Cobalt hexaamine mediated electrocatalytic voltammetry of dimethyl sulfoxide reductase: Driving force effects on catalysis

Article abstract of DOI:10.1007/s00775-010-0719-x

The bacterial molybdoenzyme dimethyl sulfoxide (DMSO) reductase from Rhodobacter capsulatus catalyzes the reduction of DMSO to dimethyl sulfide in anaerobic respiration. In its native state, DMSO reductase is reduced to its active state by a pentaheme cytochrome (DorC). Alternatively, we show that DMSO reductase catalysis may be driven electrochemically using a series of homologous coordination compounds as mediating synthetic electron donors. All mediators are macrocyclic hexaaminecobalt(II) complexes in their active form, differing principally in their redox potentials over a range of about 250 mV. Thus, each complex presents a different reductive driving force to DMSO reductase and this leads to pronounced differences in the electrocatalytic behavior as measured by cyclic voltammetry. Digital simulation of the experimental voltammetry enables the critical features of the catalytic cycle to be extracted.

Full text of DOI:10.1007/s00775-010-0719-x

J Biol Inorg Chem (2011) 16:227–234
DOI 10.1007/s00775-010-0719-x
ORIGINAL PAPER
Cobalt hexaamine mediated electrocatalytic voltammetry
of dimethyl sulfoxide reductase: driving force effects on catalysis
Kuan-I. Chen
Paul V. Bernhardt
•
Alastair G. McEwan
•
Received: 2 September 2010 / Accepted: 11 October 2010 / Published online: 27 October 2010
Ó SBIC 2010
Abstract The bacterial molybdoenzyme dimethyl sulf-
oxide (DMSO) reductase from Rhodobacter capsulatus
catalyzes the reduction of DMSO to dimethyl sulfide in
anaerobic respiration. In its native state, DMSO reductase
is reduced to its active state by a pentaheme cytochrome
DMSO
MV
Dimethyl sulfoxide
Methyl viologen
NHE
sep
Normal hydrogen electrode
Sepulchrate
(
DorC). Alternatively, we show that DMSO reductase
catalysis may be driven electrochemically using a series of
homologous coordination compounds as mediating syn-
thetic electron donors. All mediators are macrocyclic
hexaaminecobalt(II) complexes in their active form, dif-
fering principally in their redox potentials over a range of
about 250 mV. Thus, each complex presents a different
reductive driving force to DMSO reductase and this leads
to pronounced differences in the electrocatalytic behavior
as measured by cyclic voltammetry. Digital simulation of
the experimental voltammetry enables the critical features
of the catalytic cycle to be extracted.
Introduction
Electrochemically mediated enzyme catalysis is a versatile
and potentially useful way of exploiting the selectivity and
sensitivity of an oxidoreductase enzyme in detecting low
concentrations of redox-active analytes present in complex
mixtures containing other potentially interfering species
[1]. There are already commercially developed examples
of such enzyme biosensors, the most celebrated being the
glucose oxidase biosensor [2, 3], which utilizes synthetic
electron acceptors such as ferrocenium salts or organic
quinones to shuttle electrons generated by enzymatic glu-
cose oxidation to a working electrode for amperometric
detection.
Keywords Molybdenum ꢀ Enzyme ꢀ Voltammetry
Abbreviations
AMMEsar Aminomethylsarcophagine
One of the criteria for a successful mediator is that it
must not irreversibly modify the enzyme nor inhibit its
catalytic mechanism. As the mechanism involves outer-
sphere electron transfer between the mediator and the
enzyme, the respective redox potentials of the two species
will have a bearing on both the rate and the equilibrium
position of the reaction. In the case of reductase enzymes,
one can, in principle, employ mediators with much lower
redox potentials than the active site to ensure complete and
rapid electron transfer. However, an excessive driving
force makes the system prone to interference from other
species that may be reduced either directly at the electrode
or by the mediator (but typically not by the enzyme).
diammac
DMS
6,13-Dimethyl-1,4,8,11-
tetraazacyclotetradecane-6,13-diamine
Dimethyl sulfide
Electronic supplementary material The online version of this
article (doi:10.1007/s00775-010-0719-x) contains supplementary
material, which is available to authorized users.
K.-I. Chen ꢀ A. G. McEwan ꢀ P. V. Bernhardt (&)
School of Chemistry and Molecular Biosciences,
Centre for Metals in Biology,
University of Queensland,
Brisbane, QLD 4072, Australia
e-mail: p.bernhardt@uq.edu.au
123

2
28
J Biol Inorg Chem (2011) 16:227–234
One potential interferant is dioxygen if the biosensor is not
operated in an inert atmosphere.
employing a series of homologous mediators that vary
III/II
principally in their Co
redox potentials while retaining
The molybdoenzyme dimethyl sulfoxide (DMSO)
reductase (Rhodobacter capsulatus) reduces DMSO as the
terminal electron acceptor during anaerobic respiration of
the bacterium [4]. The enzyme bears a single redox-active
cofactor, the Mo active site (Fig. 1) [5–7], which cycles
the same structural, molecular weight, and charge proper-
ties. The four complexes illustrated in Fig. 2 (including the
3
previously investigated [Co(trans-diammac)] ) fit these
?
III/II
criteria and their Co
redox potentials span a range of
about 250 mV. In separate investigations, these complexes
have been shown to be effective mediators to a number of
proteins in potentiometric redox titrations [13] and to date
we have found them to be benign compounds that do not
inhibit, denature, or in any way destabilize or deactivate any
protein we have studied.
IV
VI
between its Mo and Mo oxidation states during catalysis,
a typical feature of most molybdoenzymes [8]. As illustrated
in Fig. 1, the O atom of DMSO is transferred to the Mo ion
during the reaction, releasing dimethyl sulfide (DMS) as
the product and resulting in a two-electron oxidation of
the metal. To restore the enzyme to its active form, a
two-electron reduction of the enzyme must occur and in
nature this is achieved by the pentaheme cytochrome DorC
Materials and methods
[
9].
To effect the same catalytic reaction electrochemically
we have employed the synthetic coordination compound
Materials
2
?
[Co(trans-diammac)] (generated electrochemically from
its resting trivalent oxidation state; diammac is 6,13-dime-
[Co(trans-diammac)](ClO4)3 [11], [Co(cis-diammac)]
(ClO ) [14], [Co(AMMEsar)]Cl ꢀHCl (where AMMEsar is
4
3
3
thyl-1,4,8,11-tetraazacyclotetradecane-6,13-diamine) as the
III/II
aminomethylsarcophagine) [15], and [Co(sep)]Cl (where
3
mediator [10]. The Co
3
redox potential of [Co(trans-di-
[-555 mV vs. the normal hydrogen electrode
sep is sepulchrate) [16] were synthesized from published
procedures. DMSO reductase was purified from R. capsul-
atus strain H123, as previously described [17]. All other
reagents were obtained commercially and were of analytical
grade purity.
?/2?
ammac)]
VI/V
(
NHE)] [11] lies well below that of the enzyme (Mo
V/IV
?
83 mV and Mo
?30 mV vs. NHE at pH 8) [12]. Our
aim here was to systematically lower this driving force by
Fig. 1 The catalytic cycle of
dimethyl sulfoxide (DMSO)
reductase (active site shown).
DMS dimethyl sulfide
DMSO
H O
2
Me
e-, 2H+
Me
S
2
O-Ser
O
MoIV
O
O-Ser
O
S
O
MoVI
NH
N
O
O
S
HN
N
S
HN
NH
N
NH2
S
HN
S
HN
NH
NH
S
NH2
S
NH
N
NH
S
H N
2
NH
O
O
H N
2
NH
O
O
DMS
3+
Fig. 2 Co complexes used as
mediators in this work and their
NH2
3+
3+
N
3+
III/II
Co
redox potentials (vs. the
H N
2
HN NH2NH
Co
normal hydrogen electrode,
NHE)
NH
Co
NH
NH
NH
Co
HN
HN
HN
HN
HN
HN
NH
NH
Co
HN
NH
HN NH
HN
NH
H N
2
NH2
N
[
Co(trans-diammac)]3+
555 mV
[Co(sep)]3+
[Co(AMMEsar)]3+
380 mV
[
Co(cis-diammac)]3+
-
-296 mV
-
-437 mV
1
23

J Biol Inorg Chem (2011) 16:227–234
Electrochemistry
229
partner the pentaheme cytochrome DorC (Em,7 in the range
-
34 to -276 mV vs. NHE) [9], the methyl viologen
?
ꢀ
2?/?ꢀ
o
Electrochemical measurements were carried out with a
BAS100B/W workstation. A three-electrode system was
employed comprising a glassy carbon disk working elec-
(MV) radical monocation (MV ) (MV
vs. NHE) [19–21], and the coordination compound
E -440 mV
2
?
III/II
o
E -555 mV vs. NHE,
[Co(trans-diammac)]
(Co
2
trode (0.052 cm ), a Pt wire counter electrode, and a
produced electrochemically) [10]. Given the disparate
chemical forms of these three electron donors, the rates at
which they react with DMSO reductase will depend not
only on their redox potentials but also on the site where
each molecule docks with the enzyme and the distance
between the two reacting centers.
Ag/AgCl reference electrode (?196 mV vs. the NHE). The
working electrode was always polished with diamond paste
and rinsed with distilled water before measurements. The
electrochemical cell (volume about 1 mL) contained vari-
III
ous concentrations of DMSO reductase, DMSO, and Co
complex as described in the text and the supporting elec-
trolyte was 50 mM tris(hydroxymethyl)aminomethane–
HCl buffer (pH 8.0). All experiments were performed
under an Ar atmosphere.
The overall catalytic cycle for the present electro-
chemically driven system is illustrated in Scheme 1. As
shown, the cycle may be separated into two parts: one that
involves only the DMSO reductase–DMSO reaction
comprising DMSO binding, turnover, and DMS release
(k /k , k /k , and k /k ) and the other that only
Digital simulation
1
-1
2
-2
3 -3
II
involves the DMSO reductase–Co reaction (k /k and
4
-4
Experimental cyclic voltammograms were simulated with
the program DigiSim (version 3.03b) [18]. The experi-
mental parameters (redox potentials, homogeneous and
heterogeneous rate constants, diffusion coefficients) were
obtained as described in [10] by systematic variations of
the enzyme, mediator, and substrate concentrations and the
k4 /k-4 ). Ideally, the rate coefficients k /k , k /k-2, and
0 0
1 -1 2
k3/k-3 should be mediator-independent and thus it was our
aim to fine-tune electrochemically mediated DMSO
II
reductase catalysis by only varying the Co –DMSO
reductase electron transfer rates (k /k and k4 /k-4 ). The
0 0
4
-4
structural homology of the mediators in Fig. 2 was
intended to avoid any differences in charge, molecular
weight, or H-bonding properties that might lead to
divergent interactions between the enzyme and the
mediator and hence the only remaining differences across
the series were the redox potentials.
III/II
o
sweep rate. The Co
redox potentials (E ), diffusion
coefficients (D ), and standard heterogeneous rate con-
Co
-
stants (k = 0.01 cm s in all cases) were obtained by
1
0
simulation of the voltammetry in the absence of DMSO.
Sweep rate variations at high DMSO concentrations yiel-
II
ded accurate values of the Co –DMSO reductase cross-
reaction (k ). Voltammetry at low DMSO concentrations
4
The catalytic waveform
was very sensitive to the DMSO–DMSO reductase binding
(
on) rate constant (k ), whereas at high DMSO concentra-
1
The Co complexes in Fig. 2 all exhibit reversible one-
III/II
tions, the turnover number (k ) was the limiting factor. All
2
electron Co
couples. An example of this is shown for
other parameters were as described previously [10] unless
otherwise stated. All mediators afford totally reversible
single-electron redox processes in the absence of enzyme
or DMSO [13]. All redox potentials are cited versus NHE
by addition of 196 mV to the potentials measured relative
to the Ag/AgCl reference electrode.
DMS
k-3
k3
CoII
k4
khet
e
MoVI
-
VI
Mo :DMS
CoIII
Results and discussion
MoV
k2 k-2
electrode
Mechanism
IV
Mo :DMSO
CoII
MoIV
k4'
DMSO reductase (R. capsulatus) is activated upon reduc-
VI/V
and
IV
tion to its Mo oxidation state (Fig. 1). The Mo
CoIII
k1
khet
k-1
V/IV
-
e
Mo
redox potentials have been determined by EPR
DMSO
redox potentiometry to be ?83 and ?30 mV vs. NHE at
pH 8, respectively [12]. We and others have shown that
reduction of the enzyme may be achieved by a number of
different electron donors, including its natural electron
mediator independent mediator dependent
Scheme 1 DMS dimethyl sulfide
123

2
30
J Biol Inorg Chem (2011) 16:227–234
0
0
.2
.0
(enzyme, mediator, and substrate) in the reaction layer is
time-dependent. These cases are discussed later.
The relationship between the observed catalytic current,
the applied potential (E), and the bulk concentrations of Co
complex, enzyme, and DMSO is complicated (Eq. 1a)
-
-
-
-
-
0.2
0.4
0.6
0.8
[
23, 24]. The last factor in the equation is potential-
dependent and defines a sigmoidal function but this also
contains a term (r) that is dependent on the concentration
of the Co complex and the rates of the DMSO reductase–
DMSO reaction. At high overpotential (n ? ?) the cur-
rent reaches a (potential-independent) plateau ilim. The
magnitude of the limiting catalytic current in Fig. 3
increases with the square root of the enzyme concentration
as predicted by Eq. 1b, this value only appearing once in
the equation.
no enzyme
200 nM enzyme
5
1
00 nM enzyme
μM enzyme
1.0
-
800 -700 -600 -500 -400 -300 -200 -100
E (mV vs NHE)
0
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
3?
iðEÞ ¼ ꢁFA½Coꢂ 2k D ½DMSO reductaseꢂ
4
Co
Fig. 3 Cyclic voltammetry of [Co(cis-diammac)] (100 lM) in the
presence of 1 mM DMSO and in increasing concentrations of DMSO
reductase (0–1 lM). Sweep rate 2 mV s , pH 8
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ꢀ
ꢁ
u 0
1
-1
u
r
ln 1 þ
u2
1
1þexpðꢁnÞ
r
t @
A; ð1aÞ
ꢃ
ꢁ
r 1 þ expðꢁnÞ
3
?
Co(cis-diammac)] (Fig. 3, uppermost curve), where a
[
classic transient (peak-shaped) cyclic voltammogram is
seen. The presence of 1 mM DMSO has no effect on the
electrochemistry, i.e., DMSO is electroinactive at these
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ilim ¼ ꢁFA½Coꢂ 2k D ½DMSO reductaseꢂ
4
Co
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ꢂ
ꢃ
2
lnð1 þ rÞ
2
?
potentials and does not react with [Co(cis-diammac)] in
solution.
ꢃ
1 ꢁ
;
ð1bÞ
r
r
ꢂ
ꢃ
As the concentration of DMSO reductase increases, two
significant changes are observed. Firstly, the cathodic cur-
rent is amplified owing to a catalytic electrochemical reac-
1
KM
r ¼
n ¼
þ
k4½Coꢂ;
ð1cÞ
ð1dÞ
k2 k2½DMSOꢂ
3
?
F
o
tion being established, i.e., [Co(cis-diammac)] is recycled
at the electrode surface by reaction of its divalent form with
oxidized DMSO reductase (generated by DMSO turnover).
Secondly, the waveform is transformed from a peak-shaped
transient voltammogram to a sigmoidal steady-state vol-
tammogram [1]. The gradual transition is apparent in Fig. 3
as the enzyme concentration increases. In each case the
potential at which the wave is seen is the same and charac-
teristic of the mediator.
ðE ꢁ EÞ:
RT
However, the appearance of the bulk concentration of Co
complex in different parts of Eq. 1b leads to a nonlinear
2
dependence of ilim on Co concentration. This is illustrated
in
Fig. 4
for
increasing
concentrations
of
3
?
[Co(AMMEsar)] . In the absence of any mediator
(Fig. 4, uppermost curve) no current is observed as the
enzyme alone is electroinactive at a glassy carbon electrode.
Direct (unmediated) electrochemistry of DMSO reductase
only occurs at specially prepared pyrolytic graphite
When the enzyme and substrate concentrations are high
(
e.g.,1 lM enzyme and 1 mM DMSO, lowermost voltam-
3
?
mogram in Fig. 3), there is ample supply of oxidized
enzyme to sustain a steady state of Co complex in the
reaction layer. The sigmoidal waveform matches that of a
electrodes [25]. Addition of [Co(AMMEsar)] to the cell
leads to an immediate sigmoidal steady-state current which
rises with each subsequent addition of mediator. In this case
the half-wave potential coincides with that of the
1
reversible steady-state voltammogram, i.e., the current at
3
?/2?
any point on the voltammogram is dependent only on the
applied potential and the diffusion layer thickness is con-
stant throughout the experiment [22]. More complicated
voltammetry emerges at lower concentrations of enzyme
[Co(AMMEsar)]
couple.
(
e.g., 200 nM enzyme in Fig. 3), where the concentration
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
of more than one component of the ternary complex
ꢀ
ꢁ
2
2
r
lnð1þrÞ
The
1 ꢁ
trations and only then is the current linearly related to Co
term in Eq. 1b vanishes at low Co concen-
r
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1
RT
ilimꢁi
E ¼ E
1
þ nF ln
:
2
i
concentration, i.e., ilim ¼ ꢁFA½Coꢂ 2k4DCo½DMSO reductaseꢂ:
1
23

J Biol Inorg Chem (2011) 16:227–234
231
independent of DMSO concentration). A more quantitative
analysis of this is given later.
0.0
The Michaelis constant (K ), an intrinsic property of
M
the enzyme, typically places an upper limit on the sensi-
tivity of the system to substrate concentration. Michaelis–
Menten kinetics predicts that the reaction velocity will be
close to saturation (about 90% Vmax) at a substrate con-
centration of 10K . The actual K value of DMSO
-
-
-
-
0.2
0.4
0.6
0.8
M
M
reductase at pH 8 determined from enzyme assays is
no Co
10 lM [26], so the fact that the observed electrochemical
2
4
6
8
0 μM Co
0 μM Co
0 μM Co
0 μM Co
current in Fig. 5d is still sensitive to DMSO concentra-
tions exceeding 30KM illustrates that the dynamic range of
this ternary system is not limited in this case by the
intrinsic characteristics of the enzyme. To quantitatively
probe the differences apparent in Fig. 5, we turned to
electrochemical simulation of the voltammograms under
various experimental conditions.
-
600
-500
-400
E (mV vs NHE)
-300
-200
-100
Fig. 4 Cyclic voltammetry of DMSO reductase (1 lM) and DMSO
1 mM) in the presence of increasing concentrations of [Co(AMME-
sar)] (0–80 lM). Sweep rate 2 mV s , pH 8
(
3?
-1
Electrochemical simulation
Driving force effects on catalysis: qualitative features
Sweep rate dependence
Altering the potential of the mediator has a profound effect
on the catalytic behavior of the system. Cyclic voltam-
mograms of the four mediators in Fig. 2 in the presence of
similar concentrations of DMSO reductase (10 lM) and
DMSO are shown in Fig. 5. The previously published [10]
The voltammetric sweep rate is a powerful variable in
elucidating the mechanisms of electrochemical processes
coupled to chemical reactions. It is apparent from
Scheme 1 that at high sweep rates, heterogeneous reduc-
III
tion and reoxidation of the Co mediator at the electrode
II
(k ) will be too fast to enable the Co complex to par-
3
?
results from the [Co(trans-diammac)] system are inclu-
ded for comparison.
het
ticipate in homogeneous electron transfer (k , k ) with the
0
4 4
3
?
The most obvious outlier is [Co(sep)] (Fig. 5a), for
which no significant change in the waveform is seen upon
enzyme, even when the substrate is present in high con-
centrations. This is illustrated for the case of DMSO
reductase catalysis mediated by electrochemically reduced
addition of DMSO to the solution. This mediator, bearing
III/II
2?
the highest Co
redox potential of the series, is evidently
[Co(cis-diammac)] (Fig. 6) in the presence of 1.4 mM
DMSO.
an ineffective electron donor to DMSO reductase. The
other three mediators all lead to steady-state voltammetry
but the DMSO concentrations at which the voltammograms
reach a steady state are clearly different. For [Co(AM-
-
1
At slow sweep rates (below 10 mV s ) an electro-
chemical steady state is established characterized by a
sigmoidal voltammogram [22, 24] and sweep-rate-inde-
3
?
-1
MEsar)] the voltammograms at DMSO concentrations of
pendent current. The 1 and 2 mV s voltammograms are
2
8 lM and above are identical (Fig. 5b). For the [Co(cis-
diammac)] system (Fig. 5c), a steady state is achieved at
essentially the same in both shape and amplitude, i.e.,
III
3
?
Co is reduced at the same rate at which it is reoxidized by
-1
the enzyme. At sweep rates of 10 mV s and above, a
3
?
about 150 lM DMSO, whereas for [Co(trans-diammac)]
Fig. 5d) the voltammograms are still changing up until
00 lM DMSO. The reasons for discontinuities in the
(
transient (peak-shaped) voltammogram emerges and the
current increases proportional to the square root of the
sweep rate (Fig. S1). In this region no significant mediation
of the enzyme reaction is occurring.
4
voltammograms at low DMSO concentrations have been
discussed before [10]. Briefly, when the rate at which
DMSO is consumed by the enzyme exceeds the rate at
which it can diffuse to the enzyme active site, the elec-
trochemical steady state breaks down and the voltammetry
abruptly reverts to that of the mediator alone. The position
at which this occurs is highlighted by vertical arrows in
Fig. 5d. It is worth noting that up until these points, all of
the voltammograms can be overlaid exactly (they are
Simulation of the voltammograms at the slower sweep
rates (Fig. 6, broken lines) led to accurate values for the
II
Co –DMSO reductase cross-reactions: k /k and k4
III/II
redox potentials
0
/k-4
0
4
-4
(Table 1). Given the fact that the Co
VI/V
V/IV
are much lower than the Mo
and Mo
potentials, the
II
Co –DMSO reductase reaction is essentially irreversible
(k » k-4) and the value of k-4 had no bearing on the
4
123

2
32
J Biol Inorg Chem (2011) 16:227–234
Fig. 5 Cyclic voltammograms
of the four mediators:
A
2
1
0
1
2
B
3?
0.0
0.2
0.4
0.6
0.8
a [Co(sep)] (1 mM),
3?
b [Co(AMMEsar)] 3?(20 lM),
c [Co(cis-diammac)]
-
-
-
-
(
60 lM), and d [Co(trans-
-
-
3?
diammac)] (60 lM) in the
presence of 1.87 lM DMSO
reductase and various
concentrations of DMSO.
-3
no DMSO
-1
Sweep rate 2 mV s , pH 8.0.
The arrows in d indicate the
points at which the voltammetry
diverges from steady-state
behavior at each concentration
of DMSO
7 μM DMSO
14 μM DMSO
28 μM DMSO
42 μM DMSO
56 μM DMSO
-
-
4
5
no DMSO
0 μM DMSO
7
168 μM DMSO
2
66 μM DMSO
-6
-800 -600 -400 -200
E (mV vs NHE)
0
200
-600 -500 -400 -300 -200 -100
E (mV vs NHE)
C
D
0
.0
0
.0
0.5
1.0
1.5
-
-
0.5
1.0
-
-
-
no DMSO
-1.5
-2.0
-2.5
2
4
8
1
2
1 μM DMSO
2 μM DMSO
4 μM DMSO
54 μM DMSO
24 μM DMSO
no DMSO
28 μM DMSO
84 μM DMSO
154 μM DMSO
406 μM DMSO
-700 -600 -500 -400 -300 -200 -100
-900 -800 -700 -600 -500 -400 -300 -200 -100
E (mV vs NHE)
E (mV vs NHE)
Table 1 Kinetic and thermodynamic parameters used in electro-
chemical simulations
5
μA
-
-
1
1
2?
1
2
mV s
mV s
[Co(trans-
diammac)]
[Co(cis-
diammac)]
[Co(AMMEsar)]
2?
2?
o
E (mV vs. NHE) -555
-437
-380
-
1
-1
6
5
4
k
k
k
k
4
(M
s
1
)
-1
2.0 9 10
3.2 9 10
2.0 9 10
3.2 9 10
1.0 9 10
9.0 9 10
-
-5
6
-7
-7
1.5 9 10
-4 (M
s
)
1.7 9 10
1.0 9 10
1.7 9 10
-
-
1
1
-
1
-1
5
4
1
2
0 mV s
0 mV s
0
(M
4
s
1
)
-1
9.0 9 10
-
-3
-7
-7
1.5 9 10
0
(M
-
s
)
-4
1
-1
)
5
k1 (M
s
5 9 10
-
-1 (s
1
k
k
k
k
k
)
0.8
-1
)
2
(s
60
-
-2 (s
1
)
360
-1
)
-
800
-600
-400
-200
0
3
(s
560
-
1
-1
s
5
E (mV vs NHE)
-3 (M
)
1.3 9 10
Fig. 6 Experimental (solid lines) and simulated (broken lines) cyclic
See Scheme 1 for definitions of the rate constants
3?
voltammograms of [Co(cis-diammac)] (600 lM), DMSO reductase
1.87 lM), and DMSO (1.4 mM) at different sweep rates
diammac 6,13-dimethyl-1,4,8,11-tetraazacyclotetradecane-6,13-dia-
mine, AMMEsar aminomethylsarcophagine, NHE normal hydrogen
electrode
(
VI
simulation. Also, it was found that k (reduction of Mo )
2?
approach was taken with [Co(AMMEsar)] as the electron
4
V
(reduction of Mo ) could be assigned the same
and k4
value despite the slightly different driving forces. A similar
0
donor. Simulation of the experimental voltammetry
(Fig. S2) as a function of sweep rate led to slightly smaller
1
23

J Biol Inorg Chem (2011) 16:227–234
233
concentrations, with constant concentrations of DMSO
0
.5 μA
reductase (1 lM) and DMSO (1 mM) (Fig. 8).
Analysis of thermodynamic and kinetic parameters
no DMSO
The first-order, long-range electron transfer rate (k ) within
et
II
the DMSO reductase–Co (hexaamine) outer-sphere com-
3
plex should follow Marcus theory predictions, i.e., the rate
1
2
4 μM DMSO
8 μM DMSO
2
should increase with driving force (log k µ -DG ) [27].
et
However, in this case the CoꢀꢀꢀMo separation (r), deter-
mined by the site at which the mediator docks with the
enzyme, is unknown and the reorganization energy (k) of
the reaction is also uncertain. Qualitatively the increase in
2
?
k by a factor of about 20 from [Co(AMMEsar)] to
4
2
?
Co(trans-diammac)]
increase in driving force assuming all other things being
the same.
is consistent with the 0.175 eV
-
700 -600 -500 -400 -300 -200 -100
E (mV vs NHE)
Fig. 7 Experimental (solid lines) and simulated (broken lines) cyclic
The mediator-independent parameters (Table 1) are
essentially the same as those published for [Co(trans-
3?
voltammograms of [Co(cis-diammac)] (60 lM) and DMSO reduc-
-
1
,
tase (10 lM) at low concentrations of DMSO. Sweep rate 2 mV s
2
diammac)] alone [10]. The DMSO complexation rate
?
pH 8.0. The slight offset is due to uncorrected sloping baselines of the
experimental voltammograms. The relevant simulation parameters are
in Table 1
constant (k ) was lowered slightly in this work to fit all
1
three systems. As has been mentioned, kinetic control can
switch between the various steps in the catalytic cycle
(Scheme 1), leading to the observed voltammogram being
independent of most parameters except those associated
with the rate-limiting step.
values of k /k
4
0
as expected from the somewhat smaller
4
driving force for the cross-reaction.
DMSO concentration dependence
There are ten independent homogeneous rate constants
in Scheme 1 plus one redox potential and one heteroge-
neous rate constant. All of these in principle may be refined
simultaneously but this is impractical and to obtain
meaningful fits, some parameters were adopted from pre-
viously published steady-state or pre-steady-state kinetic
studies [19, 28, 29] of DMSO reductase and were not
allowed to refine. The data in Table 1 remain consistent
with the known properties of the enzyme and indicate that
the synthetic complexes used as mediators in this work do
not alter the catalytic properties of the enzyme.
As indicated before, ideally the same set of enzyme-
dependent parameters should be able to be transferred
between different mediators and also used to model dif-
ferent limiting conditions. The dependence of the current
on DMSO concentration is complex (see Eq. 1b), but
moreover at DMSO concentrations that are similar to those
of the enzyme, the DMSO reductase–DMSO reaction
effectively becomes stoichiometric, i.e., all of the substrate
is consumed by the enzyme, culminating in an abrupt halt
in catalysis. This behavior is illustrated in Fig. 7 for
2
?
[
Co(cis-diammac)] -mediated catalysis (and also in Fig. 5
for all mediators).
Conclusions
3
?
The [Co(AMMEsar)]
system was less remarkable
and, as seen in Fig. 5b, the voltammograms became
insensitive to DMSO at very low concentrations (about
DMSO reductase may be activated electrochemically by a
series of homologous mediating Co hexaamine complexes
through outer-sphere electron transfer. The same set of
enzyme-specific parameters obtained from electrochemical
simulation may be transferred across the series, with the
only variables from one mediator to another being the
28 lM), with all subsequent voltammograms exhibiting
steady-state (sigmoidal) behavior.
Mediator and enzyme concentration dependence
II
outer-sphere electron transfer rates between the Co com-
With use of the same set of rate constants (Table 1) derived
from the sweep rate and DMSO concentration dependent
systems described above, the voltammetry at various
concentrations of mediator was also reproduced. An
plexes and the enzyme (k , Scheme 1). The trends in these
4
3
2
log ket ¼ 13 ꢁ 0:6ðr ꢁ 3:6Þ ꢁ 3:1ðDG þ k Þ=k where
r is the
10
˚
MoꢀꢀꢀCo separation (A), k is the reorganization energy (eV), and
3
?
example is shown for [Co(AMMEsar)] at three different
DG is the driving force (eV).
123

2
34
J Biol Inorg Chem (2011) 16:227–234
Fig. 8 a Cyclic voltammetry of
[
A
B
3
?
Co(AMMEsar)] at three
0
.5 μA
different concentrations, with
constant concentrations of
DMSO reductase (1 lM) and
DMSO (1 mM). b Cyclic
voltammetry of
1
μA
no enzyme
30 μM Co
60 μM Co
3
?
[
Co(AMMEsar)] (30 lM) at
2
4
00 nM enzyme
00 nM enzyme
four different concentrations of
DMSO reductase, with constant
concentration of DMSO
-
1
(
1 mM). Sweep rate 2 mV s
pH 8. The experimental
voltammograms are shown as
solid lines and the simulations
Table 1) are shown as broken
,
90 μM Co
7
00 nM enzyme
(
lines. No compensation for the
sloping baselines has been made
-600
-500
-400
-300
-200
-100
-600
-500
-400
-300
-200
-100
E (mV vs NHE)
E (mV vs NHE)
rate constants followed Marcus theory predictions. The
13. Bernhardt PV, Chen K-I, Sharpe PC (2006) J Biol Inorg Chem
11:930–936
2
?
highest potential complex studied [Co(sep)] was unable
to sustain catalysis, whereas the three remaining complexes
were capable of reducing DMSO reductase at a sufficient
rate to establish steady-state catalysis.
1
4. Bernhardt PV, Comba P, Hambley TW (1993) Inorg Chem
2:2804–2809
3
15. Geue RJ, Hambley TW, Harrowfield JM, Sargeson AM, Snow
MR (1984) J Am Chem Soc 106:5478–5488
1
1
1
1
2
2
2
2
2
2
2
2
2
2
6. Creaser II, Geue RJ, Harrowfield JM, Herlt AJ, Sargeson AM,
Snow MR, Springborg J (1982) J Am Chem Soc 104:6016–6025
7. Bennett B, Benson N, McEwan AG, Bray RC (1994) Eur J
Biochem 225:321–331
8. Rudolf M, Feldberg SW (2004) DigiSim version 3.03; Bioana-
lytical Systems, Inc., IN, USA
9. Adams B, Smith AT, Bailey S, McEwan AG, Bray RC (1999)
Biochemistry 38:8501–8511
0. Abo M, Dejima M, Asano F, Okubo A, Yamazaki S (2000)
Tetrahedron Asymmetr 11:823–828
1. Abo M, Ogasawara Y, Tanaka Y, Okubo A, Yamazaki S (2003)
Biosens Bioelectron 18:735–739
2. Bard AJ, Faulkner LR (2001) Electrochemical methods: funda-
mentals and applications. Wiley, New York
3. Bourdillon C, Demaille C, Moiroux J, Saveant J-M (1996) Acc
Chem Res 29:529–535
4. Limoges B, Moiroux J, Saveant J-M (2002) J Electroanal Chem
Acknowledgments Financial support of the project from the
Australian Research Council (DP0880288) is gratefully acknowledged.
References
1
2
3
. Bernhardt PV (2006) Aust J Chem 59:233–256
. Wang J (2008) Chem Rev 108:814–825
. Fragkou V, Turner APF (2009) Handbook of optical sensing of
glucose in biological fluids and tissues, CRC Press, Boca Raton,
FL, USA, pp 41–64
4
5
6
7
. McEwan AG, Wetzstein HG, Ferguson SJ, Jackson JB (1985)
Biochim Biophys Acta Bioenerg 806:410–417
. McAlpine AS, McEwan AG, Shaw AL, Bailey S (1997) J Biol
Inorg Chem 2:690–701
. Schneider F, Loewe J, Huber R, Schindelin H, Kisker C,
Knaeblein J (1996) J Mol Biol 263:53–69
. McAlpine AS, McEwan AG, Bailey S (1998) J Mol Biol
5
21:1–7
5. Aguey-Zinsou K-F, Bernhardt PV, McEwan AG (2002) J Biol
Inorg Chem 7:879–883
6. Ridge JP, Aguey-Zinsou K-F, Bernhardt PV, Hanson GR, McE-
wan AG (2004) FEBS Lett 563:197–202
7. Moser CC, Keske JM, Warncke K, Farid RS, Dutton PL (1992)
Nature 355:796–802
8. Bray RC, Adams B, Smith AT, Bennett B, Bailey S (2000)
Biochemistry 39:11258–11269
9. Bray RC, Adams B, Smith AT, Richards RL, Lowe DJ, Bailey S
2
. Hille R (1996) Chem Rev 96:2757–2816
. Shaw AL, Hochkoeppler A, Bonora P, Zannoni D, Hanson GR,
McEwan AG (1999) J Biol Chem 274:9911–9914
75:613–623
8
9
1
1
1
0. Chen K-I, McEwan AG, Bernhardt PV (2009) J Biol Inorg Chem
4:409–419
1. Bernhardt PV, Lawrance GA, Hambley TW (1989) J Chem Soc
Dalton Trans 1059–1065
2. Benson N (1994) Biochemical and spectroscopic studies of
1
(
2001) Biochemistry 40:9810–9820
2
Me SO reductase from Rhodobacter capsulatus. PhD thesis,
University of East Anglia, UK
1
23