Tài liệu Báo cáo khoa học: Critical roles of Asp40 at the haem proximal side of haem-regulated phosphodiesterase from Escherichia coli in redox potential, auto-oxidation and catalytic control doc

Critical roles of Asp40 at the haem proximal side of haem-regulated phosphodiesterase from Escherichia coli in redox potential, auto-oxidation and catalytic control Miki Watanabe, Hirofumi Kurokawa, Tokiko Yoshimura-Suzuki, Ikuko Sagami and Toru Shimizu Institute of Multidisciplinary Research for Advanced Materials Tohoku University, Sendai, Japan In haem-regulated phosphodiesterase (PDE) from Escheri- chia coli (Ec DOS), haem is bound to the PAS domain, and the redox state of the haem iron regulates catalysis by the PDE d omain. We generated mutants of Asp40, which forms a hydrogen bond with His77 (a proximal haem axial ligand) via two water molecules, and a salt bridge wit h Arg85 at the protein surface. The redox potential of haem was marked ly increased from 67 mV vs. the stan dard hydrogen e lectrode in the wild-type enzyme t o 95 mV and 114 mV in the Ala and Asn mutants, r espectively. Additionally, the auto-oxidation rate of Ec DOS PAS was significantly increased from 0.0053 to 0.051 and 0.033 min )1 , respectively. Interestingly, the catalytic activities of the A sp40 mut ants were a bolished completely. T hus, Asp40 appears to play a critical role in the electronic structure of the h aem iron and redox-dependent catalytic control of the PDE domain. In this report, we discuss the mechanism of catalytic control o f Ec DOS, based on the physico-chemical characteristics of the Asp40 mutants. Keywords: auto-oxidation; h aem axial ligand; haem sensor; phosphodiesterase; redox potential. Haem-regulated phosphodiesterase (PDE) from Escheri- chia coli (Ec DOS) is a haem sensor enzyme composed of two functional domains: an N-terminal haem-bound sensor domain and a C-terminal PDE catalytic domain [1,2]. Catalysis by this enzyme is regulated by the haem redox state in that PDE is functional in the Fe(II) haem- bound enzyme, but not the F e(III) haem-bound enzyme [2,3]. The crystal structure of the haem-bound domain revealed a typical PAS structure [4,5]. PAS proteins display a characteristic three-dimensional structure with a glove-like fold consisting of five j uxtaposed b-sheets and flanking a-helices [6–9]. The characteristic three-dimen- sional structure of the PAS domain is commonly used for discussing the signal transduction mechanism of numerical signal transducing enzymes [6–9]. The crystal structures of both the Fe(II) and Fe(III) forms of the isolated haem-bound PAS domain (Ec DOS PAS) disclose that haem axial ligand switching from His77/ hydroxide anion to the His77/Met95 ligand pair occurs upon haem reduction. Moreover, haem ligand switching induces conformational changes in the FG loop region and movement of t wo subunits. These structural changes may play critical roles in c atalytic regulation of the PD E domain [4]. Structures of the haem-bound PAS domain have been reported under various conditions, and structure–function relationships are well documented [8–14]. FixL is an oxygen sensor enzyme with a haem-bound PAS domain. Specific- ally, O 2 association/dissociation to/from the haem switches off/on catalysis [8,9]. Global structural changes at the haem distal side are i nduced upon O 2 binding to FixL, and these changes contribute significantly to intramolecular s ignal transduction [9–11]. For Ec DOS, site-directed mutations at Met95, the axial ligand at the distal side in the Fe(II) complex, induced significant changes i n t he redox potential of the haem, rate constants o f CO, O 2 and CN b inding to haem, and CD spectra in the Soret region [15–18], suggesting that this residue is critical for maintaining the electronic states of haem and ligand access channel. However, the catalytic activities of Met95 mutants were comparable to those of wild-type [14], implying no direct involvement in the catalytic control of Ec DOS. A number of site-directed mutagenesis [2,3,16] and crystallographic [4,5] s tudies show that His77 is the proximal axial ligand of haem. Therefore, a residue interacting with H is77 may p lay an important role in regulation of catalysis. Ec DOS forms hydrogen networks at the haem proximal side (Fig. 1) including the Asp40 residue that interacts w ith His77 via two water molecules and Arg85 at the protein surface [4]. In the present study we generated Asp40Ala and Asp40Asn mutants and analysed their physico-chemical Correspondence to T. Shimizu, Institute of M ultidisciplinary Research for Advanced Materials, Tohoku University, 2-1-1 Katahira, Aoba-ku, Sendai 980-8577, Japan. Fax: +81 22 217 5604/5390, Tel.: +81 22 217 5604/5605. E-mail: shimizu@tagen.tohoku.ac.jp Abbreviations: PDE, phosphodiesterase; Ec DOS, full-length haem- bound phosphodiesterase from Escherichia coli or a redox sensor from Escherichia coli; PAS, an acronym formed from the following names: Drosophila period clock protein (PER), vertebrate aryl hydrocarbon receptor nuclear translocator (ARNT) and Drosophila single-minded protein (SIM); FixL, an oxygen sensor haem p rotein from Rhizobia melilori;ant-cAMP, 3¢,5¢-cyclic monophosphate, 2¢-O-anthraniloyl; Ec, DOS PAS, isolated haem-bound PAS domain of Ec DOS; SHE, standard hydrogen electrode. (Received 20 June 2004, revised 9 Au gust 2004, accepted 12 August 2004) Eur. J. Biochem. 271, 3937–3942 (2004) Ó FEBS 2004 doi:10.1111/j.1432-1033.2004.04331.x characteristics, includ ing optical absorptio n spectra, cyanide binding, redox potential and auto-oxidation rates of Ec DOS PAS, and catalytic activities of full-length enzymes. We found that the mutations at Asp40 markedly altered t he redox potential and auto-oxidation rate. Fur- thermore, PDE activity was completely abolished in the Asp40 mutants. To our knowledge, this is the first report showing that mutations in the haem environment substan- tially influence the catalytic activity of haem-bound Ec DOS. Experimental procedures Materials The expression vector, pET28a(+) was from Novagen. E. coli competent cells, XL1-blue (for cloning), and BL21 (for pr otein expression) were purchased from Stratagene. Site-directed mutagenesis was performed using the Quikchange Site-Directed Mutagenesis Kit TM (Stratagene) with the following oligonucleotides: Asp40Ala: 5¢-TGTTAATTAACGAAAATGC TGAAGTGATGTTTT TC-3¢ (forward); 3¢-GAAAAACATCACTTC AGCATTT CGTTAATTAAC-5¢ (reverse); Asp40Asn: 5¢-GGTGTT AATTAACGAAAATAAC GAAGTGATGTTTTTCA AC-3¢ (forward); 3¢-GTTGAAAAACATCACTTCGTTA TTTTCGTTAATTAACA-5¢ (reverse). Mutation sites are shown in italics. Oligonucleotides were synthesized by the Nihon Gene Research Laboratory (Sendai, Japan). Restric- tion and m odification enzymes were from Takara Bio (Otsu, Japan), Toyobo, New England Biolabs and Nippon Roche K.K. The fluorescence substrate, adenosine 3 ¢,5¢-cyclic monophosphate, 2 ¢-O-anthraniloyl (ant-cAMP) was from Calbiochem. Calf intestine alkaline phosphatase was from Takara Bio. DEAE Sephadex was from Amersham Biosciences. Other chemicals were from Wako Pure Chemicals. Expression and purification of Ec DOS wild-type and Asp40 mutant proteins Expression and purification procedu res were performed as described previously [2,3]. Purified proteins were more than 95% homogenous, as verified by SDS/PAGE. Yields of Ec DOS and Ec DOS PAS from 1 L E. coli culture were 210 n mol and 610 nmol, respectively, in terms of haem absorbance at 417 n m [2,3]. Optical absorption spectra Spectral experiments were performed under aerobic condi- tions on Shimadzu UV-1650, UV-2500 and Hitachi U-2010 spectrophotometers maintained at 25 °C using a tempera- ture controller [2,3,15,17,18]. Anaerobic spectral experi- ments were conducted in the glove box of a Shimadzu UV-160 A spectrophotometer. Following reduction of the haem b y sodium dithionite, excess dithionite was removed in the glove box by using a Sephadex G25 column. To ensure that the temperature of the solution was maintained at 25 °C, the reaction mixture was i ncubated for 10 min prior to spectroscopic measurements. Cyanide binding The association rate of cyanide to haem was observed by monitoring changes at 417 nm in spectra of the haem protein in 50 m M Tris/HCl pH 7.5, as described earlier [17]. Redox potential Anaerobic spectral experiments on Ec DOS PAS proteins were performed i n the glove box on a Shimadzu UV-160 A spectrophotometer and a n O RION Research (Tokyo, Japan) Model 701 digital p H meter equipp ed with a TOA (Tokyo, Japan) ORP g old/calomel combination microelec- trode. Redox potentials were measured using the same apparatus. 2,3,5,6-Tetramethyl phenylenediamine (5 l M ), N-ethyl phenazonium ethosulfate (5 l M ) a nd 2-hydroxy- 1,4-naphthaquinone (5 l M ) were added as m ediators to the wild-type protein solution before titration [2,15]. The concentration o f h aem protein used was 15 l M . Spectral changes in i ntensity at 563 nm accompanying redox changes were monitored, as dye absorption hampers the d etection o f Soret spectral changes. To ensure that the appropriate temperature o f the solution was maintained, the r eaction mixture was incubated for 10 min prior to spectroscopic measurements. Titration experiments were re peated at least three times for ea ch complex. Auto-oxidation rate To measure auto-oxidation rates, Ec DOS PAS proteins were reduced with sodium dithionite in 50 m M Tris/HCl pH 7.0 containing 1 m M EDTA. Excess dithionite was removed in the glove box using a Sephadex G-25 column. After removal from the glove box, the mutant wasdilutedto1mLwith50m M potassium phosphate pH 7.0 containin g 1 m M EDTA. The auto-oxidation rates of Asp40Ala and Asp40Asn mutants were meas- ured at 25 °C. Oxidation of t he sample was observed b y Fig. 1. Structure of ferric Ec DOS PAS (PDB code 1V9Y). Asp40 interacts with the imidazole r ing o f His77 via t wo wate r molecules (W3, W4).Asp40additionallyformsasaltbridgewithArg85atthesurface of the m olecu le. T he h ydrogen bonds and fl exible FG loop are depicted by black dotte d and cyan broken lines, respectively. The figure was obtained using PYMOL [20]. 3938 M. Watanabe et al. (Eur. J. Biochem. 271) Ó FEBS 2004 recording entire visible spectra or monitoring the decrease in absorbance at a single wavelength (567 nm) on a Hitachi U-2010 spectrometer. All auto-oxidation reac- tions were characterized by only two spectral species with sharp i sosbestic points, and the time course patterns revealed a simple first-order process under each set of conditions. Enzymatic assay Full-length Ec DOS protein was incubated at 37 °Cwith ant-cAMP in a reaction mixture of 500 lL containing 50 m M Tris/HCl pH 8.5 a nd 2 m M MgCl 2 in a glove box under a nitrogen atmosphere with an O 2 concentration < 50 p.p.m., as described previously [2,3]. At least five experiments were conducted to obtain e ach value. Results Optical absorption spectra Figure 2 depicts the optical absorption spectra of Fe(III), Fe(II) and Fe(II)-CO complexes of the Asp40Ala and Asp40Asn mutants. Table 1 summarizes the spectral absorption maxima of these complexes. Despite subtle differences in the spectral maxima in the visible region of the Fe(III) species, spectra of the Asp40 mutants were essentially similar to that of the wild-type protein. Additionally, spin states of the Fe(III), Fe(II) and Fe(II)-CO complexes of Asp40 mutants were similar to those of the wild-type enzyme in that all mutant proteins were in the six-coordinated low-spin state. T herefore, we propose that Asp40 mutations do not significantly affect the h aem e nvironment or coordination structures. More- over, these mutants are suitable for further spectral and catalytic characterization. Cyanide binding Kinetic and e quilibrium studies on cyanide binding provide valuable inform ation on t he haem distal structure o f Ec DOS PAS [17]. As the structure of the cyanide-bound complex of FixL is similar to that of its oxygen-bound complex [10], a cyanide binding study should facilitate elucidation of the structure and catalytic mechanism of Ec DOS. Cyanide-bound Fe(III) complexes of t he Asp40 mutants displayed optical absorption spectra containing a sharp Soret peak at 421 nm and a broad visible band around 540 nm (data not shown), analogous to that of the wild-type enzyme [17]. Optical absorption changes observed for both the Asp40 mutants upon cyanide binding were composed of only one phase. The first-order rate co nstants f or cyanide binding to Asp40 mutants were dependent on the cyanide concentration. The rates of cyanide association to t he Asp40Ala and Asp40Asn mutants were 0.065 and 0.073 m M )1 Æs )1 , respectively, which are only slightly higher than that (0.045 m M )1 Æs )1 ) of the wild-type protein (Table 2). The data i ndicate that mutation of Asp40 a t the haem proximal side alters the cyanide b inding property only slightly, a nd may not significantly affect the exogenous ligand binding access channel a t t he haem distal side and/or structure in t erms of cyanide binding in the Fe(III) complex. Redox potentials The redox state of haem is related to the PDE activity of Ec DOS, since it is a haem redox-sensing enzyme [2,3]. Thus, it is important to determine the redox potentials of the Asp40 mutants of Ec DOS. Asp40 m utant proteins w ere converted from the Fe(III) complex to the Fe(II) complex by reductive titration, accompanied by clear isosbestic points, similar to previo usly documented data for the wild- type and Met95 mutant proteins [2,15]. Electrochemical reductive titration of the mutant protein is depicted in Fig. 3. The redox potential values of the m utants as well as wild-type protein are summarized in Table 2. Marked increases in the redox potential value from 67 m V vs. the standard hydrogen electrode (SHE) (wild-type) up to 95 and 114 mV were observed f or the Asp40Ala and Asp40Asn mutants, respectively. This tendency is opposite to that observed on mutating Met95 at the haem distal site. Specifically, Met95 mutations led to significant decreases in redox potential [15]. O xidative titration experiments were additionally conducted. No remarkable differences were detected in the potentials between reductive and oxidative titration. Accordingly, we propose that Asp40 is located close enough to the haem iron or axial ligand to influence the haem electronic state of Ec DOS. Fig. 2. Soret and visible optical absorption spectra of Fe(III) (solid line), Fe(II) (broken line) and Fe(II)-CO (dotted line) complexes of the Asp40Ala (upper) and Asp40Asn (lower) mutants of Ec DOS P AS. The small peak seen around 670 nm is occasionally appears when we purified mutant enzymes. It is likely to be a minor component of a denatured form. Ó FEBS 2004 Haem proximal side of a haem-regulated phosphodiesterase (Eur. J. Biochem. 271) 3939 Auto-oxidation rates As Ec DOS was initially identified as an O 2 sensor enzyme [1], we examined the auto-oxidation rates of the Asp40 mutants. As shown in Fig. 4, s emi-logarithmic t ime- dependent changes in optical absorption spectra were linear, and composed of only one phase. Auto-oxidation rates and half-lives of the O 2 -bound Fe(II) complexes of Asp40 mutants are summarized in Table 2 . Marked increases in the auto-oxidation rate (from 0 .0053 min )1 for the wild- type enzyme up to 0.033 and 0.051 min )1 ) were observed i n the Asp40 mutants. PDE activity Electronic states of haem, such as redox potential and autoxidation rate, largely regulate the PDE activity of full- length Ec DOS enzyme. Interestingly, no PDE activity was detected for the two Asp40 mutants analysed in this study. This finding is in contrast with data on mutants of Met95 at the haem distal side, which disclosed no effect on PDE activity [15]. Discussion The present study reveals an interesting aspect of the structural and functional relationships of Ec DOS. Muta- tions at Asp40 did not essentially affect the optical absorption spectra of this enzyme. Howeve r, Asp40 muta- tions at the haem proxima l side significantly altered the redox potential values, auto-oxidation rates, and PDE activities. These results aid in elucidating the transduction mechanism of this enzyme. Recent c rystal structure analyses of t he isolated haem- bound PAS domain indicate that in the Fe(III) complex, a hydroxide anion (or w ater molecule) is an axial ligand tra ns to His77, the e ndogenous (proximal) axial ligand [4]. U pon haem reduction, ligand switching from the hydroxide anion to the side chain of Met95 i s evident at the haem distal s ide. Met95, in turn, becomes the direct axial ligand f or the F e(II) complex trans to His77 [4,5]. The haem redox potential value was decreased from +67 mV vs. SHE (wild-type) to )26, )1, and )122 mV in t he Met95Ala, M et95Leu and Met95His mutants, respectively [15]. The auto-oxidation rate was altered from 0.0058 to 0.0013, 0.0017 and 0.018 min )1 in the Met95Ala, Met95Leu and Met95His mutants, respectively [18]. Mutations at Met95 also signi- ficantly affect the binding of exogenous ligands, such as the cyanide anion, to the Fe(III) complex [17], and O 2 and CO to the Fe(II) complex [18]. H owever, the PDE activities of these mutants are comparable to that of the wild-type enzyme. B ased on the data, we propose that Met95 Table 2. Cyanide binding rates, r edox potentials and auto-oxidation rates of the Asp40Ala and Asp40Asn mutants of Ec DOS PAS. Cyanide binding rate, experimental errors are within 20%; k ox , autoxidation rate; t 1/2 , Half life. Ec DOS PAS CN (m M – 1Æs )1 ) Redox potential (mV vs. SHE) k ox (min )1 ) t 1/2 (min) Wild-type 0.045 67 (n ¼ 0.93) 0.0053 132 Asp40Ala 0.065 95 ( n ¼ 0.92) 0.051 14 Asp40Asn 0.073 114 (n ¼ 0.96) 0.033 21 Table 1. Optical absorption spectral maxima (nm) of the Asp40Asn and As p40Ala mutants of Ec DOS PAS. Fe(III) Fe(II) Fe(II)-CO Soret baSoret baSoret ba Wild-type 416 530 564 427 532 563 423 540 570 Asp40Ala 415 535 566 428 533 563 424 539 565 Asp40Asn 417 535 565 427 532 563 423 541 572 Fig. 3. Electrochemical reductive titrations of the Asp40Asn m utant of Ec DOS PAS. Changes in a bsorption intensity were monitored at 563 nm. Fig. 4. Time-dependent optical absorption changes of the Fe(II)-O 2 complexe s o f Ec DOS PAS wild-type (solid line, ·), Asp40Ala (broken line, 4) and Asp40Asn (dotted line, s) mutants monitored at 578 nm. 3940 M. Watanabe et al. (Eur. J. Biochem. 271) Ó FEBS 2004 modulates the redox potential to relatively high values (+61 mV), but is not directly involved in catalytic control and/or interactio ns with the PDE domain, which are critical in catalysis. In contrast to data obt ained with Met95Ala and Met95Leu mutants [15–18], Asp40 mutations increased the redox poten tials and auto-oxidation rates. The effects of Asp40 mutations are opposite to those of Met95 mutations. Asp40 mutants are more easily reduced than the wild-type enzyme, and favour the Fe(II) s tate to the Fe(III) state. Accordingly, we suggest that Asp mutations alter the haem environment to a more cationic state, and the F e(II) state is more stabilized. The crystal structur e of the PAS domain d iscloses two water molecules between His77 and Asp40 (Fig. 1). There- fore, the hydrogen bond ing network involving His77, two water molecules and Asp40 should function in regulating the electronic state affecting the redox potential and auto- oxidation rate. A signal transduction mechanism for the PAS domain has b een proposed by Cr osson et al. [19] to explain catalytic c ontrol. The investigators suggested that a well- conserved salt bridge on the surface of PAS proteins is important in the signal transduction mechanism. Although a conserved salt bridge exists between Glu59 and Ly s104 in Ec DOS [ 4], the roles of these amino acids remain to be elucidated. In Ec DOS, Asp40–Arg85 salt bridges (Fig. 1) appear to be important for catalytic control. Breakage of th e salt b ridge by m utations at Asp40 abolishes PDE activity. In a previous report, we proposed that movement of the FG loop in Ec DOS accompanying haem reduction regulates the catalytic switch [4]. The FG loop is rigidified in the Fe(II) haem protein, but very flexible in t he Fe(III) haem protein. Ec DOS is active only in the Fe(II) h aem form. Notably, Asp40 forms a salt bridge with Arg85, which is located near the FG loop (Fig. 1) [4]. Mutations at Asp40 should break the salt bridge w ith Arg85. It is possible that these mutations maintain flexibility of the FG loop, even when the haem is the Fe(II) state, and thus lead to inactive enzyme ( Fig. 1). Thus, switching the FG loop from the ÔorderedÕ to ÔdisorderedÕ form by substitu- tions at Asp40 may explain the loss of catalytic activity of these mutant p roteins. Our recent study showed that mutations at Trp residues markedly changed fluorescence intensities, but did not significantly alter the environment of the haem [21]. A single-Trp containing mutant, in which only one Trp residue is located near or a t the FG loop region, may be u seful to analyse flexibility of the FG loop in Asp40 mutants. Our proposal could also be substantiated by limited proteolysis t o see a change in signals upon haem r eduction. These studies remain to be carried out. In summary, mutations at Asp40 markedly alter the redox pot entials, auto-oxidation ra tes and cataly tic activities of Ec DOS. Breakage of the salt bridge between Asp40 and Arg85, and t he hydrogen bond network consisting of Asp40, two water molecules and His77, appear to be critical for these significant changes. In the mutants, the FG loop region may become disordered, even in the Fe(II) state, resulting in a marked decrease in catalysis. Acknowledgements This work was supported in part by a Grand-in-Aid from the Ministry of Culture, Education, Science, Sp orts and Technology of Japan to H.K. References 1. Delgado-Nixon, V.M., Gonzalez, G. & Gilles-Gonzalez, M.A. (2000) Dos, a heme-binding PAS Protein from Escherichia coli,isa direct oxygen sensor. Biochemistry 39 , 2685–2691. 2. Sasakura, Y., Hirata, S., Sugiyama, S., Suzuki, S., Taguchi, S., Watanabe, M., Matsui, T., Sagami, I. & Shimizu, T. (2002) Characterization of a direct oxygen sensor heme proteins from Escherichia coli: e ffects of the heme redox states and mutations at the heme-binding site on catalysis and structure. J. Biol. Ch em. 277, 23821–23827. 3. Yoshimura, T., Sagami, I., Sasakura, Y. & Shimizu, T. (2003) Relationships b etween heme incorpo ration, tetramer formation, and c atalysis of a heme-regulated phosphodiesterase from Escherichia coli. A study of deletion and site-directed mutants. J. Biol. Chem. 278, 53105–53111. 4. Kurokawa, H., Lee, D S., Wa tanabe, M., Sagam i, I., Mikami, B., Raman, C.S. & S himizu, T. (2004) A redox-controlled molecular switch revealed by the crystal structure of a bacterial heme PAS sensor. J. Biol. Chem. 279, 20186–20193. 5. Park, H., Suquet, C., Satterlee, J.D. & Kang, C. (2004) Insights into signal transduction involving PAS domain oxygen-sensing heme proteins from the X-ray crystal structure of Esche richia c oli DOShemedomain(Ec DOSH ). Biochemistry 43, 2738–2746. 6. Amezcua, C.A., Harper, S.M., Rutter, J. & G ardner, K .H. (2002) Structure and interactions of PAS kinase N-terminal PAS domain: model for intramoleculr kinase recognition. Structure 10, 1349– 1361. 7.Rutter,J.,Michnoff,C.H.,Harper,S.M.,Gardner,K.H.& McKnight, S.L. (2001) PAS kinase: an evolutionarily conserved PAS domain-regulat ed serine/t hreonine k inase. Proc.NatlAcad. Sci. USA 98, 8991–8996. 8. Jain, R. & Chan, M.K. (2003) Mec hanisms of ligan d discrimina- tion by heme proteins. J. Biol. Inorg. Chem. 8, 1–11. 9. Chan, M.K. ( 2001) R ecent a dvances in heme-p rotein sensors. Curr. Opin. Chem. B iol. 5, 216–222. 10.Hao,B.,Isaza,C.,Arndt,J.,Sotis,M.&Chan,M.K.(2002) Structure-based mechanism of O 2 sensin g and ligand discrimin a- tion by the FixL h eme domain of Bradyrhizobium japonium. Biochemistry 41, 12952–12958. 11. Dunham, C.M., Dioum, E.M., Tuckermann, J.R., Gonzalez, G., Scott, W.G. & Gillez-Gonzalez, M.A. (2003) A distal a rgine in oxygen-sensing heme-PAS domain is essential to ligand b inding, signal transduction, a nd structure. Biochemistry 42 , 7701–7708. 12. Miyatake,H.,Mukai,M.,Park,S Y.,Adachi,S.,Tamura,K., Nakamura,H.,Nakamura,K.,Tsuchiya,T.,Iizuka,T.&Shiro,I. (2000) Sensory mechanism of oxygen sensor FixL from Rhizobium meliloti: crystallographic, mutagenesis and resonance Raman spectroscopic studies. J. Mol. Biol. 301, 415–431. 13. Mukai, M., N akamura, K., N akamura, H., Iizuka, T. & Shiro, Y. (2000) Roles of Ile209 and Ile210 on the heme pocket structure and regulation of histidine kinase activity of o xygen sensor FixL from Rhizo bi um m el ilo t i. Bioc hemistry 39, 13810–13816. 14. Miyatake, H., Mukai, M., Adachi, S., Nakamura, H., Tamura, K., Iizuka, T., Shiro, Y., Strange, R.W. & Hasnain, S.S. (1999) Iron coordination structures of oxygen sensor FixL characterized by Fe K-edge extended X-ray absorption fine structure and resonance raman spectroscopy. J. Biol. Chem. 274, 23176–23184. 15. Hirata, S., Matsui, T., Sasakura, Y., Sugiyama, S., Yoshimura, T., Sagami, I. & Shimizu, T. (2003) Characterization of Met95 Ó FEBS 2004 Haem proximal side of a haem-regulated phosphodiesterase (Eur. J. Biochem. 271) 3941 Mutants of a heme-regulated phosphodiesterase from Escherichia coli. Optical absorption, magnetic circular dich roism, circular di- chroism, and redox poten tials. Eur. J. Biochem. 270, 4771–4779. 16. Sato, A., Sasakura, Y., Sugiyama, S., Sagami, I., Shimizu, T., Mizutani, Y. & Kitagawa, T. (2002) Stationary and time-resolved resonance raman spe ctra of His77 and Met95 mutants of the isolated heme domain of a direct oxygen s ensor from Escherichia coli. J. Biol. Chem. 277, 3 2650–32658. 17. Watanabe, M., Ma tsui, T., Sasakura, Y., Sagami, I. & Shimizu, T. (2003) Unusual cyanide binding to a heme-regulated phospho- diesterase from Escherichia coli: effect of Met95 mutations. Bio- chem. Biophys. Res. Commun. 299, 169–172. 18. Taguchi, S., Matsui, T., Igarashi, J., Sasakura, Y., Araki, Y., I to, O., Sugiyama, S., Sagami, I. & Shimizu, T. (2004) Binding of oxygen an d carbon m onoxide to a heme-regulated phosphodies- terase from Escherichia coli: kinetics and infrared spectra of the full-length wild-type enzyme, isolated PAS domain and Met95 mutants. J. Biol. C hem. 279, 3340–3347. 19. Crosson, S., Rajagopal, S. & Moffat, K. (2003) The LOV dom ain family: photoresponsive signaling modules coupled to diverse output domains. Biochemistry 42, 2–10. 20. DeLano, W.L. (2002) The PyMOL molecular graphics System. DeLano Scientific, San Carlos, CA, USA. 21. Hirata, S., Kurokawa, H., Sagami, I. & Shimizu, T. (2004) Fluorescence spectra of Trp53Phe and Trp110Ile mutations of a heme-regulated phosphodiesterase from Escherichia coli. Chem. Lett. 33, 870–871. 3942 M. Watanabe et al. (Eur. J. Biochem. 271) Ó FEBS 2004 . Critical roles of Asp40 at the haem proximal side of haem- regulated phosphodiesterase from Escherichia coli in redox potential, auto-oxidation and catalytic. networks at the haem proximal side (Fig. 1) including the Asp40 residue that interacts w ith His77 via two water molecules and Arg85 at the protein surface