Formation and catalytic roles of compound I in the hydrogen peroxide-dependent oxidations by His64 myoglobin mutants

Article abstract of DOI:10.1021/ja9914846

A His64 → Asp mutant of sperm whale myoglobin (Mb), H64D Mb, has been prepared to mimic the active site of chloroperoxidase from the marine fungus Caldariomyces fumago, in which distal glutamic acid is suggested to enhance compound I formation by H_2<

Full text of DOI:10.1021/ja9914846

9952
J. Am. Chem. Soc. 1999, 121, 9952-9957
Formation and Catalytic Roles of Compound I in the Hydrogen
Peroxide-Dependent Oxidations by His64 Myoglobin Mutants
Toshitaka Matsui,† Shin-ichi Ozaki,†,§ and Yoshihito Watanabe*,†,‡
Contribution from the Institute for Molecular Science, Myodaiji, Okazaki 444-8585, Japan, and
Department of Structural Molecular Science, The Graduate UniVersity for AdVanced Studies,
Myodaiji, Okazaki 444-8585, Japan
ReceiVed May 5, 1999
Abstract: A His64 f Asp mutant of sperm whale myoglobin (Mb), H64D Mb, has been prepared to mimic
the active site of chloroperoxidase from the marine fungus Caldariomyces fumago, in which distal glutamic
acid is suggested to enhance compound I formation by H2O2. The H64D mutant allows us to see the accumulation
of compound I in the reaction of Mb with H2O2 for the first time. The successful observation of compound I
is due to at least 50-fold improvement in the formation rate of compound I as well as its stabilization upon the
His64 f Asp replacement. Catalytic activity of wild-type Mb and a series of His64 Mb mutants (H64A,
H64S, H64L, and H64D Mb) are examined for one-electron oxidation and oxygenation by using H2O2 as an
oxidant. The H64D mutant is the best catalyst among the myoglobins and shows 50-70-fold and 600-800-
fold higher activity than the wild type in the one-electron oxidations and peroxygenations, respectively. The
origin of the varied activity upon the mutations is discussed on the basis of the formation rate and stability of
compound I.
Introduction
Incubation of Mb with H2O2 causes slow conversion (∼102 M-1
s-1) of the ferric heme to a ferryl heme (FeIVdO Por), similar
to compound II in horseradish peroxidase (HRP).8,15,16 Com-
pound II has only one oxidizing equivalent above the ferric state
despite two oxidizing equivalents of H2O2. One more oxidizing
equivalent in Mb is found as a protein radical, leading to protein
dimerization (Figure 1).16,17 On the contrary, peroxidases and
catalases readily react with H2O2 (∼107 M-1 s-1) to give a two-
electron oxidized heme (compound I), which is normally a ferryl
porphyrin cation radical (FeIVdO Por+•).18,19 Compound I in
Mb has never been observed for the native and wild-type
proteins. While partial homolysis of the peroxide bond can form
compound II (Figure 1), heterolysis has been suggested to be a
major process.20 Thus, the absence of Mb compound I appears
to be due to its rapid decay to compound II.
Myoglobin (Mb),1 a carrier of molecular oxygen, is one of
the most extensively examined hemoproteins, and is a good
model for elucidating the role of active site residues in the
interaction of hemoproteins with small molecules such as
dioxygen, carbon monoxide, and cyanide.2-5 The reaction
between Mb and H2O2 also has been studied over three decades
from such perspectives as a possible source of myocardial
reperfusion injury,6,7 a model for the reactions catalyzed by
peroxidase and catalase,8-10 and, like cytochrome P-450, a
catalyst for the oxygenation of olefins and sulfides.11-14
* To whom correspondence should be addressed. E-mail: yoshi@ims.ac.jp.
† Institute for Molecular Science.
§ Present address: Faculty of Education, Yamagata University, Kojiraka-
wa, Yamagata 990-8560, Japan.
‡ The Graduate University for Advanced Studies.
A histidine residue in a distal heme pocket (His64 in sperm
whale Mb) has been shown as one of the radical sites of Mb
treated by H2O2.21,22 Recently, we have reported that His64
mutants (H64A, H64S, and H64L mutants) of sperm whale Mb
(1) Abbreviations used: Mb, myoglobin; HRP, horseradish peroxidase;
CPO, chloroperoxidase from the marine fungus Caldariomyces fumago;
compound I, a ferryl porphyrin cation radical; compound II, a ferryl
porphyrin; mCPBA, m-chloroperbenzoic acid; ABTS, 2,2′-azino-bis(3-
ethylbenzothiazoline-6-sulfonic acid).
(2) Olson, J. S.; Phillips, G. N. J. Biol. Chem. 1996, 271, 17593-17596.
(3) Springer, B. A.; Egeberg, K. D.; Sligar, S. G.; Rohlfs, R. J.; Mathews,
A. J.; Olson, J. S. J. Biol. Chem. 1989, 264, 3057-3060.
(4) Brancaccio, A.; Cutruzzola, F.; Allocatelli, C. T.; Brunori, M.;
Smerdon, S. J.; Wilkinson, A. J.; Dou, Y.; Keenan, D.; Ikeda-Saito, M.;
Brantley, R. E., Jr.; Olson, J. S. J. Biol. Chem. 1994, 269, 13843-13853.
(5) Dou, Y.; Olson, J. S.; Wilkinson, A. J.; Ikeda-Saito, M. Biochemistry
1996, 35, 7107-7113.
(13) Matsui, T.; Nagano, S.; Ishimori, K.; Watanabe, Y.; Morishima, I.
Biochemistry 1996, 35, 13118-13124.
(14) Adachi, S.; Nagano, S.; Ishimori, K.; Watanabe, Y.; Morishima, I.;
Egawa, T.; Kitagawa, T.; Makino, R. Biochemistry 1993, 32, 241-252.
(15) King, N. K.; Winfield, M. E. J. Biol. Chem. 1963, 238, 1520-
1528.
(16) King, N. K.; Looney, F. D.; Winfield, M. E. Biochim. Biophys. Acta
1967, 133, 65-82.
(6) Fantone, J.; Jester, S.; Loomis, T. J. Biol. Chem. 1989, 264, 9408-
9411.
(17) Tew, D.; Ortiz de Montellano, P. R. J. Biol. Chem. 1988, 263,
17880-17886.
(7) Grisham, M. B. J. Free Radicals Biol. Med. 1985, 1, 227-232.
(8) Yonetani, T.; Schleyer, H. J. Biol. Chem. 1967, 242, 1974-1979.
(9) Krishna, M. C.; Samuni, A.; Taira, J.; Goldstein, S.; Mitchell, J. B.;
Russo, A. J. Biol. Chem. 1996, 271, 26018-26025.
(10) Sievers, G.; Ro¨nnberg, M. Biochim. Biophys. Acta 1978, 533, 293-
301.
(11) Ortiz de Montellano, P. R.; Catalano, C. E. J. Biol. Chem. 1985,
260, 9265-9271.
(12) Rao, S. I.; Wilks, A.; Ortiz de Montellano, P. R. J. Biol. Chem.
1993, 268, 803-809.
(18) Dunford, H. B.; Stillman, J. S. Coord. Chem. ReV. 1987, 19, 187-
251.
(19) Schonbaum, G. R.; Chance, B.; The Enzymes, 3rd ed.; Boyer, P.
D., Ed.; Academic Press: New York, 1976; Vol. 13, pp 363-408.
(20) Allentoff, A. J.; Bolton, J. L.; Wilkis, A.; Thompson, J. A.; Ortiz
de Montellano, P. R. J. Am. Chem. Soc. 1992, 114, 9744-9749.
(21) Phillips, G. N., Jr.; Arduini, R. M.; Springer, B. A.; Sligar, S. G.
Proteins: Struct. Funct. Genet. 1990, 7, 358-365.
(22) Fenwick, C. W.; English, A. M. J. Am. Chem. Soc. 1996, 118,
12236-12237.
10.1021/ja9914846 CCC: $18.00 © 1999 American Chemical Society
Published on Web 11/03/1999

Compound I of Myoglobin
J. Am. Chem. Soc., Vol. 121, No. 43, 1999 9953
activity than the wild type in the one-electron oxidations and
peroxygenations, respectively. The origin of the varied activity
upon the mutations is discussed on the basis of the formation
rate and stability of compound I.
Experimental Procedures
Materials. The mutant genes for H64D, H64A, H64S, and H64L
sperm whale Mb were constructed by cassette mutagenesis. The cassette
including the desired His-64 substitution and a new silent HpaI
restriction site was inserted between the BglII and HpaI sites. The
expression and purification of the mutants were performed according
to the method described by Springer et al.3 All the chemicals were
obtained from Wako and Nakalai Tesque, and used without further
purification. The buffers used for the reactions were 50 mM sodium
phosphate (pH 7.0) and sodium acetate (pH 5.3).
Figure 1. Proposed pathway for the reaction of Mb with H2O2.
Reaction with mCPBA and Hydrogen Peroxide. All the spectral
changes were monitored on a Hi-Tech SF-43 stopped-flow apparatus
equipped with a MG 6000 diode array spectrophotometer. An optical
filter (360 nm) was used for avoiding possible photoreduction of
compound I by UV light. The reactions with mCPBA were carried out
at pH 5.3 for observing a complete accumulation of compound I and
at pH 7.0 for determining the decay rate of compound I to compound
II.23 The reactivity of compound I with substrates was determined by
means of a double-mixing rapid scan technique at 5.0 °C and pH 5.3.
The first mixing of ferric Mb with a slight excess of mCPBA (1.3 mol
equiv) resulted in approximately 80% accumulation of compound I,
which was subsequently mixed with at least 10-fold excess of substrates.
The bimolecular rate constants were calculated from observed rates at
more than four different concentrations of the substrates. The reaction
with H2O2 was performed at 20 °C in the absence and presence of
0.87 mM styrene. The compound I intermediate was best observed at
pH 7.0 with H2O2 as an oxidant.
Assay of the One-Electron Oxidations. Activity for one-electron
oxidation of guaiacol and ABTS was measured at 20 °C and pH 7.0
on a Shimadzu UV-2400 spectrophotometer. The formation rate of the
guaiacol oxidation product was determined from the increase in the
absorbance at 470 nm (ꢀ ) 3.8 × 103 M-1 cm-1).24 The 1 mL final
assay volume contained 2 mM guaiacol, variable amounts of H2O2
(0.2-5 mM), and 1 µM Mb except for the H64D mutant (0.5 µM).
The formation of the ABTS cation radical was monitored at 730 nm (ꢀ
) 1.4 × 104 M-1 cm-1).24 The reaction mixture contained 1 mM ABTS
and variable amounts of H2O2 (0.1-2 mM). Final concentrations of
Mb were 0.5 µM for wild type, H64A, and H64S Mb, 1 µM for H64L
Mb, and 10 nM for H64D Mb.
Assay of the Peroxygenation Reactions. Peroxygenations of
thioanisole and styrene were performed at 25 °C and pH 7.0. The
reaction mixture contained 10 µM Mb, 1 mM H2O2, and either 1 mM
thioanisole or 8.7 mM styrene. For the thioanisole sulfoxidation,
acetophenone was added as an internal standard, and the mixture was
extracted with dichloromethane for HPLC analysis on a Dicel OD
chiral-sensitive column installed on a Shimadzu SPD-10A spectropho-
tometer equipped with a Shimadzu LC-10AD pump system. For the
styrene oxidation, an internal standard was 2-phenyl-2-propanol, and
the dichloromethane extracts were analyzed by GC (Shimadzu GC-
14B) equipped with a Chiraldex G-TA capillary column.
To determine total yields of the mCPBA-supported styrene oxidation,
100 µM mCPBA was added to a solution containing 10 µM Mb and
8.7 mM styrene at pH 5.3. Incubation time was 5 min for wild-type
Mb and 0.5-1 min for the His64 mutants. The exhaustion of mCPBA
was confirmed by the addition of excess potassium iodide before the
extraction with dichloromethane.
Figure 2. Proposed roles of distal histidine in the formation of
compound I.
react with organic peracids including m-chloroperbenzoic acid
(mCPBA) to accumulate compound I.23 Thus, the instability of
Mb compound I can be attributed to the rapid reduction by the
distal histidine, which is located very close to the heme center.21
In the reaction with H2O2, however, Mb compound I has never
been observed, even in the His64 mutants.23 Although we have
attributed the absence of compound I to its rapid reduction by
H2O2 (Figure 1), there is no direct evidence for the compound
I formation in the reaction with H2O2. The direct observation
of Mb compound I requires its faster formation and slower
reduction by H2O2 as well as its stabilization. The enhancement
of the compound I formation, which is often a rate-determining
step in oxidation reactions catalyzed by Mb, is of particular
interest from the viewpoint of constructing an efficient oxidation
enzyme.24
In this paper, we report the preparation of a mutant of sperm
whale Mb where the distal histidine is replaced by aspartic acid
(H64D Mb), to mimic the active site of chloroperoxidase from
the fungus Caldariomyces fumago (CPO). While classical
peroxidases and catalases accelerate the formation of compound
I by utilizing distal histidine, probably as a general acid-base
catalyst (Figure 2), distal glutamic acid is suggested to be crucial
for the rapid formation of compound I in CPO.25-27 The H64D
Mb mutant allows us to observe, for the first time, partial
accumulation of compound I in the reaction with H2O2. The
reactivity of ferric H64D Mb with H2O2 is suggested to be at
least 50-fold higher than that of wild-type Mb. We have also
examined catalytic activity of wild-type Mb and a series of
His64 mutants (H64A, H64S, H64L, and H64D Mb) for one-
electron oxidations and oxygenations by using H2O2 as an
oxidant. The H64D mutant is the best catalyst among the
myoglobins and shows 50-70-fold and 600-800-fold higher
(23) Matsui, T.; Ozaki, S.; Watanabe, Y. J. Biol. Chem. 1997, 272,
32735-32738.
Kinetic Measurements for Association of Cyanide The association
rate of ferric Mb with cyanide was measured at 20 °C and pH 7.0 on
a Hi-Tech stopped-flow apparatus. The kinetic traces at 408 nm were
used for determining pseudo-first-order rates. The association rate
constants were given by the slope of a plot of the observed rates versus
cyanide concentration.
Reaction with Cumene Hydroperoxide The reaction mixture
containing 10 µM Mb and 270 µM cumene hydroperoxide was
incubated for 50 min at 20 °C and pH 7.0. Aliquots of the mixture
(24) Matsui, T.; Ozaki, S.; Liong, E.; Phillips, G. N.; Watanabe, Y. J.
Biol. Chem. 1999, 274, 2838-2844.
(25) Poulos, T. L.; Freer, S. T.; Alden, R. A.; Edwards, S. L.; Skogland,
U.; Takio, K.; Eriksson, B.; Xuong, N.-h.; Yonetani, T.; Kraut, J. J. Biol.
Chem. 1980, 255, 575-580.
(26) Murthy, M. R.; Reid, T. J. d.; Sicignano, A.; Tanaka, N.; Rossmann,
M. G. J. Mol. Biol. 1981, 152, 465-99.
(27) Sundaramoorthy, M.; Terner, J.; Poulos, T. L. Structure 1995, 3,
1367-1377.

9954 J. Am. Chem. Soc., Vol. 121, No. 43, 1999
Matsui et al.
Figure 4. Soret absorbance changes during the reaction of compound
I in H64D Mb, prepared by mCPBA, with guaiacol and styrene at pH
5.3 and 5.0 °C. (A) The reaction with guaiacol was monitored before
mixing (broken line), at 2, 8, 20 ms (solid lines), and at 1, 3, 5 s (dotted
lines) after mixing. (B) The reaction with styrene was monitored before
mixing (broken line), and at 0.1, 0.3, 0.5, and 0.7 s after mixing. The
directions of absorbance changes are indicated by arrows.
Figure 3. Rapid-scan absorption spectra of H64D Mb in the reaction
with mCPBA in 50 mM sodium acetate buffer (pH 5.3) at 5.0 °C.
Broken line represents the spectrum of the ferric mutant and solid lines
are spectra recorded at 50, 100, 200, and 300 ms after mixing. Directions
of the spectral change are indicated in the figure by arrows.
Table 1. Bimolecular Rate Constants for One- and Two-Electron
were analyzed by a Shimadzu HPLC system equipped with a Shimadzu
CR-6A Data Module on a GL Sciences ODS-80A column. Phenethyl
alcohol was employed as an internal standard. The column was eluted
with 50% water/50% methanol at a flow rate of 1.0 mL/min and the
effluent was monitored at 210 nm. Assignment of the components was
based on the retention time of authentic samples.
a
Reduction of Compound I (M-1 s-1
)
c
myoglobin
guaiacolb
thioanisolec
styrenec
H2O2
H64A
H64S
H64D
3.1 × 106
1.6 × 106
8.8 × 105
1.5 × 106
1.5 × 106
2.2 × 105
2.2 × 104
2.6 × 104
2.1 × 104
5.9 × 103
6.2 × 103
1.8 × 104
a Determined in 50 mM sodium acetate buffer (pH 5.3) at 5.0 °C.
b Reduction rates of compound I to compound II. c Reduction rates of
compound I to the ferric form.
Results
Stability and Reactivity of Compound I in H64D Mb.
Figure 3 shows absorption spectral change during the reaction
of ferric H64D Mb with m-chloroperbenzoic acid (mCPBA).
Upon mixing, Soret absorption decreased to less than half and
a broad visible band having a peak around 648 nm appeared.
The spectral change is typical for the conversion of a ferric heme
to a ferryl porphyrin cation radical (FeIVdO Por+•), equivalent
of compound I in HRP.18 Compound I in the H64D mutant
spontaneously decayed to a ferryl heme (FeIVdO Por) and a
protein radical.15,28 Since wild-type Mb directly formed the ferryl
heme, similar to compound II in peroxidases, coupled with a
protein radical under the same condition,15 the H64D mutation
as well as H64A, H64S, and H64L mutations23 appear to prolong
the lifetime of compound I. The decay rate of H64D compound
I (1.2 s-1) was almost the same as those in the other His64
mutants (0.9-1.2 s-1),23 indicating a subtle effect of Asp64 on
the stability of compound I. The elimination of the histidine
residue is crucial for the stabilization as suggested earlier.23
The reactions of compound I, prepared by mCPBA, with
substrates were examined by using a double mixing stopped-
flow technique, in which all the His64 mutants showed similar
spectral changes. Addition of guaiacol resulted in sequential
reduction of compound I to compound II and the ferric state
(Figure 4A). On the contrary, thioanisole, styrene, and H2O2
reduced compound I directly to the ferric state (Figure 4B).
Thus, compound I in the Mb mutants appears to be capable of
one-electron oxidations, oxygenations, and oxidation of H2O2
to O2 as catalyzed by peroxidase, P450, and catalase, respec-
tively. The reduction rates of compound I summarized in Table
1 are essentially the same among the mutants except for 7-fold
decrease in the thioanisole sulfoxidation by H64D compound
I.
Figure 5. Spectral changes of H64D Mb in the reaction with 1 mM
H2O2 at pH 7.0 and 20 °C. The spectra were recorded before mixing
(broken line) and at 10, 30, 60, 100, and 150 ms (solid line) and 1, 2,
4, and 6 s (dotted line) after mixing. Directions of absorbance changes
are indicated by arrows. For clarity, Soret and visible regions are
expanded and magnified, respectively.
mediate species prior to the compound II formation. The initial
Soret decrease coupled with increase in visible absorption
appears to be due to partial accumulation of compound I
(approximately 40% based on the Soret absorbance). In fact,
incubation with guaiacol and H2O2 resulted in the direct
formation of compound II, which is consistent with the rapid
reduction of compound I to compound II by guaiacol (Figure
4A, Table 1). The addition of thioanisole or styrene also
prevented us from observing the compound I-like intermediate
and significantly retarded the compound II formation. This
observation can be rationalized by the rapid reduction of
compound I to the ferric state by thioanisole and styrene (Figure
4B, Table 1) by a two-electron process. Thus, it is now
convincing that the observed intermediate is compound I, and
this is the first observation of Mb compound I by use of H2O2
as an oxidant.
Direct Observation of H64D Mb Compound I in the
Reaction with H2O2. In the reaction with H2O2, ferric H64D
Mb was oxidized to compound II with a biphasic spectral change
(Figure 5), which clearly indicates accumulation of an inter-
(28) Ozaki, S.; Matsui, T.; Watanabe, Y. J. Am. Chem. Soc. 1997, 119,
6666-6667.

Compound I of Myoglobin
J. Am. Chem. Soc., Vol. 121, No. 43, 1999 9955
Table 2. Slopes of the One-Electron Oxidation Activity of Mb
a
against the Concentration of H2O2
myoglobin
guaiacol
ABTS
wild type
H64L
H64A
H64S
38
0.12
6.8
44
0.73
7.0
8.9
9.5
H64D
1920
3050
a All the assays were performed in 50 mM sodium phosphate buffer
(pH 7.0) at 20 °C. The values for the wild type were the rates at [H2O2]
) 0.2 mM divided by 0.2 mM. The unit is nmol product/nmol
Mb‚min‚[H2O2] mM.
Table 3. H2O2-Dependent Sulfoxidation of Thioanisolea
Figure 6. Initial oxidation rates of ABTS as a function of the
concentration of H2O2 at 20 °C in 50 mM sodium phosphate buffer
(pH 7.0).
myoglobin
rate/min-1
ee/R %
18O from H218O2/%
wild type
H64L
H64A
H64S
0.25
0.072
0.74
1.7
25
27
6
9
6
92
89
98
99
99
Compound I of the H64A, H64S, and H64L mutants could
not be prepared by H2O2, possibly because its formation is
slower than its reduction by H2O2 (Table 1).23 The rates of
compound I reduction by H2O2 do not greatly differ between
the H64D mutant and the others (Table 1). Therefore, the
formation rate of H64D compound I should be much higher
than those in other His64 mutants. Unfortunately, we have failed
in kinetic analysis of the reactions between the His64 mutants
and H2O2 due to the complex reaction mechanism (Figure 1).
The exact reaction rates of ferric Mb with H2O2 could be
determined only for the wild type as 5.1 × 102 M-1 s-1, where
compound I formation was kinetically negligible.24 Since the
40% accumulation of H64D compound I requires that the
formation rate of compound I is similar to its reduction rate by
H2O2 (1.8 × 104 M-1 s-1, Table 1), the His64 f Asp
replacement is strongly suggested to promote the formation of
compound I by H2O2.
One-Electron Oxidation of Guaiacol and ABTS. Catalytic
one-electron oxidation (peroxidation) of guaiacol and ABTS was
examined for the wild type and His64 mutants at 20 °C and pH
7.0 in the presence of H2O2. A possible oxidation of H2O2 by
compound I is negligible during the peroxidation because of
the higher reactivity of guaiacol and ABTS than H2O2 (Table
1). With increasing concentrations of H2O2, all the His64 Mb
mutants exhibited a linear increase in the peroxidation activity
(Figure 6). Thus, the major rate-determining step of the mutants
is the reaction of ferric Mb with H2O2. Wild-type Mb showed
hyperbolic dependency due to the transition of the rate-
determining step as reported earlier.24 In the presence of a small
amount of H2O2, the slowest step for the wild type is also the
reaction of the ferric Mb with H2O2. Therefore, the slopes of
the [H2O2]-rate plots (for the wild type, determined at [H2O2]
) 0.2 mM) mainly depend on the reactivity of ferric Mb with
H2O2. Each Mb afforded similar slopes for guaiacol and ABTS
(Table 2).
H64D
145
a Determined in 50 mM sodium phosphate buffer (pH 7.0) at 20 °C.
Table 4. Peroxide-Ddependent Styrene Oxidation Catalyzed by
Mb
H2O2
mCPBAb
a
rate/
SO:
ee(SOc)/
R %
18O in SOc
yield/
%
d
myoglobin min-1 PAA:BAc
from H218O2/%
wild type
H64L
0.022 67:22:11
9
34
75
75
88
20
73
87
86
95
6.7
93
111
104
102
0.023
0.49
0.71
87:6:7
15:80:5
17:79:4
19:79:2
H64A
H64S
H64D
18
a Determined in 50 mM sodium phosphate buffer (pH 7.0) at 20 °C.
b Determined in 50 mM sodium acetate buffer (pH 5.3) at 20 °C.
c Abbreviations: SO, styrene oxide; PAA, phenylacetoaldehyde; BA,
benzaldehyde. d nmol total products/nmol mCPBA added.
of one molecule of H2O2. Any incomplete fit to the assumption
affords lower values than real rate constants. The ABTS
oxidation activity gives a reasonable value for the wild type,
3.7 × 102 M-1 s-1, which is roughly the same as that directly
determined from the formation rate of compound II (5.1 × 102
M-1 s-1). An estimated rate constant for the H64D mutant is
2.5 × 104 M-1 s-1, and thus, ferric H64D Mb is found to react
with H2O2 at least 50-fold faster than wild-type Mb.
Peroxygenation of Thioanisole and Styrene. The H2O2-
dependent oxygenation (peroxygenation) of thioanisole and
styrene was examined at 20 °C and pH 7.0 (Tables 3 and 4).
The H2O2 oxidation by compound I is also negligible due to
the higher reactivity or higher concentration of thioanisole and
styrene than H2O2 (Table 1). As reported earlier, thioanisole
was exclusively converted to the corresponding sulfoxide, and
styrene was oxidized to styrene oxide, phenylacetaldehyde, and
benzaldehyde.13,28-30 A major product for the styrene oxidation
was styrene oxide in wild type and H64L Mb, and phenyl-
acetaldehyde in the other His64 mutants (Table 4).
The peroxidation activity appears to be affected by the
polarity of the residues at position 64; i.e., the apolar replace-
ment by Leu resulted in more than 100-fold decrease in activity,
compared to approximately 6- and 4-fold decrease in activity
in H64A and H64S Mb, respectively, but 50-70-fold increase
in activity in H64D Mb (Table 2). The results indicate that the
His64 f Asp replacement significantly improves the activation
rate of H2O2 by ferric Mb while the other His64 mutants are
less reactive with H2O2 than the wild type.
The peroxygenation activity of the mutants is lowest with
H64L Mb and highest with H64D Mb as in the peroxidations
(Tables 2-4). Relative activities of the mutants versus wild-
type Mb, however, are much higher than those in the peroxi-
dations. For example, the H64D replacement caused 50-70-
The rate constant for the H2O2 activation can be calculated
from the peroxidation activity assuming that (1) compound I
formation is the sole rate-determining step and (2) two molecules
of the substrate are oxidized by one electron upon the activation
(29) Levinger, D. C.; Stevenson, J.-A.; Wong, L.-L. J. Chem. Soc., Chem.
Commun. 1995, 2305-2306.
(30) Ozaki, S.; Matsui, T.; Watanabe, Y. J. Am. Chem. Soc. 1996, 118,
9784-9785.

9956 J. Am. Chem. Soc., Vol. 121, No. 43, 1999
Matsui et al.
Table 5. Association Rate Constants of Cyanide and Product
Analysis of the Reaction with Cumene Hydroperoxidea
cumylalcohol/
acetophenone
myoglobin
kCN/mM-1 s-1
wild type
H64L
H64A
H64S
0.32
3.3
3.3
3.7
4.7
4.0
0.002
0.075
0.41
H64D
0.034
a Determined in 50 mM sodium phosphate buffer (pH 7.0) at 20 °C.
for the wild type was only 7% (Table 4), indicating that most
of the reactive intermediate formed is not utilized for the
oxidation of styrene. Thus, the low peroxygenation activity of
wild-type Mb is due to the low efficiency of compound I (or
its equivalents) to oxidize external substrates.
Figure 7. Relationships between ABTS oxidation activity and per-
oxygenation activity of wild-type Mb and the His64 mutants. All the
values were normalized by comparison with corresponding rates of the
wild type.
General Acid-Base Functions of the Residues at Position
64. Finally, we have examined roles of Asp64 in H64D Mb in
the activation of H2O2. The ferric H64D mutant was found to
react with H2O2 at least 50-fold faster than the wild type and
the other mutants. As proposed for distal histidine in classical
peroxidases and distal glutamate in chloroperoxidase,27,31 Asp64
may facilitate the reaction as a general acid-base catalyst. To
evaluate the capability of Asp64 as a general base, association
rates of cyanide (kCN) to the ferric heme were measured (Table
5). At neutral pH, cyanide is dominantly protonated (HCN, pKa
∼ 9) and the crucial step for cyanide association has been shown
to be the deprotonation of HCN in the distal heme pocket as
suggested for the binding of H2O2 (pKa ) 11.6).4,5,31 The His64
f Asp replacement, however, retarded the cyanide association
by 10-fold, indicating less basicity of Asp64 than His64.
fold rate increase in the peroxidations but 600-800-fold increase
in the peroxygenations. Figure 7 depicts plots of the peroxy-
genation activity against ABTS oxidation activity compared with
that of the wild type. The His64 mutants exhibited roughly linear
correlation of the peroxidation to the peroxygenation activity
(slope: 0.89 for thioanisole and 0.74 for styrene). As noted in
the previous section, the ABTS oxidation activity is a good
measure of the activation rate of H2O2 by Mb. Thus, a major
rate-determining step of the peroxygenation by the His64
mutants should be the reaction of ferric Mb with H2O2. In
contrast, wild-type Mb showed 15- and 50-fold lower activity
for the thioanisole and styrene oxidation, respectively, than those
expected from the lines (Figure 7). The large deviation suggests
a different peroxygenation mechanism of the wild type from
those of the mutants.
FeIIIPor + PhC(CH3)2OOH heterolysis8
OdFeIVPor+• + PhC(CH3)2OH (1)
In the incubation with 18O-labeled H2O2, the His64 mutants
mainly incorporated 18O into the sulfoxide and epoxide (Tables
3 and 4). High enantioselectivity of the mutants in the styrene
epoxidation ruled out the involvement of hydroxy radical (Table
4). Thus, the His64 mutants appear to perform the peroxygen-
ation through a ferryl oxygen transfer mechanism. The reactive
species in H64D Mb is unambiguously compound I because
compound I of the mutant, which was clearly observed in the
reaction with H2O2 (Figure 5), was not accumulated under the
catalytic condition (data not shown). In contrast, styrene
epoxidation by the wild type resulted in only 20% incorporation
of the peroxide oxygen although the oxygen source of the
sulfoxidation was the peroxide (Tables 3 and 4). A major
epoxidation process of wild-type Mb has been shown to be the
incorporation of an oxygen atom of molecular oxygen mediated
by a protein radical (co-oxidation mechanism),12 which is
consistent with the instability of compound I in the wild-type
protein. It is likely that compound I in the wild type oxidizes
amino acids rather than the exogenous substrates.
To examine the peroxygenation efficiency, the styrene
oxidation was performed at pH 5.3 by employing mCPBA as
an oxidant. Under the conditions used, mCPBA was consumed
within 2 min by the wild type and within less than a few seconds
by the mutants. The peroxygenation efficiency could be easily
determined from total yields of the products after the exhaustion
of mCPBA. It should be noted that a similar product distribution
was obtained for mCPBA- and H2O2-supported peroxygenation
(data not shown), suggesting the same mechanism for the two
peroxides. The product yields for the His64 mutants were 100%
within experimental errors (Table 4). On the contrary, the yield
FeIIIPor + PhC(CH3)2OOH homolysis8
OdFeIVPor + PhC(CH3)2O• + H+ (2)
PhC(CH3)2O• f PhCOCH3 + CH3
(3)
•
The ability of Asp64 to function as a general acid was
estimated from product analysis in the reaction of ferric Mb
with cumene hydroperoxide (Table 5).24 Heterolytic and ho-
molytic cleavage of the O-O bond of the peroxide affords
cumyl alcohol and acetophenone, respectively (eqs 1-3).32 The
acid catalyst is expected to selectively enhance the heterolysis
and raise the ratio of heterolysis over homolysis (cumyl alcohol/
acetophenone). The H64D mutant showed a slightly higher ratio
than the wild type (Table 5); however, the improvement does
not appear to be large enough for rationalizing the increased
reactivity of H64D Mb with H2O2. The H64S mutant exhibited
the highest ratio among the myoglobins but was less reactive
with H2O2 than wild type and H64D Mb. Thus, we conclude
that the residues at position 64 including Asp64 are not an
effective acid catalyst in the reaction with cumene hydroper-
oxide.
Discussion
Reaction with Hydrogen Peroxide. The His64 f Asp
substitution enables us to observe compound I of Mb in the
(31) Poulos, T. L.; Kraut, J. J. Biol. Chem. 1980, 255, 8199-8205.
(32) Meehan, E. J.; Kolthoff, I. M.; Auerbach, C.; Minato, H. J. Am.
Chem. Soc. 1961, 83, 2232-2234.

Compound I of Myoglobin
J. Am. Chem. Soc., Vol. 121, No. 43, 1999 9957
reaction with H2O2 for the first time (Figure 5). The successful
observation is due to at least 50-fold faster formation of
compound I as well as stabilization of compound I. The other
His64 mutants (H64A, H64S, and H64L) are less reactive with
H2O2 than the wild type. It also has been reported that His64
f Gly, Thr, Val, Phe, and Tyr substitutions in Mb depress the
reaction with H2O2.33,34 Therefore, the carboxylate group of the
distal Asp is crucial for the facile formation of compound I in
H64D Mb. This result strongly supports the importance of a
distal glutamic acid in CPO. Although the distal glutamate in
CPO has been suggested to be a critical residue for the rapid
formation of compound I,27 its catalytic importance has not been
examined yet using CPO mutants. The enhancement by a distal
carboxylate group was also pointed out from a mutagenesis
study on horseradish peroxidase (HRP).35
H2O2. During the catalytic oxidation of styrene and thioanisole,
the H64D mutant did not give detectable amounts of compound
I (data not shown), which is easily observable in the absence
of the substrates (Figure 5).
The plots of peroxygenation activity against peroxidation
activity (Figure 7) suggest that the factors controlling peroxy-
genation activity of Mb are the formation rate and stability of
compound I. In fact, under the assay conditions, the reduction
rates of compound I by styrene and thioanisole calculated from
the corresponding bimolecular rate constants (Table 1) are much
higher than the formation rates of compound I even in the H64D
mutant. Furthermore, while the efficiency of the mCPBA-
supported styrene oxidation was only 7% in the wild type due
to its very short lifetime of compound I, the His64 mutants
completely transferred the oxidizing equivalents to styrene
(Table 4). The His64 f Asp replacement improves the two
critical factors to promote the catalytic activity for both
peroxidation and peroxygenation reactions (Tables 2-4).
The present study provides a guide to construct further active
enzymes based on Mb. Despite the high reactivity of the H64D
mutant with H2O2, the rate-determining step in its catalytic cycle
is the formation process of compound I. Thus, additional
mutations on H64D Mb to help compound I formation will
produce more efficient oxidation enzymes. Active site residues,
which can directly perturb the reaction between heme and H2O2,
are strong candidates to be replaced. Recently, Wan et al.36
showed on the basis of random mutagenesis study on horse heart
Mb that residues at the outer sphere of the active site can also
influence the reactivity of Mb with H2O2. Thus, as proposed in
their paper, the combined application of the rational design of
the active site and random mutagenesis technique may produce
a truly effective oxidation enzyme.
In summary, we have prepared the H64D mutant of sperm
whale myoglobin to mimic the active site of chloroperoxidase.
The His64 f Asp replacement is found to significantly improve
the reactivity of ferric Mb with H2O2 as well as the stability of
compound I. The two advances in H64D Mb allowed us to
observe compound I for the first time in the reaction of Mb
with H2O2, and to identify compound I as the reactive
intermediate in the H2O2-dependent oxidations. The H64D
mutant showed 50-70-fold higher peroxidation activity and
600-800-fold higher peroxygenation activity than the wild-type
protein. These changes in the oxidation activity are rationalized
by the enhanced stability and formation rate of compound I.
Of particular interest is the role of the distal carboxylate group
in helping the reaction with H2O2. We examined the capability
of Asp64 in H64D Mb as a general acid-base catalyst in the
reactions with cyanide and cumene hydroperoxide (Table 5) and
concluded that Asp64 was not an efficient general acid-base
catalyst in these reactions. Although the role of Asp64 in the
mutant remains unclear at this point, we might have to consider
functions other than general acid-base. The reactivity with H2O2
appears to be affected by polarity of the residues at position 64
rather than their steric bulkiness (Table 2). It might be possible
that the highly polar distal heme pocket of the H64D mutant
raises the affinity of H2O2 to the pocket and/or Asp64 fixes
H2O2 through a hydrogen bonding to have a preferable position
to bind to the heme iron.
Catalytic Activity and Mechanism. Wild-type and native
myoglobins have been shown to catalyze H2O2-dependent
epoxidation of styrene mainly through a co-oxidation mecha-
nism, in which the polypeptide radical at a protein surface
mediates incorporation of one oxygen atom of molecular
oxygen.11 This is in contrast to the ferryl oxygen transfer
mechanism suggested for oxygenation reactions by cytochrome
P-450. Ortiz de Montellano and colleagues reported that a His64
f Val Mb mutant dominantly incorporated an oxygen atom of
H2O2 in forming an epoxide.12 Although the result implies a
ferryl oxygen transfer mechanism for the epoxidation catalyzed
by H64V Mb, the reactive species was not identified.12 We have
shown that the replacement of His64 by unoxidizable amino
acids stabilizes compound I of Mb, and that the Mb compound
I generated with mCPBA oxidizes styrene and thioanisole to
the corresponding oxygenated products.23,28 The H64D Mb
mutant in this study provides the first direct evidence for the
reactive intermediate being compound I in the reaction with
Acknowledgment. This work was supported by Grant-in-
Aid for Priority Areas, Molecular Biometallics (Y.W.), and
Grant-in-Aid for Scientific Research (No. 10680575) (S.O.).
T.M. is supported by Research Fellowships of the Japan Society
for the Promotion of Science for Young Scientists.
(33) Brittain, T.; Baker, A. R.; Butler, C. S.; Little, R. H.; Lowe, D. J.;
Greenwood, C.; Watmough, N. J. Biochem. J. 1997, 326, 109-115.
(34) Khan, K. K.; Mondal, M. S.; Padhy, L.; Mitra, S. Eur. J. Biochem.
1998, 257, 547-555.
(35) Tanaka, M.; Ishimori, K.; Mukai, M.; Kitagawa, T.; Morishima, I.
Biochemistry 1997, 36, 9889-9898.
JA9914846
(36) Wan, L. L.; Twitchett, M. B.; Eltis, L. D.; Mauk, A. G.; Smith, M.
Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 12825-12831.