Antioxidant and Cytoprotective Responses to Redox Stress
Joanne Mathers
1,4
, Jennifer A. Fraser
1,4
, Michael McMahon
1
, Robert D.C. Saunders
2
,
John D. Hayes
1
and Lesley I. McLellan
1,3
1
Biomedical Research Centre,
University of Dundee,
Ninewells Hospital and Medical School,
Dundee, DD1 9SY, UK
2
Department of Biological Sciences,
The Open University,
Walton Hall, Milton Keynes, MK7 6AA, UK
3
To whom correspondence should be addressed
Tel: 01382 496618
Fax: 01382 669993
Email: lesley.mclellan@cancer.org.uk
4
These authors made an equal contribution to this article and should be considered as
joint first authors

2Abstract
Aerobic cells produce reactive oxygen species as a consequence of normal cellular
metabolism, and an array of antioxidant systems are in place to maintain the redox
balance. When the redox equilibrium of the cell is upset by pro-oxidant environmental
stimuli, adaptive responses to the redox stress take place, which can result in up-
regulation of antioxidant proteins and detoxication enzymes. Over the past few years,
it has become apparent that members of the cap ‘n’ collar (CNC)-bZip family of
transcription factors are principal mediators of defensive responses to redox stress. In
mammals, the CNC family members Nuclear Factor – Erythroid 2 p45 – related factors 1
and 2 (Nrf1 and Nrf2) have been shown to be involved in the transcriptional up-
regulation of cytoprotective genes including those encoding glutamate cysteine ligase,
NAD(P)H:quinone oxidoreductase, glutathione S-transferases and aldo-keto
reductases. An evolutionarily conserved system exists in Caenorhabditis elegans,
and it is possible that Drosophila melanogaster may also utilize CNC transcription
factors to induce antioxidant genes in response to pro-oxidant chemicals. The advent
of microarray and proteomic technologies has advanced our understanding of the gene
batteries regulated by oxidative insult, but has highlighted the complexity of gene
regulation by environmental factors. This review focuses on the antioxidant response
to environmental stress, and the impact that microarrays and proteomics have made in
this field.
Introduction
Reactive oxygen species (ROS) are formed as by-products from normal cellular
metabolism, and the redox balance is maintained by a range of intracellular antioxidants.
ROS are also produced in response to environmental stimuli such as UV light, g-rays,
pathogens and drugs. In addition, extracellular signalling molecules such as the
cytokines interleukin 1b and tumour necrosis factor a (TNFa) generate ROS [1]. The

3importance of ROS as sub-cellular messengers is becoming increasingly apparent.
Changes in the redox status of the intracellular environment can activate multiple
signalling pathways, altering gene transcription, the cell cycle and apoptosis. It is,
however, unclear how specificity is achieved in redox-dependent signalling, and the
exact mechanisms by which oxidants regulate this process are only just starting to be
elucidated. The role of ROS in cell signalling pathways is the subject of several
excellent recent reviews [1-6], and here we will concentrate primarily on the gene-
regulatory effects of thiol-reactive chemicals, which disturb the antioxidant status of
the cell.
Reactive oxygen species, thiol-reactive agents and cellular antioxidants
It is not within the scope of this article to give a full description of all antioxidant systems
and types of reactive oxygen species. This is covered comprehensively by Halliwell
and Gutteridge [7], but a brief outline of some key points is given below.
Oxygen metabolism in aerobic organisms is associated with the generation of numerous
ROS that can cause extensive damage to major cellular constituents such as membrane
lipids, DNA and proteins [7]. Reactive oxygen species encompass a variety of diverse
chemical species including superoxide (O
2
• –
) the extremely unstable and reactive
hydroxyl radical (OH
•
) and the relatively more long-lived hydrogen peroxide (H
2
O
2
). The
majority of intracellular ROS production occurs in the mitochondria, and it has been
estimated that this organelle converts 1-2% of O
2
molecules consumed into superoxide
anions [5]. Superoxide can also be generated by NADPH oxidase and cytochrome P450
reductase; the former is attracting increasing interest for its role in cell signalling [1, 2,
8]. Superoxide dismutases convert O
2
• –
to H
2
O
2
, which in turn can be reduced to water
by several different antioxidant systems (discussed below). Alternatively, in the
presence of transition metals, H
2
O
2
can be converted to the highly reactive and
damaging hydroxyl radical.

4Normally, endogenously-produced ROS are largely scavenged by a diverse and
multifaceted antioxidant defense system that functions to eliminate oxidative insult.
Superoxide dismutases (SOD) catalyse the dismutation of O
2
• –
to H
2
O
2
and oxygen [7].
The superoxide dismutases use various metal ions at their active-sites in the
dismutation of O
2
• –
. As their name suggests, the copper-zinc SOD isoenzymes (CuZn-
SOD), including the predominantly cytoplasmic homdimeric SOD1 and homotetrameric
extracellular SOD (EC-SOD or SOD3), have a copper and a zinc atom in each active-
site. Whilst the copper acts directly in the dismutation of O
2
•–
by undergoing alternate
oxidation and reduction, the zinc functions to stabilise the enzyme. Another type of
SOD has manganese at the active-site. Homotetrameric manganese SOD (Mn-SOD or
SOD2) is found predominantly in the mitochondria and catalyses the same reaction as
CuZn-SODs. The importance of Mn-SOD in normal mitochondrial function is illustrated
by the fact that mice with a targeted deletion of Mn-SOD die within 3 weeks post-
partum, and show severe mitochondrial damage in cardiac tissue [9].
Hydrogen peroxide produced by the superoxide dismutases can be detoxified by
catalases or peroxidases. In mammals, catalase activity appears to be usually confined
to the peroxisomes, suggesting that it is unlikely that catalase contributes to the
detoxication of H
2
O
2
generated at extra-peroxisomal sites such as the mitochondria [7].
Thiol-dependent peroxidases play a major role in H
2
O
2
and organic peroxide reduction.
These enzymes include the selenium-dependent glutathione peroxidase family and the
peroxiredoxins [10, 11]. Certain members of the glutathione S-transferase (GST) super
family of enzymes can also reduce organic peroxides, although the major detoxication
role for GST is in the conjugative metabolism of electrophilic agents [10].
Glutathione is the most abundant non-protein thiol in most aerobic cells and makes a
major contribution to the redox status of the cell. It is an important scavenger of free
radicals, as well as serving as a substrate for glutathione peroxidases and GST.
Treatment of cells with thiol-reactive agents (e.g. diethyl maleate, DEM) depletes

5glutathione and upsets the redox balance. This results in oxidative stress (defined by
Sies [12] as a disturbance in the prooxidant-antioxidant balance in favour of the former,
leading to potential damage), although not necessarily the same kind of oxidative stress
experienced by cells subjected to an increase in superoxide or hydrogen peroxide. It is
worth noting here that agents such as DEM can also react with unprotonated cysteine
residues in certain proteins (e.g. reference [13]). This has the potential to generate a
kind of cellular stress distinct from that of the redox balance, and generates a layer of
complexity in deciphering the mechanisms underlying the adaptive responses to thiol-
reactive agents.
Adaptive responses to oxidative stress
Adaptive responses to oxidative stress are complex and differ depending on cell type
and the nature of the stress. Important cellular factors in determining the response to
pro-oxidant stimuli will be the presence or absence of redox sensors, signalling relay
systems or transcription factors that mediate responses. The antioxidant status of the
cell, as well as the levels of detoxication or metabolic enzymes, are also likely to have
an impact on the response to stress. The type of oxidant will influence the response,
as will the dose and duration of stimulus.
Oxidative stimuli impact on phosphorylation-dependent signalling pathways, and can
modulate the activity of redox-sensitive transcription factors such as AP-1 or NF-kB [5].
Although the mechanisms by which alterations in the redox equilibrium stimulate
different cellular responses are not precisely understood, recent evidence suggests
that oxidants can impinge on signalling by altering enzyme activity through direct
oxidation or by influencing redox-dependent protein:protein interactions [1]. The former
is exemplified by finding that protein phosphatases which contain a reactive cysteine at
the catalytic site (protein tyrosine phosphatases or dual specificity phosphases) can be
oxidised and inactivated [14-17]. Redox-dependent inactivation of phosphatases has
the potential to impact on down-stream phosphorylation events, and it has recently

6been shown that oxidative stress activates phosphoinositide 3-kinase (PI 3-kinase)-
dependent signalling by the inactivation of the protein tyrosine phosphatase PTEN [18].
The redox-dependent regulation of protein:protein interactions also appears to have an
important role in the cellular response to oxidative stress [6]. For example, apoptosis
signal-regulating kinase (ASK) 1, which activates c-Jun N-terminal protein kinase (JNK)
and p38 signalling cascades, can be inhibited through its interaction with thioredoxin
[19]. The interaction between thioredoxin and ASK1 was found to be dependent on the
redox status of thioredoxin, where thioredoxin only bound to ASK1 under reducing
conditions. The model proposed is that oxidative stress can induce dissociation of
thioredoxin from ASK1, which would permit activation of ASK1-dependent signalling
pathways.
Glutathione S-transferase P1 has also been shown to be involved in the regulation of
JNK activity by redox-dependent protein:protein interactions. It was shown that
association of GSTP1 with JNK occured in non-stressed cells and inhibited JNK activity
[20]. Treatment of cells with H
2
O
2
caused dissociation of GSTP1 and activation of JNK
activity.
More recently, it has been shown that the nuclear localization of the Yap1 transcription
factor in Saccharomyces cerevisiae is regulated by its interaction with a thioredoxin
peroxidase, Gpx3 [21, 22]. In S. cerevisiae, Yap1 is a critical mediator of the adaptive
response to H
2
O
2
, and regulates the expression of stress-response genes, which
include those encoding antioxidant enzymes. Yap1 contains a nuclear export signal,
which is sterically obstructed under conditions of oxidative stress due to
conformational changes in the protein caused by disulphide bond formation. This
causes accumulation of Yap1 in the nucleus. Delaunay et al. [21] showed that the
formation of the disulphide bond is mediated by Gpx3 in response to hydrogen
peroxide-treatment. It is unclear whether similar regulatory mechanisms could occur in
mammalian cells.

7As discussed earlier, cells can respond to oxidative stress in different ways. Of
particular interest to us is the adaptation to redox stress that involves the induction of
antioxidant and detoxication enzymes. This antioxidant response to sub-lethal stress is
relevant to cancer chemoprevention, as well as prevention of other degenerative
diseases associated with ageing [23, 24]. The principal players in the transcriptional
up-regulation of cytoprotective genes in response to environmental stress are members
of the cap ‘n’ collar (CNC) family of transcription factors, and the cytoplasmic regulator
Keap1.
The role of the antioxidant responsive element and CNC transcription factors
in regulation of antioxidant and detoxication genes
The antioxidant responsive element: A significant contribution to cell defense against
oxidative insult is mediated through transcriptional activation of antioxidant and
detoxication genes via a cis-acting enhancer known as the antioxidant responsive
element (ARE). This element was discovered by Pickett and co-workers more than a
decade ago [25], and since then, ARE consensus sequences have been identified in
the 5’ flanking regions of many antioxidant and cytoprotective genes [23]. The binding
consensus sequence of the ARE has been elucidated (5’-
A
/
G
TGA
C
/
T
NNNGC
A
/
G
-3’) and
comparison with other sequences reveals that it shares similarity with the DNA
recognition sequences of Nuclear Factor - Erythroid 2 (NF-E2, 5’-
A
/
G
TGA
C
/
G
TCAGC
A
/
G
-
3’) and v-Maf (5’-TGCTGACTCAGCA-3’) [23]. NF-E2 belongs to the CNC basic leucine
zipper (bZIP) family of transcription factors which includes NF-E2 p45 related factors 1,
2 and 3 (Nrf1 Nrf 2 and Nrf3) as well as the more distantly related transcriptional
repressors Bach1 and 2 (reviewed in [23, 26]). Unlike NF-E2, which is expressed
specifically in haematopoietic tissues, Nrf 1 and Nrf2 expression is more ubiquitous and
both are widely expressed in many tissues [27]. The widespread tissue distribution
profile of Nrf1 and Nrf2, and the striking similarity of the NF-E2 and v-Maf DNA binding

8motif with the ARE led to the hypothesis that CNC transcription factors may regulate
antioxidant and cytoprotective gene transcription through the ARE [23].
The majority of studies have implicated Nrf2 as the key mediator of ARE dependent
activation of antioxidant gene expression. Transient transfection of Nrf2 into HepG2,
RL34 or Hepa1c1c7 cells induces reporter gene activity from constructs derived from
the human and mouse NAD(P)H:quinone oxidoreductase (NQO) 1 gene promoters [28,
29]. By contrast, expression of dominant negative Nrf2 inhibits endogenous Nrf2
function and blocks both basal and tert-butylhydroquinone (tBHQ)-inducible activity of
NQO1 ARE driven reporter activity in astrocytes [30]. Moreover, Nrf2 binding to ARE –
containing promoters from a variety of antioxidant genes has been demonstrated [29,
31, 32].
Unequivocal evidence for the role of Nrf2 in regulation of cellular protection against
oxidative stress has come from targeted disruption of the mouse Nrf2 gene. Whilst
mice lacking Nrf2 are viable and fertile, western blot analyses of hepatic and intestinal
tissues from these mice showed diminished levels of NQO1 and Alpha and Mu classes
of GST [27, 32]. This was correlated with a reduction in GST and NQO1 enzyme
activity [27]. Further analyses showed an impaired ability of Nrf2-deficient mice to
induce NQO1 and Alpha and Mu classes of GST following administration of several
agents capable of mediating ARE-dependent gene induction [27]. The data suggest that
expression of antioxidant and cytoprotective genes under normal and oxidatively-
stressed conditions is regulated by Nrf2 via the ARE.
Glutathione levels are significantly reduced in the livers of Nrf2-deficient mice [33].
Comparison of fibroblasts from wild type and Nrf2
-/-
mice showed that the reduction in
glutathione was correlated with a dramatic reduction in levels of glutamate cysteine
ligase (GCL) transcripts. GCL catalyses the first, and rate-limiting, step in glutathione
synthesis and is composed of a catalytic (GCLC) and regulatory subunit (GCLM). The
5' region of human GCLC contains 4 ARE consensus sequences. The ARE positioned

93.1 kb upstream of the transcriptional start site, known as ARE4, has been shown to be
important for basal and inducible expression of GCLC in response to b-naphthoflavone
[34]. Indeed, studies have shown that Nrf2 in combination with small Maf proteins was
capable of binding the ARE4 under basal and inducible conditions [31]. GCLM contains
an ARE consensus sequence close to the transcriptional start site [35, 36], although
work in our laboratory did not find that it was a functional ARE in HepG2 cells treated
with tBHQ [37]. This was in contrast to work from Mulcahy’s group [38], who found
this consensus sequence to function as an ARE in HepG2 cells treated with b-
naphthoflavone. Mulcahy and colleagues had performed a different type of mutational
analysis to that of Galloway and McLellan [37], and they later went on to show that the
conflicting data could be explained by the fact that GCLM contains a variant ARE in its
promoter region [39]. The regulation of GCLM by a Nrf2-dependent mechanism is also
supported by in vivo studies. Induction and basal expression of both GCLM and GCLC
is reduced in tissues of Nrf2
-/-
mice [27, 40]. Futhermore, endogenous expression of
both GCLM and GCLC can be rescued by transient expression of Nrf2 in Nrf2
-/-
mouse
embryonic fibroblasts [33] indicating that Nrf2 is a relevant factor in transcription of
glutathione-associated genes. Reduced resynthesis of glutathione in Nrf2 deficient
macrophages has also been correlated with decreased activation of the cysteine x
c
-
transporter [41]. This transporter maintains the intracellular cysteine pool and therefore
can modulate glutathione synthesis. In the absence of Nrf2, macrophages were unable
to induce x
c
-
activity and maintain GSH synthesis following DEM-mediated oxidative
stress. Nrf2-deficient macrophages also displayed loss of basal and inducible haem
oxygenase 1, stress protein A170 and peroxiredoxin (Prx) MSP23 expression in
response to oxidative insult from a wide variety of agents [41].
Oligonucleotide microarray analyses have been used to identify the global changes in
gene expression mediated by Nrf2 by comparing the expression patterns of genes in
Nrf2
+/+
and Nrf2
-/-
cells. The basal expression of 137 genes was altered in Nrf2-

10
deficient cortical astrocytes [30] and of the 207 genes induced following treatment with
tBHQ, 181 were dependent on the presence of Nrf2. Nrf2-dependent genes were
categorised as detoxification genes (including those encoding NQO1, GST and
aldehyde dehydrogenase), antioxidants and genes involved in regulating the reducing
potential, growth and defense/immune response and inflammation. Nrf2 was found to
regulate the expression of several glutathione- or thioredoxin-dependent genes
including the GCL subunits, GST family members, glutathione peroxidase 4, glutathione
reductase and Prx 1. The other antioxidant genes shown to be under the regulation of
Nrf2 (including SOD, catalase and thioredoxin) function at different levels in
cytoprotection. Regulation of several enzymes, including glucose 6-phosphate
dehydrogenase (G6PDH), involved in the supply of reducing equivalents were also
found to be Nrf2-dependent. From this analysis, it is clear that Nrf2 is capable of
upregulating several sets of genes in a coordinated manner, to prevent the
accumulation of compounds capable of cellular damage and to redress the cellular
redox status. A more extensive microarray analysis was carried out on the livers of
Nrf
+/+
and N r f
-/-
mice [42] following treatment with the chemopreventive agent,
dithiolethione. This study identified several more clusters of genes regulated by Nrf2.
Here it was shown that genes encoding enzymes involved in xenobiotic metabolism
(e.g. cytochrome P450s, aldehyde and carbonyl reductases), chaperone and stress
responsive proteins and the ubiquitin/proteosomal system are regulated by Nrf2. A
separate microarray study showed that other genes involved in xenobiotic metabolism
including multidrug resistance protein 1, UDP-glucuronosyl transferases, aflatoxin
aldehyde reductase and aldo-keto reductases are also regulated by Nrf2 in mouse
intestine [43]. Taken together, these studies indicate that Nrf2 regulation extends
beyond the primary control of cellular stress into secondary protective mechanisms
such as recognition, repair and degradation of damaged cellular components. These
findings imply that Nrf2 provides a more global mechanism of cytoprotection which
incorporates cellular response to a variety of forms of stress. Indeed recent findings

11
show loss of Nrf2 sensitises macrophages to cell death following exposure to ER
stress and the unfolded protein response [44].
Nrf1 and Nrf2 are ubiquitously expressed, although their expression patterns are not
identical [27]. It is therefore possible that in tissues where Nrf1 is predominantly
expressed over Nrf2, the former is responsible for driving basal and inducible
antioxidant gene expression. Indeed analysis of fibroblasts obtained from mice with a
targeted disruption of Nrf1 revealed that the absence of Nrf1 increased cell sensitivity
to oxidative stress induced by paraquat, cadmium and diamide [45]. The increase in
sensitivity was associated with increased oxidative burden due to reduced intracellular
glutathione content. The reduction in intracellular glutathione was correlated with
lowered basal expression of Gclm and glutathione synthetase transcripts [45], and
sensitivity to oxidative stress was associated with a decreased ability to induce
transcription of Gclm and glutathione synthetase following administration of paraquat.
Overexpression of Nrf1 in COS cells was shown to elevate intracellular glutathione
levels and transactivate a luciferase reporter construct containing 4.2 kb of the 5'
flanking region of human GCLC [46]. Electrophoretic mobility shift assays have shown
that the ARE4 in this region can also be bound by combinations of Nrf1 and small Maf
proteins G and K [46]. These analyses suggest Nrf1 may also contribute to cellular
protection from oxidative stress by regulating cellular capacity to synthesise glutathione
and antioxidant gene transcription via the ARE. Nrf1 and Nrf2 are, however,
functionally distinct as targeted disruption of Nrf1 causes embryonic lethality [47], and
recent evidence suggests that Nrf2 is the more important protein in ARE-driven gene
expression [48].
It is clear that Nrf2, and probably Nrf1, are key factors in the regulation of many genes
involved in cellular defense against oxidative stress. In the absence of Nrf2, the lack of
induction of protective enzymes sensitises cells and tissues to the cytotoxic effects of
agents such as paraquat and DEM. The importance of Nrf2 is underscored by the

12
dramatic increase in sensitivity of Nrf2
-/-
mice to gastric tumours following administration
of benzo(a)pyrene and the inability of chemopreventative agents such as oltipraz to
protect against benzo(a)pyrene-induced chemical carcinogenesis in mice lacking Nrf2
[49].
Regulation of Nrf2 activity: The actin binding protein Keap1 negatively regulates Nrf2 by
controlling its stability and sub-cellular localisation [50, 51]. In the unstressed cell, Nrf2
interacts with the Kelch domain of Keap1 through its amino terminal Neh2 domain [51]
and as a result, Nrf2 is sequestered in the cytoplasm. Following administration of
sulphydryl reactive chemicals such as DEM, and in response to oxidative stress, the
Keap1:Nrf2 complex is disrupted. Dissociation with Keap1 causes the rapid
accumulation of Nrf2 in the nucleus [51] where it associates with small Maf
transcription factors and mediates ARE-dependent gene expression. Keap1 is a
cysteine rich protein and it has been proposed that by virtue of its sensitivity to
oxidative modification, it may act as a redox sensor so that changes in the intracellular
reducing environment could be transduced to Nrf2 via modification of Keap1. It has
been suggested that oxidative modification may cause conformational changes in
Keap1 which disrupt the interaction with Nrf2, thus negating Keap1 repression [52]. A
recent study has shown that oxidative modification of Keap1 in response to treatment
with indoles could be mediated by activation of NADPH oxidase activity [53].
Targeted deletion of Keap1 showed that in Keap1
-/-
mice, Nrf2 constitutively
accumulates in the nucleus and drives high level expression of both GCL subunits,
NQO1 and Prx [54]. Furthermore, macrophages from these animals were unresponsive
to oxidative insult and expression of these genes could not be elevated in response to
treatment with DEM. These data show that without Keap1, regulation of Nrf2-
dependent gene expression is largely abolished, and the ability of cells to further
upregulate antioxidant and detoxication genes in response to oxidative stress is
diminished.

13
Several protein kinase-dependent signalling pathways have also been reported to
regulate ARE-driven gene expression, and include the mitogen activated protein kinases
(MAPK), PI 3-kinase, protein kinase C (PKC) and the endoplasmic reticulum resident
kinase, PERK (Figure 1). In several instances, this has been demonstrated through the
use of chemical inhibitors of these pathways. For example, pretreatment with the PI 3-
Kinase inhibitor, wortmannin, reduced binding to the ARE in nuclear extracts from H4IIE
cells and prevented induction of rat GSTA2 mRNA [55]. Transient over-expression of
members of the MAPK kinase pathways has been shown to trigger transcription of
ARE-dependent reporter activity in HepG2 cells [56]. This is thought to occur in an Nrf2-
dependent manner as a dominant negative form of Nrf2 abolished this effect and
prevented MAPK-dependent induction of endogenous haem oxygenase. The precise
contribution of the various MAPK pathways to ARE-dependent gene regulation is,
however, unclear as there are conflicting reports in the literature. For example, p38
was shown to negatively regulate basal and tBHQ-inducible expression of NQO1 [56],
but positively regulate inducible expression of GCL subunits in response to pyrrolidine
dithiocarbamate [57]. Exactly how members of the MAPK family transmit signals to
modulate ARE transcription is unknown, but it should be borne in mind that certain
experimental strategies such as over-expression of MAPK family members may cause
cellular dysfunction, with oxidative stress as a possible side effect. Nrf2, however,
has been shown to be directly phosphorylated by PKC and PERK in vitro [44, 58].
Phosphorylation was correlated with increased nuclear localisation of Nrf2 in response
to stress and was prevented by catalytically inactive PERK or inhibition of PKC using
staurosporin. PKC phosphorylation occurs at the Neh2 domain of Nrf2, which is known
to mediate interactions with Keap1 [59]. It is thought that phosphorylation disrupts the
association of Nrf2 and Keap1, as it was shown that Keap1 is less able to co-
immunoprecipitate with PKC-phosphorylated Nrf2 than unmodified Nrf2 [59]. Keap1 also
contains several putative phosphorylation sites but it is unknown whether it is also
targeted by kinases during oxidative stress. In addition, Bach1, a negative regulator of

14
ARE-driven gene expression, appears to be regulated by phosphorylation [60]. It is
clear that a variety of different signals stimulated by various forms of stress converge
on the Nrf2:Keap1 complex. What determines the specificity and importance of each
signalling pathway is not yet known but it is likely to be cell type-specific and depend on
the relative abundance of the various factors in the signalling networks.
Evolutionary conservation of CNC:Keap 1 gene regulation: Although most of the work
on the Nrf2:Keap1 pathway has been performed in mammalian systems, recent
publications suggest that similar pathways may exist in other organisms. Orthologues
of both Nrf2 and Keap1 have been identified in Danio rer io (zebrafish) and
characterised in vitro by Kobayashi et al [61]. Treatment of zebrafish larvae with tBHQ
induced the expression of several antioxidant and detoxication genes (gstp1, nqo1 and
gclc) [61]. Their expression was proposed to be Nrf2-dependent as over-expression
of nrf2 in D. rerio embryos enhanced expression levels of gstp1, nqo1 and gclc in the
absence of oxidative stress. Futhermore, Nrf2 was also shown to be essential for the
inducible expression of gstp1, as targeted knock-down of nrf2 abolished gstp
expresion [61]. Like the mammalian Nrf2 protein, the zebrafish Nrf2 interacts with and
is repressed by Keap1. Repression of Nrf2 by Keap1 is mediated by cytoplasmic
sequestration and occurs by binding through the Neh2 and Kelch domains. Indeed
specific mutation of the ETGE motif in the Neh2 domain of Nrf2 from zebrafish abolished
Keap1 mediated repression of Nrf2 and permitted nuclear accumulation of the mutant
Nrf2 form [61]. Moreover, overexpression of Keap1 was unable to repress gstp1
expression in zebrafish embryos co-overexpressing the E82G mutant Nrf2 [61]. These
findings highlight the importance of Nrf2:Keap1 in regulating cellular response to
oxidative stress in zebrafish and indicate that Nrf2:Keap1 is a highly conserved
regulator of cellular protection in vertebrates.
An Nrf2-like signalling pathway also appears to be conserved in nematodes. Recent
reports suggest that the transcription factor SKN-1 is important in regulation of oxidant

15
signalling in C. elegans in a manner similar to Nrf2 [62]. SKN-1 is a distant relative of
Nrf2 and both share homology at the CNC domain in the NH
2
terminal. Furthermore, the
consensus sequence of the SKN-1 DNA binding motif is similar to that of Nrf2. The
SKN-1 DNA-binding sites have been identified in the 5' flanking regions of several C.
elegans antioxidant genes including gcl, gst, nqo and sod [62]. In C. elegans,
expression of a gcl promoter-driven reporter construct was dramatically increased in
response to paraquat. This was abolished in SKN-1 deficient worms as was basal
expression of the gclc reporter gene, suggesting that SKN-1 mediates both constitutive
and inducible expression of gclc. Moreover SKN-1 mutants are more sensitive to
oxidative stress than wild type worms [62]. These findings suggest that SKN-1 may
contribute to cellular protection from oxidative stress by regulating antioxidant gene
transcription in a manner similar to Nrf2 and the ARE. Unlike Nrf2, SKN-1 binds DNA as
a monomer and does not require additional factors such as small Maf proteins. A
functional homologue of Keap1 has not yet been identified in C. elegans. SKN-1 does,
however, share sequence similarity with a portion of the Neh2 domain of Nrf2 and is
shown to accumulate in the nucleus following oxidative stress in a manner similar to
Nrf2 [62].
Bioinformatic and genetic analyses suggest that a system analogous to the Nrf2:Keap 1
antioxidant regulators may also be present in D. melanogaster. The cap’n’collar locus
(CG17894) encodes three transcripts and protein isoforms, Cnc-A, Cnc-B and Cnc-C
[63]. The cnc-A and cnc-C transcripts are expressed ubiquitously in Drosophila
embryos whilst expression of the cnc-B transcript is restricted to head and mandibular
structures. The cnc-A, cnc-B and cnc-C transcripts encode proteins containing 533,
805 and 1296 amino acids respectively and are 100% identical at the carboxyl terminal.
This region is also highly homologous to the CNC-bZIP domain of NE-F2 p45. Work has
shown that Cnc-B acts a transcriptional repressor during Drosophila development and
mutation of cnc-B leads to lethal head and mandibular defects [63]. A Maf orthologue
(Maf-S) was identified as a Cnc-B binding protein via yeast two-hybrid analysis [64]

16
and mutation of Maf-S recapitulates the embryonic defects formed by cnc-B mutations
suggesting Maf-S is a functional partner of Cnc-B. The DNA-binding element of the
Cnc-B/Maf-S heterodimer has been mapped [64] and comparison shows it is identical to
the DNA binding motif of TCF11/Nrf1/MafG, and very similar to the consensus binding
site of Nrf2 and SKN-1. Moreover, Cnc-A and Cnc-B can both bind this element in
combination with Maf-S.
Drosophila appear to posses regulatory systems capable of inducing antioxidant gene
expression as GCLC levels can be increased in Drosophila S2 cells following exposure
to oxidative insult generated by tBHQ [65] and DEM (Figure 2). Although the biological
function of Cnc-A and Cnc-C has not been characterised, it is possible that they may
regulate antioxidant gene expression in vivo. Alignment of several Nrf2 sequences
reveals that the ETGE motif and highly hydrophobic domain of the Neh2 domain of Nrf2
are present in the amino terminal region of Cnc-C [61] but not Cnc-A and Cnc-B. This
motif has been shown to be essential for Keap1 mediated sequestration of Nrf2 in the
cytoplasm. This, in combination with the identification of a hypothetical gene encoding a
Drosophila Keap1 orthologue in the Drosophila genome sequence (CG3962), and the
observation that Cnc proteins form complexes with Maf-S which bind NF-E2 - like
elements is highly suggestive that Cnc forms (in particular Cnc-C) may mediate Nrf2 -
like functions in vivo. Functional analyses of the Cnc protein isoforms will be required
to confirm this and such analyses are currently underway in our laboratory.
DNA microarrays and proteomics to study the antioxidant response to
stress
As detailed earlier in this review, redox regulation is now known to be an important
mechanism in signalling and the activation and regulation of transcription factors [1]. To
fully understand the complexity of cellular redox regulation, it is essential the targets of
such pathways are identified both at the gene and protein level. Over the past decade

17
this goal has become more feasible with the emergence of the ‘omics’ technologies of
genomics [66] and proteomics [67].
The era of genomics, the study of complete genomes, has exploded following the
completion of the entire genome sequences of many experimental animal systems
including budding yeast (S. cerevisiae) [68], the nematode (C. elegans) [69] and the
fruit fly (D. melanogaster) [70]. With the completion of the Human Genome Project
imminent [71] there is now a wealth of genetic knowledge freely available in the public
domain; this information has been greatly exploited in microarray technology, which is
emerging as one of the most powerful techniques for analysing gene regulation and
function. In this technique, mRNA extracted from cells or tissue is reverse transcribed
to cDNA with the incorporation of a fluorescent tag. This is then hybridised to a
complementary single stranded DNA sequence immobilised on an array chip.
Quantification of the amount of each mRNA is then determined from the intensities of
fluorescence observed. Living organisms constantly sense and adapt to their redox
environment, inducing batteries of genes to maintain their redox homeostasis. As
microarray technology allows quantitative comparison between treated and untreated
samples (e.g. between normoxic and oxidatively stressed samples) it has proven itself
a valuable technique in providing an insight into cellular redox regulation.
Alternative gene splicing and post-translational modification of proteins means that a
single gene can give rise to a diversity of functional protein products. This level of
complexity is beginning to be addressed in the field of ‘proteomics’, the analysis of
genomic complements of proteins. In particular, two-dimensional polyacrylamide-gel
electrophoresis (2D-PAGE), used in conjunction with mass spectrometry for protein
identification [72, 73] has emerged as a widely used technology in the analysis of
oxidative insult on global protein expression.
Several recent studies have employed either a genomic microarray or a proteomic 2D-
PAGE approach to unravel the complex picture of cellular responses to oxidative stress

18
[43, 74-79]. Numerous stress-generating agents including H
2
O
2
, diamide, tBHQ and
paraquat have been used to treat a wide range of species including yeast (S.
cerevisiae), nematodes (C. elegans), flies (D. melanogaster) and mammalian cell-lines
(see Table 1). From these studies, it appears that certain categories of genes and
proteins are consistently induced in response to redox stress.
Perhaps rather unsurprisingly, one category consistently exhibiting enhanced levels of
expression following oxidative stress treatment are the antioxidant defense enzymes.
Over the numerous studies carried out, virtually every known class of antioxidant is
represented, including thioredoxins, peroxiredoxins, superoxide dismutases, catalases,
glutathione peroxidase, glutathione synthetase and GCL. Detoxifying enzymes,
including numerous members of the GST family, were also found to be induced in
several studies (see Table 1). The induction of GST occurred in all species investigated
including yeast, Drosophila and mammals and with all oxidative agents tested.
It is important to note that most studies only investigate the expression profile of one
oxidative stress-inducing agent in a single cell-type. Relatively few groups have
attempted comparative studies to investigate whether distinct oxidative stress
treatments evoke common or disparate gene expression profiles. One such
comparative microarray analysis in MCF7 cells exposed to either H
2
O
2
, menadione or
tBHQ concluded that both common and distinct genes were induced by the different
oxidants. Common to all three treatments was the upregulation of antioxidant enzymes
including glutathione peroxidase and GCLM. Distinctions were, however, apparent in
induction of heat shock protein (hsp) genes. Menadione or tBHQ caused induction of
Hsp-coding genes to a similar extent, whilst comparable levels of induction were not
observed in H
2
O
2
treated cells [75].
Stimulus-specific responses are known to occur in Escherichia co l i following
exposure to either H
2
O
2
or superoxide radicals (O
2
• –
). This distinction is explained by
two partially overlapping regulons, one responsive to H
2
O
2
and the other to O
2
• –
. The

19
differential response to oxidants is regulated by the redox-sensitive transcription
factors OxyR and SoxRS, which are regulated by H
2
O
2
and O
2
• –
respectively [80, 81].
In yeast, yAP-1 transcription factors play an important role in regulating the response to
oxidative stress [82, 83], and the use of 2D-PAGE allowed Godon et al. [79] to identify
the ‘H
2
O
2
stimulon’ in S. cerevisiae. This study identified 115 proteins that are induced
by H
2
O
2
in S. cerevisiae [79]. Subsequent work showed that Yap1 mediates the
induction of at least 32 proteins in response to oxidative stress [76]. Of these, 15 were
shown to require the presence of the Skn7 transcription factor.
Another category of enzymes consistently induced by oxidative insult are metabolic
enzymes including NAD(P)H regenerating enzymes. NAD(P)H functions as an indirect
antioxidant by regenerating reduced glutathione (GSH) from its oxidized form (GSSG) in
a process catalysed by NADPH-dependent glutathione reductase. Recently, however,
it has been proposed that NAD(P)H may also be capable of acting as a direct
antioxidant [84]. A major endogenous source of NADPH is the pentose phosphate
pathway, which acts to generate appropriate levels of ribose 5-phosphate and
erythrose 4-phosphate, precursors of purines and aromatic amino acids respectively.
Within the oxidative branch of this pathway, NADPH is generated as a by-product of
reactions catalysed by G6PDH and gluconate 6-phosphate dehydrogenase, and
mutations in these enzymes cause sensitivity to H
2
O
2
in yeast [85]. This antioxidant
function is conserved in mammals as mouse embryonic stem cells lacking G6PDH
displayed an elevated sensitivity to hydrogen peroxide [86]. The link between metabolic
pathways and ROS dynamics is further substantiated by the finding that switching
carbon sources in yeast from glucose to oleate dramatically increases transcription of
the antioxidant enzymes, thioredoxin, glutathione peroxidase and cytosolic SOD [87].
Cellular redox state is an important factor in maintaining protein structure and function.
The accumulation of abnormal or denatured proteins in a cell in response to various
experimental treatments, including oxidative stress, can potentially be toxic to the cell.

20
Hsps function as molecular chaperons to promote protein renaturation, prevent further
aggregation and denaturation and to facilitate proteolytic degradation of damaged
proteins [88]. It is therefore not surprising that numerous proteomic and microarray
investigations have observed induction of heat shock proteins in response to oxidative
insults (see Table 1). This response appears to be highly conserved in all species
investigated including bacteria, yeast and mammals and occurs in response to most
oxidative stress generating agents tested.
Hsp activity is not only regulated at a transcriptional level. It appears that the redox
environment can directly regulate activity. Under reducing conditions in vitro the
reactive cysteine residues of Hsp33 are co-ordinated to a zinc atom rendering the
enzyme inactive. However, under oxidizing conditions the zinc atom is released,
intramolecular disulphide bonds are formed and the protein exhibits potent chaperone
activity [89].
Our own work using both a microarray approach and a 2D-PAGE proteomics approach
to investigate the impact of the thiol-depleting agent DEM on gene and protein
expression in Drosophila S2 cells has observed induction of genes and proteins
belonging to each of the major categories discussed above (unpublished data).
Post-translational modifications such as phosphorylation, acetylation and glycosylation
modulate the activity of many eukaryotic proteins. Analysis of such modifications can
provide clues about the biological function and regulation of a protein [90]. Proteomic
studies together with mass spectrometry are beginning to emerge as common
methodologies to investigate redox-dependent post-translational modifications of
proteins, providing an indispensable insight into the mechanistic basis behind redox
signalling. For example, phosphorylation changes the charge of a protein, which is then
apparent on a 2D-gel.

21
In a recent study, non-reducing two-dimensional electrophoresis was used to identify
proteins in human T-lymphocytes that are glutathionylated following oxidative stress
(diamide-treatment or H
2
O
2
-exposure) [91]. Glutathionylation, the formation of mixed
disulfides between glutathione and the cysteines of some proteins has been postulated
as a mechanism through which protein function can be regulated by redox status [92].
It can also occur by reaction of proteins with S-nitrosoglutathione, which is generated
from nitric oxide [92, 93]. Proteins which have been reported to be glutathionylated
include redox enzymes such as Prx I, heat shock proteins Hsp60 and Hsp70 and a
regulator of protein folding, protein disulfide isomerase [91].
The Prx antioxidants are a family of thiol-dependent peroxidases that reduce H
2
O
2
and
alkyl hydroperoxides to water and alcohol. Using 2D-PAGE to compare proteins in
human umbilical endothelial vein cells before and after exposure to H
2
O
2
variants of Prx
I and Pr III were generated within thirty minutes of exposure to 100 mM H
2
O
2
. These
variants exhibited altered migration consistent with a decreased isoelectric point,
suggestive of oxidation [78]. Subsequent 2D-PAGE analyses in combination with mass
spectroscopy studies determined that the shift in isoelectric point of Prx I following
exposure to H
2
O
2
[94] or TNFa [95] was due to oxidation of the catalytic cysteine
residue to sulphenic acid, resulting in inactivation of the enzyme. More recently it has
been shown using 2D-PAGE in combination with activity assays that the inactivated
sulphinic form of Prx I produced in HeLa, A594 and W136 cells following exposure to
H
2
O
2
can be reduced back to the catalytically active thiol form [96].
It cannot be disputed that both genomic microarray and proteomic 2D-PAGE has allowed
considerable advances to be made in the understanding of the complex cellular
response to redox status. The next challenge must lie in furthering these advances to
understand the mechanisms behind the redox regulation of signalling pathways.

22
References
1 Finkel, T. (2003) Curr. Opin. Cell Biol. 15, 247-254
2 Finkel, T. (2000) FEBS Lett. 476, 52-54
3 Allen, R. G. and Tresini, M. (2000) Free Radic. Biol. Med. 28, 463-499
4 Hancock, J. T., Desikan, R. and Neill, S. J. (2001) Biochem. Soc. Trans. 29, 345-
350
5 Kamata, H. and Hirata, H. (1999) Cell Signal. 11, 1-14
6 Adler, V., Yin, Z., Tew, K. D. and Ronai, Z. (1999) Oncogene 18, 6104-6111
7 Halliwell, B. and Gutteridge, J. M. C. (1999) in Free Radicals in Biology and
Medicine, Oxford University Press, Oxford
8 Irani, K. and Goldschmidt-Clermont, P. J. (1998) Biochem. Pharmacol. 55, 1339-
1346
9 Lebovitz, R. M., Zhang, H., Vogel, H., Cartwright, J., Jr., Dionne, L., Lu, N.,
Huang, S. and Matzuk, M. M. (1996) Proc. Natl. Acad. Sci. USA 93, 9782-9787
10 Hayes, J. D. and McLellan, L. I. (1999) Free Radic. Res. 31, 273-300
11 Wood, Z. A., Schroder, E., Robin Harris, J. and Poole, L. B. (2003) Trends
Biochem. Sci. 28, 32-40
12 Sies, H. (1991) in Oxidative Stress II. Oxidants and antioxidants, Academic
Press, London
13 Castillo, E. A., Ayte, J., Chiva, C., Moldon, A., Carrascal, M., Abian, J., Jones, N.
and Hidalgo, E. (2002) Mol. Microbiol. 45, 243-254
14 Xu, D., Rovira, I.I. and Finkel, T. (2002) Dev. Cell 2, 251-252
15 Meng, T. C., Fukada, T. and Tonks, N. K. (2002) Mol. Cell 9, 387-399
16 Savitsky, P. A. and Finkel, T. (2002) J. Biol. Chem. 277, 20535-20540
17 Lee, S. R., Yang, K. S., Kwon, J., Lee, C., Jeong, W. and Rhee, S. G. (2002) J.
Biol. Chem. 277, 20336-20342
18 Leslie, N. R., Bennett, D., Lindsay, Y. E., Stewart, H., Gray, A. and Downes, C.
P. (2003) EMBO J. 22, 5501-5510
19 Saitoh, M., Nishitoh, H., Fujii, M., Takeda, K., Tobiume, K., Sawada, Y.,
Kawabata, M., Miyazono, K. and Ichijo, H. (1998) EMBO J. 17, 2596-2606
20 Adler, V., Yin, Z., Fuchs, S. Y., Benezra, M., Rosario, L., Tew, K. D., Pincus, M.
R., Sardana, M., Henderson, C. J., Wolf, C. R., Davis, R. J. and Ronai, Z. (1999)
EMBO J. 18, 1321-1334
21 Delaunay, A., Pflieger, D., Barrault, M. B., Vinh, J. and Toledano, M. B. (2002)
Cell 111, 471-481
22 Georgiou, G. (2002) Cell 111, 607-610
23 Hayes, J. D. and McMahon, M. (2001) Cancer Lett. 174, 103-113
24 Finkel, T. and Holbrook, N. J. (2000) Nature 408, 239-247
25 Rushmore, T. H., Morton, M. R. and Pickett, C. B. (1991) J. Biol. Chem. 266,
11632-11639
26 Motohashi, H., O'Connor, T., Katsuoka, F., Engel, J. D. and Yamamoto, M. (2002)
Gene 294, 1-12

23
27 McMahon, M., Itoh, K., Yamamoto, M., Chanas, S. A., Henderson, C. J., McLellan,
L. I., Wolf, C. R., Cavin, C. and Hayes, J. D. (2001) Cancer Res. 61, 3299-3307
28 Venugopal, R. and Jaiswal, A. K. (1996) Proc. Natl. Acad. Sci. U S A 93, 14960-
14965
29 Nioi, P., McMahon, M., Itoh, K., Yamamoto, M. and Hayes, J. D. (2003) Biochem J
374, 337-348
30 Lee, J. M., Calkins, M. J., Chan, K., Kan, Y. W. and Johnson, J. A. (2003) J. Biol.
Chem. 278, 12029-12038
31 Wild, A. C., Moinova, H. R. and Mulcahy, R. T. (1999) J. Biol. Chem. 274, 33627-
33636
32 Itoh, K., Chiba, T., Takahashi, S., Ishii, T., Igarashi, K., Katoh, Y., Oyake, T.,
Hayashi, N., Satoh, K., Hatayama, I., Yamamoto, M. and Nabeshima, Y. (1997)
Biochem. Biophys. Res. Commun. 236, 313-322
33 Chan, J. Y. and Kwong, M. (2000) Biochim. Biophys. Acta 1517, 19-26
34 Mulcahy, R. T., Wartman, M. A., Bailey, H. H. and Gipp, J. J. (1997) J. Biol. Chem.
272, 7445-7454
35 Galloway, D. C., Blake, D. G., Shepherd, A. G. and McLellan, L. I. (1997)
Biochem. J. 328, 99-104
36 Galloway, D. C., Blake, D. G. and McLellan, L. I. (1999) Biochim. Biophys. Acta
1446, 47-56
37 Galloway, D. C. and McLellan, L. I. (1998) Biochem. J. 336, 535-539
38 Moinova, H. R. and Mulcahy, R. T. (1998) J. Biol. Chem. 273, 14683-14689
39 Erickson, A. M., Nevarea, Z., Gipp, J. J. and Mulcahy, R. T. (2002) J. Biol. Chem.
277, 30730-30737
40 Chanas, S. A., Jiang, Q., McMahon, M., McWalter, G. K., McLellan, L. I., Elcombe,
C. R., Henderson, C. J., Wolf, C. R., Moffat, G. J., Itoh, K., Yamamoto, M. and
Hayes, J. D. (2002) Biochem. J. 365, 405-416
41 Ishii, T., Itoh, K., Takahashi, S., Sato, H., Yanagawa, T., Katoh, Y., Bannai, S.
and Yamamoto, M. (2000) J. Biol. Chem. 275, 16023-16029
42 Kwak, M. K., Wakabayashi, N., Itoh, K., Motohashi, H., Yamamoto, M. and
Kensler, T. W. (2003) J. Biol. Chem. 278, 8135-8145
43 Thimmulappa, R. K., Mai, K. H., Srisuma, S., Kensler, T. W., Yamamoto, M. and
Biswal, S. (2002) Cancer Res. 62, 5196-5203
44 Cullinan, S. B., Zhang, D., Hannink, M., Arvisais, E., Kaufman, R. J. and Diehl, J.
A. (2003) Mol. Cell. Biol. 23, 7198-7209
45 Kwong, M., Kan, Y. W. and Chan, J. Y. (1999) J. Biol. Chem. 274, 37491-37498
46 Myhrstad, M. C., Husberg, C., Murphy, P., Nordstrom, O., Blomhoff, R., Moskaug,
J. O. and Kolsto, A. B. (2001) Biochim. Biophys. Acta 1517, 212-219
47 Chan, J. Y., Kwong, M., Lu, R., Chang, J., Wang, B., Yen, T. S. B. and Kan, Y.
W. (1998) EMBO J. 17, 1779-1787
48 Leung, L., Kwong, M., Hou, S., Lee, C. and Chan, J. Y. (2003) J. Biol. Chem. in
the press
49 Ramos-Gomez, M., Kwak, M. K., Dolan, P. M., Itoh, K., Yamamoto, M., Talalay, P.
and Kensler, T. W. (2001) Proc. Natl. Acad. Sci. U S A 98, 3410-3415

24
50 McMahon, M., Itoh, K., Yamamoto, M. and Hayes, J. D. (2003) J. Biol. Chem. 278,
21592-21600
51 Itoh, K., Wakabayashi, N., Katoh, Y., Ishii, T., Igarashi, K., Engel, J. D. and
Yamamoto, M. (1999) Genes Dev. 13, 76-86
52 Dinkova-Kostova, A. T., Holtzclaw, W. D., Cole, R. N., Itoh, K., Wakabayashi, N.,
Katoh, Y., Yamamoto, M. and Talalay, P. (2002) Pro. Natl. Acad. Sci. USA 99,
11908-11913
53 Sekhar, K. R., Crooks, P. A., Sonar, V. N., Friedman, D. B., Chan, J. Y., Meredith,
M. J., Starnes, J. H., Kelton, K. R., Summar, S. R., Sasi, S. and Freeman, M. L.
(2003) Cancer Res. 63, 5636-5645
54 Wakabayashi, N., Itoh, K., Wakabayashi, J., Motohashi, H., Noda, S., Takahashi,
S., Imakado, S., Kotsuji, T., Otsuka, F., Roop, D. R., Harada, T., Engel, J. D. and
Yamamoto, M. (2003) Nature Genet. in the press
55 Kang, K. W., Ryu, J. H. and Kim, S. G. (2000) Mol. Pharmacol. 58, 1017-1025
56 Yu, R., Chen, C., Mo, Y. Y., Hebbar, V., Owuor, E. D., Tan, T. H. and Kong, A. N.
(2000) J. Biol. Chem. 275, 39907-39913
57 Zipper, L. M. and Mulcahy, R. T. (2000) Biochem. Biophys. Res. Commun. 278,
484-492
58 Huang, H. C., Nguyen, T. and Pickett, C. B. (2000) Proc. Natl. Acad. Sci. U S A
97, 12475-12480
59 Huang, H. C., Nguyen, T. and Pickett, C. B. (2002) J. Biol. Chem. 277, 42769-
42774
60 Suzuki, H., Tashiro, S., Sun, J., Doi, H., Satomi, S. and Igarashi, K. (2003) J. Biol.
Chem. in the press
61 Kobayashi, M., Itoh, K., Suzuki, T., Osanai, H., Nishikawa, K., Katoh, Y., Takagi,
Y. and Yamamoto, M. (2002) Genes Cells 7, 807-820
62 An, J. H. and Blackwell, T. K. (2003) Genes Dev. 17, 1882-1893
63 McGinnis, N., Ragnhildstveit, E., Veraksa, A. and McGinnis, W. (1998)
Development 125, 4553-4564
64 Veraksa, A., McGinnis, N., Li, X., Mohler, J. and McGinnis, W. (2000)
Development 127, 4023-4037
65 Fraser, J. A., Saunders, R. D. and McLellan, L. I. (2002) J. Biol. Chem. 277, 1158-
1165
66 Zhou, H., Roy, S., Schulman, H. and Natan, M. J. (2001) Trends Biotechnol. 19,
S34-39
67 Phizicky, E., Bastiaens, P. I., Zhu, H., Snyder, M. and Fields, S. (2003) Nature
422, 208-215
68 Goffeau, A., Barrell, B. G., Bussey, H., Davis, R. W., Dujon, B., Feldmann, H.,
Galibert, F., Hoheisel, J. D., Jacq, C., Johnston, M., Louis, E. J., Mewes, H. W.,
Murakami, Y., Philippsen, P., Tettelin, H. and Oliver, S. G. (1996) Science 274,
546, 563-547
69 The C. elegans Sequencing Consortium (1998) Science 282, 2012-2018
70 Adams, M. D., Celniker, S. E., Holt, R. A., Evans, C. A., Gocayne, J. D.,
Amanatides, P. G., Scherer, S. E., Li, P. W., Hoskins, R. A., Galle, R. F., George,
R. A., Lewis, S. E., Richards, S., Ashburner, M., Henderson, S. N., Sutton, G. G.,
Wortman, J. R., Yandell, M. D., Zhang, Q., Chen, L. X., Brandon, R. C., Rogers,

25
Y. H., Blazej, R. G., Champe, M., Pfeiffer, B. D., Wan, K. H., Doyle, C., Baxter, E.
G., Helt, G., Nelson, C. R., Gabor, G. L., Abril, J. F., Agbayani, A., An, H. J.,
Andrews-Pfannkoch, C., Baldwin, D., Ballew, R. M., Basu, A., Baxendale, J.,
Bayraktaroglu, L., Beasley, E. M., Beeson, K. Y., Benos, P. V., Berman, B. P.,
Bhandari, D., Bolshakov, S., Borkova, D., Botchan, M. R., Bouck, J., Brokstein, P.,
Brottier, P., Burtis, K. C., Busam, D. A., Butler, H., Cadieu, E., Center, A.,
Chandra, I., Cherry, J. M., Cawley, S., Dahlke, C., Davenport, L. B., Davies, P., de
Pablos, B., Delcher, A., Deng, Z., Mays, A. D., Dew, I., Dietz, S. M., Dodson, K.,
Doup, L. E., Downes, M., Dugan-Rocha, S., Dunkov, B. C., Dunn, P., Durbin, K. J.,
Evangelista, C. C., Ferraz, C., Ferriera, S., Fleischmann, W., Fosler, C.,
Gabrielian, A. E., Garg, N. S., Gelbart, W. M., Glasser, K., Glodek, A., Gong, F.,
Gorrell, J. H., Gu, Z., Guan, P., Harris, M., Harris, N. L., Harvey, D., Heiman, T. J.,
Hernandez, J. R., Houck, J., Hostin, D., Houston, K. A., Howland, T. J., Wei, M.
H., Ibegwam, C., et al. (2000) Science 287, 2185-2195
71 Burley, S. K., Almo, S. C., Bonanno, J. B., Capel, M., Chance, M. R., Gaasterland,
T., Lin, D., Sali, A., Studier, F. W. and Swaminathan, S. (1999) Nature Genet. 23,
151-157
72 Govorun, V. M. and Archakov, A. I. (2002) Biochemistry (Moscow) 67, 1109-
1123
73 Cutler, P. (2003) Proteomics 3, 3-18
74 Zou, S., Meadows, S., Sharp, L., Jan, L. Y. and Jan, Y. N. (2000) Proc. Natl.
Acad. Sci. U S A 97, 13726-13731
75 Chuang, Y. Y., Chen, Y., Gadisetti, Chandramouli, V. R., Cook, J. A., Coffin, D.,
Tsai, M. H., DeGraff, W., Yan, H., Zhao, S., Russo, A., Liu, E. T. and Mitchell, J.
B. (2002) Cancer Res. 62, 6246-6254
76 Lee, J., Godon, C., Lagniel, G., Spector, D., Garin, J., Labarre, J. and Toledano,
M. B. (1999) J. Biol. Chem. 274, 16040-16046
77 Mitsumoto, A., Nakagawa, Y., Takeuchi, A., Okawa, K., Iwamatsu, A. and
Takanezawa, Y. (2001) Free Radic. Res. 35, 301-310
78 Mitsumoto, A., Takanezawa, Y., Okawa, K., Iwamatsu, A. and Nakagawa, Y.
(2001) Free Radic. Biol. Med. 30, 625-635
79 Godon, C., Lagniel, G., Lee, J., Buhler, J. M., Kieffer, S., Perrot, M., Boucherie,
H., Toledano, M. B. and Labarre, J. (1998) J. Biol. Chem. 273, 22480-22489
80 Storz, G. and Imlay, J. A. (1999) Curr. Opin. Microbiol. 2, 188-194
81 Pomposiello, P. J. and Demple, B. (2001) Trends Biotechnol. 19, 109-114
82 Toone, W. M. and Jones, N. (1999) Curr. Opin. Genet. Dev. 9, 55-61
83 Toone, W. M., Morgan, B. A. and Jones, N. (2001) Oncogene 20, 2336-2346
84 Kirsch, M. and De Groot, H. (2001) FASEB J. 15, 1569-1574
85 Juhnke, H., Krems, B., Kotter, P. and Entian, K. D. (1996) Mol. Gen. Genet. 252,
456-464
86 Pandolfi, P. P., Sonati, F., Rivi, R., Mason, P., Grosveld, F. and Luzzatto, L.
(1995) EMBO J. 14, 5209-5215
87 Koerkamp, M. G., Rep, M., Bussemaker, H. J., Hardy, G. P., Mul, A., Piekarska, K.,
Szigyarto, C. A., De Mattos, J. M. and Tabak, H. F. (2002) Mol. Biol. Cell. 13,
2783-2794
88 Ruddock, L. W. and Klappa, P. (1999) Curr. Biol. 9, R400-402

26
89 Jakob, U., Muse, W., Eser, M. and Bardwell, J. C. (1999) Cell 96, 341-352
90 Mann, M. and Jensen, O. N. (2003) Nature Biotechnol. 21, 255-261
91 Fratelli, M., Demol, H., Puype, M., Casagrande, S., Eberini, I., Salmona, M.,
Bonetto, V., Mengozzi, M., Duffieux, F., Miclet, E., Bachi, A., Vandekerckhove, J.,
Gianazza, E. and Ghezzi, P. (2002) Proc. Natl. Acad. Sci. U S A 99, 3505-3510
92 Klatt, P. and Lamas, S. (2000) Eur. J. Biochem. 267, 4928-4944
93 Padgett, C. M. and Whorton, A. R. (1995) Am. J. Physiol. 269, C739-749
94 Yang, K. S., Kang, S. W., Woo, H. A., Hwang, S. C., Chae, H. Z., Kim, K. and
Rhee, S. G. (2002) J. Biol. Chem. 277, 38029-38036
95 Rabilloud, T., Heller, M., Gasnier, F., Luche, S., Rey, C., Aebersold, R.,
Benahmed, M., Louisot, P. and Lunardi, J. (2002) J. Biol. Chem. 277, 19396-
19401
96 Woo, H. A., Chae, H. Z., Hwang, S. C., Yang, K. S., Kang, S. W., Kim, K. and
Rhee, S. G. (2003) Science 300, 653-656

27
Legends to Figures
Figure 1: Potential pathways leading to the activation of Nrf2
Nrf2 regulates the expression of several antioxidant genes. Upon treatment with
agents which generate environmental stress and perturb the intracellular redox status,
the interaction between Keap1 and Nrf2 becomes disrupted permitting Nrf2 to
translocate to the nucleus and enhance transcription. The exact molecular mechanism
triggering the disruption of Nrf2 and Keap1 has not been defined, although several
studies have suggested that Nrf2 may become phosphorylated at serine and threonine
residues by certain kinases including, PI 3-kinase, PKC, PERK and members of the
MAPK family. Keap1 is also thought to be a target in this response, and oxidation or
covalent modification of cysteine residues within Keap1 has been proposed to
destabilise its interaction with Nrf2 and promote dissociation of the Nrf2:Keap1 complex
Figure 2: Treatment of Drosophila S2 cells with diethyl maleate DEM induces
expression of the catalytic subunit of glutamate cysteine ligase
Drosophila S2 cells were treated with diethyl maleate (or DMSO as control) for 16 h.
Total cell-lysates (30 mg protein) were subject to western blotting using antiserum
raised against the catalytic subunit of Drosophila glutamate cysteine ligase.

28
Table 1 Genes and proteins induced by redox stress
Experimental Organism Inducing Agent Examples of Genes/ References
approach Proteins Induced
Microarray Drosophila Paraquat GSTD1, GST6b [74]
Arsenite translocating ATPase
Microarray MCF7 cells H
2
O
2
GPX2, GLRX, Hsp70, [75]
Menadione ubiquitin conjugation and
tBHQ ligase enzymes
Microarray Mouse Sulforaphane NQO1, GST, GCL, UGT [43]
G6PDH, Hsp40,
6-phosphogluconate
dehydrogenase
Microarray Mouse tBHQ NQO1, various GSTs, GCLC, [30]
Prx isoforms, G6PDH, Nrf2
2D-PAGE HUEVC H
2
O
2
/ tBHQ Prx I – III, Hsp27, [78]
glyceraldehyde-3 phosphate
dehydrogenase
2D-PAGE Yeast H
2
O
2
Cytochrome c peroxidase, [76,79]
SOD, thioredoxin1/2,
TRX reductase, glutathione
reductase, various Hsps,
proteosome subunits, G6PDH,
enolase, alcohol dehydrogenase
2D-PAGE HUEVC Paraquat Prx I, Prx II, Prx III, DnaJ [77]
Abbreviations: GPX, glutathione peroxidase; GLRX, glutaredoxin; UGT, UDP-glucuronosyltransferase; NQO, NADPH:quinone
oxidoreductase; HUEVC ,human umbilical endothelial vein cells.

29
Acknowledgements
The Biotechnology and Biological Sciences Research Council (JM and JF, Grant
Numbers ERA16290 and G15091) and The Association for International Cancer
Research (MMcM) are gratefully acknowledged for financial support.

GSTA-2
NQO1
GCLC
GCLM
HO-1
SS
Keap1
SRSR
S T
Nrf2
Keap1
SHSH
SHSH
PI3K
PERK
JNK
p38
ERK
Environmental stress
Protein
unfolding
ER
Akt
PKC
ARE
Maf
S T
Nrf2
P
P
S T
Nrf2
P
P
perturbation of redox homeostasis
Figure 1