [5], the hydrophobic tail of the Okamoto et al.’s T243N mutant might not be located at the original position, possibly affecting sterically and / or electrostatically the heme pocket structure. The present crystallographic analysis clearly showed structure at mutated site of the T243N enzyme we prepared, as shown in Fig. 2, and the N-terminal structure is completely the same as that of the native enzyme. Therefore, we will discuss the effects of the Thr243 mutation on the structure and function of P450nor on the basis of the observations that we present in this study.

E. Obayashi et al. / Journal of Inorganic Biochemistry 82 (2000) 103 – 111

106

3.2. Resonance Raman spectra of ferric–NO complexes of Thr243 mutants of cytochrome P450 nor The resonance Raman spectra of their ferric–NO and ferrous–CO complexes of the Thr243 mutants were collected, in order to characterize their heme environmental structures. The stretching and bending frequencies of the Fe–ligand in the enzymes are compiled in Table 2. The Fe–NO stretching is observed at 530 cm 21 for all the enzymes. Since the Fe–NO stretching is sensitive to steric effects on the iron-bound NO [8,20,22], this result indicates that the environment in the vicinity of the iron-bound NO is sterically the same between the WT and the mutant enzymes. The Fe–CO stretching is located in the region of 490–487 cm 21 . The Fe–CO stretching serves as a marker of the hydrophobicity or the electronic field in the vicinity of the iron-bound CO [20]. The chemical environments of the iron-bound CO are subtly different between the WT and the mutant enzymes. Correspondingly, the Fe–C–O bending frequency, which is observed in the region of 557–559 cm 21 , is also subtly changed as the result of Thr243 mutation.

3.3. Crystal structures of ferric–NO complexes of Thr243 mutants of cytochrome P450 nor

Fig. 2. (a) Structures of the side chains of the 243 residues of the WT and the Thr243 mutant enzymes of P450nor. (b) Structures of the I helices of the mutants around the 243 residue: red, WT; green, T243N; yellow; T243A; and blue, T243V.

In the crystal structures of the mutant enzymes of P450nor, the mutated sites (Asn243, Val243 and Ala243) are well characterized, as illustrated in Fig. 2a. However, the overall structure of the enzymes and topologies of the secondary structures were not significantly altered as a result of the Thr243 mutation. The distortion of the I helix was also observed in all Thr243 mutants, as is the case for the WT enzyme (Fig. 2b). Fig. 3 shows the final models of the heme environments of the Thr243 mutants, including a map of the electron densities of the water molecules present in the heme pockets. In Fig. 4, schematic representations of the active site of the final refined models for the Thr243 mutants are shown, and compared with that of the WT enzyme [8]. ˚ away from the iron-bound NO Wat74 is present about 3 A in all the structures. This water molecule is hydrogenbonded to the hydroxyl group of Ser286, and is then connected with the solvent through the Asp393 carboxylate

Table 2 The Fe–NO and Fe–CO stretching, and the Fe–C–O bending frequencies (cm 21 ) of the WT and the Thr243 mutant enzymes of cytochrome P450nor

WT T243N T243A T243V

nFe–NO

nFe–CO

dFe–C–O

530 530 530 530

488 490 487 488

557 559 557 557

Fig. 2. (continued)

E. Obayashi et al. / Journal of Inorganic Biochemistry 82 (2000) 103 – 111

group and some water molecules. The Thr243 mutation does not disturb the Ser286 /Asp393 hydrogen-bonding network. Therefore, the active sites of the Thr243 mutants of P450nor as well as the WT enzyme are connected with the solvent through the Ser286 /Asp393 hydrogen bonding networks. In the case of the WT enzyme, we conclude that the Wat74 is a direct proton donor to the iron-bound NO, and that the Ser286 /Asp393 hydrogen bonding network serves as a proton delivery pathway from the solvent to the active site [8]. In the WT enzyme of P450nor, the hydroxyl group of Thr243 is hydrogen-bonded with the main-chain carbonyl of Ala239 through a water molecule (Wat46), and is also involved in the hydrogen-bonding network (the Thr243 hydrogen-bonding network: Thr243–Wat65–Ala285– Wat37–Asn246–solvent). The active site of the enzyme is connected with the solvent through the Thr243 hydrogenbonding network. In the case of the T243N mutant (Fig. 4b), these structural characteristics are also observed, since the amide group of Asn243, as well as the hydroxyl group of Thr243, is capable of entering into hydrogen-bonding interactions. In T243A mutant (Fig. 4c), since the methyl group of Ala243 does not have the hydrogen-bonding ability, the hydrogen-bonding pattern at this site is different from that of the WT enzyme. Since new water molecules (Wat540, Wat494) are present in the hydrogen-bonding interactions, the active site of the T243A mutant is also connected with the solvent through a hydrogen-bonding network (Wat46– Wat65–Wat540–Wat494–Wat37–Asn246–solvent). Similarly, in the T243V mutant (Fig. 4d), new water molecules, Wat233 and Wat283, take part in the hydrogen-bonding that connects the active site of the enzyme with the solvent. Although the Ala243 or Val243 residues in each mutant are isolated from the solvent region, the active site of the mutant enzymes is not.

3.4. Role of Thr243 in P450 nor reaction The structural characteristics of the Thr243 mutants discussed above are basically similar to that of the WT enzyme of P450nor in the ferric–NO form. In spite of the considerable structural similarities, the mutants are barely able to reduce NO to N 2 O. Therefore, the ‘conserved’ Thr243 must play a crucial role in the enzymatic cycle of P450nor. However, the role might be different from those of other P450s, due to structural differences at the conserved Thr residue between them. Roles of the ‘conserved’ Thr in the P450 reactions have been extensively discussed on the basis of comprehensive results on mutational and / or crystallographic experiments [10–19]. Their OH group does not face to the sixth iron coordination site, therefore making a hydrophobic environment. In cryo-crystal structure of the oxy-complex of d-camphor hydroxylase P450cam reported by Schlichting

107

et al., one ordered water molecule, Wat901, is found to be hydrogen-bonded with both of the hydroxyl group of conserved Thr252 and the iron-bound molecular oxygen [19]. The hydrogen-bonding network is extended to the protein surface through some amino acid residues, and the only pathway connecting the active site of P450cam with the solvent water. Indeed, mutation of the residues involved in this hydrogen-bonding network, such as Thr252, Asp251 and so on, dramatically effect on the activity of the enzyme. It is now considered that the hydrogen-bonding network including the conserved Thr252 in P450cam stabilizes the direct proton donor, Wat901, and serves as a proton delivery pathway when the O–O bond is cleaved. On the other hand, the heme environment of P450nor is quite hydrophilic, and a couple of water molecules are located in the distal heme pocket. The OH group of Thr243 points to the iron-bound ligand, and is hydrogen-bonded with two water molecules, Wat46 and Wat65. The hydrogen-bonding network is extended to the protein surface, as the case for P450cam. However, Wat46 and Wat65 as well as the OH group of Thr243 of P450nor do not directly interact with the iron-bound NO ligand, and, therefore, the hydrogen-bonding network including Thr243 does not connect the active position of the enzyme with the solvent water. On the basis of the observation, we ruled out a possibility of the proton delivery pathway as the role of Thr243 of P450nor. Since the hydroxyl group of Ser286 is hydrogen-bonded to Wat74 just adjacent to the iron-bound NO, the Ser286 /Asp393 hydrogen-bonding network is now postulated as the proton delivery pathway in the NO reduction by P450nor [5,8]. Unfortunately, we cannot conclude the role of the conserved Thr243 in the NO reduction by P450nor on the basis of present crystallographic studies, which only pointed out its importance in the enzymatic reaction. One possibility is that Thr243 might be related to the electron transfer from the electron donor, NADH, to the heme site, or to the NADH binding site to the enzyme. Therefore, we tried to compare structures of NADH-bound form of the enzyme between the WT and the Thr243 mutants of P450nor.

3.5. Attempts to crystallize the NADH complex of cytochrome P450 nor As the results of our previous studies, we proposed a molecular mechanism for the reduction of NO by P450nor (Eqs. (1)–(3)). In this mechanism, we assumed the formation of a ternary complex of the enzyme, NO and NADH. This assumption is based on the observation that the reduction of NO by P450nor does not require any redoxproteins to mediate the electron-transfer from NADH to the heme site [1], and that the ferric–NO complex of P450nor is reduced with two electrons from NADH, to yield the transient intermediate, ]I [7]. In addition, the

108

E. Obayashi et al. / Journal of Inorganic Biochemistry 82 (2000) 103 – 111

Fig. 3. Stereo view. Structures of the active site of the ferric–NO complexes of the WT and the Thr243 mutant enzymes of P450nor: (a) WT, (b) T243N, (c) T243A and (d) T243V.

denitrifying fungal P450nor’s discriminate between NADH and NADPH as an electron donor; the enzyme from F. oxysporum shows a specificity toward NADH, while that from Cylindrocarpon tonkinense has a specificity for NADPH [23]. Thus, it has been suggested that the NADH binding-site is present in P450nor itself [21]. However, the

crystal structure of the NADH-bound form of P450nor has not yet been determined. We have attempted to obtain crystals of the NADHbound form under a variety of conditions: e.g., the diffusion of NADH into the crystals of the ferric and ferrous–CO forms of P450nor, at a variety of NADH

E. Obayashi et al. / Journal of Inorganic Biochemistry 82 (2000) 103 – 111

109

Fig. 3. (continued)

concentrations (10, 50, 100 mM) and co-crystallization of the enzyme with NADH (3–200 mM). We also tried to use NADH analogs such as nicotine amide, ADP and NAD 1 , but have not yet been successful in obtaining crystals. No additional electron density from NADH and its analogs was observed in the map (data not shown). One of the possible reasons for this is that the structure of the NADH-

bound form of P450nor might be dramatically different from that of the NADH-free enzyme. However, this possibility seems unlikely, because the turnover number of the catalytic reaction by P450nor is so high (.1000 s 21 ) [1,6] that a large conformational change is not possible during the catalytic reaction. Another possibility is that NADH might weakly interact with the enzyme during the

110

E. Obayashi et al. / Journal of Inorganic Biochemistry 82 (2000) 103 – 111

Fig. 4. Schematic representation of the top view of the heme pocket of the WT and the Thr243 mutant enzymes of P450nor, showing the hydrogen-bonding networks: (a) WT, (b) T243N, (c) T243A and (d) T243V.

catalytic reaction. Since the addition of NADH to the enzyme solution barely changes the optical, Raman and IR spectra of P450nor, we have not yet obtained the dissociation constant, Kd , of NADH. In the enzymatic analyses of the NO reduction by P450nor (,10 nM), the Michaelis– Menten constant, Km , of NADH was estimated to be about 0.2 mM [1]. The weak interaction of NADH might permit the high turnover of the catalytic reaction of P450nor, but

might not lead to a single crystal of the NADH-bound enzyme.

Acknowledgements We thank Dr. Shin-ichi Adachi for his assistance in the X-ray diffraction measurements at BL44B2 (RIKEN BL2)

E. Obayashi et al. / Journal of Inorganic Biochemistry 82 (2000) 103 – 111

of SPring-8. This work was supported, in part, by a grant-in-aid for Scientific Research on Priority Areas (Molecular Metallics) from the Ministry of Education, Science, Culture, and Sports of Japan (to H. S. and Y. S.).

References [1] K. Nakahara, K. Tanimoto, K. Hatano, K. Usuda, H. Shoun, J. Biol. Chem. 268 (1993) 8350–8355. [2] H. Kizawa, D. Tomura, M. Oda, A. Fukamizu, T. Hoshino, O. Gotoh, T. Yasui, H. Shoun, FEBS Lett. 266 (1991) 10632–10637. [3] H. Shoun, Y. Sudo, Y. Seto, T. Beppu, J. Biochem. 94 (1983) 1219–1229. [4] H. Shoun, T. Tanimoto, J. Biol. Chem. 266 (1991) 11078–11082. [5] S.-Y. Park, H. Shimizu, S. Adachi, A. Nakagawa, I. Tanaka, K. Nakahara, H. Shoun, H. Nakamura, T. Iizuka, Y. Shiro, Nat. Struct. Biol. 4 (1997) 827–832. [6] Y. Shiro, M. Fujii, M.T. Iizuka, S. Adachi, S.K. Tsukamoto, K. Nakahara, H. Shoun, J. Biol. Chem. 270 (1995) 1617–1623. [7] E. Obayashi, S. Takahashi, Y. Shiro, J. Am. Chem. Soc. 120 (1998) 12964–12965. [8] H. Shimizu, E. Obayashi, Y. Gomi, H. Arakawa, S.-Y. Park, H. Nakamura, S. Adachi, H. Shoun, Y. Shiro, J. Biol. Chem. 275 (2000) 4816–4826. [9] N. Okamoto, Y. Imai, H. Shoun, Y. Shiro, Biochemistry 37 (1997) 8839–8847. [10] J.A. Peterson, S.E. Graham-Lorence, in: P.R. Ortiz de Montellano (Ed.), Cytochrome P450. Structure, Mechanism, and Biochemistry, 2nd Edition, Plenum Press, New York, 1995, pp. 151–180.

111

[11] T.L. Poulos, J. Cupp-Vickery, H. Li, in: P.R. Ortiz de Montellano (Ed.), Cytochrome P450. Structure, Mechanism, and Biochemistry, 2nd Edition, Plenum Press, New York, 1995, pp. 125–150. [12] E.J. Mueller, P.J. Loida, S.G. Sligar, in: P.R. Ortiz de Montellano (Ed.), Cytochrome P450. Structure, Mechanism, and Biochemistry, 2nd Edition, Plenum Press, New York, 1995, pp. 83–124. [13] M. Imai, H. Shimada, Y. Watanabe, Y. Matsushima-Hibiya, R. Makino, H. Koga, T. Horiuchi, Y. Ishimura, Proc. Natl. Acad. Sci. USA 86 (1989) 7823–7827. [14] S.A. Martinis, W.M. Atkins, P.S. Stayton, S.G. Sligar, J. Am. Chem. Soc. 111 (1989) 9252–9253. [15] R. Raag, S.A. Martinis, S.G. Sligar, T.L. Poulos, Biochemistry 30 (1991) 11420–11429. [16] Y. Kimata, H. Shimada, T. Hirose, Y. Ishimura, Biochem. Biophys. Res. Commun. 208 (1995) 96–102. [17] N.C. Gerber, S.G. Sligar, J. Biol. Chem. 269 (1994) 4260–4266. [18] M. Vidakovic, S.G. Sligar, H. Li, T.L. Poulos, Biochemistry 37 (1998) 9211–9219. [19] I. Schlichting, J. Berendzen, K. Chu, A.M. Stock, S.A. Maves, D.E. Benson, R.M. Sweet, D. Ringe, G.A. Petsko, S.G. Sligar, Science 287 (2000) 1615–1622. [20] E. Obayashi, K. Tsukamoto, S. Adachi, S. Takahashi, N. Nomura, T. Iizuka, H. Shoun, Y. Shiro, J. Am. Chem. Soc. 119 (1997) 7807– 7816. [21] H. Shimizu, S.-Y. Park, D.-S. Lee, H. Shoun, Y. Shiro, J. Inorg. Biochem. 81 (2000) 191–205. [22] U.P. Singh, E. Obayashi, S. Takahashi, T. Iizuka, H. Shoun, Y. Shiro, Biochim. Biophys. Acta 1384 (1998) 103–111. [23] N. Toritsuka, H. Shoun, U.P. Singh, S.-Y. Park, T. Iizuka, Y. Shiro, Biochim. Biophys. Acta 1338 (1997) 93–99.