High-resolution crystal structures of the solubilized domain of porcine cytochrome b5
aQuantum Beam Science Center, Japan Atomic Energy Agency, 2-4 Shirakata, Tokai, Ibaraki 319-1195, Japan, and bDepartment of Biomolecular Functional Engineering, Faculty of Engineering, Ibaraki University, 4-12-1 Nakanarusawa, Hitachi, Ibaraki 316-8511, Japan
*Correspondence e-mail: firstname.lastname@example.org
Mammalian microsomal cytochrome b5 has multiple electron-transfer partners that function in various electron-transfer reactions. Four crystal structures of the solubilized haem-binding domain of cytochrome b5 from porcine liver were determined at sub-angstrom resolution (0.76–0.95 Å) in two crystal forms for both the oxidized and reduced states. The high-resolution structures clearly displayed the electron density of H atoms in some amino-acid residues. Unrestrained refinement of bond lengths revealed that the protonation states of the haem propionate group may be involved in regulation of the haem redox properties. The haem Fe coordination geometry did not show significant differences between the oxidized and reduced structures. However, structural differences between the oxidized and reduced states were observed in the hydrogen-bond network around the axial ligand His68. The hydrogen-bond network could be involved in regulating the redox states of the haem group.
Mammalian microsomal cytochrome b5 is anchored to the membrane of the endoplasmic reticulum and is involved in various electron-transfer reactions, such as lipid unsaturation, cholesterol synthesis and drug metabolism. Because of its diversity of function, cytochrome b5 has been reported to interact with multiple electron-transfer partners including NADH-cytochrome b5 reductase (b5R; Spatz & Strittmatter, 1973), C4-methyl sterol oxidase (Fukushima et al., 1981), cytochrome c (Ren et al., 2004), NADH-cytochrome P450 reductase and cytochrome P450 (Porter, 2002). Mammalian microsomal cytochrome b5 consists of 134 residues, in which the N-terminal ∼100 residues contain the haem-binding region located at the cytoplasmic side of the endoplasmic reticulum membrane and the C-terminal ∼30 residues contain the membrane-anchoring region and the signal sequence targeting the endoplasmic reticulum membrane (Mitoma & Ito, 1992). The haem Fe atom is coordinated by two axial ligands, His44 and His68, and the haem group is not bound covalently to the protein moiety. The numbering of amino-acid residues in this study is based on the cytochrome b5 coding region and is different from the numbering of the protease-solubilized cytochrome b5 from bovine liver reported previously (Durley & Mathews, 1996). For example, His44 and His68 in this study are His39 and His63 in the protease-solubilized bovine cytochrome b5.
The crystal structure of cytochrome b5 has been obtained at 1.5 Å resolution using the protease-solubilized haem-binding region from bovine liver (Durley & Mathews, 1996). The solution structures of the solubilized domain of cytochrome b5 (b5) have also been reported for several species including cattle (Zhang et al., 2004), rat (Banci et al., 1997; Arnesano et al., 1998) and rabbit (Banci et al., 2000). The solution structures of b5 have revealed heterogeneity in the orientation of the haem group. The haem group of rat b5 has two orientations related by a 180° rotation about the porphyrin α,γ-meso axis (Rivera et al., 1992; Altuve et al., 2004), although one of the two orientations is predominantly observed in the haem group of bovine b5. Val28 of rat b5, as opposed to Leu28 of bovine b5, is located at the bottom of the haem-binding crevasse. Differences in the haem orientations are considered to be caused by steric repulsion between the hydrophobic residues and the vinyl group of the haem. However, it is still unclear how the orientation affects the electron-transfer reaction (Walker et al., 1988). Solution structures of rat b5 have been solved in both the oxidized and reduced forms (Banci et al., 1997; Arnesano et al., 1998). A comparison of the average structures of both redox forms suggests that mobility around the haem group is important to optimize interactions with the electron-transfer partners.
Structural information on bovine b5 has been utilized for extensive mutation studies related to protein stability, haem redox potential and interaction with electron-transfer partners. Protein stability is decreased by the mutation of valine residues (Val50 and Val66) that are located at the haem edge and restrict the solvent-accessibility of the haem group (Gan et al., 2002; Wu et al., 2000). Stability is also decreased by the mutation of Phe63 involved in a π-stacking interaction with the axial ligand His68 (Wang et al., 2006). The redox potential is decreased by the mutation of Phe40 located near the haem plane (Yao et al., 2002) and by the mutation of Ser69 interacting with the haem propionate group (Funk et al., 1990). The mutation of Val50 and Val66 also influences the redox potential of the haem (Gan et al., 2002; Wu et al., 2000). Mutations of the surface acidic residues (Glu49, Glu53, Glu61 and Asp65) decrease the association constant between b5 and cytochrome c (Sun et al., 1999; Qian et al., 2001; Wu et al., 2001). Electrostatic interactions between these acidic residues and basic residues on the surface of cytochrome c have been thought to participate in formation of the b5–cytochrome c complex.
We have recently reported crystal structures of porcine b5R in both fully oxidized and reduced forms (Yamada et al., 2013). The oxidized structure determined at 0.78 Å resolution includes H atoms that elucidate a hydrogen-bonding network around FAD. The reduced structure determined at 1.68 Å resolution shows a slight shift in the relative arrangement of the FAD- and NADH-binding domains compared with the oxidized structure. Although structural studies have been performed for several mammalian b5s, structural information has not yet been obtained for porcine b5. In addition, X-ray crystal structures determined at sub-angstrom resolution have not been obtained for both the oxidized and reduced forms of b5. The high-resolution structures of both redox forms are important for understanding the structural features involved in regulation of the electronic properties of the haem group. In this work, we report the crystal structures of porcine b5 determined at sub-angstrom resolutions in both the oxidized and reduced forms. The crystal structures were obtained for two crystal forms of both redox forms. A structural comparison was performed between the two redox forms and also between the two crystal forms. The comparison using high-resolution structures revealed that structural differences around the haem group may be important for regulating the electronic properties of the haem.
The N-terminal solubilized domain of porcine cytochrome b5, consisting of amino-acid residues 1–94, was overexpressed in Escherichia coli BL21 cells containing plasmid pCPb5, as described previously (Kimura et al., 2003). The purification procedure for porcine b5 has also been reported previously (Kimura et al., 2003), but anion-exchange chromatography was additionally performed. The b5 fraction after size-exclusion chromatography was purified using a Resource Q column (GE Healthcare). Final samples were obtained in the oxidized form as deduced from the UV–Vis absorption spectrum and were concentrated to 30 mg ml−1 as calculated using an extinction coefficient of ∊413 = 1.13 × 105 M−1 cm−1. The reduced form of b5 was obtained by adding sodium dithionite to the oxidized form of b5 to a final concentration of 5 mM. To monitor the autoxidation of b5, UV–Vis absorption spectra were measured 0, 5, 10, 25, 60, 120 and 240 min after adding sodium dithionite under aerobic conditions at room temperature. The spectral measurements were performed three times.
Initial screening of crystallization conditions was performed using the oxidized form of b5 with Crystal Screen, Crystal Screen 2 (Hampton Research) and Precipitant Synergy (Molecular Dimensions). Crystals were obtained under several conditions containing polyethylene glycol (PEG) as a precipitant. After optimization of the crystallization conditions, two crystal forms were obtained by the hanging-drop vapour-diffusion method at 293 K. Form 1 crystals were obtained by mixing 1 µl protein solution with 1 µl reservoir solution consisting of 24%(w/v) PEG 1500, 4%(v/v) 2-propanol, 0.1 M calcium chloride, 0.1 M HEPES–NaOH pH 7.5. The final pH of the reservoir solution was 7.43 at 298 K. Form 2 crystals were obtained by mixing 1 µl protein solution with 1 µl reservoir solution consisting of 15%(v/v) PEG 400, 3%(w/v) PEG 3350, 0.1 M sodium acetate pH 5.5. The final pH of the reservoir solution was 5.52 at 298 K. Crystallization of the reduced form of b5 was performed in an anaerobic chamber (Anaero Box ANX-3, Hirasawa). The reduced form of b5 was prepared by adding sodium dithionite to the oxidized form in the anaerobic chamber to a final concentration of 5 mM. Crystals of the reduced form of b5 were obtained under the same conditions as the oxidized form of b5.
Before diffraction data collection, crystals were transferred to cryoprotectant conditions. Form 1 crystals of the oxidized form were soaked in a cryoprotectant solution consisting of 35%(w/v) PEG 1500, 4%(v/v) 2-propanol, 0.1 M calcium chloride, 0.1 M HEPES–NaOH pH 7.5. Form 2 crystals of the oxidized form were soaked in a cryoprotectant solution consisting of 30%(v/v) PEG 400, 3%(w/v) PEG 3350, 0.1 M sodium acetate pH 5.5. Form 1 crystals of the reduced form were soaked in a cryoprotectant solution consisting of 35%(w/v) PEG 1500, 4%(v/v) 2-propanol, 0.1 M calcium chloride, 0.1 M HEPES–NaOH pH 7.5, 10 mM sodium dithionite in the anaerobic chamber. Form 2 crystals of the reduced form were soaked in a cryoprotectant solution consisting of 30%(v/v) PEG 400, 3%(w/v) PEG 3350, 0.1 M sodium acetate pH 5.5, 10 mM sodium dithionite in the anaerobic chamber. Crystals of the oxidized form were flash-cooled in a nitrogen-gas stream at 100 K just before data collection. Crystals of the reduced form were cooled in liquid nitrogen in the anaerobic chamber and stored until diffraction data collection.
Diffraction data sets were collected on the BL5A and BL17A beamlines at Photon Factory (PF), Tsukuba, Japan. Diffraction intensities were measured using ADSC Q315 (BL5A) and ADSC Q270 (BL17A) CCD detectors. Two data sets (high resolution and low resolution) were separately collected from a single crystal at different positions. For the high-resolution data sets, the positions at which the crystals were exposed to the X-rays were changed in order to restrict the maximum dose to 3.0 × 105 Gy for each exposure position. Dose calculation was performed using RADDOSE (Paithankar et al., 2009). For the low-resolution data sets, the crystals were exposed to attenuated X-rays to avoid overloading the CCD detectors. Diffraction data sets were integrated and scaled using the HKL-2000 program package (Otwinowski & Minor, 1997). The high- and low-resolution data sets were merged to obtain a complete data set.
The crystal structure of porcine b5 was determined using the molecular-replacement method with the crystal structure of bovine b5 (PDB entry 1cyo ; Durley & Mathews, 1996) without the haem group or solvent molecules as a search model. One molecule was found in the asymmetric unit in both the form 1 and form 2 crystals. Structure refinement started with rigid-body refinement in the resolution range 8.0–1.2 Å using PHENIX (Adams et al., 2010). Positional and isotropic displacement parameters were refined in the resolution range 50–1.2 Å. Manual model fitting and incorporation of the haem group were performed using Coot (Emsley et al., 2010). Ordered water molecules were automatically included in the model when peak heights in the Fo − Fc map were detected at the 4.0σ level. Other solvent molecules were manually incorporated into the model. Anisotropic displacement parameters were then refined in the whole resolution range using SHELXL (Sheldrick, 2008). Restraints on bond angles and lengths were removed for the single-conformation residues and the haem group only when they converged to reasonable geometries after conjugate-gradient least-squares refinement. H atoms were incorporated in the model when they were observed in the hydrogen OMIT map contoured at the 1.5σ level. The bond lengths and angles of the H atoms were constrained using the riding model in SHELXL. The estimated standard deviations of bond angles and lengths were obtained by one cycle of full-matrix least-squares refinement. Figures were prepared using PyMOL (DeLano, 2002).
An autoxidation experiment showed that 50% of the reduced form of b5 was converted to the oxidized form following a 30 min incubation (Supplementary Fig. S1). Therefore, the crystallization and cooling of crystals were performed in the anaerobic chamber for the reduced form. Oxidized and reduced forms of b5 were crystallized in two crystal forms, both of which belonged to the orthorhombic space group P212121, with unit-cell parameters a = 32.3, b = 39.3, c = 61.0 Å (oxidized form 1), a = 33.6, b = 51.2, c = 55.1 Å (oxidized form 2), a = 33.6, b = 38.9, c = 60.6 Å (reduced form 1) and a = 33.7, b = 51.3, c = 55.3 Å (reduced form 2). The crystal structures were determined at resolutions of 0.83 Å (oxidized form 1), 0.93 Å (oxidized form 2), 0.76 Å (reduced form 1) and 0.95 Å (reduced form 2). Diffraction data and refinement statistics are given in Table 1.
All structures included residues Ala8–Lys94 in the final models; the remaining seven residues at the N-terminus were not observed in the 2Fo − Fc electron-density maps. The final models contained 20–28 multiple conformations of amino-acid residues. Many multiple conformations were observed for the N-terminal (Ala8–Tyr11) and C-terminal (Lys91–Lys94) residues. 48% (oxidized form 1), 27% (oxidized form 2), 49% (reduced form 1) and 17% (reduced form 2) of the H atoms in the amino-acid residues could be included in the models. In the form 1 structures, electron density derived from H atoms was observed for some amino-acid residues. Electron density for H atoms bound to N atoms was observed as well as for those bound to C atoms. In the form 2 structure, H atoms were primarily added to atoms with low Beq values. The final models of both crystal forms also contained some of the H atoms of the haem group (Table 1). Two Ca ions and one acetate ion were found in the structures of the form 1 and 2 crystals, respectively (Table 1).
The orientation of the haem group was clearly confirmed by a haem OMIT map. Figs. 1(a)–1(d) show the OMIT maps for the haem molecules in the structures of oxidized form 1, oxidized form 2, reduced form 1 and reduced form 2, respectively. The haem orientation in porcine b5 corresponds to the major isomer observed in the rat b5 structure. The residue located at the bottom of the haem-binding crevasse is Leu28 in porcine b5, which is the same as that in bovine b5 (Supplementary Fig. S2).
All four structures exhibited high similarity to each other (Figs. 2a–2d). Superposition of all Cα atoms showed root-mean-square distances between the form 1 and form 2 structures of 0.51–0.67 Å, which were larger than those between the oxidized and reduced structures in the same crystal form (0.11 and 0.31 Å) (Supplementary Table S1). When the structures of the two crystal forms were superposed on their haem-plane atoms, both the oxidized and reduced structures showed a highly similar structure around the haem groups (Figs. 2e and 2f) and structural changes were observed in the N- and C-terminal regions.
The 7-propionate group forms hydrogen bonds to the side-chain hydroxyl and main-chain amide of Ser69 in both crystal forms (Fig. 3a). Strong peaks in the haem OMIT maps clearly indicated the atomic positions of the carboxyl groups of the 7-propionate (Fig. 1). Restraints on bond angles and lengths were removed for the carboxyl groups of the 7-propionate in SHELXL refinement. The C—O distances of the carboxyl groups (C—O1 and C—O2) are listed in Table 2. In the form 1 structures, the 6-propionate group makes a contact with the Nζ atom of Lys10* (Fig. 4a; residues of the neighbouring molecule in the crystal are hereafter indicated with an asterisk). In the form 2 structures, the 6-propionate group interacts with the Nδ1 atom of His32* (Fig. 4b). Superposition on the porphyrin-ring atoms shows that the haem-plane conformations are quite similar among the four structures (Supplementary Fig. S3). The 7-propionate groups also show a quite similar conformation, while large differences are observed in the conformations of the 6-propionate groups.
The distances and angles involved in the haem iron coordination do not show significant changes between the oxidized and reduced forms (Table 3). Only a small difference of 0.03 Å is observed in the distance from Fe to His68 N∊2 between the oxidized form 1 and reduced form 1 structures. The differences in the other iron-coordination distances are less than 0.03 Å between the oxidized and reduced structures. In a comparison of the form 1 and form 2 structures, most of the iron-coordination distances and angles also show small differences, but relatively large changes of an ∼8° rotation are observed in the angles between the imidazole plane of His68 and the NII—NIV axis of the haem plane.
The Nδ1 atoms of the two axial ligands His44 and His68 form hydrogen bonds to the main-chain carbonyl O atoms of Gly47 and Phe63, respectively (Fig. 3). Supplementary Figs. S4 and S5 show the hydrogen OMIT maps of the axial ligands His44 and His68, respectively. The hydrogen-bond distances from His44 Nδ1 to Gly47 O are 2.852 (7) Å in the oxidized form 1 structure and 2.868 (6) Å in the reduced form 1 structure. The hydrogen-bond distance from His68 Nδ1 to Phe63 O is 2.893 (6) Å in the oxidized form 1 structure and 3.097 (5) Å in the reduced form 1 structure. If the final models did not contain the H atoms of His44 and His68, the H atoms in the imidazole rings of His44 and His68 except for the H∊2 atoms were generated by the riding model of SHELXL to estimate the standard deviations for hydrogen-bond distances from His44 to Gly47 and from His68 to Phe63.
Two Ca2+ ions are found in the form 1 structures (Fig. 5a). The calcium-binding sites are located on the protein surface, where acidic residues are clustered around the exposed haem propionate groups (Fig. 5b). One of the Ca2+ ions (Ca-1) is coordinated by the side-chain carboxyl groups of two glutamic acid residues, Glu42 and Glu48, located around the axial ligand His44. Ca-1 is also coordinated by the side-chain amide O atom of Gln54* and the main-chain carbonyl O atom of Asp58*, both of which are residues of the neighbouring molecule in the crystal. In addition, two water molecules are involved in coordination of the Ca-1 ion. Another Ca ion (Ca-2) is coordinated by the main-chain carbonyl O atom of Glu42 and water molecules; Ca-2 makes no direct interaction with the neighbouring molecules in the crystal. The interaction mediated by Ca-1 changes the side-chain structure around His44, but His44 does not show significant changes in the haem iron coordination between the form 1 and form 2 structures (Fig. 5c). Structural comparison between the form 1 and form 2 structures shows a 2.2° rotation of the aromatic ring of Phe40 and a 0.45 Å shift of the Cα atom of Val50.
In the form 2 structures, Arg89* of the neighbouring molecule in the crystal interacts with both Val66 and Gly67 next to the axial ligand His68 (Fig. 4b). The interaction changes the conformation of Gly67, and the movement of the main-chain carbonyl O atoms is 2.80 Å between the form 1 and form 2 structures (Fig. 5d). The position of Val66 is also changed by the interaction with Arg89*. The movement of the Cα atoms of Val66 is 1.38 Å between the form 1 and form 2 structures (Fig. 5d). The positions of the aromatic ring of Phe63 are also changed by the interaction with the neighbouring molecule in the crystal. The C∊1 atoms, which are proximal to the haem Fe atom in the aromatic ring of Phe63, are located at distances of 4.99 and 5.25 Å from the haem Fe atoms in the form 1 and form 2 structures, respectively. Although the packing interaction induces structural changes around the axial ligand His68, there are small changes in the iron-coordination distance of His68 (Table 3).
The overall structure of porcine b5 is very similar to the b5 structures from other species, especially regarding the haem-binding region. Superposition of the crystal structures of porcine b5 (oxidized form 2) and bovine b5 (PDB entry 1cyo ; Durley & Mathews, 1996) shows a root-mean-square distance of 0.45 Å for 84 Cα atoms (Supplementary Fig. S6). In the solution structures of mammalian microsomal b5, the N- and C-terminal regions appear as highly flexible structures (Banci et al., 1997, 2000; Arnesano et al., 1998; Zhang et al., 2004). In the crystal structures of porcine b5, flexible structures appear as several multiple conformations observed in the N- and C-terminal residues, and the seven disordered residues at the N-terminus also indicate flexibility. The high-resolution structures of porcine b5 reveal multiple conformations that were not observed in the lower resolution structure of bovine b5.
Form 1 and 2 crystals were obtained at pH 7.4 and 5.5, respectively. In a pulse voltammetry experiment using bovine b5, the values of the redox potential were decreased at increasing pH values (Qian et al., 2002). Structural changes between the different crystal forms are observed near the axial ligand His68 (Fig. 5d) and in the conformation of the 6-propionate group (Fig. 4). The structure around His68 is affected by interactions with the neighbouring molecules in the form 2 crystal (Fig. 4b). For the 6-propionate group, restrained refinement was performed for the oxidized form 2, reduced form 1 and reduced form 2 structures. The 6-propionate groups showed multiple conformations in the oxidized form 2 and reduced form 1 structures (Supplementary Fig. S3), and the electron density of the 6-propionate group was ambiguous in the reduced form 2 structure. From these results, it is difficult to estimate pH-dependent changes from the axial ligands and the 6-propionate group. On the other hand, the 7-propionate groups show a highly similar conformation in the crystal structures of porcine and bovine b5s (Supplementary Figs. S3 and S6b). Hydrogen bonds to Ser69 fix the conformation of the 7-propionate, although the 7-propionate group could form an extended conformation in the four crystals because there was sufficient space between adjacent molecules. The sub-angstrom data for porcine b5 allowed unrestrained refinement of the bond angles and lengths of the atoms in the 7-propionate group (Table 2). The C—O bond distances of the carboxyl groups of the 7-propionate indicate that C—O1 is a double bond in the reduced form 1 structure (the typical C=O double-bond distance is 1.21 Å). However, the C—O2 distance is ∼0.07 Å shorter than the typical distance for a protonated C—OH bond (1.32 Å). Therefore, the protonation state in the reduced form 1 structure is unclear. On the other hand, the uniform C—O distances in the oxidized form 1 structure indicate that the 7-propionate group is in a charged state. Unrestrained refinement was also applied to the 6-propionate group in the oxidized form 1 structure. The C—O1 [1.253 (9) Å] and C—O2 [1.246 (10) Å] distances are consistent with the results for the 7-propionate group. In the reduced form 2 structure, the C—O bond distances of the 7-propionate indicate that the O1 atom is possibly protonated. Form 2 crystals were obtained at pH 5.5, which is lower than the experimentally measured pKa values of 5.7 and 5.9 for oxidized and reduced haem, respectively (Das & Medhi, 1998). On the other hand, the carboxyl group in the oxidized form 2 structure is in a charged state as deduced from the C—O bond distances of the 7-propionate. Based on the results of electrostatics calculations, Mao et al. (2003) reported that haem oxidation is coupled to some proton loss. The protonation states in the form 2 structures (determined at pH 5.5) are consistent with this result.
Structural changes depending on the redox state have been investigated in haem-binding proteins. X-ray absorption spectroscopy showed that the bond distance from the haem Fe to the proximal His changed by 0.15 Å between the oxidized and reduced forms of myoglobin (Longa et al., 2003). On the other hand, apparent structural changes were not observed between the oxidized and reduced forms in solution structures of cytochrome c (Feng et al., 1990). In the high-resolution structures of porcine b5, significant changes could not be detected in the haem iron-coordination geometry when the oxidized and reduced structures were compared in the same crystal form (Table 3). The stabilization of b5 was decreased by mutation of the axial ligand (Wang et al., 2003). This result suggests that structural changes of the axial ligands are highly restricted in b5. The hydrogen-bond distance from His44 to Gly47 also exhibited a small difference of ∼0.02 Å between the form 1 structures. However, there was an ∼0.20 Å difference in the hydrogen-bond distance from His68 Nδ1 to the main-chain carbonyl O atom of Phe63 between the form 1 structures. The residues from Phe63 to His68 make a few interactions with the neighbouring molecules in the form 1 crystal. Because there is a relatively large space between adjacent molecules, the residues show flexible structures, as observed from the multiple conformations of Glu64–Gly67 (Fig. 6). In the form 2 structures, Arg89* of the neighbouring molecule in the crystal interacts with both Val66 and Gly67 (Fig. 4b). The interactions with Arg89* and other interactions with the neighbouring molecules in the crystal fix the structure from Phe63 to His68, and small spaces exist between the adjacent molecules. In addition, water molecules were found near the main-chain atoms of Phe63 and His68 in the form 1 structures (Fig. 6). In the reduced form 1 structure, the water molecules show an alternate conformation (Fig. 6b). The water molecules form hydrogen bonds to both the main-chain carbonyl of Phe63 and the main-chain amide of His68. The hydrogen-bond distances in the oxidized form 1 structure are shorter (0.2–0.4 Å) than those in the reduced form 1 structure. These results indicate that the hydrogen-bond network around His68 might be involved in regulating the redox state of the haem if the residues from Phe63 to His68 have flexible structures. The changes in the hydrogen-bond network around the axial ligand are consistent with the structural and mutation studies of haem proteins, which suggested hydrogen bonding to the axial ligand as an important factor in controlling the haem redox potential (Poulos, 1996).
Divalent cations (Ca2+ or Mg2+) have been reported to reduce the rate of b5 reduction by b5R (Tamura et al., 1988). In the form 1 structure the Ca2+ ions are coordinated by the glutamic acid residues Glu42 and Glu48 located on the surface near the haem propionate group (Fig. 5b). These glutamic acid residues are conserved in mammalian b5 (Supplementary Fig. S2). Studies using the b5–b5R docking model have indicated that the acidic surface residue Glu48 of b5 interacts with the basic surface residue Lys97 near the FAD- and NADH-binding sites of b5R (Nishida & Miki, 1996). Therefore, the binding of Ca ions may prevent the electrostatic interaction between acidic surface residues of b5 and basic surface residues of b5R.
We are grateful to the beamline staff at Photon Factory for their help in diffraction data collection. We also thank Ms Hiroko Kawamata and Ms Yuka Tachihara for their help in sample preparation. This work was supported in part by Ministry of Education, Culture, Sports, Scientific and Technology Grant-in-Aid for Scientific Research (C) 25440037 (to TT).
Adams, P. D. et al. (2010). Acta Cryst. D66, 213–221. Web of Science CrossRef CAS IUCr Journals Google Scholar
Altuve, A., Wang, L., Benson, D. R. & Rivera, M. (2004). Biochem. Biophys. Res. Commun. 314, 602–609. CrossRef PubMed CAS Google Scholar
Arnesano, F., Banci, L., Bertini, I. & Felli, I. C. (1998). Biochemistry, 37, 173–184. CrossRef CAS PubMed Google Scholar
Banci, L., Bertini, I., Ferroni, F. & Rosato, A. (1997). Eur. J. Biochem. 249, 270–279. CrossRef CAS PubMed Google Scholar
Banci, L., Bertini, I., Rosato, A. & Scacchieri, S. (2000). Eur. J. Biochem. 267, 755–766. Web of Science CrossRef PubMed CAS Google Scholar
Das, D. K. & Medhi, O. K. (1998). J. Inorg. Biochem. 70, 83–90. CrossRef CAS PubMed Google Scholar
DeLano, W. L. (2002). PyMOL. https://www.pymol.org . Google Scholar
Della Longa, S., Arcovito, A., Benfatto, M., Congiu-Castellano, A., Girasole, M., Hazemann, J. L. & Lo Bosco, A. (2003). Biophys. J. 85, 549–558. Web of Science CrossRef PubMed CAS Google Scholar
Durley, R. C. E. & Mathews, F. S. (1996). Acta Cryst. D52, 65–76. CrossRef CAS Web of Science IUCr Journals Google Scholar
Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486–501. Web of Science CrossRef CAS IUCr Journals Google Scholar
Feng, Y., Roder, H. & Englander, S. W. (1990). Biochemistry, 29, 3494–3504. CrossRef CAS PubMed Google Scholar
Fukushima, H., Grinstead, G. F. & Gaylor, L. (1981). J. Biol. Chem. 256, 4822–4826. CAS PubMed Google Scholar
Funk, W. D., Lo, T. P., Mauk, M. R., Brayer, G. D., MacGillivray, R. T. A. & Mauk, A. G. (1990). Biochemistry, 29, 5500–5508. CrossRef CAS PubMed Web of Science Google Scholar
Gan, J.-H., Wu, J., Wang, Z.-Q., Wang, Y.-H., Huang, Z.-X. & Xia, Z.-X. (2002). Acta Cryst. D58, 1298–1306. Web of Science CrossRef CAS IUCr Journals Google Scholar
Kimura, S., Kawamura, M. & Iyanagi, T. (2003). J. Biol. Chem. 278, 3580–3589. CrossRef PubMed CAS Google Scholar
Mao, J., Hauser, K. & Gunner, M. R. (2003). Biochemistry, 42, 9829–9840. CrossRef PubMed CAS Google Scholar
Mitoma, J. & Ito, A. (1992). EMBO J. 11, 4197–4203. PubMed CAS Google Scholar
Nishida, H. & Miki, K. (1996). Proteins, 26, 32–41. CrossRef CAS PubMed Google Scholar
Otwinowski, Z. & Minor, W. (1997). Methods Enzymol. 276, 307–326. CrossRef CAS Web of Science Google Scholar
Paithankar, K. S., Owen, R. L. & Garman, E. F. (2009). J. Synchrotron Rad. 16, 152–162. Web of Science CrossRef CAS IUCr Journals Google Scholar
Porter, T. D. (2002). J. Biochem. Mol. Toxicol. 16, 311–316. CrossRef PubMed CAS Google Scholar
Poulos, T. L. (1996). J. Biol. Inorg. Chem. 1, 356–359. CrossRef CAS Google Scholar
Qian, C., Yao, Y., Ye, K., Wang, J., Tang, W., Wang, Y., Wang, W., Lu, J., Xie, Y. & Huang, Z. (2001). Protein Sci. 10, 2451–2459. CrossRef PubMed CAS Google Scholar
Qian, W., Wang, Y.-H., Wang, W.-H., Yao, P., Zhuang, J.-H., Xie, Y. & Huang, Z.-X. (2002). J. Electroanal. Chem. 535, 85–96. CrossRef CAS Google Scholar
Ren, Y., Wang, W.-H., Wang, Y.-H., Case, M., Qian, W., McLendon, G. & Huang, Z.-X. (2004). Biochemistry, 43, 3527–3536. CrossRef PubMed CAS Google Scholar
Rivera, M., Barillas-Mury, C., Christensen, K. A., Little, J. W., Wells, M. A. & Walker, F. A. (1992). Biochemistry, 31, 12233–12240. CrossRef PubMed CAS Google Scholar
Sheldrick, G. M. (2008). Acta Cryst. A64, 112–122. Web of Science CrossRef CAS IUCr Journals Google Scholar
Spatz, L. & Strittmatter, P. (1973). J. Biol. Chem. 248, 793–799. CAS PubMed Google Scholar
Sun, Y.-L., Wang, Y.-H., Yan, M.-M., Sun, B.-Y., Xie, Y., Huang, Z.-X., Jiang, S.-K. & Wu, H.-M. (1999). J. Mol. Biol. 285, 347–359. CrossRef CAS PubMed Google Scholar
Tamura, M., Yubisui, T. & Takeshita, M. (1988). Biochem. J. 251, 711–715. CAS PubMed Google Scholar
Walker, F. A., Emrick, D., Rivera, J. E., Hanquet, B. J. & Buttlaire, D. H. (1988). J. Am. Chem. Soc. 110, 6234–6240. CrossRef CAS PubMed Google Scholar
Wang, W.-H., Lu, J.-X., Yao, P., Xie, Y. & Huang, Z.-X. (2003). Protein Eng. Des. Sel. 16, 1047–1054. CrossRef CAS Google Scholar
Wang, L., Sun, N., Terzyan, S., Zhang, X. & Benson, D. R. (2006). Biochemistry, 45, 13750–13759. CrossRef PubMed CAS Google Scholar
Wu, J., Gan, J.-H., Xia, Z.-X., Wang, Y.-H., Wang, W.-H., Xue, L.-L., Xie, Y. & Huang, Z.-X. (2000). Proteins, 40, 249–257. CrossRef PubMed CAS Google Scholar
Wu, Y., Wang, Y., Qian, C., Lu, J., Li, E., Wang, W., Lu, J., Xie, Y., Wang, J., Zhu, D., Huang, Z. & Tang, W. (2001). Eur. J. Biochem. 268, 1620–1630. CrossRef PubMed CAS Google Scholar
Yamada, M., Tamada, T., Takeda, K., Matsumoto, F., Ohno, H., Kosugi, M., Takaba, K., Shoyama, Y., Kimura, S., Kuroki, R. & Miki, K. (2013). J. Mol. Biol. 425, 4295–4306. CrossRef CAS PubMed Google Scholar
Yao, P., Wu, J., Wang, Y.-H., Sun, Y.-H., Xia, Z.-X. & Huang, Z.-X. (2002). Eur. J. Biochem. 269, 4287–4296. CrossRef PubMed CAS Google Scholar
Zhang, Q., Cao, C., Wang, A.-Q., Wang, Y.-H., Wu, H. & Huang, Z.-X. (2004). Protein Sci. 13, 2161–2169. CrossRef PubMed CAS Google Scholar
This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.