Journal of Theoretical Biology 226 (2004) 6568

New equations for redox and nano-signal transduction

John T. Hancock
a,
*, Radhika Desikan
a
, Steven J. Neill
a
, Andrew R. Cross
b
a
Centre for Research in Plant Science, University of the West of England, Bristol, Coldharbour Lane, Bristol BS16 1QY, UK
b
Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, USA
Received 10 April 2003; accepted 4 August 2003
Abstract
Cells maintain redox potentials (E
h
) in intracellular compartments, sometimes referred to as redox environments. These potentials
are often very reducing, for example in the cytoplasm, but throughout the cell different potentials are maintained, commensurate
with the functionality of that particular part of the cell. Furthermore, within a simple cellular compartment, hot-spots of redox
poise may be maintained. However, despite this complexity, the quantication of such redox potentials has been attempted, and
there is indeed a need to accurately assess such potentials, and to monitor how they might change with time. Changes in intracellular
potentials may control the oxidation or reduction of protein residues, such as cysteine, which would alter the conformation of those
proteins and so modulate their function. Although there are several methods for estimating the intracellular redox potential, the
most accessible technique is the measurement of intracellular concentrations of GSH and GSSG, and the calculation of E
h
using the
Nernst equation. However, using this equation shows that the E
h
imposed by the glutathione couple is dependent on the total
concentration of glutathione present, and therefore values of E
h
obtained may be erroneous. Here, we suggest new equations that
can be used to calculate the redox environments of cells.
r 2003 Published by Elsevier Ltd.
Keywords: Apoptosis; Glutathione; Nano-switches; Nernst equation; Redox potentials
1. Introduction
The Nernst equation, developed in the late 19th
Century by Walter H. Nernst, has been used for several
decades in calculations to demonstrate the thermody-
namic feasibility of electron transfer pathways. Only
recently has its use for an understanding of modulation
of protein function been appreciated, especially for its
implications for cell signalling processes. Many proteins
contain residues with side chains that can undergo redox
cycling, in which functionality is bestowed upon the
protein via the structural changes that result from
electron transfer, and not the presence of electrons
per se. Such proteins, or protein domains, usually
containing thiol groups, have been described as nano-
switches or nano-transducers (Schafer and Buettner,
2001; Cooper et al., 2002). The role of such nano-
switches can be dened only if the mid-point potential
for the redox mechanism within the protein can be
determined, and if the redox state of the proteins
environment can be measured accurately.
There are various methods of estimating the redox
environment (E
h
) in cells, including the use of micro-
injection of glutathione reductase crystals (Keese et al.,
1999), the use of engineered uorescent proteins
(stergaard et al., 2001), or the measurement of the
concentrations of reduced and oxidized glutathione
(GSH and GSSG) in cells (Meyer et al., 2001; Ridnour
et al., 1999). These concentrations can then be used in
the Nernst equation to calculate E
h
; as discussed below.
Therefore, it is possible to determine the E
h
of cells
before and after stimulation, or during physiological
responses such as apoptosis where the intracellular E
h
has been suggested to become more oxidizing by
approximately 60 mV (Kirlin et al., 1999; Cai and Jones,
1998).
Profound changes in the concentration of reduced
glutathione have also been noted following treatments
that might induce the onset of apoptosis (Smith et al.,
2000). However, even though GSH depletion led to a
ARTICLE IN PRESS
*Corresponding author. Tel.: +44-117-344-2475; fax: +44-117-344-
2904.
E-mail addresses: john.hancock@uwe.ac.uk (J.T. Hancock),
radhika.desikan@uwe.ac.uk (R. Desikan), steven.neill@uwe.ac.uk
(S.J. Neill), scross@scripps.edu (A.R. Cross).
0022-5193/$ - see front matter r 2003 Published by Elsevier Ltd.
doi:10.1016/j.jtbi.2003.08.003
release of cytochrome c from mitochondria, apoptosis
was not an inevitable consequence (Ghibelli et al., 1999),
and therefore the exact ramications of GSH depletion
have yet to be fully understood. Using the Nernst
equation it would be assumed that the E
h
inside the cell
will alter following glutathione depletion. However, it
might be the change in GSH concentration per se which
is critical, rather than an alteration in E
h
: Therefore,
robust equations are required to unravel the exact
nature of the role of glutathione in such processes.
Here, we derive new equations for determination of
the redox environment, which will be instrumental for
an understanding of the role of glutathione and nano-
switches in cell signalling, and will be crucial for a full
understanding of the role of redox in processes such as
apoptosis.
2. Results and discussion
A key signalling process operating in all cells involves
rapid and reversible changes in protein structure,
recognized and acted upon by the next component in a
signalling cascade. For many years phosphorylation and
dephosphorylation have been thought of as the foremost
mechanism for such rapid alterations in protein
structure, but recent suggestions also highlight changes
in the thiol status of proteins as having regulatory
importance (Cooper et al., 2002; Rhee et al., 2000;
Schafer and Buettner, 2001). A thiol group will undergo
reduction or oxidation only if its reduction mid-point
potential is appropriate to accept or donate electrons
from or to likely donors and acceptors respectively.
Therefore, the likelihood of a redox reaction of a
particular thiol group proceeding is partly due to the
three-dimensional structure of the protein and partly
determined by the solution in which the protein resides,
the latter being termed the redox environment, dened
by Schafer and Buettner (2001) by the equation:
Redox environment
X
E
h
reduced species:
This is the summation of the product of the reduction
potential (E
h
) of the redox couples found in solution and
the concentration of reduced species of the redox
couples. Clearly, within any cell there will be many
redox couples contributing to this environment, many
with a very negative value for E
h
: Using the equation
above, every couple added to the sum would, in theory,
make the cell environment more negative, which clearly
can not be the case. Secondly, both the reduced and
oxidized species of couples will contribute to the E
h
imposed by the couples. Thirdly, the units derived from
the above equation do not allow easy interpretation of
the results of the calculation, being mVmM (Schafer
and Buettner, 2001). Therefore, we suggest that the
redox environment should be calculated as a mean of
the redox contributions of these couples and that the
equation should be modied to:
Redox environment
P
E
h
total concentration of couple
P
total concentration of couple
:
Values of E
h
which are needed for the above equations
are calculated using the Nernst equation:
E
h
E
m

RT
nF
2:303 log
red
ox
;
R is the gas constant (8.314 J K
1
mol
1
), T is the
temperature in Kelvin, F is the Faraday constant
(9.6485 10
4
Cmol
1
), while n is the number of
electrons involved in the redox of the couple. E
m
is the
mid-point potential, that is, the electrical potential at
which the concentration of the reduced species and
concentration of the oxidized species are equal for any
given redox couple, and therefore the concentrations at
which E
h
E
m
:
In order to calculate E
h
; a known value of E
m
is
required. Experimentally, the voltage of the solution, or
E
h
value, can be chemically modied, and the concen-
tration of reduced and oxidized species determined.
Therefore, the Nernst equation can be used to derive E
m
:
Although there are undoubtedly many couples to be
considered for the determination of the redox environ-
ment of any particular part of the cell, the one that
is used extensively is that of the glutathione couple
(Cai and Jones 1998; Kirlin et al., 1999; Schafer and
Buettner, 2001). Here the intracellular concentrations of
reduced glutathione (GSH) and oxidized glutathione
(GSSG) are measured, entered into the Nernst equation
and the value of intracellular E
h
calculated. Glutathione
exists in cells at a relatively high concentration, reported
to be 111 mM (Meyer et al., 2001; Smith et al., 1996),
and undergoes the redox chemistry:
GSSG2H

2e

-2GSH:
Therefore, to calculate the contribution to the redox
environment of glutathione, the [red] function within the
Nernst equation has to be squared, giving:
E
h
E
m

RT
nF
2:303 log
GSH
2
GSSG
:
Using this equation, with an E
m
value of zero, a
theoretical contribution of the reduced:oxidized ratio to
the value of E
h
can be obtained as shown in Fig. 1. It is
clear that the mid-points of the redox curves, giving a
true value for E
m
; depend on the concentration of total
glutathione. In order for the redox of the solution to
remain the same as the concentration of total glu-
tathione alters, the proportion of the GSH/GSSG in
the oxidized or reduced state would also need to alter,
following the curve shown in Fig. 1 (inset). The
proportion of GSH/GSSG for a xed redox potential
should remain the same (as the Nernst equation suggests
for most other redox couples), and not alter with
ARTICLE IN PRESS
J.T. Hancock et al. / Journal of Theoretical Biology 226 (2004) 6568 66
concentration. Thus, in this case E
m
is not known, and
E
h
cannot be determined even if [GSH] and [GSSG] are
known. Fig. 1 shows that the dependency of E
m
on the
concentration of total glutathione is particularly pro-
nounced over the range that has been determined for
intracellular levels of glutathione, i.e., 111 mM (Meyer
et al., 2001; Smith et al., 1996). This suggests that in
the case of the GSH/GSSG couple the Nernst equation
returns values of E
h
which are false by a factor
of730 mV, and if E
m
can not be determined, the Nernst
equation must be modied in order to compensate for
this.
Expressing the theoretical values for E
m
over the
range 0.110 mM of total glutathione concentration, the
relationship shown in Fig. 2A is seen. A log
10
derivation
(Fig. 2B), shows that the change of E
m
with glutathione
concentration is linear, and the slope can now be used to
re-adjust the value of E
m
back to that predicted from
the Nernst equation. Therefore, we suggest that for the
calculation of redox environments in cells from mea-
surements of GSH:GSSG concentrations the following
equation should be used:
E
h
E
m

RT
nF
2:303 log
GSH
2
GSSG

RT
nF
2:303 logtotal glutathione;
which can be condensed to:
E
h
E
m

RT
nF
2:303
log
GSH
2
GSSG
logtotal glutathione

and shortened further to:
E
h
E
m

RT
nF
2:303
log
GSH
2
GSSG
total glutathione

:
The value for E
m
for the glutathione couple at pH 7
(E
m7:0
) has been taken as approximately 240 mV
(Schafer and Buettner, 2001; Kirlin et al., 1999). This
was determined using the catalysis of GSSG-reductase
(Rost and Rapoport, 1964), but was calculated at
T=40

C, while most calculations using the Nernst

equation assume T=25

C. This new derivation of the

Nernst equation now allows the determination of E
m
over a range of physiological temperatures and a range
of total glutathione concentrations, and can be used for
the true measurements of intracellular E
h
:
It should be noted that the calculation of E
h
should
include a factor for pH, again as a squared relationship,
although intracellular pH is not always easy to estimate.
The pH factor has been omitted here for clarity, but
needs to be added for accurate determination of the
redox environment (see Schafer and Buettner, 2001).
The accurate determination of intracellular E
h
values
in different parts of the cell and organelles is extremely
important. Only by such correct measurements can the
roles and impact of the oxidation and reduction of thiols
in the modulation of protein structures in nano-switches
be assessed. Fig. 3 shows the alteration in redox state of
both one and two electron couples, calculated using the
Nernst equation but for simplication with an assumed
E
m
of zero. It is clear that a change of redox potential
of730 mV will have a profound effect on the ratio of
ARTICLE IN PRESS
0
10
20
30
40
50
60
70
80
90
100
-80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80
mV
%
G
S
S
G
0
0 2 4 6 8 10
20
40
60
80
100
max [GSH]
%

G
S
S
G
Fig. 1. The Nernst redox curves for glutathione. The theoretical plots for the ratio of GSH/GSSG against the voltage of the solution over a range of
concentrations of maximum GSH (i.e. total glutathione in a solution), assuming that the E
m
function of the Nernst equation is zero. =10 mM;
}=8 mM; =6 mM; m=4 mM; =2 mM; E=1 mM; =0.5 mM; J=0.25 mM; n=1 mM. Inset: the alteration in GSH/GSSG ratio that
would be needed to maintain the redox of the solution as the concentration of total glutathione alters.
J.T. Hancock et al. / Journal of Theoretical Biology 226 (2004) 6568 67
oxidized and reduced redox species and could determine
whether a nano-switch is on or off, i.e. propagating a
cellular signal or not. It has already been noted that
during apoptosis cellular E
h
becomes more positive (Cai
et al., 1998; Kirlin et al., 1999; Schafer and Buettner,
2001) and been suggested that this might be important
for the progression of apoptosis (Hancock et al., 2001).
Redox signalling may determine many cellular events,
including whether a cell lives or dies, and needs to be
studied in a robust manner.
References
Cai, J., Jones, D.P., 1998. Superoxide in apoptosis: mitochondrial
generation triggered by cytochrome c loss. J. Biol. Chem. 273,
1140111404.
Cooper, C., Patel, R.P., Brookes, P.S., Darley-Usmar, V.M., 2002.
Nanotransducers in cellular redox signaling: modication of thiols
by reactive oxygen and nitrogen species. Trends Biochem. Sci. 27,
489492.
Ghibelli, L., Coppola, S., Fanelli, C., Rotilio, G., Civitareale, P.,
Scovassi, A.I., Ciriolo, M.R., 1999. Glutathione depletion causes
cytochrome c release even in the absence of cell commitment to
apoptosis. FASEB J. 13, 20312036.
Hancock, J.T., Desikan, R., Neill, S.J., 2001. Does the redox status of
cytochrome c act as a fail-safe mechanism in the regulation of
programmed cell death? Free Rad. Biol. Med. 31, 697703, doi
0891-5849/01/S0891-5849(01)00646-3.
Keese, M.A., Saffrich, R., Dandekar, T., Becker, K., Schirmir, R.H.,
1999. Microinjected glutathione reuctase crystals as indicators of
the redox status in living cells. FEBS Lett. 447, 135138.
Kirlin, W.G., Cai, J., Thompson, S.A., Diaz, D., Kavanagh, T.J.,
Jones D.P., 1999. Glutathione redox potential in response to
differentiation and enzyme inducers. Free Rad. Biol. Med. 27,
12081218, doi: 0891-5849/99/S0891-5849(99)00145-8.
Meyer, A.J., May, M.J., Fricker, M., 2001. Quantitative in vivo
measurement of glutathione in Arabidopsis cells. Plant J. 27, 6778.
stergaard, H., Henriksen, A., Hansen, F., Winther, J.R., 2001.
Shedding light on disulde bond formation: engineering a redox
switch in green uorescent protein. EMBO J. 20, 58535862.
Rhee, S.G., Bae, Y.S., Lee S.-R., Kwon, J., 2000. Hydrogen peroxide:
a key messenger that modulates protein phosphorylation through
cysteine oxidation. Sciences Signal Transduction Knowledge
Environment, http://stkesciencemag.org/cgi/content/full/OC sigtrans;
2000/53/pe1.
Ridnour, L.A., Winters, R.A., Ercal, N., Spitz, D.R., 1999. Measure-
ment of glutathione, glutathione disulde, and other thiols in
mammalian cell and tissue homogenates using high-performance
liquid chromatography separation of N-(1-pyrenyl)maleimide
derivatives. Methods Enzymol. 299, 258267.
Rost, J., Rapoport, S., 1964. Reduction-potential of glutathione.
Nature 201, 185.
Schafer, F.Q., Buettner, G.R., 2001. Redox environment of the cell as
viewed through the redox state of the glutathione disulde/
glutathione couple. Free Rad. Biol. Med. 30, 11911212, doi:
0891-5849/01/S0891-5849(01)00480-4.
Smith, J.D., ONeill, A., Brady, H.R., Fitzpatrick, J.M., Watson,
R.W.G., 2000. Cellular glutathione extrusion is an integral part of
neutrophil apoptosis. FASEB J. 14, A194.
Smith, C.V., Jones, D.P., Guenthner, T.M., Lash, L.H., Lauterburg,
B.H., 1996. Contemporary issues in toxicology. Compartmentation
of glutathione: implications for the study of toxicity and disease.
Toxicol. Appl. Pharmacol. 140, 112.
ARTICLE IN PRESS
0
10
20
30
40
50
60
70
80
90
100
-200 -100 0 100 200
mV
%

o
x
i
d
i
s
e
d

Fig. 3. The Nernst redox curve. The theoretical plot for the
concentration of reduced species in a redox couple against the voltage
of the solution. This plot assumes E
m
=0. E: n=1; : n=2.
-40
-30
-20
-10
0
0 2 4 6 8 10
10
20
30
40
[GSH] mM
m
V
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
-30 -20 -10 10 20 30 0
mV
l
o
g

[
g
l
u
t
a
t
h
i
o
n
e

(
m
M
)
]
(a)
(b)
Fig. 2. The variance of glutathione E
m
. (a) Variance from 0 mV of the
calculated E
m
for glutathione against the concentration of total
glutathione (maximum concentration of GSH in the solution). (b) A
linearized log
10
plot for the variance of calculated E
m
for glutathione
against the concentration of total glutathione.
J.T. Hancock et al. / Journal of Theoretical Biology 226 (2004) 6568 68