How does methylviologen cation radical supply two electrons to the formate dehydrogenase in the catalytic reduction process of CO2to formate?

Article abstract of DOI:10.1039/d0cp02665d

Formate dehydrogenase from Candida boidinii (EC.1.2.1.2; CbFDH) is a commercially available enzyme and can be easily handled as a catalyst for the CO2 reduction to formate in the presence of NADH, single-electron reduced methylviologen (MV+) and so on. It was found that the formate oxidation to CO2 with CbFDH was suppressed using the oxidized MV as a co-enzyme and the single-electron reduced MV (MV+) was effective for the catalytic activity of CbFDH for the CO2 reduction to formate compared with that using the natural co-enzyme of NADH [Y. Amao, Chem. Lett., 2017, 46, 780-788]. The CO2 reduction to formate catalyzed by CbFDH requires two molecules of the MV+. In order to clarify the two-electron reduction process using MV+ in the CO2 reduction to formate catalyzed with CbFDH, we attempted enzyme reaction kinetics, electrochemical and quantum chemical analyses. Kinetic parameters obtained from the enzymatic kinetic analysis metric revealed an index of affinity of MV+ for CbFDH in the CO2 reduction to formate. From the results of the electrochemical analysis, it was predicted that only one molecule of MV+ was bound to CbFDH, and the MV bound to CbFDH was to be necessarily re-reduced by the electron source outside of CbFDH to supply the second electron in the CO2 reduction to formate. From the results of docking simulation and density functional theory (DFT) calculations, it was indicated that one molecule of MV bound to the position close to CO2 in the inner part of the substrate binding pocket of CbFDH contributed to the two-electron CO2 reduction to formate.

Full text of DOI:10.1039/d0cp02665d

PCCP
View Article Online
View Journal
PAPER
How does methylviologen cation radical supply
two electrons to the formate dehydrogenase
Cite this:DOI: 10.1039/d0cp02665d
in the catalytic reduction process of CO to
2
formate?
a
bc
Akimitsu Miyaji* and Yutaka Amao
*
Formate dehydrogenase from Candida boidinii (EC.1.2.1.2; CbFDH) is a commercially available enzyme
and can be easily handled as a catalyst for the CO reduction to formate in the presence of NADH,
single-electron reduced methylviologen (MV ) and so on. It was found that the formate oxidation to
CO with CbFDH was suppressed using the oxidized MV as a co-enzyme and the single-electron
reduced MV (MV ) was effective for the catalytic activity of CbFDH for the CO
compared with that using the natural co-enzyme of NADH [Y. Amao, Chem. Lett., 2017, 46, 780–788].
2
+
ꢀ
2
+
ꢀ
2
reduction to formate
+
ꢀ
The CO
the two-electron reduction process using MV in the CO
we attempted enzyme reaction kinetics, electrochemical and quantum chemical analyses. Kinetic para-
2
reduction to formate catalyzed by CbFDH requires two molecules of the MV . In order to clarify
+
ꢀ
2
reduction to formate catalyzed with CbFDH,
+
ꢀ
meters obtained from the enzymatic kinetic analysis metric revealed an index of affinity of MV for
CbFDH in the CO reduction to formate. From the results of the electrochemical analysis, it was predicted
that only one molecule of MV was bound to CbFDH, and the MV bound to CbFDH was to be necessarily
re-reduced by the electron source outside of CbFDH to supply the second electron in the CO reduction
to formate. From the results of docking simulation and density functional theory (DFT) calculations, it was
indicated that one molecule of MV bound to the position close to CO in the inner part of the substrate
binding pocket of CbFDH contributed to the two-electron CO reduction to formate.
2
+
ꢀ
Received 16th May 2020,
Accepted 22nd July 2020
2
DOI: 10.1039/d0cp02665d
2
rsc.li/pccp
2
reduction has received considerable attention, CbFDH is used
as a catalyst for the visible-light-driven redox system consisting
Introduction
Formate dehydrogenases (FDHs) are a set of enzymes that of an electron donor (D), a photosensitizer (P), an electron carrier
+
catalyze formate oxidation to CO
second substrate, and the CO reduction to formate requires directly reduced to NADH with a metal complex-based photo-
electron donation from a second substrate.
2
, donating the electrons to a (C) and CbFDH as shown in Fig. 1. In general, NAD is not
2
1
–6
Among these sensitizer. By incorporating a rhodium complex as a catalyst for
FDHs, FDH from Candida boidinii (EC.1.2.1.2; CbFDH) is hydride production into this system, it is possible to architect
+
commercially available and can be easily handled as a catalyst a visible-light-driven NAD to NADH reduction system and
7
–14
for the CO2 reduction to formate. CbFDH catalyzes formate accomplish CO reduction to formate using CbFDH.
2
oxidation to CO and CO reduction to formate with the redox
coupling of NAD /NADH as shown in the following equation.
Here, let us focus on the catalytic activity of CbFDH for
formate oxidation or CO2 reduction with NAD /NADH redox
2
+
2
+
coupling. For example, the kinetic parameters for the Michaelis
CbFDH
ꢁ
+
CO2 þNADH þHþ
HCO2H þNADþ
m
ꢁ
!
constant (K ) values of NAD and NADH for CbFDH in the
2
formate oxidation to CO and the reverse reaction are deter-
+
Thus, CbFDH is an attractive biocatalyst for CO reduction mined. The Km values for NAD in the formate oxidation and
2
to formate. For example, as solar fuel production based on CO
2
2
NADH in the CO reduction to CbFDH are estimated to be 50
6
,15,16
and 2087 mM, respectively.
by the lower concentration of NAD , compared with that of
NADH, and the affinity of NAD for CbFDH is higher than that
of NADH. Even if the effective NAD reduction to NADH with
Thus, CbFDH can be activated
a
+
School of Materials and Chemical Technology, Tokyo Institute of Technology,
4
259 G1-14, Nagatsuda, Midori-Ku, Yokohama 226-8502, Japan
+
b
c
Graduate School of Science, Osaka City University, 3-3-138 Sugimoto,
Sumiyoshi-ku, Osaka 558-8585, Japan. E-mail: amao@ocarina.osaka-cu.ac.jp
Research Centre of Artificial Photosynthesis (ReCAP), Osaka City University,
+
+
the photoredox system could be achieved, NAD is a very
3
-3-138 Sugimoto, Sumiyoshi-ku, Osaka 558-8585, Japan
expensive biological material and, it is necessary to improve
This journal is © the Owner Societies 2020
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
0
Fig. 3 Chemical structures of methylviologen (MV) and 1,5-bis (4 -methyl-
4
5
,4-bipyridin-4-yl)pentane (MVC VM).
+
ꢀ
molecules of the single-electron reduced BP (BP ). In addition
+
ꢀ
to the kinetic parameters, elucidation of the BP to CbFDH will
contribute to the design and the architecture of the highly
active co-enzyme for CbFDH.
Fig. 1 Visible-light-driven CO
2
reduction to formate with the redox
In this article, to clarify the two-electron supply mechanism
from single-electron reduced BP to CbFDH, we studied the
system consisting of an electron donor (D), a photosensitizer (P), an
electron carrier (C) and CbFDH. Cox and Cred indicate the oxidized and
single-electron reduced electron carriers, respectively.
extended enzymatic kinetic analysis for the CO reduction to
2
formate catalyzed by CbFDH using the dithionite-induced
+
ꢀ
single-electron reduced methylviologen (MV ) and a dimeric
+
0
0
its usage and turnover of the redox coupling of NAD /NADH. As 4,4 -BP, 1,5-bis(4 -methyl-4,4-bipyridin-4-yl)pentane (MVC
5
VM;
+
long as the redox coupling of NAD /NADH is used, the affinity chemical structure is shown in Fig. 3). We also investigated
+
2
between NAD or NADH and CbFDH does not change, thus, the mechanism of CO reduction to formate using electro-
CbFDH catalytic activity cannot be controlled in the photoredox chemically reduced MV and CbFDH that are free of the reducing
system. As the produced NADH acts as a sacrificial reagent and agent sodium dithionite. Furthermore, the interaction between
+
+
ꢀ
is consumed, moreover, the redox coupling of NAD /NADH is MV and CbFDH was clarified by docking simulation, and
not suitable for use in the photoredox system.
electron transfer ability was evaluated based on density func-
In contrast, the visible-light-driven CO reduction to formate tional theory (DFT) calculations.
2
systems with the coupling of the photoreduction of various
0 0 0 0
4
,4 - or 2,2 -bipyridinium salts (4,4 - or 2,2 -BPs) with ruthenium
Results and discussion
polypyridyl complex, water soluble zinc porphyrin or chlorophyll-a
17–32
and CbFDH has been reported.
Furthermore, the effects of Analysis of enzymatic kinetics based on the Michaels–Menten
0
the chemical structures and redox potentials of various 2,2 -BPs and Hill equations
on the CO reduction to formate were reported.26 In this reaction,
2
First, Michaels–Menten-type enzymatic kinetics for the
0
0
the single-electron reduced 4,4 - or 2,2 -BPs act as a co-enzyme for
CbFDH-catalyzed CO
reduced MV and MVC
kinetics, the CO
2
reduction to formate using dithionite-
VM was analyzed. For analysis of enzyme
0
CbFDH instead of NADH. Representative example of 4,4 -BP
5
single-electron reduction of methyl viologen (MV) with sodium
dithionite produces a radical species as shown in Fig. 2.
+
ꢀ ꢀ ꢀ
+
+
2
reduction to formate with MV or MV C VM
5
and CbFDH was conducted under the following conditions.
Recently, we reported the kinetic parameters of various
Sample solution consisted of MV or MVC VM, sodium dithionite
5
0
single-electron reduced 4,4 -BPs for CO
2
reduction to formate
(
160 mM) and FDH (7.5 mM) in CO -saturated 50 mM sodium
2
with CbFDH by using enzymatic kinetic analysis for the first
time. We also elucidated that the formate oxidation to CO with
CbFDH is suppressed using oxidized 4,4 - or 2,2 -BP and the
pyrophosphate buffer (pH 7.0). The concentration of formate
was measured for 1 min to estimate the initial rate (v ) of the CO
2
0
2
0
0
reduction to formate. The amount of formate production in this
system was measured by ionic chromatography. Fig. 3 shows the
relationship between the initial rate of formate production and
0
0
single-electron reduced 4,4 - or 2,2 -BP was effective in the
catalytic activity of CbFDH for CO reduction to formate, as
2
3–38
3
shown in the following equation.
+
ꢀ
+
ꢀ
ꢀ
+
concentration of MV and MV C
5
VM.
CbFDH
CO2 þ2BPþꢀþ2Hþ ƒƒƒƒ!HCO2H þ2BP
Vmax½Viologen radicalꢂ
Km þ½Viologen radicalꢂ
n ¼
On the other hand, as shown in the reaction equation, the
CO
2
reduction to formate catalyzed by CbFDH requires two
The initial rate of formate production increased with
+
ꢀ
+
ꢀ
+
ꢀ
increasing concentrations of MV or MV C VM and then
5
reached constant values; each plot obeyed the following
Michaelis–Menten equation.
Here, Vmax and apparent Km are maximum velocity and
m
Michaelis constant, respectively. The Vmax and apparent K values
were determined by the curve fitting using the Michaelis–Menten
equation. From the results of the curve fitting, the apparent
Fig. 2 Redox scheme for methylviologen (MV).
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2020

View Article Online
PCCP
and Vmax for the MV were estimated to be 223 mM and
Paper
+
ꢀ
K
m
ꢁ
1
1
8.2 mM min , respectively. In contrast, the apparent K
m
+
ꢀ
+
ꢀ
and Vmax for MV C VM were estimated to be 191 mM and
5
ꢁ
1
17.7 mM min , respectively. The turnover number termed kcat
is estimated by dividing V by the CbFDH concentration.
max
5
Fig. 6 Redox scheme for MVC VM with two viologen moieties.
+
ꢀ
+
ꢀ
+
ꢀ
Thus, kcat values for MV and MV C
5
VM were estimated to
be 2.43 and 2.36 min , respectively. The values of catalytic
ꢁ1
+
ꢀ
+
ꢀ
+
ꢀ
efficiency termed kcat/K
estimated to be 0.011 and 0.012 mM min , respectively. Here,
m
for MV and MV C
5
VM were
ꢁ
1
ꢁ1
+
ꢀ
2
two molecules of MV are required in the CO reduction to
+
ꢀ
formate catalyzed by CbFDH. When two molecules of MV are
simultaneously bound to CbFDH, the Michaelis–Menten equa-
tion can be approximately regarded as the following equation.
0
:5Vmax½Viologen radicalꢂ
n ¼
Kmþ½Viologen radicalꢂ
Fig. 7 Possible mechanism for viologen cation radical complex bound to
From the results of the curve fitting using this equation, the the active site of CbFDH.
apparent K and Vmax for the single-electron reduced MV were
estimated to be 223 mM and 36.4 mM min , respectively. From
m
ꢁ
1
+
ꢀ
forms a complex between viologen cation radical sites in the
molecule, and it is expected that the viologen cation radical
complex site binds to the active site of CbFDH as shown in Fig. 7.
By using sodium dithionite as a reducing agent, each of the
this result, the k and kcat/Km values for MV were estimated
cat
ꢁ
49
1
ꢁ1
ꢁ1
to be 4.85 min and 0.022 mM min , respectively. The
reaction mechanism for CO reduction to formate catalyzed
2
+
ꢀ
by CbFDH with MV is shown in Fig. 5.
+
ꢀ
5
39–48 two viologen sites in MVC VM is reduced by single electrons,
It is known that MV forms a complex with two molecules,
and considering the mechanism of Fig. 5, it is expected to bind
to the active site of CbFDH as an MV complex. Here, the
dithionite reduction scheme in MVC VM with two viologen
moieties is shown in Fig. 6. It is also known that MV C VM
+
ꢀ
+
ꢀ
resulting in two radicals in the molecule (MV C
the reaction mechanism for CO reduction to formate catalyzed
by CbFDH with MV C VM is shown in Fig. 8.
5
VM). Thus,
+
ꢀ
2
+
ꢀ
+
ꢀ
5
5
+
+
ꢀ
ꢀ
value of MV C VM for
5
+
ꢀ
+
5
ꢀ
It is expected that the apparent K
m
+
ꢀ
CbFDH will be half of that of MV in the CO reduction as shown
2
in Fig. 8. However, there is no significant difference in K
m
value for
reduction.
reduction to
formate using CbFDH does not proceed with only one molecule of
+
ꢀ
+
ꢀ
CbFDH between MV C VM and MV in the CO
5
+
2
ꢀ
+
ꢀ
Therefore, even when MV C VM is used, the CO
5
2
+
ꢀ
+
ꢀ
+
ꢀ
+
ꢀ
+
ꢀ
MV C
5
VM, and as with MV , two molecules of MV C
5
VM are
required. Kinetic parameters obtained from the enzymatic kinetic
+
ꢀ
analysis metric revealed an index of affinity of MV
or
+
ꢀ
+
ꢀ
MV C VM for CbFDH in the CO reduction to formate.
5
2
ꢀ
+
When two molecules of MV are sequentially bound to
CbFDH, the reaction mechanism for CO reduction to formate
catalyzed by CbFDH with MV is shown in Fig. 9.
2
+
ꢀ
+
ꢀ
Fig. 4 The relationship between the concentration of MV (blue) or
In this reaction mechanism, the CO
obeyed the following Hill equation.
2
reduction to formate
+
ꢀ
+
ꢀ
MV
C
5
VM (red) and the initial rate of formate production (v
0
). Blue and
red lines indicate the curve fitting using the Michaels–Menten equation.
Orange line indicates the curve fitting using the Hill equation. Reaction
2
Vmaxð½Viologen radicalꢂ=KmÞ
5
conditions: MV or MVC VM, sodium dithionite (160 mM) and FDH (7.5 mM) in
n ¼
2
CO -saturated 50 mM sodium pyrophosphate buffer (pH 7.0).
2
1þð½Viologen radicalꢂ=KmÞ
Fig. 5 Reaction mechanism for CO
2
reduction to formate catalyzed by
Fig. 8 Reaction mechanism for CO
2
reduction to formate catalyzed by
+
ꢀ
+
ꢀ
+
ꢀ
CbFDH with two molecules of MV
.
CbFDH with MV
C
5
VM.
This journal is © the Owner Societies 2020
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
2
Fig. 9 Reaction mechanism for CO reduction to formate catalyzed by
+
ꢀ
CbFDH with two sequentially bound molecules of MV
.
The orange line in Fig. 4 indicates the curve fitting using Hill
+
ꢀ
equation for MV . From the results of the curve fitting using Fig. 10 Cyclic voltammograms for MV (blue) and MV + CbFDH (red) in
+
ꢀ
this equation, the apparent K
m
and Vmax for MV in the CO
2
2
CO -saturated 0.1 M phosphate buffer (pH 6.8).
reduction to formate with CbFDH were estimated to be 226 mM
ꢁ
1
and 15.8 mM min , respectively.
Fig. 10 shows the cyclic voltammograms (CV) for MV (blue)
and MV coexisting with CbFDH (MV + CbFDH; red) in CO
saturated 0.1 M phosphate buffer (pH 6.8). For CV measure-
ment, grassy carbon, platinum wire and Ag/AgCl electrodes
were used as working, counter and reference electrodes, respec-
tively. The first reduction potential for MV was estimated to be
However, in this reaction system, sodium dithionite is
excessive as a reducing agent. The following control experi-
ments were conducted to investigate the possibility of direct
2
reduction of CO
sodium dithionite (concentration: 0, 50, 100 and 150 mM)
was added to the CO -saturated 50 mM sodium pyrophosphate
2
to formate with sodium dithionite. First,
2
ꢁ
660 mV. On the other hand, a sigmoidal and steady-state
buffer (pH 7.0) and then incubated at 30.5 1C for 1 h. No
formate production was observed under the condition of
sodium dithionite concentration (0–150 mM). Moreover, MV
cathodic wave was observed in CO saturated 0.1 M phosphate
2
buffer containing MV and CbFDH. From the results of Fig. 10,
in contrast, the peak current of MV coexisting with CbFDH was
larger than that of MV in the CV measurement. Similar catalytic
current observation has been reported for CV measurement of a
(
concentration: 0–3.5 mM) and sodium dithionite (160 mM), in
the absence of CbFDH, were added to the CO -saturated 50 mM
sodium pyrophosphate buffer (pH 7.0) and then incubated at
0.5 1C for 1 h. Also, no formate production was observed under
2
5
0
solution containing MV, bicarbonate, and W-containing FDH.
3
Thus, the increase in current corresponds to the catalytic
these conditions. In all the cases, it was confirmed that the
concentrations of CO2 and bicarbonate ion in the solution
measured by ionic chromatography were unchanged during
the incubation. These results suggest that direct reduction of
current based on CO reduction in the presence of CbFDH.
2
Next, the CO2 reduction to formate was investigated by
applying a constant potential (ꢁ660 mV vs. Ag/AgCl) to a CO2
saturated-buffer containing MV and CbFDH. In constant
CO
In order to investigate the possibility of direct electron
transfer from sodium dithionite to CbFDH, CO reduction to
2
to formate with sodium dithionite did not proceed.
2
potential CO reduction to formate using CbFDH and MV, an
H-type electrochemical cell as shown in Fig. 11 was used.
Platinum wire, carbon fabric paper (CFP) (TGP-H-030 obtained
2
formate in sodium pyrophosphate buffer (pH 7.0) without MV
was attempted. The sample solution consisted of sodium
dithionite and FDH (7.5 mM) in CO -saturated 50 mM sodium
pyrophosphate buffer (pH 7.0). In this test, the sodium dithio-
nite concentration was varied from 0 to 150 mM. After 1 h
incubation at 30.5 1C, no formate production was observed and
2
from TORAY INDUSTRIES, INC.; electrode area: 3.75 cm ) and
Ag/AgCl electrodes were used as an anode, a cathode and a
2
reference electrode, respectively. The CO saturated phosphate
2
buffer solution (pH 6.8) was used as an anolyte. The
CO -saturated phosphate buffer solution (pH 6.8) containing
2
MV (100 mM) and CbFDH (7.5 mM) was used as a catholyte. The
amount of formate production in this system was measured by
ionic chromatography. Furthermore, for determination of the
2
the concentrations of CO and bicarbonate ion in the solution
were unchanged during the incubation.
It is still possible that MV bound to CbFDH is again reduced by
sodium dithionite and finally supplies two electrons to CbFDH.
Therefore, the analysis of the reaction using dithionite-reduced
MV does not lead to the elucidation of the two-electron supply
+
ꢀ
mechanism from MV to CbFDH.
CO reduction to formate catalyzed with CbFDH using
2
electrochemically single-electron-reduced MV
In order to eliminate the influence of excess sodium dithionite in
the reaction system, the CO reduction to formate catalyzed by
2
CbFDH was investigated using electrochemically single-electron-
2
reduced MV. First, the electrochemical analysis for CO reduction
to formate catalyzed by CbFDH using MV is investigated.
2
Fig. 11 Schematic representation for the electrochemical CO reduction
+
ꢀ
to formate using CbFDH and MV.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2020

View Article Online
PCCP
Paper
Fig. 13 Proposed mechanisms for the CO
2
reduction to formate with
+
ꢀ
CbFDH and two molecules of MV
.
Fig. 12 Time-dependence of the current changes in the H-type electro-
chemical cell consisting of CO
2
-saturated phosphate buffer as an anolyte
and CO -saturated phosphate buffer containing MV and CbFDH as a
2
From the results of electrochemical experiments, in addition
to the mechanism in Fig. 5, two mechanisms were proposed for
catholyte under a continuously applied potential at ꢁ660 mV vs. Ag/AgCl.
The inset shows the change of the catholyte under continuous potential
application.
+
ꢀ
the CO reduction to formate with CbFDH and MV as shown
2
in Fig. 13(a) and (b). The major difference among the three
mechanisms is that, in the mechanism in Fig. 5, two molecules
+
ꢀ
+
ꢀ
MV concentration produced from the UV-visible absorption of MV are bound to CbFDH simultaneously, in the mecha-
+
ꢀ
spectrum (SHIMADZU, MultiSpec-1500), an H-type cell with a nism in Fig. 13(a), MV is sequentially bound to CbFDH,
+
ꢀ
1
cm quartz cell on the cathodic side was used. MV concen- and in the mechanism in Fig. 13(b), MV bound to CbFDH is
tration was determined using the absorbance at 605 nm with a re-reduced by external electron injection. Each mechanism
ꢁ
1
ꢁ1 51
molar absorption coefficient e605 = 13 500 M cm
.
process according to the previous report based on the photo-
5
2
Fig. 12 shows the time-dependence of current changes in the electrochemical studies, is as follows.
H-type electrochemical cell consisting of CO -saturated phos-
2
Mechanism in Fig. 5
phate buffer as an anolyte and CO -saturated phosphate buffer
2
+
ꢀ
containing MV and CbFDH as a catholyte under continuously (1) Tow molecules of MV simultaneously bind to the catalytic
+
ꢀ
+
applied potential at ꢁ660 mV vs. Ag/AgCl.
active site of CbFDH to form the CbFDH–2MV complex.
ꢀ
(2) The two electrons transfer from two MV s to the CO
2
As shown in Fig. 12, an average current of ꢁ600 mA was
constantly observed under a continuously applied potential at binding site of CbFDH in the complex.
ꢁ
ꢁ
660 mV vs. Ag/AgCl. The inset photograph in Fig. 12 also showed
(3) CO reduces to formate with CbFDH –2MV.
(4) Tow MVs dissociate from the CbFDH–2MV complex.
2
that the color of the solution changed from colorless to blue due to
+
ꢀ
MV production. After 1 h of continuously applied potential,
.3 mM of formate production was observed. Considering that
Mechanism (a)
5
+
ꢀ
+
ꢀ
the MV concentration by electrochemical reduction (ca. 1.0 mM) (1) MV binds to the catalytic active site of CbFDH to form the
+
ꢀ
is higher than the apparent K
kinetic analysis, the formate production rate (0.09 mM min ) was
lower than that of the system of dithionite-reduced MV and of CbFDH in the complex.
m
value calculated from enzymatic CbFDH–MV complex.
ꢁ
1
+
ꢀ
(2) The electron transfers from MV to the CO binding site
2
ꢁ
1
+
ꢀ
CbFDH (ca. 15 mM min at 1.0 mM of MV ) shown in Fig. 4.
(3) MV dissociates from the CbFDH–MV complex.
(4) Second MV binds to the catalytic active site of CbFDH.
(5) The electron transfers from MV to the CO binding site
+
ꢀ
+
ꢀ
The Faraday efficiency calculated from the amount of MV
produced and the current value was estimated to be ca. 85%.
Next, the CO saturated phosphate buffer containing MV as of CbFDH in the complex.
+
ꢀ
2
2
a catholyte was electrochemically reduced in advance under a
(6) CO reduces to formate with CbFDH–MV.
2
continuously applied potential at ꢁ660 mV vs. Ag/AgCl for 1 h,
(7) MV dissociates from the CbFDH–MV complex.
2
and then CbFDH was added to investigate the CO to formate.
Mechanism (b)
However, no formate production is observed in this condition.
+
ꢀ
Finally, a constant potential at ꢁ660 mV vs. Ag/AgCl was (1) MV binds to the catalytic active site of CbFDH to form the
+
ꢀ
applied to the CO
and CbFDH as a catholyte for 1 h, and then sodium dithionite
was added in an attempt to reduce CO2 to formate. After of CbFDH in the complex.
2
-saturated phosphate buffer containing MV CbFDH–MV complex.
+
ꢀ
(2) The electron transfers from MV to the CO binding site
2
addition of sodium dithionite for 1 h, 189 mM of formate was
(3) Accepting electrons from the electrode, the MV moiety of
+
ꢀ
produced.
the CbFDH–MV complex is re-reduced to form CbFDH–MV .
+
ꢀ
+
ꢀ
When MV is used as a co-enzyme for CbFDH, CO
2
reduction
(4) The electron transfers from MV to the CO binding site
2
+
ꢀ
to formate with CbFDH requires two molecules of MV . of CbFDH in the complex.
This journal is © the Owner Societies 2020
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
+
ꢀ
(
5) CO
2
reduces to formate with CbFDH–MV.
MV to CbFDH and subsequent electron transfer from sodium
dithionite.
(6) MV dissociates from the CbFDH–MV complex.
The CO reduction to formate did not proceed even when
In addition, the possibility that double electron-reduced MV
2
0
MV was electrochemically reduced to produce a large excess of (MV ) functions as a co-enzyme for CbFDH in the CO2
+
ꢀ
MV with respect to CbFDH. That is, it is shown that two reduction to formate was investigated using an electrochemical
+
ꢀ
molecules of MV simultaneously bind to CbFDH and then the method. The second reduction potential for MV was estimated
CO
reduction to formate (mechanism shown in Fig. 5) does to be ꢁ1.05 V vs. Ag/AgCl. The phosphate buffer containing MV
not proceed. External electron injection is required for the CO as a catholyte was electrochemically reduced in advance under
reduction to formate, based on the results of catalytic current continuously applied potential at ꢁ1.05 V vs. Ag/AgCl for 1 h,
observation in CV measurement shown in Fig. 10 and formate and then CO gas was introduced into a catholyte. Finally,
2
2
2
production under constant potential application (ꢁ660 mV vs. CbFDH was added to investigate the CO reduction to formate.
2
Ag/AgCl) in the CO -saturated buffer solution containing MV However, no formate production is observed under this condi-
2
and CbFDH. From the experimental result of adding CbFDH tion. Thus, it shows that the CO reduction to formate does not
2
+
ꢀ
0
after electrochemically producing excess MV , the mechanism proceed by donating two electrons from MV directly to CbFDH.
+
ꢀ
of sequentially binding MV to CbFDH as shown in Fig. 13(a)
2
Finally, the effect of CO concentration on the formate
was also rejected. When the solution containing MV and production catalyzed by CbFDH with MV is mentioned. In
CbFDH was electrochemically treated and then sodium dithionite many previous studies, bicarbonate has been considered as a
was added, formate production after 1 h incubation (189 mM) CO equivalent substrate for CbFDH-catalyzed formate produc-
2
was increased compared with that under constant potential tion. We reported that formate production with CbFDH in the
application (ꢁ660 mV vs. Ag/AgCl) (5.3 mM). These results presence of NADH was investigated in solutions with different
+
ꢀ
53
suggest that the CO reduction to formate by MV and CbFDH ratios of CO , bicarbonate and carbonate. The reaction rate of
2
2
depends on the external electron injection source. In other formate production with CbFDH increased in proportion to the
words, it is presumed that the process for the external electron concentration of CO in the reaction solution. On the other
injection to MV bound to CbFDH contributes to the reduction hand, the formate production with CbFDH was suppressed in
rate of CO to formate. Thus, it is suggested that the CO proportion to the concentration of bicarbonate or carbonate in
reduction to formate with CbFDH and MV proceeds with the reaction solution. Thus, CbFDH was found to catalytically
mechanism (b), as shown in Fig. 13. reduce only CO to formate among the three types of carbonate
2
2
2
+
ꢀ
2
To qualitatively prove this process, the CbFDH-catalyzed CO2 species. On the other hand, unfortunately, it is technically
reduction to formate in the presence of sodium dithionite below difficult to control the CO concentration in the buffer solution
2
MV concentration was investigated. Fig. 14 shows the Line- containing sodium dithionite, and the concentration-dependence
+
ꢀ
weaver–Burk plot for CbFDH-catalyzed CO
in the system containing MV, CbFDH (7.5 mM) under various the could not be mentioned. Since the focus of this study is on the
2 2 2
reduction to formate of CO on CbFDH-catalyzed CO reduction to formate with MV
+
ꢀ
concentrations of sodium dithionite (100, 500 mM and 160 mM). interaction between MV and sodium dithionite with CbFDH,
+
ꢀ
No change in the apparent K
m
value of MV for CbFDH in the CO concentration is in a condition of large excess (ca. 75 mM)
2
the CO2 reduction to formate was observed at any sodium with respect to CbFDH. Moreover, no formate production was
dithionite concentration. However, the gradient of the Line- observed under the condition of MV, sodium dithionite and
weaver–Burk plot tended to increase as the concentration of CbFDH without CO2.
added sodium dithionite decreased. This trend in the gradient
change indicates a typical sequential process for binding of
+
ꢀ
Docking simulation for MV binding to CbFDH
+
ꢀ
The binding mode of MV to CbFDH was estimated by using
+
ꢀ
the Autodock simulation. The MV , whose structure was
optimized by DFT calculations, was docked into the substrate
binding pocket of CbFDH (PDB ID: 5DN9), in which an azide
ion was replaced with a CO molecule. From the simulation,
2
47 clusters (total of 200 conformations) ranked by the lowest
binding energy were obtained. The binding energy distribution
in each cluster is shown in Fig. 15. Among these binding
+
ꢀ
modes, MV s in 5 clusters (total of 68 conformations) were
bound within 10 Å from CO in CbFDH, while the others were
around the entrance of the substrate binding pocket.
2
+
ꢀ
The structures of MV in the 5 clusters are shown in Fig. 16.
+
ꢀ
The lowest binding energy (LBE) of MV and its distance
+
ꢀ
between the methyl-group of MV and CO (D(CH –CO )) in
2
3
2
2
Fig. 14 Lineweaver–Burk plot for CbFDH-catalyzed CO reduction to
each cluster are listed in Table 1. In the simulation results,
structures similar to the binding state of Run 167, in which
MV binds to the inner part of the substrate binding pocket of
formate in the system containing MV, CbFDH (7.5 mM) and sodium
dithionite. The concentrations of sodium dithionite: 100 mM: green,
7
+
ꢀ
5
00 mM: red, and 160 mM: blue.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2020

View Article Online
PCCP
Paper
+
ꢀ
Fig. 15 The binding energy distribution of MV with CbFDH simulated by
Autodock. The right-side of the graph is an enlargement of the part
surrounded by the dotted line in the left one. Pink circles are the cluster
including Run 167, and blue circles are the clusters including Runs 29,
1
29, 1 and 139.
+
ꢀ
2
Table 1 MV bound within 10 Å from CO in CbFDH simulated by
Autodock
a
b
c
d
ꢁ1
e
Rank
Run
Num
LBE (kcal mol
)
3 2
D(CH –CO ) (Å)
1
2
7
8
29
129
1
167
139
18
13
13
7
ꢁ5.17
ꢁ4.57
ꢁ3.86
ꢁ3.83
ꢁ2.93
7.01
6.25
6.19
3.80
9.10
+
ꢀ
Fig. 16 Simulated MV orientation at the substrate binding site of CbFDH
+
ꢀ
by Autodock. CO2 is shown as a ball model, MV is shown as a ball and
stick model, and the substrate binding site of CbFDH is shown as the
contact surfaces.
1
1
17
a
b
Ranked by the lowest binding energy of clusters. Run number that
c
shows the lowest binding energy within each cluster. Number of
structures within each cluster. The lowest binding energy within each binding pocket of CbFDH (Fig. 16). On the other hand, E
2
cluster. Distance between methyl-groups of MV and CO .
d
SOMO
ꢁ1
e
+
ꢀ
(
ꢁ6.69 to ꢁ6.86 eV) and ETE values (503–532 kJ mol ) of
the other structures, which bind around the middle of the
ꢀ
CbFDH, were obtained. In these binding states, MV was close substrate binding pocket of CbFDH (Fig. 16), were lower and
to CO (B4 Å). Besides, 61 structures similar to the binding higher, respectively, than those of the Run 167 structure.
2
+
ꢀ
state shown as Run 29, Run 129, Run 1, and Run 139 in Fig. 16 Furthermore, ESOMO of MV in an aqueous medium was
+
ꢀ
were obtained in the state for MV binding from the middle of the ꢁ6.19 eV calculated using CPCM, when water was used as the
substrate binding pocket of CbFDH to the vicinity of the entrance. solvent. The potential level was also lower than that of the Run
+
ꢀ
+
ꢀ
In these binding states, MV was 6–10 Å away from CO
2
. These 167 structure. These results indicate that the MV bound to the
+
ꢀ
results suggest that MV binds more easily to the middle of the inner part of the substrate binding pocket of CbFDH shows
substrate binding pocket of CbFDH than to the inner part when a high electron-donating ability compared to the MV in
MV is inserted within the substrate binding pocket.
+
ꢀ
+
ꢀ
aqueous medium and bound to the entrance or the middle of
the substrate binding pocket of CbFDH.
The high ESOMO of Run 167 is due to the structural distortion
of MV . Among the five structures of MV investigated in this
ꢀ
Electron transfer ability of MV binding to CbFDH estimated by
density functional theory (DFT)
+
ꢀ
+
ꢀ
Based on the conformation in five clusters, the total energy study, the correlations between the dihedral angles of two
+
ꢀ
+
ꢀ
calculations of MV and MV were performed using density aromatic rings of MV and the ESOMO values or the ETE values
functional theory (DFT). From the calculations, the energy level were observed (Table 2). The highly distorted structure of Run
of the singly occupied molecular orbital (ESOMO) and the 167 is caused by the interaction with the surface of the
+
ꢀ
electron transfer enthalpy (ETE) of MV were estimated. These substrate binding pocket of CbFDH (Fig. 16). The space of the
are used as the indexes of electron donation and acceptor substrate binding pocket around the CO molecule is narrower
abilities. The higher the ESOMO, the higher the reactivity of than in the middle and around the entrance of the pocket.
2
+
ꢀ
the radical species. Also, the smaller the ETE value, the easier Therefore, when the MV molecule binds closely to the CO
2
+
ꢀ
the electron donation to the acceptor is. ESOMO and ETE molecule, the MV is structurally distorted due to its inter-
obtained from DFT calculations are summarized in Table 2. action with the protein surface. These results suggest that the
In the bound state of MV to CbFDH, the structure of Run substrate binding site of CbFDH controls the binding structure
+
ꢀ
+
ꢀ
1
67 showed the highest ESOMO and the smallest ETE among and electron donation ability of the MV .
The ETE of dithionite in the hydrophobic environment
18 kJ mol , respectively. The MV of Run 167 is bound at a similar to the substrate binding pocket was calculated to be
the 5 structures, which were calculated to be ꢁ5.68 EV and
ꢁ
1
+
ꢀ
4
ꢁ1
+
ꢀ
short distance (B3.8 Å) to CO
2
in the inner part of the substrate 439 kJ mol . Compared to this value, MV bound to the
This journal is © the Owner Societies 2020
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
+
ꢀ
Table 2 ETE and the HOMO level of the MV in CbFDH
Experimental
DA (1) Materials
a
ꢁ1
b
c
Rank
Run
ETE (kJ mol
)
E
SOMO (eV)
1
2
7
8
29
129
1
167
139
503
516
532
418
528
ꢁ6.69
ꢁ6.73
ꢁ6.86
ꢁ5.68
ꢁ6.81
30.8
26.9
2.5
88.7
15.3
FDH from Candida boidinii (CbFDH) was purchased from Roche
Diagnostics K.K. The molecular weight of CbFDH was estimated to
5
4
0
be 74 kDa. MV dichloride, 4,4 -bipyridine, 1,5-dibromopentane
and methyl iodide were purchased from Tokyo Kasei Co. Ltd. The
1
1
s
a
ꢁ
+
ꢀ
+
ꢀ
proton permeable film, Nafion 115 film, was obtained from
ETE = H(MV) + H(e ) ꢁH(MV ), where H(MV), and H(MV ) are the
ꢀ
total enthalpy values of MV and MV , which were estimated from the Sigma-Aldrich Co. Ltd. Carbon fabric paper (CFP) (TGP-H-030)
calculation using oB97XD/6-311++g(d,p) with CPCM(diethyl ether), and
was obtained from TORAY INDUSTRIES, INC. The other chemical
regents were of analytical grade or the highest grade available.
ꢁ
H(e ) is the enthalpy of electrons, which was taken from ref. 57.
b
+
ꢀ
c
Energy level of the SOMO of MV
aromatic rings.
. Dihedral angle between two
5
Preparation of MVC VM
0
1
,5-Bis(4 -methyl-4,4-bipyridin-4-yl)pentane (MVC VM) was synthe-
5
0
entrance or the middle of the substrate binding pocket of sized from 1-methyl-4,4 -bipyridinium iodide and 1,5-dibromo-
CbFDH (Run 29, 129, 1, and 139) had high ETE as shown in pentane as shown in Fig. 17.
0
Table 2, suggesting that these MVs play a role as the electron
acceptor from dithionite rather than donating electrons to CO
1-Methyl-4,4 -bipyridinium iodide was prepared from
0
0
2
.
4,4 -bipyridine and methyl iodide. 4,4 -Bipyridine (0.16 mol) and
According to the difference in ESOMO and ETE, together with methyl iodide (0.18 mol) were dissolved in 400 mL of acetone and
the distribution of the docking orientation simulated by Auto- stirred at room temperature for 24 h. A yellow precipitate was
+
ꢀ
dock, CbFDH is assumed to form a complex with a MV
collected by suction filtration and washed with acetone. The
moiety, which dominantly binds around the entrance of the desired product was recrystallized from ethanol/water and then
+
ꢀ
substrate binding pocket, and some of the MV moieties can dried under vacuum overnight. MVC
5
VM was prepared from
0
bind near CO in the inner part of the substrate binding pocket, 1-methyl-4,4 -bipyridinium iodide and 1,5-dibromopentane.
where MV donates an electron to CbFDH. In this case, the 1-Methyl-4,4 -bipyridinium iodide (0.34 mmol) was dissolved
CO
reduction is strongly influenced by the binding affinity of in 200 mL of acetonitrile, then 1,5-dibromopentane (0.13 mol)
MV to the inner part of substrate binding pocket of CbFDH was added to this solution. The reaction mixture was heated to
rather than the electron-donating property of the MV .
2
+
ꢀ
0
2
+
ꢀ
+
ꢀ
reflux for 24 h. A yellow precipitate was collected by suction
filtration, washed with acetonitrile and dried under vacuum
overnight. The desired product was recrystallized from ethanol.
The purity of MVC
nuclear magnetic resonance spectroscopy using a H-NMR
5
VM prepared was checked by proton
Conclusion
1
1
In order to elucidate the two-electron supply process using Varian GEMINI-200 system. H-NMR in D
2
O: d (ppm) 1.40
to formate (2H, quint, J = 5.5 Hz), 1.95 (4H, tt, J = 7.6, 5.5 Hz), 4.30
2
+
ꢀ
MV as a co-enzyme in the reduction process of CO
catalyzed by CbFDH, we attempted to clarify the mechanism by (6H, s), 4.52 (4H, t, J = 7.6 Hz), 8.30–8.35 (8H, 8.32 (ddd, J = 5.7,
experimental (enzyme reaction kinetics, and electrochemical 1.5, 0.5 Hz), 8.32 (ddd, J = 5.7, 1.4, 0.5 Hz)), 9.00–9.05 (8H, 9.02
methods) and quantum chemical analyses (docking simulation (ddd, J = 5.7, 2.0, 0.5 Hz), 9.02 (ddd, J = 5.7, 2.0, 0.6 Hz)).
and DFT calculation). Kinetic parameters obtained from the
CO
CbFDH
in this reaction system, sodium dithionite is excessive as a The CO reduction to formate with dithionite MV or MVC VM
2
reduction to formate with dithionite-reduced viologen and
enzymatic kinetic analysis metric revealed an index of affinity
+
ꢀ
2
of MV for CbFDH in the CO reduction to formate. However,
2
5
+
ꢀ
ꢀ
ꢀ
reducing agent. Therefore, the analysis of the reaction using and CbFDH was carried out as follows. MV or MV C VM was
5
dithionite-reduced MV does not lead to the clarification of the prepared with sodium dithionite using the following method.
+
ꢀ
two-electron supply mechanism from MV to CbFDH. From
the results of the electrochemical analysis, it was predicted that
+
ꢀ
only one molecule of MV was bound to CbFDH, and the MV
bound to CbFDH needs to be re-reduced by the electron source
outside of CbFDH to supply the second electron in the CO
reduction to formate. From the results of docking simulation
2
+
ꢀ
and DFT calculations, it was shown that one molecule of MV
bound to the position close to CO in the inner side of substrate
2
binding pocket of CbFDH, contributing to the two-electron CO2
reduction to formate. We propose that clarification of the
mechanism of two-electron CO
CbFDH and MV
2
reduction to formate with
ꢀ
gives guidelines to develop new CO
+
0
2
Fig. 17 Synthesis scheme for 1,5-bis(4 -methyl-4,4-bipyridin-4-yl)pentane
(MVC VM).
reduction systems using biocatalyst and effective co-enzymes.
5
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2020

View Article Online
PCCP
Paper
+
ꢀ
First, sodium dithionite powder was added into a quartz cell
For determination of the concentration of MV produced in
and then deaerated using a vacuum pump and then flushed the cathode side, a UV-vis spectrum was obtained with the
4
ꢁ1
ꢁ1
with CO gas. A solution containing MV or MVC VM in 5.0 mL molar extinction coefficient at 605 nm (e = 1.3 ꢃ10 M cm
)
2
5
of 50 mM sodium pyrophosphate buffer (pH 7.4) was deaerated using a spectrophotometer (SEC200 BAS Inc.). The cell volumes
by freeze–pump–thaw cycles repeated 6 times and then flushed of the chambers of anode and cathode sides were 18 and
+
ꢀ
+
ꢀ
+
ꢀ
with CO
solution, 3.0 mL of MV or MVC
a quartz cell and reacted with sodium dithionite (160 mM) under
continuously flowing CO
gas. The reaction was started by For preparing the molecular model of CbFDH for docking
injection of CbFDH solution (7.5 mM) to MV or MV C
in 3.0 mL of 50 mM CO -saturated sodium pyrophosphate buffer the Protein Data Bank (PDB) database (PDB ID: 5DN9). The
2
gas for 5 min. To prepare MV or MV C
5
VM 5.0 mL, respectively.
5
VM solution was added into
Docking simulation and DFT calculations
2
+
ꢀ
+
ꢀ
+
ꢀ
VM simulations, the crystal structure of CbFDH was obtained from
5
5
5
2
+
(
pH = 7.4). The amount of formate was determined using protein structure of 5DN9 contains NAD and N
as substrates.
3
ion chromatography (Dionex ICS-1100; electrical conductivity For constructing a CbFDH model used for docking simulation,
+
detector) with an ion exclusion column (Thermo ICE AS1; column NAD and water molecules were removed, and N
was replaced
3
+
length: 9 ꢃ150 mm; composed of a 7.5 mm cross-linked styrene/ with CO
2
because the CbFDH–NAD –N
3
complex has been
divinylbenzene resin with functionalized sulfonate groups). reported as the transition-state analogue of the formate dehydro-
56
1
.0 mM of octane sulfonic acid and 5.0 mM of tetrabutyl- genation to CO
ammonium hydroxide were used as an eluent and a regenerant, (formate anion) and a product (CO
respectively. Lowering of pH by addition of sodium dithionite with N . Besides, for preparing the molecular model of MV, the
2
by CbFDH. In the transition model, a reactant
2
) are located at the same place
3
was suppressed by using sodium pyrophosphate buffer molecular structure of MV was drawn using GaussView 5.0. For
solution (pH 7.4), indicating that the stability of FDH was the calculations, the amino and carboxyl groups were protonated
retained in the sodium dithionite-saturated solution. The and deprotonated, respectively. The MV model was optimized
concentration of formate was measured for 1 min to estimate by DFT calculation using the oB97XD functional and the
5
7
the initial rate of the CO
2
reduction to formate.
6-311++G(d,p) basis set in Gaussian 09.
By using the molecular models, docking simulation was
performed using Autodock 4.2. As the docking site, the grid
Cyclic voltammogram measurements
58
The CV curves of MV and MV + CbFDH in CO -saturated 0.1 M size was set to 40 ꢃ40 ꢃ40 points with a grid spacing of
2
ꢁ
1
phosphate buffer (pH 6.8) at a scan rate of 10 mV s (vs. Ag/ 0.608 Å centered on the original ligand in the crystal structure
AgCl (1 M KCl)) were measured using a potentiostat (Hokuto of CbFDH. The grid box included the entire substrate binding
Denko HZ-3000). Grassy carbon and platinum wire electrodes pocket of CbFDH and provided enough space for the ligand
were used as the working and counter electrodes, respectively. translational and rotational walk. To search molecular
Ag/AgCl electrode was used as the reference electrode.
dockings, the Lamarckian genetic algorithm and the pseudo-
Solis and Wets methods were applied for minimization using
default parameters. The standard docking protocol for rigid
and flexible ligand docking was performed. In the binding of
2
CO reduction to formate with electrochemically reduced MV
and CbFDH
The CO reduction to formate with electrochemically reduced MV to CbFDH, some amino acid residues composed of the
2
MV and CbFDH was carried out as follows. In constant substrate binding pocket of CbFDH possibly move and fit MV.
potential CO
2
reduction to formate using CbFDH and MV, an Thus, V93, N119, V123, R174, I175, T256, R258, H311, S313, and
H-type electrochemical cell was used. The cell volume is 18 mL. Y358 of CbFDH, which are oriented on the surface of the
Platinum wire, carbon fabric paper (CFP) (electrode area: substrate binding pocket, were selected as the flexible amino
2
3
.75 cm ) and Ag/AgCl electrodes were used as the anode, cathode acid residues. The molecular models were subjected to 200
and reference electrodes, respectively. The CO
-saturated phos- independent runs per ligand, and other parameters were set to
2
phate buffer solution (pH 6.8) and the CO -saturated phosphate the default values.
2
ꢀ
The ETE of CbFDH-bound forms of MV simulated using
buffer solution (pH 6.8) containing MV (100 mM) and CbFDH
7.5 mM) were used as the anolyte and catholyte, respectively. Autodock was calculated using the following equation.
(
The volumes of the anolyte and catholyte were 10 mL, respec-
ꢁ
ꢀ
ETE = H(MV) + H(e ) ꢁH(MV )
tively. The chambers of the anode and cathode sides were
s
ꢀ
connected via Nafion 115. Photocurrent was measured with where H(MV) and H(MV ) are the total enthalpies of MV and
ꢀ
an electrochemical analyzer (ALS1200C BAS Inc.). The amount MV at 298 K, respectively. These enthalpies were estimated
of formate was detected using ion chromatography (Dionex from the results of the single-point calculation using the DFT
ICS-1100; electrical conductivity detector) with an ion exclusion method described above. In the calculations, the solvation
column (Thermo ICE AS1; column length: 9 ꢃ 150 mm; effects were taken into account using the conductor-like polar-
composed of a 7.5 mm cross-linked styrene/divinylbenzene izable continuum model (CPCM). As solvents, diethyl ether and
resin with functionalized sulfonate groups). 1.0 mM octane water (relative permittivity = 4.3 and 78.4, respectively) were
sulfonic acid and 5.0 mM tetrabutylammonium hydroxide were chosen as the models of the hydrophobic substrate binding
used as an eluent and a regenerant, respectively.
pocket and aqueous medium, respectively. The enthalpies were
This journal is © the Owner Societies 2020
Phys. Chem. Chem. Phys.

View Article Online
Paper
PCCP
estimated by the approximation of the molecules in the gas 20 I. Willner, I. Willner and N. Lapidot, J. Am. Chem. Soc., 1990,
phase, and contained three-dimensional translation energy,
112, 6438.
three directional rotation energy, and a PV term as the ideal 21 R. Miyatani and Y. Amao, Biotechnol. Lett., 2002, 24, 1931.
gas. These values are not the same as those inside the protein, 22 R. Miyatani and Y. Amao, J. Mol. Catal. B: Enzym., 2004,
but would be exactly canceled out because H(MV) and H(MV*)
27, 121.
ꢁ
contain the same values. H(e ), the enthalpy of electrons, was 23 R. Miyatani and Y. Amao, J. Jpn. Pet. Inst., 2004, 47, 27.
5
9
taken from the literature.
24 Y. Amao, R. Abe and S. Shiotani, J. Photochem. Photobiol., A,
015, 313, 149.
5 I. Tsujisho, M. Toyoda and Y. Amao, Catal. Commun., 2006,
, 173.
6 M. Kodaka and Y. Kubota, J. Chem. Soc., Perkin Trans. 2,
999, 891.
2
2
2
2
2
2
Conflicts of interest
7
There are no conflicts to declare.
1
7 S. Ikeyama and Y. Amao, Sustainable Energy Fuels, 2017,
1, 1730.
8 S. Ikeyama and Y. Amao, Photochem. Photobiol. Sci., 2018,
Acknowledgements
This work was partially supported by a Grant-in-Aid for Scientific
Research on Innovative Areas ‘‘Artificial Photosynthesis (2406)’’,
1
7, 60.
9 R. Miyatani and Y. Amao, Photochem. Photobiol. Sci., 2014,
, 681.
‘‘Innovations for Light-Energy Conversion (4906)’’, the Fund for the
3
Promotion of Joint International Research (Fostering Joint Interna-
tional Research (B))(19KK0144), and the joint usage/research pro-
gram of the Artificial Photosynthesis, Osaka City University.
3
3
3
3
3
3
0 Y. Amao and T. Watanabe, Chem. Lett., 2004, 33, 1544.
1 Y. Amao and T. Watanabe, Appl. Catal., B, 2009, 86, 109.
2 Y. Amao and R. Kataoka, Catal. Today, 2018, 307, 243.
3 Y. Amao and S. Ikeyama, Chem. Lett., 2015, 44, 1182.
4 S. Ikeyama and Y. Amao, Chem. Lett., 2016, 45, 1259.
5 S. Ikeyama and Y. Amao, ChemCatChem, 2017, 9, 833.
Notes and references
1
2
V. I. Tishkov and V. O. Popov, Biochemistry, 2004, 69, 1252. 36 S. Ikeyama and Y. Amao, New J. Chem., 2018, 42, 15556.
R. Wichmann, C. Wandrey, A. F. Buckmann and M. R. Kula, 37 S. Ikeyama, R. Abe, S. Shiotani and Y. Amao, Chem. Lett.,
Biotechnol. Bioeng., 1981, 23, 2789.
2016, 45, 907.
3
4
W. Hummel and M. R. Kula, Eur. J. Biochem., 1989, 184, 1. 38 S. Ikeyama, R. Abe, S. Shiotani and Y. Amao, Bull. Chem. Soc.
H. Zhao and W. A. van der Donk, Curr. Opin. Biotechnol,
003, 14, 583.
J. G. Ferry, FEMS Microbiol. Rev., 1990, 7, 377.
Y. Amao, J. CO2 Util., 2018, 26, 623.
Jpn., 2018, 91, 1369.
39 P. M. S. Monk, The viologens: physicochemical properties,
2
0
5
6
7
synthesis, and applications of the salts of 4,4 -bipyridine,
Wiley, Hoboken, 1998.
H. Choe, J. C. Joo, D. H. Cho, M. H. Kim, S. H. Lee, 40 D. Small, V. Zaitsev, Y. Jung, S. V. Rosokha, M. Head-Gordon
K. D. Jung and Y. H. Kim, PLoS One, 2014, 9, e103111.
and J. K. Kochi, J. Am. Chem. Soc., 2004, 126, 13850.
J. H. Kim, D. H. Nam and C. B. Park, Curr. Opin. Biotechnol., 41 S. Suzuki, Y. Morita, K. Fukui, K. Sato, D. Shiomi, T. Takui
014, 28, 1.
and K. Nakasuji, J. Am. Chem. Soc., 2006, 128, 2530.
S. K. Kuk, R. K. Singh, D. H. Nam, R. Singh, J. K. Lee and 42 J. L. Zafra, L. Qiu, N. Yanai, T. Mori, M. Nakano, M. P.
8
9
2
C. B. Park, Angew. Chem., Int. Ed., 2017, 56, 3827.
0 H. Wu, C. Tian, X. Song, C. Liu, D. Yang and Z. Jiang, Green
Chem., 2013, 15, 1773.
Alvarez, J. T. L. Navarrete, C. J. Glmez-Garcia, M. Kertesz,
K. Takimiya and J. Casado, Angew. Chem., Int. Ed., 2016,
55, 14563.
1
1
1 R. K. Yadav, J.-O. Baeg, G. H. Oh, N.-J. Park, K.-J. Kong, 43 Z. Mou, K. Uchida, T. Kubo and M. Kertesz, J. Am. Chem.
J. Kim, D. W. Hwang and S. K. Biswas, J. Am. Chem. Soc.,
012, 134, 11455.
2 X. Ji, Z. Su, P. Wang, G. Ma and S. Zhang, ACS Nano, 2015,
, 4600.
3 R. K. Yadav, G. H. Oh, N.-J. Park, A. Kumar, K.-J. Kong and
J.-O. Baeg, J. Am. Chem. Soc., 2014, 136, 16728.
4 K. J. Shah and T. Imae, J. Mater. Chem. A, 2017, 5, 9691.
5 Y. Amao, Chem. Lett., 2017, 46, 780.
Soc., 2014, 136, 18009.
2
44 J. J. Novoa, P. Lafuente, R. E. Del Sesto and J. S. Miller,
Angew. Chem., Int. Ed., 2001, 40, 2540.
45 J. Jakowski and J. Simons, J. Am. Chem. Soc., 2003,
125, 16089.
46 J.-F. Penneau and L. L. Miller, Angew. Chem., Int. Ed. Engl.,
1991, 30, 986.
1
1
9
1
1
1
1
47 J.-M. Lg, S. V. Rosokha and J. K. Kochi, J. Am. Chem. Soc.,
2003, 125, 12161.
6 Y. Amao, Sustainable Energy Fuels, 2018, 2, 1928.
7 D. Mandler and I. Willner, J. Chem. Soc., Perkin Trans. 2, 48 M. R. Geraskina, A. S. Dutton, M. J. Juetten, S. A. Wood and
988, 997.
A. H. Winter, Angew. Chem., Int. Ed., 2017, 56, 9435.
8 I. Willner and D. Mandler, J. Am. Chem. Soc., 1989, 49 T. Sagara and H. Eguchi, Electrochim. Acta, 2019, 295, 215.
1
1
1
111, 1330.
50 K. Sakai, B.-C. Hsieh, A. Maruyama, Y. Kitazumi, O. Shirai
and K. Kano, Sens. Biosens. Res., 2015, 5, 90.
9 I. Willner, N. Lapidot, A. Riklin, R. Kasher, E. Zahavy and
E. Katz, J. Am. Chem. Soc., 1994, 116, 1428.
51 T. Watanabe and K. Honda, J. Phys. Chem., 1982, 86, 2617.
Phys. Chem. Chem. Phys.
This journal is © the Owner Societies 2020

View Article Online
PCCP
Paper
5
2 T. Ishibashi, M. Higashi, S. Ikeda and Y. Amao, ChemCatChem,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers,
K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand,
K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar,
J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene,
J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin,
R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador,
J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas,
J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox,
Gaussian 09, Revision A.1, Gaussian, Inc., Wallingford CT,
2009.
2019, 11, 6227.
53 R. Sato and Y. Amao, New J. Chem., 2020, 44, 11922.
54 N. L. Allinger, J. Am. Chem. Soc., 1977, 99, 8127.
55 Q. Guo, L. Gakhar, K. Wickersham, K. Francis,
A. Vardi-Kilshtain, D. T. Major, C. M. Cheatum and
A. Kohen, Biochemistry, 2016, 55, 2760.
5
6 J. N. Bandaria, C. M. Cheatum and A. Kohen, J. Am. Chem.
Soc., 2009, 131, 10151.
5
7 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, 58 G. M. Morris, R. Huey, W. Lindstrom, M. F. Sanner,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
R. K. Belew, D. S. Goodsell and A. J. Olson, J. Comput. Chem.,
2009, 16, 2785.
H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta, 59 J. E. Bartmess, J. Phys. Chem., 1993, 98, 5648.
This journal is © the Owner Societies 2020
Phys. Chem. Chem. Phys.