Precise design of artificial cofactors for enhancing peroxidase activity of myoglobin: Myoglobin mutant H64D reconstituted with a "single-winged cofactor" Is equivalent to native horseradish peroxidase in oxidation activity

Article abstract of DOI:10.1002/asia.201100107

H64D myoglobin mutant was reconstituted with two different types of synthetic hemes that have aromatic rings and a carboxylate-based cluster attached to the terminus of one or both of the heme-propionate moieties, thereby forming a "single-winged cofactor" and "double-winged cofactor," respectively. The reconstituted mutant myoglobins have smaller K_m values with respect to 2-methoxyphenol oxidation activity relative to the parent mutant with native heme. This suggests that the attached moiety functions as a substrate-binding domain. However, the k_cat value of the mutant myoglobin with the double-winged cofactor is much lower than that of the mutant with the native heme. In contrast, the mutant reconstituted with the single-winged cofactor has a larger k_cat value, thereby resulting in overall catalytic activity that is essentially equivalent to that of the native horseradish peroxidase. Enhanced peroxygenase activity was also observed for the mutant myoglobin with the single-winged cofactor, thus indicating that introduction of an artificial substrate-binding domain at only one of the heme propionates in the H64D mutant is the optimal engineering strategy for improving the peroxidase activity of myoglobin. Copyright

Full text of DOI:10.1002/asia.201100107

DOI: 10.1002/asia.201100107
Precise Design of Artificial Cofactors for Enhancing Peroxidase Activity of
Myoglobin: Myoglobin Mutant H64D Reconstituted with a “Single-Winged
Cofactor” Is Equivalent to Native Horseradish Peroxidase in Oxidation
Activity
Takashi Matsuo,*[b] Kazuki Fukumoto,[a] Takuro Watanabe,[a] and Takashi Hayashi*[a]
Abstract: H64D myoglobin mutant was
reconstituted with two different types
of synthetic hemes that have aromatic
rings and a carboxylate-based cluster
attached to the terminus of one or both
of the heme-propionate moieties,
thereby forming a “single-winged co-
factor” and “double-winged cofactor,”
respectively. The reconstituted mutant
myoglobins have smaller Km values
with respect to 2-methoxyphenol oxi-
dation activity relative to the parent
mutant with native heme. This suggests
that the attached moiety functions as a
substrate-binding domain. However,
the kcat value of the mutant myoglobin
with the double-winged cofactor is
much lower than that of the mutant
with the native heme. In contrast, the
mutant reconstituted with the single-
winged cofactor has a larger kcat value,
thereby resulting in overall catalytic ac-
tivity that is essentially equivalent to
that of the native horseradish perox-
idase. Enhanced peroxygenase activity
was also observed for the mutant myo-
globin with the single-winged cofactor,
thus indicating that introduction of an
artificial substrate-binding domain at
only one of the heme propionates in
the H64D mutant is the optimal engi-
neering strategy for improving the per-
oxidase activity of myoglobin.
Keywords: cofactors
·
heme
proteins · myoglobin · oxidation ·
reconstitution
Introduction
cess (decrease in Km value).[1,2] Enhancement of catalytic
steps (increase in kcat value) is related to the regulation of
reactive intermediates (their rates of formation, lifetimes,
and intrinsic reactivities). Controlling both substrate-binding
and chemical processes will be required for effective molec-
ular design of an artificial enzyme that exhibits high catalyt-
ic activity equivalent or superior to that of naturally occur-
ring enzymes.
Myoglobin (Mb), an oxygen-storage hemoprotein, is an
easily obtainable protein that is suitable for use as a scaffold
to engineer the native functions by genetic and/or chemical
modifications.[3] The native Mb contains one protoheme IX
1 at the active site (Scheme 1; see the Supporting Informa-
tion). This prosthetic group is responsible for the reversible
binding of dioxygen.[4] Although the protoheme IX of Mb is
the same as that of horseradish peroxidase (HRP), the cata-
lytic oxidation activity of Mb is much lower than that of
HRP. This is because Mb lacks a specific binding site for
substrates and a general acid–base catalytic system to
smoothly generate the oxoferryl species, a key intermediate
in the catalytic oxidation cycle. Possible strategies for intro-
ducing these factors into Mb are 1) genetic mutations to
provide an HRP-like environment at the heme distal site,
To create an artificial enzyme with high catalytic activity, we
should consider the elementary processes of enzymatic reac-
tions: substrate-binding and the chemical catalytic process.
These processes are apparently qualified by Km and kcat in
Michaelis–Menten kinetics, respectively. Having a solid un-
derstanding of the general concepts of supramolecular
chemistry is necessary to improve a substrate-binding pro-
[a] K. Fukumoto, T. Watanabe, Prof. Dr. T. Hayashi
Department of Applied Chemistry
Graduate School of Engineering
Osaka University
2-1 Yamadaoka, Suita, Osaka 565-0871 (Japan)
Fax : (+81)6-6879-7929
E-mail: thayashi@chem.eng.osaka-u.ac.jp
[b] Dr. T. Matsuo
Graduate School of Materials Science
Nara Institute of Science and Technology (NAIST)
8916-5 Takayama-cho, Ikoma, Nara 630-0192 (Japan)
Fax : (+81)743-72-6119
E-mail: tmatsuo@ms.naist.jp
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201100107.
Chem. Asian J. 2011, 6, 2491 – 2499
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2491

FULL PAPERS
tion as an effective binding site for hydrophobic substrates
such as 2-methoxyphenol. This leads to a reduction of the
Km value, thereby indicating that binding is more effective.
As described above, the introduction of an effective H2O2
activation system by genetic mutation and the attachment of
a substrate-binding domain by synthetic chemical strategy
are indeed useful to the remarkable overall enhancement of
the peroxidase activity of the native Mb. It was disclosed,
however, that the kcat values unexpectedly decreased by re-
placement of the native heme with 2 when these values for
the two H64D Mbs, MbACTHNUTRGENN(GU H64D·1) and MbACHTUNGTRNE(NUGN H64D·2), were
compared. This finding suggests that modification of the
heme propionates often causes negative effects on the chem-
ical processes in the catalytic cycle. Our independent investi-
gations have also indicated that the interactions between the
heme propionates and nearby amino acid residues in some
hemoproteins are related to the reactivity of the heme iron
as well as the thermostability of the proteins.[13] The next ap-
proach to enhance the peroxidase activity of Mb is to im-
prove the overall catalytic activity without negative effects
on chemical processes. In this paper, we describe a systemat-
ic investigation of the results of heme modification with a
“single-winged cofactor” 3, which was developed for Mb re-
constitution to obtain key insights with respect to the crea-
tion of artificial enzymes optimized for both substrate-bind-
ing and chemical processes.
Scheme 1. Native and synthetic heme cofactors.
2) attachment of an artificial substrate binding domain at
the termini of the heme propionates, and 3) replacement of
the native heme with a more reactive heme analogue. The
first approach is attained by the H64D mutant, in which the
introduced Asp residue works as a general acid–base unit to
activate H2O2.[5] In the second strategy, a variety of designed
substrate-binding domains have been suggested, including
boronic acid,[6] aromatic rings,[7] peptides,[8] and DNA,[9]
among others. The third method involves the insertion of an
iron complex of porphycene[10] or corrole[11] into the heme
pocket of Mb.
Results and Discussion
Reconstitution of the H64D Myoglobin Mutant with the
Single-Winged Cofactor and NMR Spectroscopic Study of
Heme Orientation
Single-winged cofactor 3 was prepared by partial condensa-
tion of protoporphyrin IX with aniline derivatives, followed
by iron insertion and alkaline hydrolysis (see the Supporting
Information for the synthetic scheme and the procedure).
Because the condensation at the propionate side chain
occurs nonselectively, the 1:1 mixture of regioisomers 3a
and 3b was obtained, and it is difficult to separate them.
Sperm whale H64D Mb expressed by Escherichia coli was
reconstituted with single-winged cofactor 3 to obtain Mb-
Parallel to the above-mentioned approaches, the combina-
tion of these strategies is also promising for the enhance-
ment of peroxidase activity of Mb. In our previous paper,[12]
we described a reconstituted H64D Mb with a “double-
winged here cofactor” 2 (Mb
ACHTUNGTNER(NUNG H64D·2); wherein the nota-
tion “Mb(H64D·X)” indicates H64D Mb with cofactor X).
ACHTUNGTRENNUNG
This is designated “hybrid Mb.” The aromatic moiety intro-
duced at both of the heme propionates was found to func-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
tor 3 occupies the same space in the heme pocket as the
native heme 1 (see the Supporting Information). This in-
cludes coordination of the His93 imidazole to the heme
iron. Although the two regioisomers were inserted into the
apoprotein together, the NMR spectroscopic study for the
prepared Mbs suggests that each isomer is accommodated
so that the unmodified propionate preferentially interacts
with Arg45 (vide infra).
Abstract in Japanese:
MbACTHNUGRTENUNG(H64D·3) could theoretically contain four possible
configurations, because a mixture of the two regioisomers,
3a and 3b, was employed for reconstitution of the H64D
mutant and each regioisomer has two possible orientations
2492
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2491 – 2499

Peroxidase Activity of Myoglobin
within the heme pocket (normal and reversed orientations)
as a result of a 1808 rotation of the heme about the heme a–
g meso axis (“normal orientation” is a fashion of the major
heme orientation, in which 6-propionate forms a hydrogen
bond with Arg45, see the Supporting Information). One
useful method to elucidate the cofactor orientations in the
or of H64D Mb also adopts the normal orientation observed
for wild-type Mb.[15] In particular, the chemical shift of 5-
methyl proton is affected by the existence of the hydrogen
bonding between 6-propionate and Arg45 located in the en-
trance of the heme pocket: La Mar and co-workers reported
that cyanomet Mb reconstituted with 6-methyl-6-despropio-
nate hemin, in which a hydrogen bond between 6-propio-
nate and Arg45 is lacking, has a downfield-shifted peak of
the 5-methyl protons at d=28 ppm.[14,16] The observation of
the 5-methyl proton peak at the same chemical shift (d=
26.8 ppm) supports the fact that the hydrogen bonding be-
1
heme pocket of Mb is the H NMR spectroscopic measure-
ment for cyanomet Mb because the paramagnetic shift of
the methyl protons on the periphery of the heme framework
is affected by orientation of proximal His93 and is useful in-
formation of the heme orientation in the myoglobin
matrix.[14] This methodology is applicable to mutant Mbs.[14c]
Figure 1 shows the 1H NMR spectra of the cyanomet
wild-type and H64D Mbs obtained in deuterium oxide. The
assignments of the selected heme methyl peaks were made
based on the knowledge obtained in the previous reports.[14]
The numbering of the heme framework is indicated in
Scheme 1. The assignment indicated in Figure 1c was con-
ducted with reference to the previous data obtained in the
NMR spectroscopic measurements before the heme orienta-
tion reached equilibrium.[14] The pattern of the singlet peaks
tween 6-propionate and Arg45 in Mb
ACHTUNGTRENNUNG
In contrast, it is expected that cyanomet MbACHTUNGTRENNUNG
1
shows a complicated peak pattern on the H NMR spectrum
because of the possibility to take several heme configura-
tions. However, as shown in Figure 1c, the characteristic
heme methyl peaks are relatively simple. In particular, only
one signal appeared at d=26.5 ppm. The chemical shift is
very similar to that observed for MbACTHNUTRGNE(NUG H64D·1). When the
H64D mutant was reconstituted with 6-methyl-6-despropio-
nate heme, the 5-methyl proton peak appeared at d=
28 ppm in the same manner as the wild-type Mb reconstitut-
ed with this propionate-truncated heme (see the Supporting
Information), although no peak was observed around d=
28 ppm in Figure 1c. This result indicates that the hydrogen
bonding between a propionate side chain and Arg45 is con-
for 5-, 1-, and 8-methyl protons for cyanomet MbACTHNUTRGNEUG(N H64D·1)
(Figure 1b) is similar to that observed for wild-type Mb
(Figure 1a). This suggests that hemin 1 in the protein interi-
served also in MbACTHNUTRGNEUNG(H64D·3). Therefore, the mechanism to
account for this spectrum is that cofactor 3 in the heme
pocket should take an orientation that allows interaction be-
tween the propionate and Arg45. Isomer 3b takes the re-
versed heme orientation to form hydrogen bonding with
Arg45 (see the Supporting Information).[17] In the reversed
orientation, the 8-methyl group should be located in the po-
sition that the 5-methyl group in the normal heme orienta-
tion occupies, thereby resulting in the similar chemical shift
(d=26.5 ppm for the 5-methyl proton in the normal orienta-
tion of 3a and 8-methyl proton in the reversed orientation
of 3b) and vice versa. The rather broad peak at d=
26.5 ppm also supports the existence of the two conforma-
tions.[18] In contrast, the chemical shifts of the 1-methyl and
3-methyl protons are significantly affected by the accommo-
dated position in the heme pocket because the positions of
these methyl groups are not symmetrical about the heme a–
g meso axis. These findings suggest that the hydrogen bond-
ing at Arg45 is essential for the stability of a reconstituted
Mb with the propionate-modified cofactor[19] and that the
hydrogen bonding between a heme propionate and Arg45
might indeed contribute to the regulation of the heme iron
reactivity (vide infra).
Steady-State Kinetics for 2-Methoxyphenol
Oxidation
Figure 1. 1H NMR spectra of cyanomet Mbs: a) wild-type Mb; b) Mb-
ACHTUNGTRENNUNG(H64D·1); c) MbACHTUNGTRENNUNG(H64D·3); KPi D2O buffer (10 mm, pD=7.4), 258C. The
numbering of the heme framework is indicated in Scheme 1. The designa-
tions 3a-N and 3b-R indicate H64D Mb reconstituted with cofactor 3a
in the normal orientation and 3b in the reversed orientation, respective-
ly.
For evaluation of the peroxidase activities of H64D Mbs,
the oxidation of 2-methoxyphenol (also known as guaiacol)
was investigated (Scheme 2).[20]
Chem. Asian J. 2011, 6, 2491 – 2499
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2493

T. Matsuo, T. Hayashi et al.
FULL PAPERS
infra), the kinetic parameters
for MbACHTNUTRGNEUNG(H64D·4), in which the
four-carboxylate moiety is at-
tached at one heme-propionate,
are also included.
Scheme 2. 2-Methoxyphenol oxidation mediated by H64D Mbs.
According to the previous
Figure 2 shows the time courses of the absorbance
changes at 470 nm during 2-methoxyphenol oxidation cata-
lyzed by the H64D Mbs in the presence of an excess of
H2O2. The protein that shows the most effective catalytic ac-
report,[12] the kinetic parameter for the wild-type Mb was
found to be kcat/Km =52mꢀ1 sꢀ1. When His64 in the wild-type
protein was converted into Asp, the obtained MbACTHNUTRGNEUG(N H64D·1)
showed the approximately 100-fold catalytic activity (kcat/
Km =5100mꢀ1 sꢀ1) as also reported by Watanabe and co-
workers.[5] When the further engineering of H64D mutant
was conducted by the replacement of the native heme cofac-
tor with a synthetic cofactor 2 or 3, the overall catalytic ac-
tivity is MbACHTUNGTRENNUNG(H64D·3). Table 1 lists the Michaelis–Menten pa-
rameters of 2-methoxyphenol oxidation.[21] To compare the
effects of cofactor structures on the catalytic activity, the
previously reported kinetic parameters for H64D Mb with
heme 1 (native heme) and with double-winged heme 2 are
also described.[12] For the purpose of discussing the effect of
the propionate modification on the catalytic activities (vide
tivities of the hybrid proteins, Mb
ACHTUNGTREN(NGNU H64D·2) and Mb-
AHCTUNGTREG(NNUN H64D·3), were remarkably enhanced. The main factor of
the enhancement in the catalytic activity is the decrease in
Km values. This is because the introduced aromatic moieties
at the termini of the propionate side chains function to facil-
itate the access of the hydrophobic substrate to the vicinity
of the reaction site.
On the other hand, the kcat value for Mb
ACHTUNGTREN(NUGN H64D·2) signifi-
cantly decreased from Mb(H64D·1), which is a negative
AHCTUNGTRENNUNG
effect of the aspects of the enhancement of the peroxidase
activity in Mb. The decrease in kcat observed for Mb-
AHCTUNGTREG(NNNU H64D·2) implies that the modification of both propionates
causes a decrease in the reactivities of the oxoferryl species,
a reactive intermediate responsible for a substrate oxidation
(see below for the investigation of the reaction intermedi-
ates). In contrast, the kcat value of Mb
proved from Mb(H64D·1). This tendency is also observed in
Mb(H64D·4), in which one heme propionate is unmodified
ACHTUNGTREN(NUNG H64D·3) is rather im-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
as in 3. The result provides useful insights into the design of
an ideal synthetic heme: retaining one unmodified propio-
nate is essential to prevent the kcat value from decreasing
along with enhancement of the substrate binding process.
Figure 2. Time-courses of 2-methoxyphenol oxidations catalyzed by
H64D Mbs; sodium malonate buffer (20 mm, pH 6.0), 258C; [H64D
Mb]=2.0 mm, [2-methoxyphenol]0 =0.5 mm, [H2O2]0 =15 mm. a) Mb-
A
E
G
Consequently, MbACTHNGUTERNNU(G H64D·3) has an extraordinary enhance-
ment of the overall catalytic activity (kcat/Km =85000mꢀ1 sꢀ1).
The catalytic activity of this engineered Mb is, interestingly,
almost equivalent to that of a naturally occurring peroxidase
under the same reaction conditions (for HRP, kcat/Km =
72000mꢀ1 sꢀ1 at pH 6.0, 258C, [H2O2]=1.0 mm).[21,22] This
result indicates that the combination of a point mutation
and precise design of an appropriate chemically modified
cofactor have enabled us to construct an “ultra-native” bio-
catalyst.
Table 1. Kinetic parameters for 2-methoxyphenol oxidations catalyzed
by H64D Mb mutants.[a]
Cofactor
1
2
3
4
[a] Sodium malonate buffer (20 mm, pH 6.0), 258C, [Mb]=4.0 mm,
[H2O2]0 =100 mm. [b] Reference [12].
2-Methoxyphenol Binding
To directly evaluate the effects of the attached moiety on
the substrate affinity, the binding of 2-methoxyphenol to the
engineered H64D Mbs was observed by UV/Vis spectral ti-
tration in the absence of H2O2. The calculated dissociation
constants (Kd1 and Kd2) are summarized in Table 2. The
spectral changes for the H64D Mbs are shown in Figures 3,
4, and 5.
2494
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2491 – 2499

Peroxidase Activity of Myoglobin
Table 2. Dissociation constants for 2-methoxyphenol binding to H64D
Mb mutants.[a]
Cofactor
Kd1 [mm]
Kd2 [mm]
[b]
1
2
3
0.13ꢁ0.01
–
0.019ꢁ0.002
0.014ꢁ0.002
2.6ꢁ0.3
3.3ꢁ0.3
[a] 20 mm sodium malonate buffer (pH 6.0), 258C, [Mb]=8.0 mm, [2-me-
thoxyphenol]=0–4 mm. [b] Not determined due to single-phase depend-
ency.
Figure 4. UV/Vis spectral changes occurring during the process of binding
of 2-methoxyphenol to Mb(H64D·2); sodium malonate buffer (20 mm,
AHCTUNGTRENNUNG
pH 6.0), 258C. a) Differential spectra based on the spectrum in the ab-
sence of 2-methoxyphenol. b) Dependency of the absorbance at 408 nm
on the concentrations of 2-methoxyphenol. Error bars are drawn based
on an estimation of 10% error.
Figure 3. UV/Vis spectral changes in the binding of 2-methoxyphenol to
MbACHTUNGTRENNUNG(H64D·1); sodium malonate buffer (20 mm, pH 6.0), 258C. a) Differ-
ential spectra based on the spectrum obtained in the absence of 2-me-
thoxyphenol. b) Dependency of the absorbance at 408 nm on the concen-
trations of 2-methoxyphenol. Error bars are drawn based on an estima-
tion of 10% error.
For MbACHTUNGTRENNUNG(H64D·1), the absorbance of the Soret band slight-
ly increases upon addition of 2-methoxyphenol, which could
be caused by small conformational changes in the protein
due to the access of the substrate to the vicinity of the
heme-propionates. In contrast, H64D Mbs reconstituted
with the synthetic cofactors 2 or 3 exhibit quite different
spectral changes. The variances of those changes are much
larger in the latter two H64D Mbs, thereby indicating that
significant perturbations of the p-conjugation system of co-
factors 2 and 3 are induced during the access of the organic
substrate to the heme pocket of these proteins.[23] Moreover,
Figure 5. UV/Vis spectral changes occurring during the process of binding
of 2-methoxyphenol to MbACHTUNTGRNEUNG(H64D·3); sodium malonate buffer (20 mm,
pH 6.0), 258C. a) Differential spectra based on the spectrum in the ab-
sence of 2-methoxyphenol. b) Dependency of the absorbance at 408 nm
on the concentrations of 2-methoxyphenol. Error bars are drawn based
on an estimation of 10% error.
for Mb
as a single substrate-binding model, whereas the decay of
the Soret band for Mb(H64D·2) or Mb(H64D·3) observed
ACHTUNGTRENNUNG(H64D·1), the absorbance changes can be analyzed
A
ACHTUNGTRENNUNG
during the titrimetric measurements requires two-phase
analysis. With respect to the two dissociation constants
obtained for 2-methoxyphenol (Kd1 and Kd2), the lower Kd1
values for the H64D Mbs with modified cofactor 2 or 3 are
attributed to the attached hydrophobic aromatic moieties.
These lower Kd1 values are in agreement with the low Km
values shown in Table 1.
Chem. Asian J. 2011, 6, 2491 – 2499
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2495

T. Matsuo, T. Hayashi et al.
FULL PAPERS
Kinetic Analysis of the Catalytic Cycle
Next, to evaluate the reactivities of the oxoferryl intermedi-
ates (Compounds I and II), we determined the two rate con-
stants, k1 and k2 (defined in Scheme 3). The generation of
Scheme 3. Reaction scheme for the kinetic analysis of the catalytic cycle
of the 2-methoxyphenol oxidation mediated by H64D Mb mutants.
Compound I (oxoferryl species with a porphyrin p-cation
radical) was attained by the reaction of the reconstituted
ferric H64D Mbs with a slight excess of meta-chloroperben-
zoic acid (mCPBA). The rate constants determined in the
kinetic study are summarized in Table 3, and the representa-
tive kinetic trace is shown in Figure 6.
Figure 6. Representative absorbance changes for determination of the
rate constants k1 and k2 for MbACTHNUTRGNE(NUG H64D·3); [Mb]0 =2 mm, [2-methoxyphe-
nol]0 =10 mm, [mCPBA]=1.3 equiv; sodium malonate buffer (20 mm,
pH 6.0), 158C; a) initial stage (within 120 ms, k1-step); b) later stage (6 s,
k2-step). The fitting assuming a single-phase kinetics is indicated by solid
lines.
Table 3. Rate constants for elemental processes in 2-methoxyphenol oxi-
dation catalyzed by H64D Mb mutants.[a,b]
[c]
[d]
Cofactor
k1 [mꢀ1 sꢀ1
]
k2 [mꢀ1 sꢀ1
]
Mbs investigated. This suggests that there is a relationship
between the heme propionate(s)–protein interactions and
the reactivity of Compound I. As indicated in the H NMR
1
2
3
1.1ꢁ106
0.32ꢁ106
3.6ꢁ106
1.3ꢁ104
3.5ꢁ104
1.2ꢁ104
1
spectroscopic study (vide supra), cofactor 3 is oriented in
the heme pocket of Mb so that the hydrogen-bonding inter-
action between the heme-propionate and Arg45 is retained
at the distal site. These findings suggest that the reactivity of
Compound I can be enhanced by two factors: 1) the conser-
vation of the 6-propionate–Arg45 interaction and 2) the re-
moval of the interaction between the 7-propionate and the
amino acid residues at the proximal site.[24] The first factor
contributes to the maintenance of the distal site structure
and the second factor regulates the characteristics of coordi-
nation of His93 to the heme iron. According to the reso-
nance Raman measurements of ferric Mb reconstituted with
7-methyl-7-despropionate heme, where the hydrogen bond-
ing interaction between the 7-propionate and Ser92 is dis-
[a] Sodium malonate buffer (20 mm, pH 6.0), 158C, [Mb]=2.0 mm, [2-me-
thoxyphenol]0 =10–100 mm. [b] Experimental errors are within 10%.
[c] Reduction rate of Compound I. [d] Reduction rate of Compound II.
The H64D Mb with the double-winged cofactor, Mb-
ACHTUNGTRENNUNG(H64D·2), has the smallest k1 value, thereby indicating that
the modification of both heme-propionates lowers the reac-
tivity of the Compound I intermediate. This finding may
partially account for the drop in the kcat value for the reac-
tion catalyzed by this protein. However, the k2 value, which
indicates the reactivity of Compound II, is approximately
threefold greater than that of the other two proteins. There-
fore, it is difficult to simply rationalize the catalytic activities
ꢀ
of Mb
(H64D·2) on the basis of the magnitudes of the k1 and
rupted, the Fe His93 stretching mode is slightly stronger be-
k2 values. Another possible factor that affects the catalytic
activities of the H64D Mbs will be discussed in the next sec-
tion.
cause there is no regulation of proximal ligand coordination
by the hydrogen-bonding network at the proximal site.[13a]
We expect that a similar mechanism is operating with re-
In the determination of the rate constants k1 and k2 for
spect to MbACTHNGUTERNNUG
(H64D·3).[25]
MbACHTUNGTRENNUNG(H64D·3), the absorbance changes were clearly observed
to follow single-phase kinetics. There is a negligible differ-
ence in the reactivities of Compound I and Compound II
between the proteins with the normal and reversed heme
orientations. As indicated by the largest k1 value in Mb-
Efficiency of Generation of the Compound I Intermediate
The catalytic activity of Mbs may also be controlled by the
ꢀ
process of O O bond cleavage after the formation of the
hydroperoxo species (FeIII OOH) because the Compound I
ꢀ
A
respect to the other two proteins, which coincides with the
intermediate, the highly reactive oxidizing species, is only
ꢀ
fact that this protein has the highest kcat value among the
produced by heterolytic cleavage of the O O bond. To eval-
2496
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2491 – 2499

Peroxidase Activity of Myoglobin
uate the efficiency of Compound I generation, an analysis of
the reaction between ferric Mb and cumene hydroperoxide
(CHPO) was examined.[5a] Table 4 summarizes the ratio of
ꢀ
O O bond heterolysis and homolysis, based on the product
yields of cumyl alcohol and acetophenone, respectively
(Scheme 4).
Scheme 5. Two-electron oxidations catalyzed by H64D Mbs.
Table 4. Reaction of cumene hydroperoxide (CHPO) with the ferric
H64D Mb mutants.[a]
Cofactor
Total amounts of products [mm][b]
Heterolysis/Homolysis[c]
(ꢀ)
15
7
Table 5. Turnover numbers for thioanisole and ethylbenzene oxidation
catalyzed by H64D Mb mutants.[a]
Cofactor
Thioanisole oxidation[b]
Sulfoxide[d] Sulfone[e]
[minꢀ1 [minꢀ1
Ethylbenzene oxidation[c]
2-Phenylethanol[f]
1
2
3
25
146
100
G
]
G
]
[minꢀ1
ACHTUNGTERNNUNG ]
13
25[g]
33[g]
49[g]
2.6[h]
2.0[h]
3.8[h]
n.d.[i]
n.d.[i]
0.3[j]
[a] Sodium malonate buffer (20 mm, pH 6.0), 158C, [Mb]=20 mm,
[CHPO]=200 mm. [b] Total amounts of acetophenone and cumyl alcohol
as a product produced over 5 min. [c] Calculated from [cumyl alcohol]/
[acetophenone].
1
2
3
[a] KPi buffer (100 mm, pH 7.0), 258C. [b] [Mb]=2.0 mm, [H2O2]0 =
1.0 mm, [thioanisole]0 =0.5 mm. [c] [Mb]=20 mm, [H2O2]0 =1.0 mm, [ethyl-
benzene]0 =0.5 mm. [d] Product of oxygen transfer to thioanisole.
[e] Product of oxygen transfer to methylphenyl sulfoxide. [f] Product of
hydroxylation at the benzyl position (as racemic forms). [g] Calculated
from the amount of the sulfoxide divided by [Mb]. The ratio of R and S
isomers was not determined. See Ref. [27]. [h] Calculated from the
amount of the sulfone divided by [Mb]. [i] Not determined because there
was no observation. [j] Calculated from the amount of 2-phenylethanol
produced divided by [Mb].
amined. The protein with the highest peroxygenase activity
that yielded both sulfoxide and sulfone is Mb
ACHTUNGTREN(NUNG H64D·3). In-
terestingly, catalytic hydroxylation was also observed in Mb-
AHCTUNGTREG(NNNU H64D·3), although such activity is very rare for most Mb
mutants.[26] Catalysis of hydroxylation of a C H bond by
heme enzymes in nature has only been observed for S-coor-
dinated hemoproteins such as cytochrome P450 s and chlor-
operoxidase. Therefore, it is a remarkable finding that Mb-
ꢀ
Scheme 4. Reaction of ferric Mb with cumene hydroperoxide (CHPO).
The total amounts of organic products obtained by Mb-
ACHTUNGTREN(NGNU H64D·3), the heme iron of which is coordinated by a neu-
A
E
tral imidazole, demonstrates hydroxylation activity. These
results reflect the synergetic effects derived from the en-
hanced reactivity of Compound I (increased k1) and efficient
substrate binding (decreased Km).[25–27]
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
small, thereby indicating that the efficiency of Compound I
generation is lower than that of the other engineered pro-
teins. Based on the data described in Tables 3 and 4, the
modification of both propionates gives rise to negative ef-
fects with respect to generation of Compound I.
Conclusion
In this research, it was demonstrated that the incorporation
of a single-winged cofactor into apo-H64D Mb enhances the
reactivities of the oxidizing intermediates as well as facili-
tates the binding of substrates. The replacement of His64
with Asp by genetic mutation contributes to the smooth
H2O2 activation mediated by a general acid–base catalysis of
the Asp residue (improvement of chemical processes). The
introduction of the aromatic moiety at the termini of the
heme propionate side chains is useful to create a substrate-
binding domain (improvement of substrate-binding process).
The kinetic analysis for each step of the catalytic cycle, how-
ever, suggests that retaining one unmodified propionate is
important to control the reactivity of Compounds I and II,
Two-Electron Oxidations Catalyzed by H64D Mb
Reconstituted with Single-Winged Cofactor 3
Next we observed the H2O2-dependent oxidation of thioani-
sole (oxygen-atom transfer) and ethylbenzene (hydroxyl-
ation) catalyzed by Mbs (Scheme 5).
As expected, MbACHTUNGTRENNUNG(H64D·3) also exhibits the highest cata-
lytic activity toward the two-electron oxidations among the
engineered Mbs (Table 5). In the case of oxygen transfer to
thioanisole, a four-electron oxidation product, sulfone, was
also observed for all of the reconstituted Mbs proteins ex-
Chem. Asian J. 2011, 6, 2491 – 2499
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2497

T. Matsuo, T. Hayashi et al.
FULL PAPERS
470 nm. The initial rates were calculated using the molar absorption coef-
thus leading to more effective improvement of the perox-
idase activity in Mb. The significance of one unmodified
propionate is the formation of hydrogen bonding between
the heme propionate and Arg45 and the stabilization of the
heme orientation in the heme pocket of Mb. Furthermore,
precise design of a synthetic cofactor enables us to endow
an appropriate myoglobin mutant with enhanced catalytic
activity toward two-electron oxidation through oxo transfer
[29]
ficient of the oxidation product; e=26600 mꢀ1 cmꢀ1
.
Determination of the Rate Constants for the Elemental Processes in 2-
Methoxyphenol Oxidations
The reactions were carried out in sodium malonate buffer (20 mm,
pH 6.0) at 158C with a double-mixing stopped-flow apparatus equipped
with a PDA detector. A Mb solution (8.0 mm) was mixed with mCPBA
(10.4 mm) in a first mixing shot to generate the Compound I species.
After an aging period of 700–1200 ms, the generated intermediate was
mixed with 2-methoxyphenol (20–200 mm) in a second mixing shot. The
spectral changes after the second mixing shot were monitored by collect-
ing the transient spectra every 10–100 ms. The absorbance changes at
408 nm were analyzed according to pseudo-first-order kinetics.
ꢀ
and C H bond activation. The findings demonstrated in this
paper will provide important insights into the methodologies
required for engineering of chemically and biologically at-
tractive biocatalysts.
Catalytic Activities toward Thioanisole and Ethylbenzene Oxidations
The reactions were carried out in KPi buffer (100 mm, pH 7.0) at 258C.
A buffer solution of Mb, thioanisole, and benzyl alcohol (internal stan-
dard) was incubated prior to the addition of H2O2 to initiate the reaction.
The final concentrations were: [Mb]=2.0 mm, [thioanisole]=0.5 mm,
[benzyl alcohol]=5.0 mm, and [H2O2]=1.0 mm. After a reaction period of
5 min, ether was added, and the reaction mixture was vigorously shaken
using a vortex mixer to extract the organic materials. The separated or-
ganic phase was concentrated by evaporation with streaming N2 gas, and
the residues were analyzed with a GC/FID system equipped with a DB-1
column. The oxidation of ethylbenzene was carried out according to the
same procedure.
Experimental Section
Materials and Protein Purification
All reagents and chemicals were obtained from commercial sources and
were used as received unless otherwise noted. The H64D myoglobin
mutant was expressed from E. coli. and purified by column chromatogra-
phy through CM-52 (Whatman) and Sephadex G-25 (GE Healthcare)
columns.[5a] The preparation of cofactor 2 was described in the previous
report.[7] Cofactors 3 and 4 were synthesized as described in the Support-
ing Information. The reconstituted H64D Mbs were prepared as de-
scribed in the previous report,[12] after the removal of the native heme by
Tealeꢂs 2-butanone method.[28] The reconstituted Mbs were purified with
a Sephadex G-25 column (2ꢁ50 cm, 100 mm KPi, pH 7.0, 48C) and were
found to be stable at ꢀ808C for at least one month as a frozen solution
(ꢂ1 mm).
Reaction of Myoglobins with Cumene Hydroperoxide (CHPO)
A Mb solution was treated with cumene hydroperoxide ([Mb]=20 mm,
[CHPO]=200 mm) in sodium malonate buffer (1 mL, 20 mm, pH 6.0) at
258C for 5 min. After the addition of benzyl alcohol (10 mm), the reaction
mixture was filtered with a Centricon concentrator, and the filtrate was
analyzed with an HPLC system equipped with a YMC Pro-C18 column,
150ꢁ4.6 mm at a flow rate of 0.8 mLminꢀ1 with elution by addition of a
1:1 mixture of H2O/MeOH to determine the amounts of acetophenone
and cumyl alcohol formed by cleavage of CHPO based on the intensity
ratios of these materials against that of a benzyl alcohol standard.
Instrumentation
1H NMR spectra were measured with a Bruker DPX 400 spectrometer.
The measurements of UV/Vis spectra and the titrations of 2-methoxyphe-
nol to monitor the binding of substrate to protein were carried out with a
Shimadzu UV-3210 double-beam spectrophotometer. Kinetic measure-
ments were conducted with an RSP-1000 stopped-flow system construct-
ed by Unisoku, Co., Ltd. (Osaka, Japan). The HPLC analyses were con-
ducted with a Shimadzu HPLC LC-VP system. The GC/FID measure-
ments were made with a Shimadzu GC-2014 gas chromatography system.
The mass analyses (FAB-MS and ESI-MS) were conducted with a JEOL
JMS-700 mass spectrometer.
Acknowledgements
This work was supported by a Grant-in-Aid for Science Research on In-
novative Areas (Molecular Activation Directed toward Straightforward
Synthesis) from Ministry of Education, Culture, Sports, Science and Tech-
nology, Japan
Determination of Dissociation Constants (Kd1 and Kd2)
H64D Mb (8 mm) was dissolved in KPi (20 mm; pH 6.0) and titrated with
2-methoxyphenol. The UV/Vis spectra were measured in the range 350
to 450 nm with dropwise additions of 2-methoxyphenol. The dependency
of the absorbance change at 408 nm (Dabs408) on concentrations of 2-me-
thoxyphenol was analyzed by the following equation [Eq. (1]:
[1] P. D. Beer, P. A. Gale, D. K. Smith in Supramolecular Chemistry,
Oxford University Press, New York, 2003 and reference therein.
[2] J.-M. Lehn in Supramolecular Chemistry, Wiley-VCH, Weinheim,
1995 and references therein.
2
2
ð1Þ
Dabs408 ¼ ðA ꢃ Kd2 ꢃ ½SꢄþB ꢃ ½Sꢄ Þ=ðKd1 ꢃ Kd2þKd2 ꢃ ½Sꢄþ½Sꢄ Þ
[3] a) E. L. Raven, A. G. Mauk in Advances in Inorganic Chemistry,
Vol. 51, Academic Press, San Diego, 2000, pp. 1–49; b) T. Hayashi,
Y. Hisaeda, Acc. Chem. Res. 2002, 35, 35–43; c) T. Ueno, S. Abe, N.
Yokoi, Y. Watanabe, Coord. Chem. Rev. 2007, 251, 2717–2731;
d) M. D. Toscano, K. J. Woycechowsky, D. Hilvert, Angew. Chem.
2007, 119, 3274–3300; Angew. Chem. Int. Ed. 2007, 46, 3212–3236;
e) L. Fruk, J. Mꢃller, G. Weber, A. Narvꢄez, E. Domꢅnguez, C. M.
Niemeyer, Chem. Eur. J. 2007, 13, 5223–5231; f) Y. Lu, D. K.
Garner, J.-L. Zhang, in Wiley Encyclopedia of Chemical Biology,
Vol. 1 (Eds: T. P. Begley), Wiley, Hoboken, NJ, 2008, pp. 124–133;
g) L. Fruk, C.-H. Kuo, E. Torres, C. M. Niemeyer, Angew. Chem.
2009, 121, 1578–1603; Angew. Chem. Int. Ed. 2009, 48, 1550–1574;
h) J. Steinreiber, T. R. Ward, Coord. Chem. Rev. 2008, 252, 751–756;
i) T. Matsuo, T. Hayashi, J. Porphyrins Phthalocyanines 2009, 13,
1082–1089.
in which [S] is the concentration of 2-methoxyphenol, A and B are con-
stants, and Kd1 and Kd2 are the first and second dissociation constants of
2-methoxyphenol, respectively.
Steady-State Kinetics for 2-Methoxyphenol Oxidation
Steady-state kinetic measurements for 2-methoxyphenol oxidation cata-
lyzed by H64D Mb mutants were carried out using the stopped-flow
method at 258C. A mixture of Mb and various concentrations of 2-me-
thoxyphenol in sodium malonate buffer (20 mm; pH 6.0) was rapidly
mixed with H2O2 in the same buffer after incubation at 258C in the
sample reservoirs. The final concentrations are as follows: [Mb]=4 mm,
[2-methoxyphenol]=0.025–3 mm, and [H2O2]=100 mm. The oxidation re-
actions were monitored by observing the increase in absorbance at
2498
www.chemasianj.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2011, 6, 2491 – 2499

Peroxidase Activity of Myoglobin
[4] E. Antonini, M. Brunori in Hemoglobin and Myoglobin in Their Re-
action with Ligands, North-Holland, Amsterdam, 1971.
[5] a) T. Matsui, S. Ozaki, Y. Watanabe, J. Am. Chem. Soc. 1999, 121,
9952–9957; b) T. D. Pfister, T. Ohki, T. Ueno, I. Hara, S. Adachi, Y.
Makino, N. Ueyama, Y. Lu, Y. Watanabe, J. Biol. Chem. 2005, 280,
12858–12866 and references therein.
[18] The two sets of distinct peaks with the intensity ratio of 1:1 were ob-
served immediately after the incorporation of the hemin into the
apoprotein. See Ref. [13b].
[19] Similar behavior was observed in Mb with a flavin-pendant hemin,
in which the flavin moiety was linked to one of the two propionates.
T. Matsuo, T. Hayashi, Y. Hisaeda, J. Am. Chem. Soc. 2002, 124,
11234–11235.
[20] The latest report regarding product analysis of 2-methoxyphenol ox-
idation mediated by peroxidases proposed that 3,3’-dimethoxy-4,4’-
biphenylquinone is formed as the main product. D. R. Doerge, R. L.
Divi, M. I. Churchwell, Anal. Biochem. 1997, 250, 10–17.
[21] It was confirmed that the kinetic parameters were independent of
the concentrations of H2O2 under the conditions employed. There-
fore, the kcat values shown in Table 1 directly reflect the process of
oxidation of 2-methoxyphenol.
[6] I. Hamachi, T. Nagase, Y. Tajiri, S. Shinkai, Chem. Commun. 1996,
2205–2206.
[7] a) T. Hayashi, Y. Hitomi, T. Ando, T. Mizutani, Y. Hisaeda, S. Kita-
gawa, H. Ogoshi, J. Am. Chem. Soc. 1999, 121, 7747–7750; b) T.
Hayashi, T. Matsuda, Y. Hisaeda, Chem. Lett. 2003, 32, 496–497.
[8] E. Monzani, G. Alzuet, L. Casella, C. Redaelli, C. Bassani, A. M.
Sanangelantoni, M. Gullotti, L. D. Gioia, L. Santagostini, F. Chille-
mi, Biochemistry 2000, 39, 9571–9582.
[9] S. Sakamoto, K. Kudo, Bull. Chem. Soc. Jpn. 2005, 78, 1749–1756.
[10] a) T. Hayashi, D. Murata, M. Makino, H. Sugimoto, T. Matsuo, H.
Sato, Y. Shiro, Y. Hisaeda, Inorg. Chem. 2006, 45, 10530–10536;
b) T. Matsuo, D. Murata, H. Hori, Y. Hisaeda, T. Hayashi, J. Am.
Chem. Soc. 2007, 129, 12906–12907.
[11] T. Matsuo, A. Hayashi, M. Abe, T. Matsuda, Y. Hisaeda, T. Hayashi,
J. Am. Chem. Soc. 2009, 131, 15124–15125.
[12] H. Sato, T. Hayashi, T. Ando, Y. Hisaeda, T. Ueno, Y. Watanabe, J.
Am. Chem. Soc. 2004, 126, 436–437.
[13] a) K. Harada, M. Makino, H. Sugimoto, S. Hirota, T. Matsuo, Y.
Shiro, Y. Hisaeda, T. Hayashi, Biochemistry 2007, 46, 9406–9416
and references therein; b) T. Hayashi, K. Harada, K. Sakurai, H.
Shimada, S. Hirota, J. Am. Chem. Soc. 2009, 131, 1398–1400 and
reference therein.
[22] M. I. Savenkova, J. M. Kuo, P. R. Ortiz de Montellano, Biochemistry
1998, 37, 10828–10836.
[23] The spectral changes would be caused by the change in the polarity
at the distal site.
[24] There are interactions of the heme-propionate with amino acid resi-
dues at the distal and the proximal sites in Mb
these interactions do not exist for Mb(H64D·2).
ACHTUNGTREN(NUGN H64D·1), whereas
AHCTUNGTRENNUNG
[25] The strong coordination of the proximal ligand to the heme iron is
recognized as a “push-effect” that controls the reactivities of the
high-valent species in native heme-containing oxidases. See
Ref. [3d].
[26] The intramolecular hydroxylation of the amino acid residue at the
distal site has been reported for the F43W/H64D/V68I myoglobin
mutant, see Ref. [5b].
[27] For thioanisole oxidation using H218O2, the complete incorporation
of 18O into the products was confirmed by GC-MS. The enantiomet-
ric excess value (ee) for methyl phenyl sulfoxide was not determined
in this research, because the value would be time-dependent. Both
stereoisomers of the sulfoxide product can be converted into the sul-
fone through the second oxygen transfer. The sequential oxidation
will bring about the change in the ratio of the stereoisomers in the
transiently formed sulfoxide product unless the second oxidation be-
comes quite negligible. The previous paper reported the slightly
preferential production of S isomer (6% ee) in the oxidation of thio-
anisole mediated by MbACTHNUTRGNEUGN(H64D·1) (Reference [5a] and correction:
Biochemistry 2008, 47, 2700).
[28] F. W. Teale, Biochim. Biophys. Acta 1959, 35, 543.
[29] a) G. D. DePillis, B. P. Sishta, A. G. Mauk, P. R. Ortiz de Montella-
no, J. Biol. Chem. 1991, 266, 19334–19341; b) B. Chance, A. C.
Maehly, Methods Enzymol. 1955, 2, 764–775.
[14] a) G. N. La Mar, U. Pande, J. B. Hauksson, R. K. Pandey, K. M.
Smith, J. Am. Chem. Soc. 1989, 111, 485–491; b) G. N. La Mar, H.
Toi, R. Krishnamoorthi, J. Am. Chem. Soc. 1984, 106, 6395–6401;
c) Y. Wu, E. Y. T. Chien, S. G. Sligar, G. N. La Mar, Biochemistry
1998, 37, 6979–6990 and reference therein.
[15] The X-ray structures of similar mutants, H64D/V68A and H64D/
V68S Mbs, show the normal heme orientation in the heme pocket.
H.-J. Yang, T. Matsui, S. Ozaki, S. Kato, T. Ueno, G. N. Phillips, Jr.,
S. Fukuzumi, Y. Watanabe, Biochemistry 2003, 42, 10174–10181.
[16] The propionate-truncated heme is also accommodated in the normal
orientation when equilibrium is attained. The crystal structure of the
Mb reconstituted with 6-methyl-6-despropionate heme reveals the
normal orientation of the heme. See Refs. [13a] and [14].
[17] La Mar and co-workers reported that the interaction of Arg45–6-
propionate is stronger than that of Ser92–7-propionate in sperm
whale myoglobin. Our observation that the nonsubstituted propio-
nate in 3 prefers the interaction with Arg45 will be consistent with
that observed by La Marꢂs group. See Refs. [13a] and [14].
Received: February 4, 2011
Published online: June 9, 2011
Chem. Asian J. 2011, 6, 2491 – 2499
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2499