Journal of Experimental Botany, Vol. 54, No. 384, pp. 935±941, March 2003
DOI: 10.1093/jxb/erg094
RESEARCH PAPER
Reactive oxygen species production in association with
suberization: evidence for an NADPH-dependent oxidase
Fawzi A. Razem and Mark A. Bernards1
Department of Biology, University of Western Ontario, London, Ontario N6A 5B7, Canada
Received 29 August 2002; Accepted 6 November 2002
Abstract
In response to wounding, potato tubers generate
reactive oxygen species (ROS) in association with
suberization. Immediately following wounding, an initial burst of ROS occurs, reaching a maximum within
30 to 60 min. In addition to this initial oxidative burst,
at least three other massive bursts occur at 42, 63
and 100 h post-wounding. These latter bursts are
associated with wound healing and are probably
involved in the oxidative cross-linking of suberin
poly(phenolics). The source of ROS is likely to be a
plasma membrane NADPH-dependent oxidase immunorelated to the human phagocyte plasma membrane
oxidase. The initial oxidative burst does not appear
to be dependent on new protein synthesis, but the
subsequent bursts are associated with an increase in
oxidase protein components. Oxidase activity is
enhanced in vitro by hydroxycinnamic acids and conjugates associated with the wound healing response
in potato.
Key words: NADPH-oxidase, potato, reactive oxygen
species, Solanum tuberosum, suberization, wound healing.
Introduction
When plants are wounded, they form a protective layer
next to the exposed surface to prevent dehydration and
potential penetration by opportunistic pathogens. This
physical barrier termed suberin comprises a speci®c cell
wall modi®cation characterized by both a poly(phenolic)
domain and wax-embedded poly(aliphatic) domain (reviewed in Bernards and Lewis, 1998; Bernards, 2002).
Suberin is also formed developmentally and is found in the
dermal cells of underground tissues, the Casparian band
and in the cork cells of bark tissue (Esau, 1977). The
1
oxidative coupling of the poly(phenolic) component of
suberin is thought to follow a peroxidase/H2O2 free radical
process analogous to that of ligni®cation (Kolattukudy,
1980). It was recently demonstrated in this laboratory that
H2O2 is essential for suberization in potatoes (Razem and
Bernards, 2002), but the origin of the H2O2 has remained
an open question.
The generation of reactive oxygen species (ROS), which
includes superoxide (O2. ±), hydrogen peroxide (H2O2), and
the hydroxyl radical (OH.), is ubiquitous in biological
systems, and occurs either through signal-regulated processes or as an unavoidable by-product of metabolic
reactions under both stress and normal conditions
(Bolwell, 1996). In plants, ROS are produced during the
cross-linking of cell wall components, after exposure to
high and low temperatures and light intensities, air
pollutants such as ozone, ultraviolet light, herbicides
and/or during pathogen attack and mechanical injury
(reviewed in Kuzniak and Urbanek, 2000). The rapidly
induced production of ROS under stress conditions (i.e. the
oxidative burst) initially results in the production of O2. ±,
which then disproportionates to H2O2 either spontaneously
or via superoxide dismutase. Both O2. ± and H2O2 have been
shown to act directly or indirectly in plant defence and
signal transduction (Bolwell, 1996; Kuzniak and Urbanek,
2000; Yoshioka et al., 2001; Vranova et al., 2002). The
conversion of O2. ± and H2O2 to OH., catalysed by transition
metals (Haber±Weiss reaction), accounts for the severe
toxicity of ROS in plants (Wojtaszek, 1997). Plants usually
keep the levels of ROS under tight control by the
production of scavenging enzymes and non-enzymatic
antioxidants (Wojtaszek, 1997; Kuzniak and Urbanek,
2000; Mùller, 2001; Vranova et al., 2002).
During the hypersensitive response (HR) to pathogen
attack and mechanical stress, ROS play a role in killing
pathogens and in eliminating the damaged cells (necrosis)
in a process widely considered to be analogous to the
To whom correspondence should be addressed. Fax: +1 519 661 3935. E-mail: [email protected]
Journal of Experimental Botany, Vol. 54, No. 384, ã Society for Experimental Biology 2003; all rights reserved
936 Razem and Bernards
defensive mechanism of mammalian phagocytes (Lamb
and Dixon, 1997). The key enzyme that contributes to ROS
formation in phagocytes is a plasma membrane-bound
NADPH-dependent oxidase (Segal and Abo, 1993), the
activation of which involves the assembly of at least three
cytosolic (p67phox, p47phox, p40phox) and two plasma
membrane-associated (gp91phox, p22phox) components
(Segal and Abo, 1993; Jones, 1994). Upon activation,
p67phox, p47phox, and p40phox translocate and dock with
gp91phox and p22phox at the plasma membrane. This process
involves the phosphorylation of at least one cytosolic
component (p47phox) by protein kinase C (PKC) (Jones,
1994).
A role for NADPH-dependent oxidases in the generation
of the H2O2 in plants is beginning to emerge. For example,
a plasma membrane NADPH-dependent oxidase homologous to the phagocyte oxidase has been demonstrated
during pathogen attack (Doke, 1985; Mehdy, 1994; Lamb
and Dixon, 1997; Xing et al., 1997; Yoshioka et al., 2001)
and ligni®cation (Ogawa et al., 1997; Ros BarceloÂ, 1998).
In this study the observations regarding the involvement of
an NADPH-dependent oxidase in the polymerization of the
poly(phenolic) domain of suberized potato tubers
(Bernards and Razem, 2001; Razem and Bernards, 2002)
have been extended to characterize the source and the
induction of the H2O2 generating system in wound-healing
potato tubers.
Materials and methods
Tissue wounding and plasma membrane isolation
Potato (Solanum tuberosum L. cv. Russet Burbank) tubers were cut
(i.e. wounded) transversely under sterile conditions and incubated at
25 °C for up to 7 d as described earlier (Bernards and Lewis, 1992).
Suberizing layers were collected at various times post-wounding by
gently separating them from the underlying, unsuberized tissue using
a razor blade (Bernards and Lewis, 1992). Suberized layers were
collected at the time of wounding (i.e. 0 h post-wounding (hpw))
followed by 1 h intervals up to 6 hpw and at 6 h intervals thereafter.
These time points were chosen to ensure a complete picture of the
timing of ROS production post-wounding. After collection,
suberized layers were frozen in liquid N2, ground in a pre-chilled
mortar and stored at ±20 °C until used. Microsomal fractions were
obtained by homogenizing ground tissue in extraction buffer (25
mM Tris-HCl buffer, pH 7.5 (3 ml g±1) containing 250 mM sucrose,
1.0 mM EDTA, 10 mM KCl, 1 mM MgCl2, 0.5 mM phenylmethylsulphonyl ¯uoride (PMSF), and 0.1 mM freshly prepared DTT). The
homogenate was ®ltered through four layers of cheesecloth and
centrifuged for 10 min (15000 g) at 4 °C. After buffer-exchange
using Econo-Pac 10 DG chromatography columns (BioRad) preequilibrated with extraction buffer (minus PMSF), the eluent was
centrifuged at 105 000 g for 60 min (4 °C) and the pellet used to
isolate plasma membranes by aqueous two phase partitioning (three
cycles) as described by Sandelius and Morre (1990) with minor
modi®cations by Xing et al. (1997). Membranes were reconstituted
in reconstitution buffer (10 mM potassium phosphate buffer, pH 7.8,
250 mM sucrose, and 1 mM DTT) at 10 ml g±1 initial tissue, and
either used immediately or frozen in liquid nitrogen and stored at
±20 °C until used. Cytosolic proteins were obtained from the 105
000 g supernatant using a concentration cell with a 10 kDa molecular
mass cut-off membrane, giving a ®nal protein concentration of
approximately 2 mg ml±1 as measured by the Bradford protein assay
(Bradford, 1976).
SDS-PAGE and Western blotting
Membrane and cytosolic proteins (approximately 10 mg) were
loaded on a discontinuous SDS-PAGE (10% separation gel) minigel
system (BioRad) and separated according to manufacturer's instructions. Proteins were transferred to nitrocellulose using a tankblotting chamber (BioRad), according to the manufacturer's instructions. Blots were blocked for 60 min at room temperature or
overnight at 4 °C in blocking buffer (20 mM Tris-HCl, pH 7.5, 150
mM NaCl, 0.05 Tween 20, and 5% milk powder). After washing
with washing buffer (TBS, 0.05% Tween 20), blots were incubated
with primary antibodies raised against gp91phox (1:1000), p67phox
(1:1000), p47phox (1:1000), p40phox (1:1000), (the kind gift of Dr A
Segal, University College London, UK and Dr Terence Murphy,
University of California, Davis, USA) for 60 min at room
temperature. Blots were washed four times (5 min each wash) in
washing buffer and subsequently incubated with secondary antibodies (goat:anti-rabbit conjugated with alkaline phosphatase) for 60
min. Blots were washed as above and ®nally with ddH2O (10 min). A
chemiluminescent detection system (ECLÔ Western Blotting
Detection Reagents) prepared shortly before ®lm development was
used according to the manufacturer's instructions. Film was exposed
for 30 s.
Enzyme assays
NADPH-dependent oxidase activity was measured indirectly using
the superoxide dismutase (SOD)-inhibited reduction of nitroblue
tetrazolium (NBT) as described by Van Gestelen et al. (1997). The
reaction was started by the addition of 0.1 mM NADPH and the
reduction of NBT (0.1 mM) was monitored at 530 nm at 30 °C.
Oxidase activity was calculated by taking the difference between
apparent reaction rates with and without SOD (50 units ml±1) in the
reaction mixture and using an extinction coef®cient of 12.8 mM±1
cm±1. Other additions to the reaction medium are discussed in the
text.
Peroxidase activity was measured spectrophotometrically in
reaction buffer (50 mM Tris-acetate buffer, pH 5.0, 100 mM KCl,
50 mM CaCl2) containing 30 mM H2O2, 0.1 mg tetramethylbenzidine-HCl (TMB), and 50 ml soluble protein in a 1 ml ®nal volume.
The oxidation of TMB was monitored at 450 nm.
Treatment with chloramphenicol
Thin tuber sections (0.5 cm) thick were soaked for 1 min in 1 mM
chloramphenicol before incubation at 25 °C. The browning surface
was carefully collected at the time points shown and processed as
above.
Results
.
Time-course generation of O2± post-wounding
The production of ROS was measured in wound-healing
potato tubers from 0±168 h post-wounding (hpw). The
wounding was carried out under sterile conditions to avoid
any possibility of pathogen interference and there was no
evidence of microbial contamination at any time during the
experiments (data not shown). The generation of O2. ±
sharply increased within 1 hpw (Fig. 1) and then gradually
declined. This initial oxidative burst appeared to be
ROS generation in suberizing potato tubers 937
Fig. 1. The effect of washing and treatment with chloramphenicol on
the initial oxidative
burst in wounded potato tubers. NAD(P)H.
dependent O2± generation was measured by the SOD-inhibited
reduction of NBT. Potato tubers were cut transversely into thin (0.5
cm) sections, and the browning surface carefully collected at the time
points shown. For the washed tissues (®lled circles), the discs were
soaked brie¯y (5 min total) in ddH2O prior to incubation. Unwashed
discs (open circles) were incubated directly after cutting or treated for
1 min with 1 mM chloramphenicol (®lled inverted triangles). Data
points are averages of triplicate measurements. Error bars represent 1
SD from the mean.
Fig. 2. Time-course of peroxidase and oxidase. activity in woundhealing potato tubers. NADPH-dependent O2± generation (®lled
circles) as measured by the SOD-inhibited reduction of NBT in
suberizing tissues collected at the time points shown. Peroxidase
activity
(®lled
inverted
triangles)
was
measured
spectrophotometrically (450 nm) using TMB. Data points are averages
of triplicate measurements. Error bars represent 1 SD from the mean.
independent of new protein synthesis (Fig. 1) and could be
signi®cantly inhibited by washing of the tuber tissue. The
initial oxidative burst was followed by three other, larger
bursts of O2. ± generation at approximately 42, 63, and 100
hpw (Fig. 2). The maximum activity of O2. ± generation
appeared to be at 42 hpw. By contrast, peroxidase activity
showed a continuous, gradual increase until it reached a
maximum at 100±120 hpw (Fig. 2).
.
Effect of DPI and NaN3 on in vitro generation of O2± in
wounded potato tubers
The generation of O2. ± could be reconstituted in vitro by the
addition of puri®ed plasma membranes to cytosolic protein
extracts prepared from suberizing tissues 42 hpw (Figs 3,
4). The complete system yielding maximum activity
(77.962.9 mmol g±1 tissue h±1) included plasma membrane
(PM) and cytosolic protein extracts, as well as all buffer
components. The production of O2. ± was evident with PM
alone, but at much lower rate (approximately 30%) than
with the complete system. It was not possible to measure
O2. ± production when PMs were excluded from the reaction
mixture (Fig. 3). The exclusion of NADPH from the
reaction medium decreased O2. ± generation in vitro by
more than 87% (Fig. 3). The addition of DPI, an NADPHdependent oxidase inhibitor (Segal and Abo, 1993)
decreased oxidase activity by over 80% whereas NaN3 (a
haem protein inhibitor) had no signi®cant effect on O2. ±
generation (Fig. 3). The inhibitory effect of NaN3 on
peroxidase was veri®ed in in vitro assays (data not shown).
.±
Fig. 3. Cell-free
. ± reconstitution of the O2 generating system. The
generation of O2 by puri®ed PM and cytosolic fractions of suberized
tissues prepared 42 hpw was measured by the SOD-inhibited reduction
of NBT in the presence and absence of oxidase and peroxidase
inhibitors and NADPH, as indicated. Diphenylene iodonium (DPI) (10
mM) and NaN3 (5.0 mM) were added 5 min before the reaction was
started by the addition of NADPH. Data are normalized to the activity
level of a complete assay (77.962.9 mmol g±1 tissue h±1) containing
PM proteins, cytosolic proteins and NADPH. Bars represent the
average of triplicate measurements. Error bars represent 1 SD from
the mean.
The cytosolic fraction prepared from tissues at 1 hpw could
not be substituted for that prepared from tissues at 42 hpw
(Fig. 4). Interestingly, O2. ± generation by PM prepared
from tissues at 1 hpw was not signi®cantly inhibited by
DPI (Fig. 4). The complete system yielding maximum
activity from tissues at 1 hpw (17.761.5 mmol g±1 tissue
938 Razem and Bernards
.
Fig. 4. Cell-free. reconstitution of a potato O2± generating system. The
±
generation of O2 by puri®ed PM prepared from tissue 42 hpw (a) and
1 hpw (b) reconstituted with cytosolic (cyto) extractions from tissues
collected 1 or 42 hpw in reaction buffer as indicated, was measured by
the SOD-inhibited reduction of NBT. Data are normalized to the
activity level of a complete assay for extracts containing PM proteins,
cytosolic proteins and NADPH from (a) 42 hpw (77.962.9 mmol g±1
tissue h±1) and (b) 1 hpw (17.761.5 mmol g±1 tissue h±1). In (b),
diphenylene iodonium (DPI) (10 mM) was added 5 min before the
reaction was started by the addition of NADPH. Bars represent the
average of triplicate measurements. Error bars represent 1 SD from
the mean.
.
Fig. 5. The in vitro
stimulation of O2± generation by phenolics. The
.±
generation of O2 was measured in suberized potato tubers 42 hpw by
the SOD-inibited reduction of NBT. Data are normalized to the
activity level of a complete assay (77.962.9 mmol g±1 tissue h±1)
containing PM proteins, cytosolic proteins and NADPH. Phenolics
(0.1 mM) were added 5 min before the start of the reaction. Bars
represent the average of triplicate measurements. Error bars represent
1 SD from the mean.
h±1) included PM and cytosolic protein extracts, as well as
all buffer components.
Effect of phenolics on in vitro generation of O2.± in 42
hpw suberized tissues
The addition of certain phenolics (0.1 mM) to the assay
mixture stimulated the generation of O2. ± in vitro. Of the
compounds tested, only chlorogenic acid, p-coumaroyltyramine, and ferulic acid signi®cantly increased the generation of O2. ± (Fig. 5). Pretreatment of tuber slices with 1
mM phenolics (ferulic acid, feyloylputrescine, p-coumaroyltyramine, or a mixture of the three), right after
wounding had little or no effect on the timing of the
secondary oxidative bursts (data not shown). Other compounds tested (e.g. coniferin, 2,3-dichloro-1,4-naphthoquinone, SDS, Triton X-100) had no effect on O2. ±
generation or gave inconsistent results (data not shown).
Western blotting analysis
Plasma membrane and cytosolic protein extracts from
suberizing tubers were probed with antibodies raised
against human p40phox, p47phox, p67phox, and gp91phox. The
anti- p40phox, p47phox, and p67phox antibodies (used as a
mixture) detected proteins with appropriate Mr in the
cytosolic fraction extracted at 42 hpw (Fig. 6). No proteins
in the PM fraction were recognized by these antibodies. By
contrast, anti-gp91phox antibodies recognized an appropri-
Fig. 6. Immunodetection of potato NADPH-dependent oxidase.
Aliquots of PM and cytosolic (Cyto) protein fractions extracted from
suberized tissues 42 hpw were separated by SDS-PAGE,
electrophoretically transferred onto nitrocellulose membranes and
incubated with primary antibodies raised against human plasma
membrane oxidase components. Blots were probed with either antigp91phox antibodies alone (lanes 1 and 2) or with a mixture of antip40phox, p47phox, and p67phox antibodies (lanes 3 and 4). Protein±
antibody complexes were detected using goat:anti-rabbit antibodies
conjugated with alkaline phosphatase and a chemiluminescence
detection system.
ROS generation in suberizing potato tubers 939
ate Mr protein in PM preparations of 42 hpw potato tubers,
but none in the cytosolic protein fraction.
All oxidase components tested (i.e. p40phox, p47phox,
p67phox, and gp91phox), were induced by wounding (Fig. 7),
albeit with different patterns. Whereas levels of p40phox
and p47phox increased steadily throughout the time-course,
p67phox appeared to reach a maximum at 24 hpw. By
contrast, the accumulation of gp91phox was slower than the
cytosolic components and appeared to peak at approximately 48 hpw, before declining slightly.
Discussion
Wound-induced suberization: pattern of ROS
production
In potato tubers, wounding triggers the formation of a
periderm (i.e. suberized layer) in the tissue immediately
below the site of damage. The initial deposition of wound
suberin in potato requires approximately 18 h (Lulai and
Corsini, 1998) and reaches a state in which the suberized
layer has suf®cient structural integrity to be peeled off
intact by 3 d post-wounding (Razem and Bernards, 2002).
Within 30 min of wounding, however, there is a burst in
the generation of ROS that reaches a peak in 1 h (Fig. 1).
This initial burst of ROS production probably results from
plant-derived signals, since wounding was carried out
under sterile conditions and was greatly reduced when cut
tubers were washed prior to incubation (Fig. 1). The initial
burst was not inhibited by DPI indicating that it may not
derive from an NADPH-dependent oxidase. This observation is supported by the fact that cytosolic proteins
extracted from tissue collected at 1 hpw cannot substitute
for those extracted from tissue collected at 42 hpw (Fig. 4)
and vice versa, even when added in excess (data not
shown). This is in apparent contrast to earlier reports
(Doke, 1985), but since the timing of the initial burst (i.e.
maximum at 1 hpw) is not directly relevant to suberin
macromolecular assembly, it was not studied further.
Nevertheless, it is possible that the H2O2 resulting from the
initial burst may act as a messenger molecule. The
insensitivity of the initial burst to protein synthesis
inhibitors (e.g. chloramphenicol, Fig. 1) suggests that it
is not dependent on any new protein synthesis, but rather
on pre-existing components.
After the initial oxidative burst, three subsequent, larger
bursts in ROS generation occurred at 42, 63, and 100 hpw
(Fig. 2). By contrast to the initial ROS burst, these
subsequent bursts were coincident with increases in
oxidase protein components, particularly gp91phox
(Fig. 7). Oxidase activity may be regulated by various
mechanisms including ¯uctuations in the levels of
endogenous activators and inhibitors, phosphorylation of
the cytosolic components and the subsequent assembly of
the oxidase components into a functional complex at the
plasma membrane. These regulatory mechanisms may
result in the oscillations in oxidase activity apparent in
Fig. 2, without an associated change in the amounts of
protein. The timing of the secondary ROS generation
coincides with phenolic polymerization and the establishment of structural integrity in the suberized layer. A
similar regulated generation of H2O2 has been reported for
lignifying Pinus taede cell cultures (Nose et al., 1995) and
suggests that the synthesis of H2O2 for the polymerization
of cell wall phenolics is under tight biosynthetic control
(Segal and Abo, 1993; Jones, 1994).
The origin of H2O2 in suberized potato tubers
Fig. 7. Time-course induction of potato NADPH-dependent oxidase
components. Aliquots of PM (top row) and cytosolic (bottom three
rows) protein fractions prepared from potato tissues 0±5 d postwounding and normalized on a mg g±1 tissue basis, were separated by
SDS-PAGE, electrophoretically transferred onto nitrocellulose
membranes and incubated with primary antibodies raised against
human plasma membrane oxidase components. Blots were probed
with either anti-gp91phox, p67phox, p47phox or p40phox antibodies as
indicated. Protein±antibody complexes were detected using goat:antirabbit antibodies conjugated with alkaline phosphatase and a
chemiluminescence detection system.
The origin of ROS has been a source of debate for many
years. Several mechanisms for ROS generation in plants
have been presented in the literature (see Wojtaszek, 1997,
for a review), including the NADPH-dependent oxidase
(Doke, 1985; Desikan et al., 1996; Groom et al., 1996;
Ogawa et al., 1997; Bolwell et al., 1998; Ros BarceloÂ,
1998; Tenhaken and RuÈbel, 1998; Amacucci et al., 1999;
Yoshioka et al., 2001; Simon-Plas et al., 2002), peroxidases (Bestwick et al., 1998; Bolwell et al., 1998), and
polyamine oxidase (Angelini and Federico, 1989). With
respect to H2O2 required for wound-induced suberization
in potato tubers, evidence for an NADPH-dependent
oxidase was recently provided based on the use of
inhibitors, particularly DPI (Bernards and Razem, 2001;
Razem and Bernards, 2002). Brie¯y it was demonstrated
that the production of H2O2 during wound-healing is
inhibited by DPI, a suicide inhibitor of the phagocyte
940 Razem and Bernards
plasma membrane NADPH-dependent oxidase (Segal and
Abo, 1993; Jones, 1994) at low mM levels (Bernards and
Razem, 2001). Although, DPI has also been shown to
inhibit the H2O2 generating activity of peroxidases, it is
only able to do so at higher concentrations (e.g. >1 mM)
(Bolwell et al., 1998).
The present results more clearly suggest the involvement of an NADPH-dependent oxidase in the generation of
ROS in suberizing potato tubers, and provides at least three
new lines of evidence. First, peroxidases and oxidases
follow a different time-course of induction during wound
healing (Fig. 2). Whereas peroxidase showed a relatively
linear and gradual increase in activity over 5 d postwounding, the oxidase showed three transient bursts
during the same time frame (Fig. 2). Furthermore, NaN3,
a peroxidase inhibitor, failed to block O2. ± generation by
puri®ed plasma membranes in vitro, even in the presence
of cytosolic proteins (Fig. 3). Second, antibodies raised
against phagocyte plasma membrane NADPH-dependent
oxidase proteins cross-reacted with plasma membrane and
cytosolic proteins of appropriate Mr, isolated from
suberizing potato tubers (Fig. 6). An oxidase showing
high homology with human gp91phox has recently been
cloned from tomato (Amacucci et al., 1999) and its activity
induced by fungal cell walls, arachidonic acid, and
salicylic acid in potatoes (Yoshioka et al., 2001).
Furthermore, an oxidase was shown to be the source of
ROS generation in tobacco (Simon-Plas et al., 2002), an
indication that the NADPH-dependent oxidase-mediated
generation of ROS is prevalent in the Solanaceae. Third,
the oxidase activity measured herein was found exclusively in plasma membrane preparations and maximum
activity was only achieved with complete systems containing PM and cytosolic protein fractions. Washing the
PM with high salt concentrations (e.g. 0.7 M NaCl) failed
to remove the activity (data not shown). Furthermore, the
exclusion of NADPH from the reaction medium signi®cantly decreased ROS generation (Fig. 3).
Role for hydroxycinnamic acids and associated
conjugates in ROS generation
In addition to a small amount of monolignols, the
poly(phenolic) domain of suberin contains a signi®cant
amount of hydroxycinnamic acids (Bernards et al., 1995;
Negrel et al., 1996). Preliminary metabolite pro®ling of
potatoes during suberization (Razem and Bernards, 2002),
indicates that p-coumaroyltyramine, may play a signi®cant
role in potato suberization since it accumulates in DPItreated tissues. Of the phenolic components tested in this
study, p-coumaroyltyramine had the greatest stimulatory
effect on the potato oxidase (Fig. 5). By contrast, pcoumaroylputrescine, which accumulates in wound-induced tubers whether treated with DPI or not, has no effect
on the activity of the oxidase (Fig. 5).
The activation of NADPH-dependent oxidases by
phenolic compounds is well established. For example,
ferulic acid, and caffeic acids showed a strong stimulatory
activity toward cauli¯ower in¯orescence oxidases
(Askerlund et al., 1987), while genistein, an iso¯avone
that accumulates in elicited soybean hypocotyls, is the
most potent activator of NoxII, a plasma membraneassociated oxidase of soybean (Graham and Graham,
1999). Thus a trend toward species relevant phenolics as
regulators of NADPH-dependent oxidases is emerging
and, in the case of wound-healing potato tubers, pcoumaroyltyramine is a potential candidate regulatory
molecule. While ferulic and chlorogenic acids also stimulated potato oxidase activity, the levels were much lower
(Fig. 5), even though they accumulated to the same extent
as p-coumaroyltyramine during wound healing (Razem
and Bernards, 2002).
Concluding remarks
Wounding of potato tubers triggers the massive production
of ROS that could serve different purposes. The initial
burst that follows mechanical injury is probably a spontaneous and less regulated production of ROS aimed at
providing a ®rst line of defence. It could also serve as a
second messenger for the synthesis of phenolics and
proteins. The subsequent, larger ROS bursts are probably
involved in the suberization process and occur several
hours apart from each other. It was possible to reconstitute
the ROS generating system in vitro and to stimulate its
activity by wound-induced hydroxycinnamic acids. Based
on these data, and earlier observations (Bernards and
Razem, 2001; Razem and Bernards, 2002), it is believed
that the source of ROS during wound-healing in potato
tubers is probably a plasma membrane-associated
NADPH-dependent oxidase that is immunorelated to
human plasma membrane oxidase. The failure of cytosolic
proteins isolated from tissue collected at 1 hpw to
compensate for those isolated from tissue collected at 42
hpw (Fig. 3), coupled with the apparent accumulation of
p40phox, p47phox and p67phox over the same time frame
strongly supports the conclusion that the generation of
H2O2 during wound-induced suberization is under strict
regulatory control.
References
Angelini R, Federico R. 1989. Histochemical evidence of
polyamine oxidation and generation of hydrogen peroxide in
the cell wall. Journal of Plant Physiology 135, 212±217.
Amacucci E, Gaschler K, Ward JM. 1999. NADPH oxidase
genes from tomato (Lycopersicon esculentum) and curly leaf
pondweed (Potamogeton crispus). Plant Biology 1, 524±528.
Askerlund P, Larsson C, Widell S, Mùller IM. 1987. NAD(P)H
oxidase and peroxidase activities in puri®ed plasma membranes
from cauli¯ower in¯orescences. Physiologia Plantarum 71, 9±19.
ROS generation in suberizing potato tubers 941
Bernards MA. 2002. Demystifying suberin. Canadian Journal of
Botany 80, 227±240.
Bernards MA, Lewis NG. 1992. Alkyl ferulates in wound healing
potato tubers. Phytochemistry 31, 3409±3412.
Bernards MA, Lewis NG. 1998. The macromolecular aromatic
domain in suberized tissue: a changing paradigm. Phytochemistry
47, 915±933.
Bernards MA, Lopez ML, Zajicek J, Lewis NG. 1995.
Hydroxycinnamic acid-derived polymers constitute the
polyaromatic domain of suberin. Journal of Biological
Chemistry 270, 7382±7386.
Bernards MA, Razem FA. 2001. The poly(phenolic) domain of
potato suberin: a non-lignin cell wall bio-polymer.
Phytochemistry 57, 1115±1122.
Bestwick CS, Brown IR, Mans®eld JW. 1998. Localized changes
in peroxidase activity accompany hydrogen peroxide generation
during the development of a nonhost hypersensitive reaction in
lettuce. Plant Physiology 118, 1067±1078.
Bolwell GP. 1996. The origin of the oxidative burst in plants.
Biochemical Society Transactions 24, 438±442.
Bolwell GP, Davies DR, Gerrish C, Auh C-K, Murphy TM.
1998. Comparative biochemistry of the oxidative burst produced
by rose and French bean cells reveals two distinct mechanisms.
Plant Physiology 116, 1379±1385.
Bradford MM. 1976. A rapid and sensitive method for the
quanti®cation of microgram quantities of protein utilizing the
principle of protein-dye binding. Analytical Biochemistry 72,
248±254.
Desikan R, Hancock JT, Coffey MJ, Neill SJ. 1996. Generation of
active oxygen in elicited cells of Arabidopsis thaliana is mediated
by a NADPH oxidase-like enzyme. FEBS Letters 382, 213±217.
.
Doke N. 1985. NADPH-dependent O 2± generation in membrane
fraction isolated from wounded potato tubers inoculated with
Phytophthora infestans. Physiological Plant Pathology 27, 311±
322.
Esau K. 1977. Anatomy of seed plants. New York: John Wiley and
Sons, 221±223.
Graham TL, Graham MY. 1999. Role of hypersensitive cell death
in conditioning elicitation competency and defense potentiation.
Physiological and Molecular Plant Pathology 55, 13±20.
Groom QJ, Torres MA, Fordham-Skelton AP, HammondKosack KE, Robinson NJ, Jones JDG. 1996. rbohA, a rice
homoloque of the mammalian gp91phox respiratory burst oxidase
gene. The Plant Journal 10, 515±522.
Jones OTG. 1994. The regulation of superoxide production by the
NADPH oxidase of neutrophils and other mammalian cells.
Bioassays 16, 919±923.
Kolattukudy PE. 1980. Biopolyester membranes of plants: cutin
and suberin. Science 208, 990±1000.
Kuzniak E, Urbanek H. 2000. The involvement of hydrogen
peroxide in plant responses to stresses. Acta Physiologiae
Plantarum 22, 195±203.
Lamb C, Dixon RA. 1997. The oxidative burst in plant disease
resistance. Annual Review of Plant Physiology and Plant
Molecular Biology 48, 251±275.
Lulai EC, Corsini DL. 1998. Differential deposition of suberin
phenolic and aliphatic domains and their roles in resistance to
infection during potato tuber (Solanum tuberosum L.) woundhealing. Physiological and Molecular Plant Pathology 53, 209±
222.
Mehdy MC. 1994. Active oxygen species in plant defense against
pathogens. Plant Physiology 105, 467±472.
Mùller IM. 2001. Plant mitochondria and oxidative stress: electron
transport, NADPH turnover, and metabolism of reactive oxygen
species. Annual Review of Plant Physiology and Plant Molecular
Biology 52, 561±591.
Negrel J, Pollet B, Lapierre C. 1996. Ether-linked ferulic acid
amides in natural and wound periderms of potato tuber.
Phytochemistry 43, 1195±1199.
Nose M, Bernards MA, Furlan M, Zajicek J, Eberhardt TL,
Lewis NG. 1995. Towards the speci®cation of consecutive steps
in macromolecular lignin assembly. Phytochemistry 39, 71±79.
Ogawa K, Kanematsu S, Asada K. 1997. Generation of
superoxide anion and localization of CuZn-superoxide
dismutase in the vascular tissue of spinach hypocotyls: their
association with ligni®cation. Plant Cell Physiology 38, 1118±
1126.
Razem FA, Bernards MA. 2002. Hydrogen peroxide is required
for poly(phenolic) domain formation during wound-induced
suberization. Journal of Agricultural and Food Chemistry 50,
1009±1015.
Ros Barcelo A. 1998. The generation of H2O2 in the xylem of
Zinnia elegans is mediated by an NADPH-oxidase-like enzyme.
Planta 207, 207±216.
Sandelius AS, Morre DJ. 1990. Plasma membrane isolation. In:
Larsson C, Mùller IM, eds. The plant plasma membrane,
structure, function, and molecular biology. Berlin: SpringerVerlag, 44±75.
Segal AW, Abo A. 1993. The biochemical basis of the NADPH
oxidase of phagocytes. Trends in Biochemical Sciences 18, 43±
47.
Simon-Plas F, Elmayan T, Blein JP. 2002. The plasma membrane
oxidase NtrbohD is responsible for AOS production in elicited
tobacco cells. The Plant Journal 31, 137±147.
Tenhaken R, RuÈbel C. 1998. Cloning of putative subunits of the
soybean plasma membrane NADPH oxidase involved in the
oxidative burst by antibody screening. Protoplasma 205, 21±28.
Van Gestelen P, Asard H, Caubergs RJ. 1997. Solubilization and
separation of a plant plasma membrane NADPH-superoxide
.
(O 2±)-synthase from other NAD(P)H-oxidoreductases. Plant
Physiology 115, 543±550.
Vranova E, Inze D, Breusegem FV. 2002. Signal transduction
during oxidative stress. Journal of Experimental Botany 53,
1227±1236.
Wojtaszek P. 1997. Oxidative burst: an early plant response to
pathogen infection. The Biochemical Journal 322, 681±692.
Xing T, Higgins VJ, Blumwald E. 1997. Race-speci®c elicitors of
Cladosporium fulvum promote translocation of cytosolic
components of NADPH oxidase to the plasma membrane of
tomato cells. The Plant Cell 9, 249±259.
Yoshioka H, Sugie K, Park H-J, Maeda H, Tsuda N, Kawakita
K, Doke N. 2001. Induction of plant gp91phox homolog by fungal
cell wall, arachidonic acid, and salicylic acid in potato. Molecular
Plant±Microbe Interactions 14, 725±736.