Electroenzymatic reactions. Investigation of a reductive dehalogenase by means of electrogenerated redox cosubstrates

Article abstract of DOI:10.1021/ja053403d

As an illustration of how cyclic voltammetry can be used to unravel the mechanisms and kinetics of redox enzymes, the reductive dechlorination of trichloroethylene and tetrachloroethylene by a typical reductive dehalogenase, the tetrachloroethene reductive dehalogenase of Sulfurospirillum multivorans (formerly called Dehalospirillum multivorans), was investigated by means of several electrochemically generated cosubstrates. They comprised the monocation and the neutral form of methylviologen, the neutral form of benzylviologen, and cobaltocene. Cyclic voltammetry is used to produce the active form of the cosubstrate under controlled potential conditions. It shows large plateau-shaped catalytic responses, which are used to measure the kinetics of the enzymatic reaction as a function of the substrate and cosubstrate concentrations. The variation of the rate constant for the cosubstrate reaction with its standard potential shows the transition between two asymptotic behaviors, one in which the reaction is under diffusion control and the other in which it is under counter-diffusion control. Simple fitting of this plot allows an estimation of the standard potential of the electron acceptor center in the enzyme (E° = -0.57 V vs NHE).

Full text of DOI:10.1021/ja053403d

Published on Web 09/08/2005
Electroenzymatic Reactions. Investigation of a Reductive
Dehalogenase by Means of Electrogenerated Redox
Cosubstrates
Gabriele Diekert,† Denitsa Gugova,† Benoˆıt Limoges,‡ Marc Robert,‡ and
Jean-Michel Save´ant*,‡
Contribution from the Institut fur Mikrobiologie, FSU Jena, Philosphenweg 12,
07743 Jena, Germany, and Laboratoire d’Electrochimie Mole´culaire, UniVersite´ de Paris
7-Denis Diderot, 2 place Jussieu,75251 Paris Cedex 05, France
Received May 25, 2005; E-mail: saveant@paris7.jussieu.fr
Abstract: As an illustration of how cyclic voltammetry can be used to unravel the mechanisms and kinetics
of redox enzymes, the reductive dechlorination of trichloroethylene and tetrachloroethylene by a typical
reductive dehalogenase, the tetrachloroethene reductive dehalogenase of Sulfurospirillum multivorans
(formerly called Dehalospirillum multivorans), was investigated by means of several electrochemically
generated cosubstrates. They comprised the monocation and the neutral form of methylviologen, the neutral
form of benzylviologen, and cobaltocene. Cyclic voltammetry is used to produce the active form of the
cosubstrate under controlled potential conditions. It shows large plateau-shaped catalytic responses, which
are used to measure the kinetics of the enzymatic reaction as a function of the substrate and cosubstrate
concentrations. The variation of the rate constant for the cosubstrate reaction with its standard potential
shows the transition between two asymptotic behaviors, one in which the reaction is under diffusion control
and the other in which it is under counter-diffusion control. Simple fitting of this plot allows an estimation
of the standard potential of the electron acceptor center in the enzyme (E° ) -0.57 V vs NHE).
Scheme 1
Introduction
In recent years, several reductive dehalogenases that couple
the reductive dehalogenation of tetrachloroethylene and other
chlorinated hydrocarbons to energy conservation (dehalorespi-
ration)1 have been isolated from anaerobic bacteria.2,3 Interesting
biological strategies are thus offered for remediation of systems
polluted by halohydrocarbons.4 Most of the reductive dehalo-
genases that mediate dehalogenation reactions contain a cor-
rinoid as a cofactor, as well as two FeS clusters (Scheme 1).
This is the case for the tetrachloroethylene reductive dehalo-
genase of Sulfurospirillum multiVorans.2a,5 This enzyme cata-
lyzes the reductive dehalogenation of tetrachloroethylene (PCE)
† FSU Jena.
‡ Universite´ de Paris 7-Denis Diderot.
(1) (a) Neumann, A.; Scholz-Muramatsu, H.; Diekert, G. Arch. Microbiol. 1994,
162, 295. (b) Hollinger, C.; Schumacher, W. Antonie LeeuVenhoek 1994,
66, 239. (c) Scholz-Muramatsu, H.; Neumann, A.; Mesmer, M.; Diekert,
G. Arch. Microbiol. 1995, 163, 48. (d) Wohlfarth, G.; Diekert, G. Curr.
Opin. Biotechnol. 1997, 8, 290. (e) Hollinger, C.; Wohlfarth, G.; Diekert,
G. FEMS Microbiol. ReV. 1999, 22, 383.
(2) (a) Neumann, A.; Wohlfarth, G.; Diekert, G. J. Biol. Chem. 1996, 271,
16515. (b) Maillard, J.; Schumacher, W.; Vazquez, F.; Regeard, C.; Hagen,
W. R.; Holliger, C. Appl. EnViron. Microbiol. 2003, 69, 4628. (c) Neumann,
A.; Siebert, A.; Trescher, T.; Reinhard, S. Wohlfarth, G.; Diekert, G. Arch.
Microbiol. 2002, 177, 420. (d) Kra¨utler, B.; Fieber, W.; Ostermann, S.;
Fashing, M.; Ongania, K.-M.; Gruber, K.; Kratky, C.; Mikl, C.; Siebert,
A.; Diekert, G. HelV. Chim. Acta 2003, 86, 3698.
(3) (a) Christiansen, N.; Ahring, B. K.; Wohlfarth, G.; Diekert, G. FEBS Lett.
1998, 436, 159. (b) Magnuson, J. K.; Romine, M. F.; Burris, D. R.;
Kingsley, M. T. Appl. EnViron. Microbiol. 2000, 66, 5141. (c) Anandarajah,
K.; Kiefer, P. M.; Donohoe, B. S.; Copley, S. D. Biochemistry 2000, 39,
5303. (d) Krasotkina, J.; Walters, T.; Maruya, K. A.; Ragsdale, S. W. J.
Biol. Chem. 2001, 276, 40991. (e) Thibodeau, J.; Gauthier, A.; Duguay,
M.; Villemur, R.; Lepine, F.; Juteau, P.; Beaudet, R. Appl. EnViron.
Microbiol. 2004, 70, 4532.
and trichloroethylene (TCE) to (Z)-1,2-dichloroethene with
exceptionally high specific activities. It uses as corrinoid cofactor
norpseudovitamin B12, rather than the usual vitamin B12.2d The
mechanism by which the cobalt(I) corrinoid effects the dechlo-
rination reaction currently arouses an active interest, giving rise
to several recent biomimetic studies.6
(4) (a) Castro, C. E. In Natural production of organohalogen compounds;
Gribble, G. W., Ed.; Handbook of Environmental Chemistry, Vol. 3, Part
P; Springer: Berlin, 2003; pp 235-264. (b) Lendvay, J. M.; Loeffler, F.
E.; Dollhopf, M.; Aiello, M. R.; Daniels, G.; Fathepure, B. Z.; Gebhard,
M.; Heine, R.; Helton, R.; Shi, J.; Krajmalnik-Brown, R.; Major, C. L.,
Jr.; Barcelona, M. J.; Petrovskis, E.; Hickey, R.; Tiedje, J. M.; Adriaens,
P. EnViron. Sci. Technol. 2003, 37, 1422. (c) Ferguson, J. F.; Pietari, J. M.
EnViron. Pollut. 2000, 107, 209.
(5) Formerly called Dehalospirillum multiVorans.
9
10.1021/ja053403d CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 13583-13588
13583

A R T I C L E S
Diekert et al.
Results and Discussion
The main kinetic characteristics of the enzyme have been
determined in solution using the monocation of methylviologen
(MV+) in experiments where MV+ was obtained by reduction
of MV2+ by titanium(III) citrate.2a,c The overall kinetics exhibits
a Michaelis-Menten type behavior as defined by eq 1,
TCE with MV0 as Cosubstrate. Typical cyclic voltammetric
responses obtained in the presence of the enzyme at a fixed
concentration of MV2+ with increasing concentrations of TCE
are shown in Figure 1a. In the absence of substrate, MV2+
exhibits two reversible waves that correspond to the successive
formation of MV+ and MV0. The corresponding standard
potentials are -0.44 and -0.77 V vs NHE, respectively. Upon
addition of TCE, a strong catalytic increase of the current is
observed at the level of the second wave, while there is a slight
but definite increase of the current on the first wave, though it
is insufficient for meaningful rate determinations.
1
kap
1
1
1
2ki
)
+
+
2ksubstrate[substrate] kcosubstrate[cosubstrate]
(1)
with kPCE ) 8.5 × 105 M-1 s-1, kTCE ) 7 × 105 M-1 s-1, and
kMV ) 2 × 105 M-1 s-1, while ki is of the order of 50 s-1 7
.
+
The global rate constant, ki, combines the rate constants of the
first-order steps in the substrate and the cosubstrate half-
reactions. The factors 2 in the first and third terms account for
the fact that two equivalents of cosubstrate are used for each
substrate equivalent consumed (Scheme 1) and that the reaction
is followed by observing the disappearance of the cosubstrate.
The purpose of the study reported in the following sections
was to take advantage of the capabilities of cyclic voltammetry
to extend the analysis of the enzyme kinetics, with particular
attention to its responses to several other cosubstrates of
increasing reductive power.
There are two electrochemical approaches to redox enzyme
kinetics using cyclic voltammetry. One is applicable when the
enzyme, immobilized onto the electrode surface, may exchange
electrons directly with the electrode.8 This was not the case with
our enzyme, where all attempts to promote direct electron
transfer failed, presumably because of its large size (57.3 kDa).
The other approach, consisting of using a one-electron reversible
mediator that serves as redox cosubstrate to the enzyme, proved
successful. This method has been applied satisfactorily to several
homogeneous and immobilized enzymatic systems,9-12 and
rigorous theoretical treatments are available to extract the
enzyme kinetics from the cyclic voltammetric raw data.9,13
Concerning immobilized systems, there are very few ex-
amples where the cyclic voltammetric data have been treated
rigorously so as to obtain reliable kinetic and mechanistic
information.9,12b Aside from the determination of the reductive
dehalogenase characteristics, the present study thus offers
another illustrative example of the strategies that can be used
with immobilized redox enzymes when direct electron transfer
with the electrode is not possible. It shows how, under such
unfavorable conditions, the redox characteristics of the electron-
exchanging center of the enzyme may be nevertheless deter-
mined.
Figure 1. Cyclic voltammetry of 5 µM MV2+ in TRIS buffer, in the
presence of 10 nM dehalogenase, as a function of increasing TCE
concentrations. Scan rate: 0.03 V/s. (a) Raw voltammetric data. (b) After
subtraction of the cosubstrate diffusion current. (c) Linear analysis according
to eq 3.
The second wave is plateau-shaped, as expected in the case
where the steady-state approximation applies for all forms of
the enzyme and the consumption of the substrate by the
enzymatic reaction is small as compared to its bulk concentra-
tion.13 It soon became apparent that the cyclic voltammetry
response involves essentially the enzyme adsorbed onto the
electrode surface rather than the enzyme in the solution, as will
be proved later on. Analysis of the raw data of Figure 1a
accordingly follows the theory developed for the case where
the enzyme is immobilized on the electrode.13b The current
flowing through the electrode is the sum of two contributions,
where idif is the diffusion current of the cosubstrate in the
(6) (a) Shey, J.; van der Donk, W. A. J. Am. Chem. Soc. 2000, 122, 12403.
(b) Rich, A. E.; DeGreeff, A. D.; McNeill, K. Chem. Commun. 2002, 234.
(c) McCauley, K. M.; Wilson, S. R.; van der Donk, W. A. Inorg. Chem.
2002, 41, 393. (d) McCauley, K. M.; Wilson, S. R.; van der Donk, W. A.
J. Am. Chem. Soc. 2003, 125, 4410. (e) Follett, A. D.; McNeill, K. J. Am.
Chem. Soc. 2005, 127, 12403. (f) Argu¨ello, J. E.; Costentin, C.; Griveau
S.; Save´ant, J.-M. J. Am. Chem. Soc. 2005, 127, 5049.
(7) The kinetic rate constants were obtained from the reciprocal plots of Figure
4 in ref 2a, giving the dehalogenase activity as a function of PCE, TCE, or
MV+ concentration.
(8) Leger, C.; Elliott, S. J.; Hoke, K. R.; Jeuken, L. J. C.; Jones, A. K.;
Armstrong, F. A. Biochemistry 2003, 42, 8653.
(9) Bourdillon, C.; Demaille, C.; Moiroux, J.; Save´ant, J.-M. Acc. Chem. Res.
1996, 29, 529.
(10) Ikeda, T. Chem. Rec. 2004, 4, 192.
(11) Madoz, J.; Fernandez-Recio, J.; Gomez-Moreno, C.; Fernandez, V. M.
Bioelectrochem. Bioenerg. 1998, 47, 179.
(12) (a) Dequaire, M.; Limoges, B.; Moiroux, J.; Save´ant, J.-M. J. Am. Chem.
Soc. 2002, 124, 240. (b) Limoges, B.; Save´ant, J.-M.; Yazidi, D. J. Am.
Chem. Soc. 2003, 125, 9192.
(13) (a) Limoges, B.; Moiroux, J.; Save´ant, J.-M. J. Electroanal. Chem. 2002,
521, 1. (b) Limoges, B.; Moiroux, J.; Save´ant, J.-M. J. Electroanal. Chem.
2002, 521, 8.
i ) idif + icat
(2)
absence of enzymatic reaction. At the plateau of the wave,
9
13584 J. AM. CHEM. SOC. VOL. 127, NO. 39, 2005

Reductive Dehalogenase Electrochemistry
A R T I C L E S
shaped curves are observed.14 The slope of the resulting
FSΓ0E
1
0
4
FS/ip,cat vs 1/[MV2+] straight line, 1/(kMV ΓE) ) 2.25 × 10
0
)
+
ip,cat
2ksubstrate[substrate]x)0
0
(kMV in M-1 s-1, ΓE in mol cm-2), leads, using the previously
0
0
1
1
2ki
determined value of ΓE, 0.4 × 10-12 mol cm-2, to kMV ) 9 ×
0
+
(3)
107 M-1 s-1
.
kcosubstrate[cosubstrate]x)0
0
We may now use the values of kTCE and kMV to estimate ki
from the intercepts in Figures 1c and 2c according to eq 3. ki is
thus found to be between 550 and 850 s-1. Having the various
rate constants in hand, we may now estimate the respective
contributions to the catalytic current of the enzyme in the
solution and the enzyme adsorbed on the electrode surface, and
see whether our assumption of a prevailing role of the adsorbed
enzyme is justified. In pure kinetic conditions, with negligible
consumption of the substrate in the reaction layer, homogeneous
enzyme catalysis also gives rise to plateau-shaped waves. The
plateau current is then given by13a
where F is the faraday constant, S the electrode surface area,
and Γ0E the surface concentration of the enzyme immobilized
on the electrode. [substrate]x)0 symbolizes the concentration of
the substrate at the electrode surface. Since substrate consump-
tion is negligible, [substrate]x)0 ) C0S, its bulk concentration.
[cosubstrate]x)0 stands for the concentration of the active form
of the cosubstrate (here its reduced form) at the electrode
surface. At the plateau, [cosubstrate]x)0 ) C0C, the bulk
concentration of the oxidized form.
Conversion of Figure 1a into Figure 1b constitutes the first
step of data analysis according to eq 2. The second step is
illustrated by Figure 1c, which shows the application of eq 3.
ihpom
ln(1 + σ)
2
) [MV2+] D
k
C0
1 -
MV2+ MV0
E
x
[
]
xσ
FS
σ
From the slope of the resulting straight line, 1/(2kTCEΓ0) ) 1.6
E
where C0E is the enzyme concentration and the competition
parameter σ is defined by
× 106 (kTCE in M-1 s-1, ΓE0 in mol cm-2), it follows that the
surface concentration of the enzyme is Γ0E ) 0.4 × 10-12 mol
cm-2, taking kTCE ) 7 × 105 M-1 s-1 7
.
1
1
2ki
σ ) kMV0C0
+
C
(
)
We then examined how the catalytic currents vary with the
concentration of cosubstrate for a fixed concentration of TCE.
The raw data are shown in Figure 2a. Subtraction of the
cosubstrate diffusion current leads to the curves in Figure 2b.
For the four lowest MV2+ concentrations, the expected plateau-
2kTCE[TCE]
In the adsorbed case, the current is the sum of the current
given in eq 2 and the diffusion current of the cosubstrate. The
ratio between the homogeneous and the immobilized contribu-
tions is thus obtained from
ln(1 + σ)
2
D
k
C0
1 -
ihpom
iipmm
MV2+ MV0
E
x
[
]
xσ
σ
)
(4)
kMV0Γ0
FV
E
0.446 DMV2+
+
x
RT
1 + σ
Figure 3 shows that the relative contribution of homogeneous
catalysis is below 10% in all the experiments that have been
used to extract the kinetic characteristics, thus justifying its
neglect in the preceding analyses.
Figure 3. Relative contribution of the homogeneous enzyme catalysis in
the experiments of Figures 1 and 2. The symbols are the same as in Figures
1c and 2c. A value of 5 × 10-6 cm2 s-1 was used for DMV
.
2+
Figure 2. Cyclic voltammetry of MV2+ in TRIS buffer, in the presence of
10 nM dehalogenase and 1 mM TCE, as a function of increasing MV2+
concentrations. Scan rate: 0.03 V/s. (a) Raw voltammetric data. (b) After
subtraction of the cosubstrate diffusion current. (c) Linear analysis according
to eq 3.
(14) The cyclic voltammetric response shows a flat peak at the highest
concentration, which corresponds to the fact that substrate impoverishment
at the electrode surface ceases to be negligible as it is when plateau-shaped
curves are obtained.
9
J. AM. CHEM. SOC. VOL. 127, NO. 39, 2005 13585

A R T I C L E S
Diekert et al.
PCE with MV0 as Cosubstrate. Two similar series of
experiments were carried out with PCE. Figure 4 shows a series
of cyclic voltammetric responses obtained at a fixed concentra-
tion of MV2+ with increasing concentrations of PCE. The raw
data are displayed in Figure 4a. After subtraction of the
cosubstrate diffusion current (Figure 4b), the linear plot of Figure
4c is obtained from the plateau currents. From its slope,
1/(2kPCEΓ0E) ) 1.4 × 106 (kPCE in M-1 s-1, ΓE0 in mol cm-2),
the surface concentration of the enzyme is estimated as ΓE0
)
0.4 × 10-12 mol cm-2, taking for kPCE, 8.5 × 105 M-1 s-1 7
.
Figure 5. Cyclic voltammetry of MV2+ in TRIS buffer, in the presence of
10 nM dehalogenase and 1 mM PCE, as a function of increasing MV2+
concentrations. Scan rate: 0.03 V/s. (a) Raw voltammetric data. (b) After
subtraction of the cosubstrate diffusion current. (c) Linear analysis according
to eq 3.
Figure 4. Cyclic voltammetry of 2.5 µM MV2+ in TRIS buffer, in the
presence of 10 nM dehalogenase, as a function of increasing PCE
concentrations. Scan rate: 0.03 V/s. (a) Raw voltammetric data. (b) After
subtraction of the cosubstrate diffusion current. (c) Linear analysis according
to eq 3.
We now examine how the catalytic currents vary with the
concentration of cosubstrate for a fixed concentration of PCE.
Figure 6. Relative contribution of the homogeneous enzyme catalysis in
the experiments of Figures 4 and 5. The symbols are the same as in Figures
4c and 5c. A value of 5 × 10-6 cm2 s-1 was used for DMV
.
The raw data are shown in Figure 5a. The slope of the resulting
2+
0
FS/ip,cat vs 1/[MV2+] straight line (Figure 5c), 1/(kMV ΓE) )
0
0
2.8 × 104 (kMV in M-1 s-1, ΓE in mol cm-2), leads to kMV
)
cobalticinium, Cc+) and doubly reduced benzylviologen cation
BV0 in terms of rate constants (see Figures 7 and 8). Cobalti-
cinium is a more reducing cosubstrate (the standard potential
of the Cc+/Cc couple is -0.93 V vs NHE), whereas BV0 is
less reducing. The interest of doubly reduced benzylviologen,
BV0, is that its standard potential, -0.62 V vs NHE, stands
between those of the MV2+/MV+ and MV+/MV0 couples.
The rate constants are inversely proportional to the slopes of
the linear plots in the right-hand diagrams of Figure 8 (13, 10,
and 13 cm-1 s from top to bottom respectively), leading to the
results summarized in Table 1.
0
0
8 × 107 M-1 s-1, practically the same as the rate constant
derived from the TCE experiments (9 × 107 M-1 s-1), when
one uses the previously determined value of Γ0E, 0.4 × 10-12
mol cm-2. From the intercepts in Figures 4c and 5c, ki is found
to be between 1500 and 2000 s-1
.
Coming now to the justification of the neglect of homoge-
neous enzyme catalysis in the experiments involving PCE and
MV0, we see in Figure 6 that this is indeed the case, the relative
contribution of homogeneous catalysis remaining below 15%
in all experiments.
PCE with Cobaltocene and Benzylviologen as Cosub-
strates. A series of experiments was carried out for comparing
electrochemically generated MV0 to cobaltocene (generated from
Standard Potential of the Electron-Accepting Centers. It
is remarkable that the rate constants are practically the same,
and close to 108 M-1 s-1, for the three most reducing
9
13586 J. AM. CHEM. SOC. VOL. 127, NO. 39, 2005

Reductive Dehalogenase Electrochemistry
A R T I C L E S
Figure 7. Cyclic voltammetry of cobalticinium and MV2+ (10 µM) in TRIS
buffer, in the presence of 50 nM dehalogenase. Scan rate: 0.03 V/s. Blue
dashed line, cobalticinium alone; red solid line, cobalticinium in the presence
of 1 mM PCE; green dotted line, MV2+ in the presence of 1 mM PCE.
cosubstrates (Figure 9). It is also interesting to note that the
same rate constant has also been found with TCE as substrate
and MV0 as cosubstrate. By comparison with the diffusion limit
(109 M-1 s-1) corresponding to the diffusion of the cosubstrate
toward any point on the enzyme surface, this value is compatible
with control by the diffusion of the cosubstrate to a portion of
enzyme surface, corresponding to a solid angle of 4π/10. This
is a quite reasonable conclusion in view of the fact that the
domain where the electron-accepting centers are likely to be
located is close to the enzyme surface, thus defining a cone for
the access of the cosubstrate.
Using the value found previously for MV+ (data point on
the left of the plot in Figure 9), one can estimate the standard
potential of the electron-accepting centers according to the
following procedure.15a
The rate constant for electron transfer between the cosubstrate
and the enzyme redox center may be expressed as15b
k′
dif
1
1
1
)
+
+
(5)
kcosubstrate kdif
F
kdifkact
kdif exp - (E - E0)
[
]
RT
The first term involves the bimolecular diffusion limit, kdif
(horizontal dotted green straight line in Figure 9). The second
term is the “counter-diffusion” term (ascending dotted red
straight line in Figure 9). k′ is a first-order rate constant (in
dif
Figure 8. Cyclic voltammetry of MV2+, Cc+, and BV2+ in TRIS buffer,
in the presence of 50 nM dehalogenase and 1 mM PCE, as a function of
increasing cosubstrate concentrations. Scan rate: 0.03 V/s. Left: voltam-
metric data. Right: linear analysis according to eq 3.
s-1), equal to the second-order diffusion limit (in M-1 s-1). E0
is the enzyme standard potential and E the standard potential
of the redox cosubstrate. kact is the activation-controlled rate
constant that may be expressed in the vicinity of the standard
potential (symmetry factor equal to 0.5) as
Table 1. Cosubstrate Rate Constants with PCE as Substrate
cosubstrate
a
MV+
BV0
MV0
Cc
F
2RT
kact ) k0 exp -
(E - E0)
E0 (V vs NHE)
cosubstrate (M-1 s-1
-0.44
-0.62
-0.77
-0.93
[
]
k
)
2 × 105
0.8 × 108
0.8 × 108
108
where k0 is the standard rate constant (rate constant at zero
driving force).
a With TCE as substrate the rate constant is 9.0 × 107.
As a crude approximation, E0 can be obtained from the
intersection of the first two terms of eq 5 taken separately. A
value of -0.60 V vs NHE is thus obtained. A better approxima-
tion consists of taking the first two terms of eq 5 together, still
ignoring the third term. A more correct value of -0.57 V vs
NHE is then obtained. Introducing nonzero values for the third
term does not lead to any improvement of data fitting. This
simply means that the standard activation rate constant, k0, is
very large, larger than the diffusion limit.
In the above procedure, we have assumed that the charac-
teristic rate constants of the enzyme are the same in solution
(corresponding to the first point on the plot of Figure 9) and in
the adsorbed state (corresponding to the other points on the plot
of Figure 9). Several examples have been given of denaturation
(15) (a) Andrieux, C. P.; Save´ant, J.-M. J. Electroanal. Chem. 1986, 205, 43.
(b) Marcus, R. Faraday Discuss. Chem. Soc. 1982, 74, 7.
9
J. AM. CHEM. SOC. VOL. 127, NO. 39, 2005 13587

A R T I C L E S
Diekert et al.
enzyme. The responses result prevalently from the enzyme
adsorbed on the glassy carbon electrode rather than from the
enzyme present in the solution. The rate constant for the reaction
of the cosubstrate with the enzyme could thus be derived from
the cyclic voltammetric responses for three couples of increasing
reducing power: BV+/BV0 (E° ) -0.62 V vs NHE), MV+/
MV0 (E° ) -0.77 V vs NHE), and Cc+/Cc (E° ) -0.93 V vs
NHE). It is remarkable that the rate constants for the three
cosubstrates are the same ((8-10) × 107 M-1 s-1) and do not
depend on whether the substrate is PCE or TCE. This indicates
that the reaction is under diffusion control. Comparing this
results with the rate constant, 2 × 105 M-1 s-1, obtained for
the MV2+/MV+ (E° ) -0.44 V vs NHE) allows the determi-
nation of a standard potential of -0.57 V vs NHE for the
electron-accepting center in the enzyme. This value is sufficient
for a facile electron transfer from the iron-sulfur clusters to
the cobalt(II) corrinoid and the alkyl or alkenyl cobalt(III)
corrinoid formed in the catalytic process.
Figure 9. Variations of the cosubstrate rate constant with its standard
potential. Determination of the standard potential of the electron-accepting
centers (see text).
Experimental Section
Reagents. The Sulfurospirillum multiVorans5 bacteria was grown
anoxically on a medium containing pyruvate and fumarate as described
earlier.1a The PCE reductive dehalogenase was extracted from the cells
of bacteria and purified on a series of chromatographic columns (Q-
sepharose, phenyl-superose, Superdex) as previously described.2a
The activity of the purified PCE reductive dehalogenase was
determined spectrophotometrically under a N2 atmosphere from the
decrease in absorbance at 578 nm of the radical MV+ (ꢀ ) 9.7 mM-1
cm-1) in the presence of a saturating concentration of chlorinated
hydrocarbon.2a The MV+ radical was chemically generated from
titanium(III) citrate. The enzyme concentration was indirectly deter-
mined from the amount of cobalt quantified from ICP-AES analysis.
A dehalogenase molecular weight of 57.3 kDa was selected.2a
The PCE, TCE, cis-DCE, methylviologen dichloride, and benzyl-
viologen dichloride compounds were used as received (Aldrich). The
cobalticinium was obtained from Strem Chemicals. A 0.1 M Tris buffer
(pH 7.5) containing 0.5 mM NH4Cl was systematically used. All of
the aqueous solutions were prepared using water purified by a Milli-Q
water purification system from Millipore.
of enzymes upon adsorption onto electrode surfaces. It is thus
conceivable that the rate constants corresponding to the three
right-hand points are underestimated. What would be the
consequence on E0 determination? The maximal value these rate
constants can take is the unrestricted diffusion limit, i.e., 109
M-1 s-1. Figure 10 shows an attempt to fit the data in these
conditions. The best fit corresponds to an E0 value -0.60 V vs
NHE, but it is seen that the fit is clearly poorer than in the
preceding case. We may thus conclude that -0.57 V vs NHE
is a more exact value, probably known with an accuracy better
than (10 mV.
Electrochemical Experiments. An Autolab potensiostat (EcoChe-
mie, Utrecht, The Netherlands) interfaced to a personal computer was
used for cyclic voltammetry (CV). Carbon-based screen-printed elec-
trodes (9.6 mm2 for the sensing disk area) were used as disposable
working electrodes, and they were prepared as previously described16
from a homemade carbon-based ink composed of graphite particles
(Ultra Carbon, UCP 1M, Johnson Matthey) and polystyrene, and using
a manual screen-printer (Circuit Imprime´ Franc¸ais, Bagneux, France)
equipped with a screen-stencil of 77 threads cm-1. A saturated calomel
electrode (SCE) and a platinum counter electrode were employed as a
reference and counter electrodes. The potentials indicated in the text
are referred to the NHE. To avoid O2 damage of the reductive
dehalogenase, all experiments were carried out under N2 in a glovebox
(O2 < 2 ppm, Jacomex, France), in which a water-jacketed electro-
chemical cell was maintained at 20 ( 0.5 °C with a circulating water
bath.
Figure 10. Alternative attempt to estimate the enzyme standard potential
(see text).
A value of -0.57 V vs NHE makes the electron-accepting
center sufficiently reductive to convert cobalt(II) into cobalt(I)
in the corrinoid cobalt center of the enzyme and also the cobalt-
(III) alkyl or alkenyl derivatives that are likely to be formed as
intermediates in the catalytic loop.
Acknowledgment. This project was supported in part by
grants from the European Commission (Proj. No. HPRN-CT-
2002-00195) and the German Science Foundation (DFG).
Concluding Remarks
JA053403D
Cyclic voltammetry of Sulfurospirillum multiVorans5 deha-
logenase allows a complete kinetic characterization of the
(16) Bagel, O.; Limoges, B.; Scho¨llhorn, B.; Degrand, C. Anal. Chem. 1997,
69, 4688.
9
13588 J. AM. CHEM. SOC. VOL. 127, NO. 39, 2005