Catalase Reaction by Myoglobin Mutants and Native Catalase
2004; Elsevier BV; Volume: 279; Issue: 50 Linguagem: Inglês
10.1074/jbc.m403532200
ISSN1083-351X
AutoresShigeru Kato, Takafumi Ueno, Shunichi Fukuzumi, Yoshihito Watanabe,
Tópico(s)Neonatal Health and Biochemistry
ResumoThe catalase reaction has been studied in detail by using myoglobin (Mb) mutants. Compound I of Mb mutants (Mb-I), a ferryl species (Fe(IV)=O) paired with a porphyrin radical cation, is readily prepared by the reaction with a nearly stoichiometric amount of m-chloroperbenzoic acid. Upon the addition of H2O2 to an Mb-I solution, Mb-I is reduced back to the ferric state without forming any intermediates. This indicates that Mb-I is capable of performing two-electron oxidation of H2O2 (catalatic reaction). Gas chromatography-mass spectroscopy analysis of the evolved O2 from a 50:50 mixture of H218O2/H216O2 solution containing H64D or F43H/H64L Mb showed the formation of 18O2 (m/e = 36) and 16O2 (m/e = 32) but not 16O18O (m/e = 34). This implies that O2 is formed by two-electron oxidation of H2O2 without breaking the O-O bond. Deuterium isotope effects on the catalatic reactions of Mb mutants and catalase suggest that the catalatic reactions of Micrococcus lysodeikticus catalase and F43H/H64L Mb proceed via an ionic mechanism with a small isotope effect of less than 4.0, since the distal histidine residue is located at a proper position to act as a general acid-base catalyst for the ionic reaction. In contrast, other Mb mutants such as H64X (X is Ala, Ser, and Asp) and L29H/H64L Mb oxidize H2O2 via a radical mechanism in which a hydrogen atom is abstracted by Mb-I with a large isotope effect in a range of 10–29, due to a lack of the general acid-base catalyst. The catalase reaction has been studied in detail by using myoglobin (Mb) mutants. Compound I of Mb mutants (Mb-I), a ferryl species (Fe(IV)=O) paired with a porphyrin radical cation, is readily prepared by the reaction with a nearly stoichiometric amount of m-chloroperbenzoic acid. Upon the addition of H2O2 to an Mb-I solution, Mb-I is reduced back to the ferric state without forming any intermediates. This indicates that Mb-I is capable of performing two-electron oxidation of H2O2 (catalatic reaction). Gas chromatography-mass spectroscopy analysis of the evolved O2 from a 50:50 mixture of H218O2/H216O2 solution containing H64D or F43H/H64L Mb showed the formation of 18O2 (m/e = 36) and 16O2 (m/e = 32) but not 16O18O (m/e = 34). This implies that O2 is formed by two-electron oxidation of H2O2 without breaking the O-O bond. Deuterium isotope effects on the catalatic reactions of Mb mutants and catalase suggest that the catalatic reactions of Micrococcus lysodeikticus catalase and F43H/H64L Mb proceed via an ionic mechanism with a small isotope effect of less than 4.0, since the distal histidine residue is located at a proper position to act as a general acid-base catalyst for the ionic reaction. In contrast, other Mb mutants such as H64X (X is Ala, Ser, and Asp) and L29H/H64L Mb oxidize H2O2 via a radical mechanism in which a hydrogen atom is abstracted by Mb-I with a large isotope effect in a range of 10–29, due to a lack of the general acid-base catalyst. Catalase is a heme enzyme that catalyzes the disproportionation of hydrogen peroxide to H2O and O2. In the first step, H2O2 serves as a two-electron oxidant to generate a ferryl porphyrin cation radical (O=Fe(IV) Por+·) called compound I and H2O (Equation 1). In the second step compound I serves as a two-electron oxidant of H2O2, affording O2 (catalatic reaction) accompanied by regeneration of the ferric form of catalase (Equation 2) (1Dunford H.B. Heme Peroxidases. Wiley-VCH, New York1999: 435-453Google Scholar, 2Fita I. Rossmann M.G. J. Mol. Biol. 1985; 185: 21-37Crossref PubMed Scopus (358) Google Scholar). Although the catalase reaction has been known since 1940s and it is suggested that the distal histidine is important for the deprotonation of hydrogen peroxide (1Dunford H.B. Heme Peroxidases. Wiley-VCH, New York1999: 435-453Google Scholar, 2Fita I. Rossmann M.G. J. Mol. Biol. 1985; 185: 21-37Crossref PubMed Scopus (358) Google Scholar), the detailed mechanism for its oxidation by compound I has yet to be clarified, since difficulty in the preparation of catalase compound I by a stoichiometric amount of H2O2 or alkyl peroxides has precluded direct observation of the reaction step of compound I with H2O2 (1Dunford H.B. Heme Peroxidases. Wiley-VCH, New York1999: 435-453Google Scholar, 3Chance B. Greenstein D.S. Roughton F.J.W. Arch. Biochem. Biophys. 1952; 37: 301-321Crossref PubMed Scopus (251) Google Scholar, 4Kremer M.L. Biochim. Biophys. Acta. 1970; 198: 199-209Crossref PubMed Scopus (47) Google Scholar). It also remains uncertain why chloroperoxidase (5Thomas J.A. Morris D.R. Hager L.P. J. Biol. Chem. 1970; 245: 3129Abstract Full Text PDF PubMed Google Scholar) and KatG (6Nagy J.M. Cass A.E.G. Brown K.A. J. Biol. Chem. 1997; 272: 31265-31271Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar) show catalatic activity among the heme peroxidases. Catalase Fe(III)+H2O2→k1Compound I(O=Fe(IV)Por+·)+H2O(Eq. 1) Compound I+H2O2→k2Catalase Fe(III)+H2O+O2(Eq. 2) Myoglobin (Mb), 1The abbreviations used are: Mb, myoglobin; Mb-I, compound I of Mb; mCPBA, m-chloroperbenzoic acid; KIE, kinetic isotope effect; GC-MS, gas chromatography-mass spectroscopy; BLC, beef liver catalase; MLC, M. lysodeikticus catalase. a carrier of molecular oxygen, has been studied as a structural and/or functional model for elucidating the role of active site residues in heme enzymes (7Adachi S. Nagano S. Ishimori K. Watanabe Y. Morishima I. Biochemistry. 1993; 32: 241-252Crossref PubMed Scopus (229) Google Scholar, 8Egeberg K.D. Springer B.A. Martinis S.A. Sligar S.G. Morikis D. Champion P.M. Biochemistry. 1990; 29: 9783-9791Crossref PubMed Scopus (106) Google Scholar, 9Hargrove M.S. Singleton E.W. Quillin M.L. Ortiz L.A. Phillips Jr., G.N. Olson J.S. Mathews A.J. J. Biol. Chem. 1994; 269: 4207-4214Abstract Full Text PDF PubMed Google Scholar, 10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar, 11Matsui T. Ozaki S. Watanabe Y. J. Am. Chem. Soc. 1999; 121: 9952-9957Crossref Scopus (100) Google Scholar, 12Ortiz de Montellano P.R. Catalano C.E. J. Biol. Chem. 1985; 260: 9265-9271Abstract Full Text PDF PubMed Google Scholar, 13Rao S.I. Wilks A. Ortiz de Montellano P.R. J. Biol. Chem. 1993; 268: 803-809Abstract Full Text PDF PubMed Google Scholar, 14Tschirret-Guth R.A. Ortiz de Montellano P.R. Arch. Biochem. Biophys. 1996; 335: 93-101Crossref PubMed Scopus (34) Google Scholar). Recently, we have proven that the distal histidine (His-64) in sperm whale Mb is a critical residue in destabilizing Mb compound I (Mb-I). In fact, His-64 mutants of Mb gave Mb-I as an observable species under stopped-flow conditions. More importantly, the successful observation of Mb-I has allowed us the direct observation of the oxidation step of H2O2, olefins, thioethers, N-demethylation, and an aromatic ring by Mb-I (10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar, 11Matsui T. Ozaki S. Watanabe Y. J. Am. Chem. Soc. 1999; 121: 9952-9957Crossref Scopus (100) Google Scholar, 15Goto Y. Matsui T. Ozaki S. Watanabe Y. Fukuzumi S. J. Am. Chem. Soc. 1999; 121: 9497-9502Crossref Scopus (157) Google Scholar, 16Hara I. Ueno T. Ozaki S. Itoh S. Lee K. Ueyama N. Watanabe Y. J. Biol. Chem. 2001; 276: 36067-36070Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar). For example, H64X (X is Ale, Ser, and Asp) mutants are almost completely oxidized to Mb-I by m-chloroperbenzoic acid (mCPBA), and Mb-I is completely reduced by H2O2, although the reaction is about 1000–3000-fold slower than that reported for catalase (11Matsui T. Ozaki S. Watanabe Y. J. Am. Chem. Soc. 1999; 121: 9952-9957Crossref Scopus (100) Google Scholar). The Mb mutants (F43H/H64L and L29H/H64L Mb) have also been found to dismutate H2O2 to O2 and H2O at the rate which is 50- and 5-fold faster than wild-type Mb, respectively (10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar). The crystal structures of them suggest that the reactivity is controlled by the presence or absence of distal histidine (10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar). We report herein the kinetic isotope effect (KIE) on catalatic reactions (Equation 2) of catalase and Mb mutants in H2O and D2O. The detailed kinetic studies provide valuable insight into the role of distal histidine for the H2O2 oxidation by compound I of catalase and Mb mutants. Materials—Standard chemicals were obtained from Wako Chemicals and Nacalai Tesque. 99.9% deuterium oxide was purchased from Aldrich and Cambridge Isotope Laboratories. mCPBA was obtained from Nacalai Tesque and purified as reported previously (17Groves J.T. Watanabe Y. J. Am. Chem. Soc. 1986; 108: 7834-7836Crossref PubMed Scopus (143) Google Scholar). H218O2 was prepared from 18O2 as described by Foote and Sawaki (18Sawaki Y. Foote C.S. J. Am. Chem. Soc. 1979; 101: 6292-6296Crossref Scopus (64) Google Scholar). 18O content in H218O2 was determined by GC-MS analysis of triphenylphosphine oxide formed by the oxidation of triphenylphosphine with H218O2. The concentration of peroxide in a reaction solution was determined by oxidation of potassium iodide in the presence of horseradish peroxidase as a catalyst to produce I3− (ϵ353 = 2.62 × 104m-1 cm-1) (19Cotton M.L. Dunford H.B. J. Am. Chem. Soc. 1973; 51: 582-587Google Scholar, 20Schowen K.B. Schowen R.L. Methods Enzymol. 1982; 87: 551-606Crossref PubMed Scopus (482) Google Scholar). Beef liver catalase (BLC) and Micrococcus lysodeikticus catalase (MLC) were obtained from Sigma and Nagase ChemteX, respectively. Catalase was purified by Superdex 75 size exclusion chromatography. The concentrations of BLC and MLC were determined by using a molar extinction coefficient of ϵ405 = 3.24 × 105m-1 cm-1 (21Lardinois O.M. Mestdagh M.M. Rouxhet P.G. Biochim. Biophys. Acta. 1996; 1295: 222-238Crossref PubMed Scopus (124) Google Scholar) and ϵ406 = 1.03 × 105m-1 cm-1 (22Brill A.S. Williams R.J.P. Biochem. J. 1961; 78: 253-262Crossref PubMed Google Scholar), respectively. Mutant genes of H64A, H64S, H64D, F43H/H64L, and L29H/H64L sperm whale Mb were constructed by a method of Matsui et al. (10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar, 23Matsui T. Ozaki S. Watanabe Y. J. Biol. Chem. 1997; 272: 32735-32738Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar). Expression and purification of the mutants were performed according to a method described by Springer et al. (24Springer B.A. Egeberg K.D. Sligar S.G. Rohlfs R.J. Mathews A.J. Olson J.S. J. Biol. Chem. 1989; 264: 3057-3060Abstract Full Text PDF PubMed Google Scholar). Preparation of D2O Buffer—In a typical run 50 mm sodium acetate buffer was prepared by adding an appropriate amount of acetate acid in D2O and titrated with a concentrated sodium hydroxide solution of D2O. The pD of the buffer solution was determined by using an equation of pD = pHobs + 0.3314n + 0.076n2, where n represents the fraction of deuterium in the solution (25Bishop G.R. Davidson V.L. Biochemistry. 1995; 34: 12082-12086Crossref PubMed Scopus (63) Google Scholar), and pHobs is the apparent pH reading on an HM-30V pH meter (TOA electronics Ltd., Tokyo, Japan). D2O samples were prepared as follows. To an H2O solution (0.5 ml) containing a protein (1 mm)D2O buffer (4.5 ml) was added, and the resulting mixture was concentrated by ultrafiltration. The same procedure was repeated 5 times. After the completion of H-D exchange, the solution was incubated overnight in D2O buffer at 4 °C. Spectroscopy—Electronic absorption spectra were recorded on a UV-2400 spectrophotometer (Shimadzu Co., Ltd., Kyoto, Japan). Spectral changes were monitored either on an SF-43 stopped-flow apparatus (HI-TECH Scientific Co., Wiltshire, UK) equipped with an MG 6000 diode array spectrophotometer or on an RSP-601 stopped-flow rapid-scan spectrometer (Unisoku Co., Ltd., Osaka, Japan). Reaction of MLC and H2O2—For the determination of kinetic parameters k1 and k2, the MLC compound I (MLC-I) formation and the dismutation of H2O2 were monitored by following changes in absorbance at 406 and 240 nm, respectively, by stopped-flow technique in 50 mm sodium phosphate buffer (pH 7.0) at 5 °C. To determine the apparent rate constant of MLC-I formation (kapp-MLC-I), the MLC (1.0–1.2 μm) was mixed with a small excess amount (2.0–8.9 mol eq to MLC) of H2O2. The rate constant of the H2O2 dismutation by MLC (kapp-H2O2) was determined under steady-state conditions ([MLC] = 50 nm, [H2O2] = 1.8–16 mm). Reactions of Mb Mutants and mCPBA—To obtain the authentic absorption spectra of Mb compound I (Mb-I), Mb mutants (5.5–9.7 μm) were oxidized by more than 10 eq of mCPBA. At the same time, the Mb-I formation rates by mCPBA were determined in 50 mm sodium acetate buffer (pH 5.0) at 5 °C by following the decay of absorbance at 408 nm. Bimolecular rate constants are determined by plotting observed pseudo-first-order rates versus oxidant concentration. Reaction of Mb-I and H2O2—The reaction of Mb-I with H2O2 was directly monitored by use of a double-mixing stopped-flow technique. To avoid catalytic catalase reactions (Equations 1 and 2), the first mixing of a Mb mutant (5.5–7.8 μm) to prepare Mb-I was carried out with 1.5 mol eq of mCPBA. Subsequently, at least a 100-fold excess of H2O2 was introduced to start the catalatic reaction (Equation 2) in 50 mm sodium acetate buffer (pH 5.0) at 5 °C. The catalatic reaction was monitored by following an increase in absorbance at 408 nm. Delay time for the second mixing was in a range of 0.1–1.1 s. The reproduction of Mb-I by remaining H2O2 is too slow to observe under the present experimental conditions. To examine the temperature dependence of the H64A Mb-I reaction with H2O2, the catalatic reaction was repeated at 5, 10, 15, 20, and 25 °C. Activation parameters (A and Ea) were determined by an Arrhenius plot: ln k2 = ln A - Ea/RT. All experiments were repeated at least twice. Identification of Evolved Oxygen—An Mb mutant (200 μm in 50 mm sodium acetate buffer (pH 5.0)) or BLC (25 μm in 50 mm sodium phosphate buffer (pH 7.0)) was mixed with H216O2/H218O2 (1:1) (200 mm) at room temperature under N2 for 5 min. Three different evolved molecular oxygens, 16O2 (m/e = 32), 16O18O (m/e = 34), and 18O2 (m/e = 36), were separated and quantified by a Shimadzu GC-17A/GC-MS-QP5000 (Shimadzu Co., Ltd., Kyoto, Japan) equipped with a Varian Molsieve 5A PLOT capillary column. Catalase Reaction of MLC in H2O and D2O—UV-visible spectral changes of MLC by the addition of peroxides in H2O and D2O are shown in Fig. 1. The spectral changes with hydrogen peroxide are different from those with methyl hydroperoxide. The steady-state spectra in the presence of hydrogen peroxide show that 45 and 85% of MLCs exist as MLC-I in H2O and D2O buffers, respectively (Fig. 1A). On the other hand, the spectra of MLC with methyl hydroperoxide indicate that MLC has been completely converted to MLC-I in both buffers (Fig. 1B). Fig. 2 shows the time course of absorbance change in the reactions of MLC and 2–8.9 mol eq of hydrogen peroxide. The steady-state concentration of MLC-I in each run is almost the same even though the amount of hydrogen peroxide was varied. Thus, we could determine the rate constants, k1 and k2, without considering the equilibrium steps (Equations 3 and 5) under these experimental conditions (see Supplemental Material, Appendices 1 and 2). To obtain the apparent rate constants of MLC-I formation (kapp-MLC-I), the MLC (1.0–1.2 μm) was mixed with a small excess amount (2.0–8.9 mol eq to MLC) of H2O2. The rate constants of the H2O2 dismutation by MLC (kapp-H2O2) were determined under steady-state conditions ([MLC] = 50 nm, [H2O2] = 1.8–16 mm). The apparent second-order rate constants (kapp-MLC-I) for MLC-I formation by H2O2 were determined from the Soret absorbance change at 406 nm as (2.11 ± 0.06) × 107 and (1.19 ± 0.02) × 107m-1 s-1 in H2O and D2O, respectively. In addition, the apparent rate constants (kapp-H2O2) of H2O2 dismutation by MLC-I were determined by the decrease of H2O2 absorption at 240 nm (Fig. 3) as (9.33 ± 0.10) × 106 and (2.86 ± 0.04) × 106m-1 s-1 in H2O and D2O, respectively. The apparent rate constants kapp-MLC-I and kapp-H2O are expressed as a function of k1 (Equation 1) and k2 (Equation 2): kapp-MLC-I = k1 + k2 (see Supplemental Material, Appendix 1, Equation 15) and kapp-H2O2 = 2k1k2/(k1 + k2) (Equation 20 in Appendix 2). Thus, the k1 and k2 values are determined from the kapp-MLC-I and kapp-H2O2 values, and they are listed in Tables I and II, respectively. On the other hand, the observed concentrations of MLCs in equilibrium are given by [MLC]eq = (k2/k1 + k2)[MLC]0 (Equation 8 in Appendix 1). Using this equation, the concentrations of MLC-I in H2O and D2O are calculated from the k1 and k2 values as 43.1 ± 6.8 and 74.1 ± 7.7%, respectively. These values agree with the concentrations of MLC-I estimated from the steady-state spectra in Fig. 1. Such agreement confirms the validity of our kinetic analysis. We have also determined the KIEs on k1 and k2 of MLC to be 1.0 and 4.0, respectively.MLC+H2O2⇄MLC-H2O2(Eq. 3) MLC-H2O2→MLC-I+H2O(Eq. 4) MLC-I+H2O2⇄MLC-I-H2O2(Eq. 5) MLC-I-H2O2→MLC+H2O+O2(Eq. 6) Fig. 2Absorption changes in MLC reaction with deuterium peroxide (2–8.9 μm). The reactions were carried out in 50 mm sodium phosphate buffer solutions (pH 7.0) at 5 °C. The final MLC concentration was 1.0 μm. The inset shows a plot of ([H2O2]0 - [MLC]0 + [MLC]eq)kapp-MLC-Iversus [H2O2]0 - [MLC]0 + [MLC]eq for the reaction of MLC with hydrogen peroxide.View Large Image Figure ViewerDownload (PPT)Fig. 3Time-dependent decrease of H2O2 absorption due to the catalase activity of MLC. The catalase reactions were carried out in a 50 mm sodium phosphate buffer solution (pH 7.0) at 5 °C. The final MLC concentration was 50 nm. The inset shows a plot of kapp-H2O2versus hydrogen peroxide concentration for the dismutation reactions of hydrogen peroxide by MLC.View Large Image Figure ViewerDownload (PPT)Table IRate constants (k1) of Mb and catalase compound I formation Mbs were in 50 mm sodium acetate buffer (pL 5.0) at 5.0 °C. MLCs were in 50 mm sodium phosphate buffer (pL 7.0) at 5.0 °C.MbOxidantsk1HaIn H2O bufferk1DbIn D2O bufferk1H/k1Dm-1 s-1m-1 s-1H64AmCPBA9.87 ± 0.01 × 1056.26 ± 0.01 × 1051.6H64SmCPBA6.29 ± 0.01 × 1054.42 ± 0.01 × 1051.4H64DmCPBA5.26 ± 0.01 × 1053.28 ± 0.01 × 1051.6H64DH2O212.4 ± 0.1 × 1030.80 ± 0.02 × 10315L29H/H64LmCPBA4.18 ± 0.03 × 1054.20 ± 0.05 × 1051.0F43H/H64LmCPBA6.67 ± 0.03 × 1058.77 ± 0.27 × 1050.8MLCH2O29.09 ± 1.45 × 1068.82 ± 0.92 × 1061.0a In H2O bufferb In D2O buffer Open table in a new tab Table IIRate constants (k2) of the catalatic reaction Mbs were in 50 mm sodium acetate buffer (pL 5.0) at 5.0 °C. MLCs were in 50 mm sodium phosphate buffer (pL 7.0) at 5.0 °C.Mbk2HaIn H2O bufferk2DbIn D2O bufferk2H/k2Dm-1 s-1m-1 s-1H64A Mb5.26 ± 0.06 × 1032.32 ± 0.02 × 10223H64S Mb6.91 ± 0.01 × 1032.38 ± 0.01 × 10229H64D Mb15.8 ± 0.3 × 1038.18 ± 0.02 × 10218L29H/H64L Mb33.6 ± 0.1 × 1033.39 ± 0.03 × 10310F43H/H64L Mb21.0 ± 0.5 × 10310.3 ± 0.1 × 1032.1MLC1.20 ± 1.98 × 1073.08 ± 0.32 × 1064.0a In H2O bufferb In D2O buffer Open table in a new tab Mb-I Formation in H2O and D2O—The second-order rate constants (k1) for the compound I formation of the Mb mutants by mCPBA were determined (see "Experimental Procedures") as listed in Table I (23Matsui T. Ozaki S. Watanabe Y. J. Biol. Chem. 1997; 272: 32735-32738Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 26Araiso T. Rutter R. Palcic M.M. Hager L.P. Dunford H.B. Can. J. Biochem. 1981; 59: 233-236Crossref PubMed Scopus (35) Google Scholar). The H64A mutant affords Mb-I by the addition of mCPBA as shown in Fig. 4A. At 320 ms after the mixing, Soret absorption of H64A decreased to less than half, and a broad band having a peak at 648 nm appeared. The rate constants of the Mb-I formation with mCPBA in H2O and D2O indicate small kH/kD values of 0.8–1.6. On the other hand, H64D Mb-I formation with H2O2 shows a larger k1H/k1D value (Table I). We attempted to observe the other Mb-I formation by H2O2, but the other Mb-Is could not be formed with H2O2 under the same conditions. Catalatic Reaction of Mb Mutants in H2O and D2O—To observe the stoichiometric reaction of Mb-I with H2O2 spectroscopically, we have performed double-mixing stopped-flow experiments at 5.0 °C and pH 5.0. A small excess amount (1.5 mol eq to Mb) of mCPBA was used for the completion of Mb-I formation (23Matsui T. Ozaki S. Watanabe Y. J. Biol. Chem. 1997; 272: 32735-32738Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar). For example, H64A Mb gave Mb-I up to 95% yield under these conditions (Fig. 4A). A decrease in the absorbance of the Soret band and an increase in the absorbance at 648 nm are the clear indications of Mb-I formation. Other Mb mutants are also oxidized to Mb-I under the same experimental conditions. Upon the addition of H2O2 to a H64A Mb-I solution (second mixing), Mb-I is reduced to the ferric form without forming any intermediates (Fig. 4B). Similar spectral changes are also observed for the other mutants, revealing that the Mb-I is capable of performing two-electron oxidation of H2O2. The reproduction of compound I by the remaining H2O2 in the solution is too slow to observe under the present experimental conditions. The rate constants of the catalatic reaction in H2O and D2O (k2H and k2D) were determined as described above, and the k2H and k2D values are listed in Table II. Replacement of the H2O solvent by D2O caused 10–29-fold decrease in the rates of the catalatic reactions of H64X and L29H/H64L Mbs. Such large primary kinetic isotope effects suggest the occurrence of tunneling. Accurate measurements of the primary kinetic isotope effects as a function of temperature are very important for the examination of the tunneling effect (27Kwart H. Acc. Chem. Res. 1982; 15: 401-408Crossref Scopus (241) Google Scholar). The incidence of tunneling results in the curvature of the Arrhenius plot (ln k = ln A - Ea/RT) in a wide range of temperature, which is larger for hydrogen as compared with deuterium. In the temperature range of measurements (5–25 °C), a larger AD value than AH, derived from an apparent linear Arrhenius plot, can be used as criteria to recognize the tunneling (AH/AD < 0.6) (Table III) (27Kwart H. Acc. Chem. Res. 1982; 15: 401-408Crossref Scopus (241) Google Scholar). The AH/AD values (0.07) derived from the Arrhenius plots for H64A Mb, thus, indicate that large KIEs are ascribed to the linear tunneling effect (Fig. 5) (27Kwart H. Acc. Chem. Res. 1982; 15: 401-408Crossref Scopus (241) Google Scholar). In contrast to H64A Mb, F43H/H64L Mb and MLC exhibited small KIEs of 2.1 and 4.0, respectively (we could not determine their thermodynamic parameters due to less stability and rapid reaction above 5 °C, respectively).Table IIITemperature dependence of the reaction of H64A Mb-I with H2O2ParameterH64AkH (m-1 s-1a50 mm sodium acetate buffer (pL 5.0) at 5 °C5260kD (m-1 s-1)a50 mm sodium acetate buffer (pL 5.0) at 5 °C232KIEa50 mm sodium acetate buffer (pL 5.0) at 5 °C23ΔEa(H) (kcal·mol-1)b50 mm sodium acetate buffer (pL 5.0) at 5, 10, 15, 20, and 25 °C0.47ΔEa(D) (kcal·mol-1)b50 mm sodium acetate buffer (pL 5.0) at 5, 10, 15, 20, and 25 °C1.3AH/ADb50 mm sodium acetate buffer (pL 5.0) at 5, 10, 15, 20, and 25 °C0.07ΔH1(O-H) (kcal l·mol-1)b50 mm sodium acetate buffer (pL 5.0) at 5, 10, 15, 20, and 25 °C0.33ΔH1(O-D) (kcal·mol-1)b50 mm sodium acetate buffer (pL 5.0) at 5, 10, 15, 20, and 25 °C1.2a 50 mm sodium acetate buffer (pL 5.0) at 5 °Cb 50 mm sodium acetate buffer (pL 5.0) at 5, 10, 15, 20, and 25 °C Open table in a new tab Identification of Evolved Oxygen—Preparation of 18O-labeled H2O2 has allowed us to examine the source of the evolved O2. A GC-MS spectrum of the evolved O2 from a solution containing a 50:50 mixture of H218O2/H216O2 and a Mb mutant (H64D or F43H/H64L) shows two peaks for 18O2 (m/e = 36) and 16O2 (m/e = 32) with no indication of 16O18O (m/e = 34) formation within an experimental error of ∼3%. This demonstrates that the catalase reactions by the Mb mutants proceed as the deprotonation process of catalase and chloroperoxidase (28Hager L.P. Doubek D.L. Silverstein R.M. Hargis J.H. Martin J.C. J. Am. Chem. Soc. 1972; 94: 4364-4366Crossref PubMed Scopus (92) Google Scholar, 29Jarnagin R.C. Wang J.H. J. Am. Chem. Soc. 1958; 80: 786-787Crossref Scopus (29) Google Scholar). In early studies the rate constants of compound I formation for human erythrocyte, horse liver, and MLC were reported to be in a range of 1.2–6 × 107m-1 s-1 (30Bonnichsen R.K. Chance B. Theorell H. Acta Chem. Scand. 1947; 1: 685-709Crossref Google Scholar, 31George P. Biochem. J. 1949; 44: 197-205Crossref PubMed Scopus (24) Google Scholar, 32Chance B. Biochem. J. 1950; 46: 387-402Crossref PubMed Scopus (95) Google Scholar, 33Chance B. Herbert D. Biochem. J. 1950; 46: 402-414Crossref PubMed Scopus (90) Google Scholar). On the other hand, the k1 and k2 values of BLC in the presence of 0.25–1.00 m ethanol are 5.6 × 106 and 1.2 × 107m-1 s-1, respectively (4Kremer M.L. Biochim. Biophys. Acta. 1970; 198: 199-209Crossref PubMed Scopus (47) Google Scholar). The k1 of BLC, which is smaller than those of the other catalases, indicates possible inhibition by acetaldehyde or denaturation of BLC under these conditions. Thus, we have employed MLC to determine k1 and k2 in the absence of ethanol to elucidate a detailed mechanism of the catalatic reaction. In the case of MLC, high catalatic activity has precluded the complete accumulation of MLC-I by H2O2, and the preparation of MLC-I required a large excess of the oxidant even by using methyl hydroperoxide. The rate constant (k2) of the catalatic reaction of MLC could not be determined directly due to a rapid reformation of MLC-I by H2O2 remaining in the reaction solution. Thus, we have determined k1 and k2 by the measurement of the apparent MLC-I formation and its reduction rates by hydrogen peroxide as 9.1 × 106 and 1.2 × 107m-1 s-1, respectively (see Tables I and II). In the case of MLC, a small KIE (1.0) for the formation of MLC-I has been observed. This value is very similar to that reported for horseradish peroxidase I formation (1.6) by Dunford et al. (34Dunford H.B. Hewson W.D. Steiner H. Can. J. Chem. 1978; 56: 2844-2852Crossref Google Scholar). Therefore, distal histidine in MLC could serve as a general acid-base catalyst for the formation of MLC-I as proposed for peroxidases. On the other hand, the reaction of H64D Mb and H2O2 shows a large KIE (kH/kD = 15). Because the compound I formation mechanism includes deprotonation from H2O2 to afford a Fe-OOH intermediate, we have assigned that the deprotonation of hydrogen peroxide is a great barrier for the formation of H64D Mb-I due to a lack of the distal histidine. These results suggest that the distal histidine might reduce the pKa value of H2O2 (11.6) by ∼4 pKa units in the active site of MLC. Although the reaction of MLC-I with hydrogen peroxide gave a mixture of MLC and MLC-I in the steady state because the resulting MLC is immediately converted to MLC-I by excess H2O2 remaining in the solution, we have successfully determined the deuterium isotope effect on the catalatic reaction of MLC to be 4.0 on the basis of apparent MLC-I formation and MLC-I reduction rates (Table II). F43H/H64L Mb also gave a small KIE (2.1), determined by the direct observation of the reaction of Mb-I and H2O2. These results indicate that hydrogen peroxide is easily deprotonated by the distal histidine and then reacts with compound I of MLC and F43H/H64L Mb. On the other hand, in the reactions of H2O2 and Mb-I of L29H/H64L and H64X (X = Asp, Ala, and Ser), the KIEs are extremely larger (10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar, 11Matsui T. Ozaki S. Watanabe Y. J. Am. Chem. Soc. 1999; 121: 9952-9957Crossref Scopus (100) Google Scholar, 12Ortiz de Montellano P.R. Catalano C.E. J. Biol. Chem. 1985; 260: 9265-9271Abstract Full Text PDF PubMed Google Scholar, 13Rao S.I. Wilks A. Ortiz de Montellano P.R. J. Biol. Chem. 1993; 268: 803-809Abstract Full Text PDF PubMed Google Scholar, 14Tschirret-Guth R.A. Ortiz de Montellano P.R. Arch. Biochem. Biophys. 1996; 335: 93-101Crossref PubMed Scopus (34) Google Scholar, 15Goto Y. Matsui T. Ozaki S. Watanabe Y. Fukuzumi S. J. Am. Chem. Soc. 1999; 121: 9497-9502Crossref Scopus (157) Google Scholar, 16Hara I. Ueno T. Ozaki S. Itoh S. Lee K. Ueyama N. Watanabe Y. J. Biol. Chem. 2001; 276: 36067-36070Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar, 17Groves J.T. Watanabe Y. J. Am. Chem. Soc. 1986; 108: 7834-7836Crossref PubMed Scopus (143) Google Scholar, 18Sawaki Y. Foote C.S. J. Am. Chem. Soc. 1979; 101: 6292-6296Crossref Scopus (64) Google Scholar, 19Cotton M.L. Dunford H.B. J. Am. Chem. Soc. 1973; 51: 582-587Google Scholar, 20Schowen K.B. Schowen R.L. Methods Enzymol. 1982; 87: 551-606Crossref PubMed Scopus (482) Google Scholar, 21Lardinois O.M. Mestdagh M.M. Rouxhet P.G. Biochim. Biophys. Acta. 1996; 1295: 222-238Crossref PubMed Scopus (124) Google Scholar, 22Brill A.S. Williams R.J.P. Biochem. J. 1961; 78: 253-262Crossref PubMed Google Scholar, 23Matsui T. Ozaki S. Watanabe Y. J. Biol. Chem. 1997; 272: 32735-32738Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 24Springer B.A. Egeberg K.D. Sligar S.G. Rohlfs R.J. Mathews A.J. Olson J.S. J. Biol. Chem. 1989; 264: 3057-3060Abstract Full Text PDF PubMed Google Scholar, 25Bishop G.R. Davidson V.L. Biochemistry. 1995; 34: 12082-12086Crossref PubMed Scopus (63) Google Scholar, 26Araiso T. Rutter R. Palcic M.M. Hager L.P. Dunford H.B. Can. J. Biochem. 1981; 59: 233-236Crossref PubMed Scopus (35) Google Scholar, 27Kwart H. Acc. Chem. Res. 1982; 15: 401-408Crossref Scopus (241) Google Scholar, 28Hager L.P. Doubek D.L. Silverstein R.M. Hargis J.H. Martin J.C. J. Am. Chem. Soc. 1972; 94: 4364-4366Crossref PubMed Scopus (92) Google Scholar, 29Jarnagin R.C. Wang J.H. J. Am. Chem. Soc. 1958; 80: 786-787Crossref Scopus (29) Google Scholar) than those of F43H/H64L Mb and MLC (Table II). Large KIEs are due to the tunneling effect, which is confirmed by the Arrhenius parameter ratio (Table III) (27Kwart H. Acc. Chem. Res. 1982; 15: 401-408Crossref Scopus (241) Google Scholar). In previous studies we have proven that the distal histidine (His-43) in F43H/H64L Mb serves as a general acid-base catalyst to form compound I in the reaction with H2O2. On the other hand, distal aspartic acid (Asp-64) in H64D and the distal histidine (His-29) in L29H/H64L Mb hardly participate as the catalyst (10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar, 11Matsui T. Ozaki S. Watanabe Y. J. Am. Chem. Soc. 1999; 121: 9952-9957Crossref Scopus (100) Google Scholar). The crystal structure of F43H/H64L Mb shows the distance between the Nϵ of His-43 and the ferric heme iron to be 5.7 Å, similar to structurally known peroxidases and MLCs (Fig. 6, A, C, and D) (2Fita I. Rossmann M.G. J. Mol. Biol. 1985; 185: 21-37Crossref PubMed Scopus (358) Google Scholar, 10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar, 35Poulos T.L. Freer S.T. Alden R.A. Edwards S.L. Skogland U. Takio K. Eriksson B. Xuong N. Yonetani T. Kraut J. J. Biol. Chem. 1980; 255: 575-580Abstract Full Text PDF PubMed Google Scholar). On the other hand, the distance in L29H/H64L Mb is too far from the heme iron (6.6 Å) to serve as the general acid-base catalyst (Fig. 6B) (10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar). Thus, the catalatic reaction of MLC and F43H/H64L Mb, in which the general acid-base catalyst is located at a proper position, could proceed via the ionic mechanism with a small KIE (<4) (Scheme 1A), whereas the other Mb mutants oxidize H2O2 via a mechanism that could be different from the ionic mechanism with exhibiting a large KIE (10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar, 11Matsui T. Ozaki S. Watanabe Y. J. Am. Chem. Soc. 1999; 121: 9952-9957Crossref Scopus (100) Google Scholar, 12Ortiz de Montellano P.R. Catalano C.E. J. Biol. Chem. 1985; 260: 9265-9271Abstract Full Text PDF PubMed Google Scholar, 13Rao S.I. Wilks A. Ortiz de Montellano P.R. J. Biol. Chem. 1993; 268: 803-809Abstract Full Text PDF PubMed Google Scholar, 14Tschirret-Guth R.A. Ortiz de Montellano P.R. Arch. Biochem. Biophys. 1996; 335: 93-101Crossref PubMed Scopus (34) Google Scholar, 15Goto Y. Matsui T. Ozaki S. Watanabe Y. Fukuzumi S. J. Am. Chem. Soc. 1999; 121: 9497-9502Crossref Scopus (157) Google Scholar, 16Hara I. Ueno T. Ozaki S. Itoh S. Lee K. Ueyama N. Watanabe Y. J. Biol. Chem. 2001; 276: 36067-36070Abstract Full Text Full Text PDF PubMed Scopus (35) Google Scholar, 17Groves J.T. Watanabe Y. J. Am. Chem. Soc. 1986; 108: 7834-7836Crossref PubMed Scopus (143) Google Scholar, 18Sawaki Y. Foote C.S. J. Am. Chem. Soc. 1979; 101: 6292-6296Crossref Scopus (64) Google Scholar, 19Cotton M.L. Dunford H.B. J. Am. Chem. Soc. 1973; 51: 582-587Google Scholar, 20Schowen K.B. Schowen R.L. Methods Enzymol. 1982; 87: 551-606Crossref PubMed Scopus (482) Google Scholar, 21Lardinois O.M. Mestdagh M.M. Rouxhet P.G. Biochim. Biophys. Acta. 1996; 1295: 222-238Crossref PubMed Scopus (124) Google Scholar, 22Brill A.S. Williams R.J.P. Biochem. J. 1961; 78: 253-262Crossref PubMed Google Scholar, 23Matsui T. Ozaki S. Watanabe Y. J. Biol. Chem. 1997; 272: 32735-32738Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar, 24Springer B.A. Egeberg K.D. Sligar S.G. Rohlfs R.J. Mathews A.J. Olson J.S. J. Biol. Chem. 1989; 264: 3057-3060Abstract Full Text PDF PubMed Google Scholar, 25Bishop G.R. Davidson V.L. Biochemistry. 1995; 34: 12082-12086Crossref PubMed Scopus (63) Google Scholar, 26Araiso T. Rutter R. Palcic M.M. Hager L.P. Dunford H.B. Can. J. Biochem. 1981; 59: 233-236Crossref PubMed Scopus (35) Google Scholar, 27Kwart H. Acc. Chem. Res. 1982; 15: 401-408Crossref Scopus (241) Google Scholar, 28Hager L.P. Doubek D.L. Silverstein R.M. Hargis J.H. Martin J.C. J. Am. Chem. Soc. 1972; 94: 4364-4366Crossref PubMed Scopus (92) Google Scholar, 29Jarnagin R.C. Wang J.H. J. Am. Chem. Soc. 1958; 80: 786-787Crossref Scopus (29) Google Scholar) due to a lack of the general acid-base catalyst.Scheme 1Proposed mechanisms for the catalatic reaction. A, ionic mechanism by utilizing a general acid-base catalyst. B, radical mechanism. C, role of a general acid-base catalyst on the formation of compound I.View Large Image Figure ViewerDownload (PPT) Exclusive formation of 18O2 and 16O2 from a 50:50 mixture of H216O2 and H218O2 indicates that O2 is formed by two-electron oxidation of H2O2 without breaking the O-O bond. There are two possible mechanisms on the formation of O2 in the reaction of H2O2 and compound I without showing 18O/16O scrambling as depicted in Scheme 1. Mechanism A shows an ionic reaction via initial proton abstraction with the help of the distal histidine acting as a general acid-base catalyst for MLC and F43H/H64L Mbs (Scheme 1A). A similar deprotonation process is involved in the formation of compound I (Scheme 1C) (36Ozaki S. Roach M.P. Matsui T. Watanabe Y. Acc. Chem. Res. 2001; 34: 818-825Crossref PubMed Scopus (141) Google Scholar). In mechanism B the reaction starts by a hydrogen atom transfer from H2O2 to the ferryl species to yield a radical intermediate. The hydrogen abstraction by the ferryl intermediate has been proposed for the alkane hydroxylation by cytochrome P450, its model complexes, and even in non-heme enzymes (37Neshein J.C. Lipscomb J.D. Biochemistry. 1996; 35: 10240-10247Crossref PubMed Scopus (240) Google Scholar, 38Sorokin A. Robert A. Meunier B. J. Am. Chem. Soc. 1993; 115: 7293-7299Crossref Scopus (125) Google Scholar). In these reactions large KIEs in a range of 9–29 are commonly observed due to a tunneling effect (27Kwart H. Acc. Chem. Res. 1982; 15: 401-408Crossref Scopus (241) Google Scholar). We have also observed the tunneling effect (AH/AD = 0.07) for the catalatic reaction of H64A Mb-I with KIE of 23 at 5 °C, suggesting the involvement of hydrogen abstraction through a tunneling process. Although the histidine residue at the position 43 helps the ionic oxidation of hydrogen peroxide, the k2 value of F43H/H64L Mb is virtually the same as that for H64D Mb and smaller than that of L29H/H64L (Table II). The crystal structure of H64D Mb suggests that an Asp residue at position 64 could help to incorporate hydrogen peroxide in the active site due to the enlargement of the active site as well as hydrogen bonding (39Yang H. Matsui T. Ozaki S. Kato S. Ueno T. Phillips Jr., G.N. Fukuzumi S. Watanabe Y. Biochemistry. 2003; 42: 10174-10181Crossref PubMed Scopus (31) Google Scholar). Although His at position 29 of L29H/H64L Mb is located too far from the heme iron to play as the acid-base catalyst, L29H/H64L Mb provides a larger space for the accommodation of hydrogen peroxide than the F43H/H64L mutant, and His-29 might help to stabilize the Mb-I form by a polar effect (10Matsui T. Ozaki S. Liong E. Phillips Jr., G.N. Watanabe Y. J. Biol. Chem. 1999; 274: 2838-2844Abstract Full Text Full Text PDF PubMed Scopus (177) Google Scholar). Our results suggest that important factors for the catalatic reaction are not only deprotonation but also enough space and polarity for the accommodation of hydrogen peroxide at a suitable position in the active site. In fact, distal cavity mutation of a Trp residue in catalase-peroxidase completely depressed its catalase activity but retained peroxidase activity. It has been concluded that the indole ring is involved in the binding of a H2O2 molecule (40Hillar A. Peters B. Pauls R. Loboda A. Zhang H. Mauk A.G. Loewen P.C. Biochemistry. 2000; 39: 5868-5875Crossref PubMed Scopus (95) Google Scholar, 41Regelsberger G. Jakopitsch C. Ruker F. Krois D. Peschek G.A. Obinger C. J. Biol. Chem. 2000; 275: 22854-22861Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar). In summary, we have provided solid evidence of the importance of a general acid base catalyst for the ionic catalatic reaction of MLC-I and Mb-I. The KIEs on the reaction of Mb-I mutants with H2O2 indicate that there are two different mechanisms for the catalatic reaction, i.e. ionic and radical mechanisms, depending on the presence and absence of the distal histidine acting as a general acid-base catalyst. In addition, Mb mutants indicate that the reaction of compound I with hydrogen peroxide can be accelerated not only by the acid-base catalyst but also by space and polarity of the amino acid residues in the active site. We thank Prof. Koji Tanaka and Dr. Tetsunori Mizukawa of the Institute for Molecular Science (Okazaki, Japan) for the GC-MS analysis. Download .pdf (.06 MB) Help with pdf files
Referência(s)