The Plant Cell, Vol. 12, 1849–1862, October 2000, www.plantcell.org © 2000 American Society of Plant Physiologists
Ozone-Sensitive Arabidopsis rcd1 Mutant Reveals Opposite
Roles for Ethylene and Jasmonate Signaling Pathways in
Regulating Superoxide-Dependent Cell Death
Kirk Overmyer,a Hannele Tuominen,a Reetta Kettunen,a Christian Betz,b Christian Langebartels,b
Heinrich Sandermann, Jr.,b and Jaakko Kangasjärvia,1
a Institute
of Biotechnology and Department of Biosciences, University of Helsinki, POB 56 (Viikinkaari 5 D), FIN-00014
Helsinki, Finland
b Institute of Biochemical Plant Pathology, GSF Research Center for Environment and Health, D-85764 Oberschleissheim,
Germany
We have isolated a codominant Arabidopsis mutant, radical-induced cell death1 (rcd1), in which ozone (O 3) and extracellular superoxide (O 2•⫺), but not hydrogen peroxide, induce cellular O 2•⫺ accumulation and transient spreading lesions. The cellular O 2•⫺ accumulation is ethylene dependent, occurs ahead of the expanding lesions before
visible symptoms appear, and is required for lesion propagation. Exogenous ethylene increased O 2•⫺-dependent cell
death, whereas impairment of ethylene perception by norbornadiene in rcd1 or ethylene insensitivity in the ethylene-insensitive mutant ein2 and in the rcd1 ein2 double mutant blocked O 2•⫺ accumulation and lesion propagation. Exogenous methyl jasmonate inhibited propagation of cell death in rcd1. Accordingly, the O 3-exposed
jasmonate-insensitive mutant jar1 displayed spreading cell death and a prolonged O 2•⫺ accumulation pattern.
These results suggest that ethylene acts as a promoting factor during the propagation phase of developing oxyradical-dependent lesions, whereas jasmonates have a role in lesion containment. Interaction and balance between these pathways may serve to fine-tune propagation and containment processes, resulting in alternate
lesion size and formation kinetics.
INTRODUCTION
Human activities have increased tropospheric ozone (O 3)
concentration two- to fivefold during the past 40 years (Kley
et al., 1999). O3 poses a twofold challenge to plants: photosynthesis and growth can be impaired by increased background concentrations, whereas acutely phytotoxic O 3
concentrations are additionally manifested as foliar lesions
in sensitive species and cultivars (Kangasjärvi et al., 1994;
Kley et al., 1999). Both of these modes of O 3 action potentially result in crop losses worth several million dollars annually. Historically, O3 has been regarded as a “wound stress”
that causes necrosis by oxidizing and damaging plasma
membranes (reviewed in Heath and Taylor, 1997). Recent
results, however, indicate that O3 responses resemble components of the hypersensitive response (HR) found in incompatible plant–pathogen interactions (Kangasjärvi et al.,
1 To whom correspondence should be addressed. E-mail jaakko.
[email protected]; fax 358-9-191-59-079.
1994; Sharma and Davis, 1997; Sandermann et al., 1998).
This similarity is most likely related to the occurrence of reactive oxygen species (ROS), such as superoxide anion radicals (O2•⫺) and H2O2 in the apoplast. O3-derived ROS
apparently trigger, by way of an as yet undescribed mechanism, an oxidative burst in the affected cells (Schraudner et
al., 1998; Pellinen et al., 1999; Rao and Davis, 1999). Similarly, an oxidative burst is one of the earliest responses of
plants to microbial pathogens and is an integral component
in HR-related cell death (Lamb and Dixon, 1997). Accumulation of H2O2 in response to O3 has been reported in tobacco
(Schraudner et al., 1998) and birch (Pellinen et al., 1999). In
Arabidopsis, O3-induced accumulation of both O 2•⫺ and
H2O2 has been reported (Rao and Davis, 1999). Sites of
ROS accumulation and visible lesions in these plants were
of a distinct size, suggesting that ROS production can function as a regulator of cell death not only in the HR, but also
under O3 stress.
O3 is known to activate ethylene and salicylate (SA) signal
transduction pathways leading to downstream responses,
such as antioxidant and antimicrobial defenses (Sharma et
al., 1996; Sharma and Davis, 1997; Sandermann et al.,
1850
The Plant Cell
1998). Increased ethylene emission from O 3-exposed plants
is an early, consistent marker for O 3 sensitivity (Tingey et al.,
1976; Mehlhorn and Wellburn, 1987; Wellburn and Wellburn,
1996; Tuomainen et al., 1997). O3 has been proposed to exert its toxicity through a chemical reaction with ethylene,
yielding toxic products that initiate a self-propagating lipid
peroxidation cycle (Elstner et al., 1985; Mehlhorn and
Wellburn, 1987). Alternatively, ethylene emission has been
postulated as a wounding symptom in O 3-exposed plants
(Heath and Taylor, 1997). However, ethylene has emerged
as a regulator of programmed cell death (pcd)—for example,
in pea carpel senescence (Orzáez and Granell, 1997), maize
endosperm development (Young et al., 1997), and root aerenchyma formation (Drew et al., 2000). Ethylene also triggers pcd in the accelerated cell death1 (acd1) mutant of
Arabidopsis (Greenberg and Ausubel, 1993) and is involved
in regulating cell death in plant–pathogen interactions (Bent
et al., 1992; Ciardi et al., 2000). These results suggest a potential new role for ethylene as a regulator of cell death in O 3
responses.
In O3-exposed plants, ethylene synthesis is a result of the
specific activation of 1-aminocyclopropane-1-carbocylic
acid (ACC) synthase and ACC oxidase genes (Tuomainen et
al., 1997) and is required for O3 damage (Mehlhorn and
Wellburn, 1987). Jasmonic acid (JA), on the other hand, can
protect tobacco plants from O3 when applied before O3
treatment (Örvar et al., 1997), and the O 3 sensitivity of a
poplar clone has been proposed to result from JA insensitivity (Koch et al., 1998). Thus, these two signaling pathways,
which act synergistically in induced systemic resistance
(Pieterse et al., 1998) and in regulating expression of pathogenesis-related (Norman-Setterblad et al., 2000) and woundinduced (O’Donnell et al., 1996) genes, appear to have opposing effects with regard to O3 sensitivity. However, the
detailed role or interaction of JA and ethylene signaling in
the regulation of ROS accumulation during O 3-induced or
other forms of ROS-dependent cell death has not been described.
Here, we report the identification and characterization of
an Arabidopsis mutant, rcd1 (for radical-induced cell death1),
that displays O3-inducible and O2•⫺-inducible lesion formation. We also show that O3 response mutants can be useful
in identifying interacting components of ROS-dependent
signaling pathways in plant–pathogen interactions, for example, to elucidate interaction between ROS and other
pathways that regulate the HR lesion formation in incompatible plant–pathogen interactions. Our results suggest that a
highly regulated induction of ethylene biosynthesis promotes cell death and that functional ethylene perception
and signaling are required for O2•⫺ accumulation, which is
responsible for the oxyradical-dependent cell death, ultimately leading to lesion propagation. JA signaling, on the
other hand, is involved in lesion containment. On the basis
of these results, we propose a model for the relative contribution of the different signaling pathways to ROS-driven lesion propagation and the process of lesion containment.
RESULTS
Isolation and Genetic Mapping of a Novel Arabidopsis
Mutant, rcd1,That Displays HR-like Lesion Development
in Response to O3
Approximately 14,000 individual M 2 plants, grown from ethyl
methanesulfonate–mutagenized Columbia (Col-0) seed,
were exposed for 3 days to 250 nL L ⫺1 O3 (8 hr/day) at a
density of 3000 plants m⫺2. Fifty-six individuals displaying
O3-induced lesions on rosette leaves were identified. In the
subsequent screenings at low density, four lines representing independent loci and distinct patterns and kinetics of
lesion formation were selected from 11 lines displaying consistent O3-sensitive phenotypes. When these four lines were
crossed with each other and the O 3 sensitivity was assayed
in the F1 progeny, no O3-sensitive F1 lines were detected, indicating that these mutants represent independent loci. Detailed characterization of the most sensitive mutant, rcd1, is
presented here. To determine its mode of inheritance, rcd1
was crossed with Col-0. In the F2 progeny, three phenotypes,
both parentals and an intermediate, segregated as a single
codominant Mendelian trait, 1:2:1 (36:66:46, ␹2 ⫽ 3.08, P ⫽
0.214). The F2 progeny from the rcd1 ⫻ Landsberg erecta
(Ler) were used for mapping with polymerase chain reaction–based microsatellite and cleaved amplified polymorphic sequence markers that cover all five chromosomes. Using
32 homozygous, O3-sensitive rcd1 individuals from the F2
progeny, we were able to position the locus in chromosome
1 at 53 ⫾ 2.2 centimorgans in the recombinant inbred map, 2
centimorgans from a microsatellite in bacterial artificial chromosome clone F23M19. No mutants with similar or related
phenotypes had previously been assigned to this region.
O3-induced lesions of rcd1 initiate along the leaf margins
and spread inward through intervascular tissue. Normally,
visible symptoms appear first as dark, water-soaked spots
12 hr after the onset of O3 treatment (250 nL L⫺1) on middleaged rosette leaves. When whole tissues are affected by
many coalescing lesions, tissue collapse joins medially and
progresses toward the leaf base and tip. In this case, damage is already visible at 3 hr as a loss of leaf turgor. By 24 hr,
as shown in Figure 1A, lesion progression in rcd1 had
halted, and the lesions had developed into well-defined dry
patches of dead tissue. Lesion initiation was independent of
the duration of O3 exposure (250 nL L⫺1) between 1 and 8
hr; only the extent of lesion formation increased with time.
No visible symptoms were found on Col-0 (Figure 1A). However, microscopic examination of Col-0 tissue revealed individual cells and small clusters of cells that had died as a
result of O3 exposure. Lesions in rcd1 were not triggered by
application of SA (or its active analog benzo[1,2,3]thiadiazole-7-carbothioic acid [BTH]) or ethylene, high-intensity
light, UV-B radiation, wounding, or alterations in daylength
(data not shown)—all factors known to trigger lesions in
other phenotypically related mutants.
Ethylene and Jasmonate in Oxidative Cell Death
1851
O3 Induces O2•ⴚ Accumulation in Front of the Spreading
Lesion in rcd1
Figure 1. Progression of Damage and Accumulation of O2•⫺ in
O3-Exposed Arabidopsis.
(A) Leaf damage in Col-0 and rcd1 24 hr after the onset of a 6-hr exposure to 250 nL L⫺1 O3.
(B) O2•⫺ accumulation in O3-exposed leaves. Col-0 and rcd1 were
exposed to O3 as above. Plants were removed from the chamber at
the times indicated, and detached leaves were infiltrated with nitroblue tetrazolium (NBT). The presence of the purple formazan precipitate indicates the location and extent of O2•⫺ accumulation.
(C) Effect of inhibitors on O2•⫺ accumulation. Leaves of rcd1 were exposed to 4 hr of O3 (250 nL L⫺1) or 4 hr of O3 (250 nL L⫺1) followed
by norbornadiene (NBD; 30 ␮L L⫺1), as indicated (⫺NBD or ⫹NBD).
After O3 exposure, leaves were infiltrated with NBT. The specificity
of NBT staining for O2•⫺ and the cellular origin of O2•⫺ were verified
by infiltrating O3-exposed (250 nL L⫺1) leaves with NBT in combination with superoxide disumutase (SOD) or diphenylene iodonium
(DPI) (20 ␮M), an inhibitor of flavin-containing oxidases, as indicated. The role of ethylene insensitivity in O2•⫺ accumulation was
elucidated by exposing Col-0 and ein2 to a high concentration (400
nL L⫺1) of O3. SOD- and DPI-treated leaves were stained at 4 hr, and
leaves were treated (or not) with NBD (⫹/⫺) and stained at 8 hr.
Col-0 and ein2 leaves exposed to 400 nL L⫺1 O3 were stained immediately after the 2-hr exposure.
Lesion development in rcd1 is transiently comparable with
lesion propagation in lsd1 (for lesion simulating disease
resistance1), in which extracellular O2•⫺ production precedes lesion spread (Jabs et al., 1996). This prompted us to
assay cellular O2•⫺ accumulation, as shown in Figure 1B, in
O3-exposed rcd1 by monitoring the precipitation of purple
formazan when reacting nitro blue tetrazolium (NBT) with
O2•⫺. To exclude reactions of NBT with O 3-derived O2•⫺
(Runeckles and Vaartnou, 1997), the plants were postcultivated in pollutant-free air for at least 15 min after exposure
to O3. No NBT precipitation could be detected in clean air–
grown rcd1 or Col-0 (data not shown). However, 2 to 4 hr after the onset of O3 exposure, NBT precipitation was visible
in both Col-0 and rcd1, although the distribution in each
strain differed. In Col-0, NBT precipitation was scattered
throughout the leaf, with greater amounts at the tip and margins, whereas in rcd1, staining was concentrated first (at 2 hr)
in the region in which the lesions typically initiated and later
(at 4 to 8 hr) just beyond the boundary of the collapsing tissue. No NBT staining was observed in Col-0 at 8 hr or later
time points. In rcd1, as illustrated in Figure 2, precipitation of
NBT by O2•⫺ continued in front of lesion expansion for as
long as 12 hr after the onset of the 6-hr O 3 treatment. NBT
precipitation clearly was the result of O 2•⫺ accumulation because staining was abolished when superoxide dismutase
(SOD) was coinfiltrated into the leaves (Figure 1C).
We used ion leakage, an indicator of plasma membrane
damage, as a quantitative measure of the extent of cell
death. When ion leakage was monitored in O 3-exposed
plants, marked differences were found between Col-0 and
rcd1 (Figure 3A). In rcd1, ion leakage increased 10-fold during O3 exposure and showed a further increase, even in the
absence of O3, between 6 and 12 hr, when the lesions were
spreading. In contrast, O3 induced only a small, transient increase in Col-0 during the exposure period.
Extracellular O2•–, but Not H2O2, Is Both Necessary and
Sufficient to Trigger Propagation of Cell Death in rcd1
To elucidate the role of O2•⫺ and H2O2 in lesion development, we infiltrated Col-0 and rcd1 leaves with an extracellular O2•⫺-generating system, xanthine/xanthine oxidase
(X/XO); with increasing concentrations of H 2O2; and with an
H2O2-generating system, glucose/glucose oxidase (G/GO)
(Jabs et al., 1996; Alvarez et al., 1998). O 2•⫺ production by
X/XO lasted for ⵑ3 hr and induced cell death in Col-0 and
rcd1 with different magnitudes and kinetics (Figure 3B). In
X/XO-infiltrated Col-0 plants, ion leakage increased to ⵑ25%
during the first 4 to 8 hr and remained at that value during
the experiments. Changes in ion leakage were more dramatic in rcd1, which showed a second increase to as much
as 65% between 8 and 24 hr (Figure 3B). Thus, O 3 could be
1852
The Plant Cell
Half-maximal reduction was obtained with 2 ␮M DPI, maximal reduction (ⵑ45%) with 5 ␮M. We conclude from these
data that both O3 and O2•⫺ from X/XO initiate active cellular
O2•⫺ production that continues in rcd1 after exogenous O2•⫺
ceases. Furthermore, this O2•⫺ is both necessary and sufficient to propagate cell death in rcd1.
Incompatible Bacterial Pathogen Induces Extensive Cell
Death during the HR in rcd1
Figure 2. Spatial Relationship between O2•⫺ Accumulation and
Spreading Lesions.
(A) The overall staining pattern of an O3-exposed (6 hr ⫻ 250 nL L⫺1)
rcd1 leaf stained for O2•⫺ at 8 hr. The leaf area with collapsed cells is
indicated with an asterisk.
(B) Light microscopy showing details of the boxed region in (A).
(C) UV epifluorescence of the same region as in (B), showing autofluorescent phenolic compounds (arrowhead) defining the boundary
of lesioned tissue. The O2•⫺ accumulation occurred in the healthy
tissue adjacent to the expanding lesion.
functionally replaced by O2•⫺ produced with X/XO. When
the X/XO-produced O2•⫺ was removed by coinfiltration with
MnSOD, cell death in rcd1 was not triggered to any greater
extent than in Col-0 (Figure 3C). Therefore, O 2•⫺ alone
seems sufficient to induce lesion formation. Physiological
concentrations of H2O2 did not induce cell death in rcd1. Infiltration of rcd1 leaves with an active H2O2-producing system (G/GO) did not cause increased cell death in either rcd1
or Col-0 in the range of steady state H 2O2 production between 5 and 250 ␮M when compared with buffer alone (data
not shown). Furthermore, direct treatment of excised rcd1 or
Col-0 leaves with H2O2 solutions up to 10 mM showed no
substantial increase in cell death in either strain compared
with the buffer control. Greater H2O2 concentrations (50 and
100 mM) resulted in equal increases in cell death in both
Col-0 and rcd1 (Figure 3D).
The role of endogenous O2•⫺ accumulation triggered by
O3 during the early phases of lesion formation was studied
further by inhibition experiments, specifically with diphenylene iodonium (DPI), an inhibitor of ROS accumulation in
Arabidopsis (Jabs et al., 1996; Alvarez et al., 1998). Infiltration of rcd1 leaves with DPI after a 2-hr O3 exposure prevented NBT precipitation (Figure 1C) and reduced leaf
damage in a concentration-dependent manner (Figure 4).
An oxidative burst, the active production of apoplastic O 2•⫺
and H2O2, is one of the earliest plant responses in incompatible interactions and is integrally involved in regulating hypersensitive cell death (Lamb and Dixon, 1997). To analyze
whether incompatible interaction with a bacterial pathogen
triggers spreading lesions in rcd1, we infiltrated 3-week-old
rosettes of Col-0 and rcd1 with increasing concentrations of
Pseudomonas syringae pv tomato, strain DC3000, carrying
the avirulence gene avrB on a plasmid (Bent et al., 1992).
The extent of tissue collapse caused by the HR was quantified by measuring ion leakage. As seen in Table 1, infiltration
of rosette leaves with bacterial concentrations from 10 4 to
107 colony-forming units (cfu) mL⫺1 induced substantially
higher ion leakage in rcd1, whereas in mock treatments, no
difference between Col-0 and rcd1 was visible. The difference between Col-0 and rcd1 regarding ion leakage was of
similar magnitude for every concentration used. Thus, we
concluded that spreading lesions are triggered in rcd1 also
by an incompatible interaction, which involves ROS production in the oxidative burst.
Oxygen Radicals Induce Pathogen Defense Genes but
Not Antioxidant Genes in rcd1 and Col-0
Expression of antioxidant, antimicrobial, and other stressrelated genes was analyzed by using 92 expressed sequence tags (ESTs) and cDNA clones in cDNA macroarrays.
A subset of the results is shown in Table 2. No major differences were found between Col-0 and rcd1 cultivated in
clean air except for lower amounts of catalase (CAT3) and
defensin (PDF1.2) mRNAs in rcd1 (Table 2). O3 induced major changes in the expression of PR-1, two glutathione
S-transferase genes (GST1 and GST2), and catalases 1 and
3 in both rcd1 and Col-0. O3-induced increases in GST1 and
GST2 mRNAs were markedly greater in rcd1 (26-fold) than in
Col-0 (nine- to 12-fold). CAT3 transcripts increased in both
Col-0 and rcd1 such that by 8 hr, they were approximately
equivalent. SOD isoforms showed minor responses, or their
transcript levels were decreased (chloroplastic Cu/ZnSOD)
in both Col-0 and rcd1. We concluded that because the
transcript levels for enzymes detoxifying O 2•⫺, H2O2, and
lipid hydroperoxides are approximately equal in Col-0 and
rcd1, then the radical-induced cell death in rcd1 is not a result of decreased expression of antioxidative genes.
Ethylene and Jasmonate in Oxidative Cell Death
1853
Ethylene Signaling Is Required for Accumulation of O 2•–,
Which Drives Lesion Propagation
Similar concentrations of O3 and O2•⫺ (X/XO treatment) in
Col-0 and rcd1 plants led to markedly more cell death in
rcd1 (Figures 1B and 2), suggesting that the initial amount of
ROS formation from O3 or X/XO is not the only factor involved in the initiation of lesion propagation. Several biochemical variables have been studied as the basis for O 3
sensitivity. Ethylene emission has been found to be the most
consistent response to O3 (Mehlhorn and Wellburn, 1987;
Wellburn and Wellburn, 1996; Tuomainen et al., 1997).
Figure 5 shows results of detailed studies on ethylene biosynthesis in Col-0 and rcd1. Of the five Arabidopsis ACC
synthase genes analyzed—AT-ACS1, 2, 4, 5, and 6—only
the isoform AT-ACS6 was responsive to O3. AT-ACS6 was
already highly induced in rcd1 at 30 min after the onset of
O3 exposure (Figure 5A). Transcript levels were at their maximum at 1 to 2 hr and decreased thereafter. Enhanced activation of AT-ACS6 in rcd1 was also reflected in increased
ACC concentrations (Figure 5B). Similarly, the amounts of
ACC oxidase transcripts (Table 2) and ethylene evolution
(Figure 5C) were also three- to fivefold higher in rcd1 than in
Col-0, and in contrast to Col-0, ethylene evolution in rcd1
continued past the period of O3 exposure.
It has been proposed (Elstner et al., 1985; Mehlhorn and
Wellburn, 1987) that a chemical reaction between ethylene
and O3 could produce water-soluble, highly reactive radicals
that could initiate a self-propagating peroxidative cycle,
thus causing ethylene-mediated damage without the action
of ethylene signaling. Results of the experiments addressing
this question are shown in Figure 6. rcd1 was exposed first
to O3, followed by incubation either in clean air or in the
presence of an antagonist of ethylene action, norbornadiene
Figure 3. Progression of O3-Induced and O2•⫺-Induced Leaf Damage in Col-0 and rcd1.
(A) O3-induced cell death. Plants were exposed to O3 (indicated by
the black bar) and postcultivated in pollutant-free air. The extent of
cell death was measured as relative ion leakage (percentage of total
ions ⫾SE).
(B) O2•⫺-induced cell death. Leaves from clean air–grown plants
were detached and vacuum-infiltrated with the O2•⫺-generating system xanthine and xanthine oxidase (X/XO). The black triangle under
the x axis indicates the approximate duration and intensity of O2•⫺
synthesis by X/XO. The time course of cell death, measured as relative ion leakage (⫾SE), was monitored for 24 hr.
(C) Inhibition of cell death by SOD. Leaves of Col-0 and rcd1 were
infiltrated with the O2•⫺-generating system X/XO, MnSOD (440 units
mL⫺1), or infiltration buffer, as indicated. Reagents were included (⫹)
or not (⫺) in each treatment as indicated. Cell death was measured
as ion leakage (⫾SE) after 20 hr.
(D) H2O2 and cell death. Leaves from clean air–grown plants were
detached and vacuum-infiltrated with increasing concentrations of
H2O2. Cell death in Col-0 and rcd1 was measured as relative ion
leakage (⫾SE) after 20 hr.
1854
The Plant Cell
Figure 4. Role of Endogenous O2•⫺ Production in O3-Induced Lesion Formation.
Leaves of rcd1 were exposed to 250 nL L⫺1 O3 for 2 hr and then infiltrated with indicated concentrations of the flavin oxidase inhibitor
DPI or with buffer alone. Damaged leaf area was assessed at 24 hr.
The effect of DPI is expressed as the percentage of damaged leaf
area (⫾SE) of plants infiltrated with buffer.
(NBD). After rcd1 was treated for 4 hr with O3 (250 nL L⫺1), ion
leakage had increased slightly above that of the clean air–
grown control plants (Figure 6A). In plants postcultivated in
clean air after the exposure, ion leakage showed an additional
fourfold increase. In contrast, blocking of ethylene perception
with NBD treatment during postcultivation inhibited any additional increase in ion leakage (Figure 6A) and led to disappearance of the NBT staining-detectable O2•⫺ accumulation
(Figure 1C). The effect of ethylene on O2•⫺ accumulation and
cell death appeared thus to act by way of ethylene perception, not through a chemical reaction between ethylene and O 3.
Ethylene Insensitivity Confers O 3 Tolerance
The role of ethylene signaling in O 3-induced tissue damage
was additionally addressed with the ethylene-insensitive
mutant ein2 (Guzman and Ecker, 1990). After exposure to
250 nL L⫺1 O3 for 4 hr, rcd1 displayed more cell death than
did Col-0 and ein2 (Figure 6B). When O3 was followed by 4
hr of ethylene (12 ⫾ 2 ␮L L⫺1), a marked increase in cell
death was evident in both rcd1 and Col-0 but not in ein2.
Ethylene treatment alone did not induce O 2•⫺ accumulation
or lesion formation in rcd1 (not shown). Furthermore, when
the plants were exposed to 400 nL L ⫺1 O3, leaf damage in
both Col-0 and rcd1 was extensive (Figure 6B), and predominant accumulation of O2•⫺ was apparent in Col-0 (Figure 1C). Under the same conditions, ein2 showed just minor
O2•⫺ accumulation at 2 hr (Figure 1C) and no later leaf damage. Enhanced ethylene synthesis (by two- to fourfold),
however, was evident in both Col-0 and ein2 under this regime (Figure 6B, insert), indicating that O 3 enters the leaves
and elicits a response in both strains. To address whether
O2•⫺ in combination with endogenously produced ethylene
has similar effects as O3 and exogenous ethylene, leaves of
Col-0, rcd1 and ein2 were infiltrated with X/XO in the pres-
ence and absence of the ethylene precursor ACC (Figure
6C). X/XO-induced cell death, determined as ion leakage after
20 hr, increased in Col-0 (by twofold), rcd1 (by threefold), and
ein2 (by fivefold) in comparison with buffer controls. Although
ACC alone had a negligible effect, O 2•⫺ and ACC together
had synergistic effects over O2•⫺ alone in Col-0 (3.5-fold increase) and rcd1 (4.5-fold increase) but not in ein2.
The position of rcd1 relative to the ethylene pathway was
further elucidated by crossing rcd1 with ein2 and identifying
the F2 individuals that were homozygous for both mutations.
Epistatic relationships between the loci were analyzed by
exposing the double mutant to 4 hr of O 3 followed by 4 hr of
clean air in a setup similar to that used with NBD. Figure 7
shows that no differences were detected in ion leakage in
O3-exposed Col-0 or ein2 during the course of the experiments. In O3-exposed rcd1, ion leakage increased from
control values to 9% leakage during the 4-hr O 3 exposure
and continued to increase to 16% during the subsequent 4
hr in clean air. The rcd1 ein2 double mutant showed an increase in ion leakage during the 4-hr O 3 exposure similar to
that for rcd1. However, the spread of cell death in clean air
after O3 was absent, and by 8 hr, ion leakage had decreased
back to the control value (Figure 7). Ethylene insensitivity in
the double mutant thus prevented the occurrence of
spreading cell death after O3 exposure in a manner similar to
NBD treatment (Figure 6A) without affecting the increase in
cell death during O3 exposure. Thus, we concluded that
both a highly regulated increase in ethylene evolution and
functional ethylene signaling are required for the amplification of cell death.
JA Signaling Is Involved in Containment of HR-like
Lesion Development and Counteracts the
Ethylene-Dependent, O2•–-Driven Lesion Propagation
In rcd1, cell death did not spread throughout the leaf; lesion
propagation was contained between 12 and 24 hr. We
Table 1. Avirulent Pathogen Triggers Spreading Cell Death in rcd1a
cfu/mL
Col-0
rcd1
Mock
104
105
106
107
9.4 ⫾ 0.5 ab
12.6 ⫾ 0.5 ab
17.4 ⫾ 1.3 bc
41.1 ⫾ 3.3 d
59.9 ⫾ 3.3 e
9.2 ⫾ 0.7 a
20.9 ⫾ 2.5 bc
25.5 ⫾ 2.4 c
53.2 ⫾ 1.7 e
71.9 ⫾ 3.0 f
a Cell death is measured as relative ion leakage (percentage of total
ions) from the rosette leaves of Col-0 and rcd1 at 20 hr after infiltration with varying concentrations of the avirulent bacterial pathogen P. s.
tomato DC3000 carrying the avirulence gene avrB in a plasmid.
b Means (⫾SE, n ⫽ 5) followed by the same letter are not significantly
(P ⬍ 0.05) different according to Tukey’s multiple range test.
Ethylene and Jasmonate in Oxidative Cell Death
Table 2. Stress and Defense Gene Transcript Levelsa
Geneb
EST Stock/
Accession No.
PR1
CHIb
GST1
GST2
AT-ACO2
MNSOD1
Cu/ZnSOD1
Cu/ZnSOD2
FeSOD1
APX1
GR
GPX2
catalase3
catalase1
AOS
PDF1.2
M90508
92G1T7
206N21T7
91H22T7
G2B6T7
109J19T7
2G11T7P
161I22T7
34D9T7
135D24T7
185P3T7
190H7T7
134I1T7
118M15T7
230J8T7
37F10T7
Air
Ozone
Col-0
rcd1
Col-0
rcd1
0.1c
0.1
0.4
0.6
0.3
1.6
2.5
4.5
1.0
1.9
1.1
2.0
5.3
0.3
0.8
1.3
0.1
0.1
0.4
0.5
0.5
2.3
3.3
5.1
2.2
2.2
1.7
2.2
2.2
0.3
0.6
0.4
2.0⇑d
0.2
5.0⇑
6.5⇑
0.6
2.0
3.0
2.6
0.4
1.7
0.3
0.7
12↑
2.2⇑
0.5
1.5
2.3⇑
0.2
11⇑
13⇑
1.6↑
1.6
4.2
1.6
0.2
2.7
0.8
1.9
10⇑
1.0↑
0.3
0.5
1855
counteracting or preventing the lesion propagation executed by the ethylene-dependent O 2•⫺ accumulation in
rcd1. Furthermore, this suggests that rcd1 is not JA insensitive. Table 3 shows further verification: Roots of rcd1 were
sensitive to growth inhibition by MeJA, whereas JAinsensitive jar1 was not affected. In the presence of 10 ␮M
MeJA, root growth inhibition in rcd1 was similar to that in
Col-0 (61 and 66%, respectively), whereas in jar1, no substantial inhibition occurred. Thus, based on JA inhibition of
a Expression of selected stress- and defense-related genes in wildtype Col-0 and the rcd1 mutant in ambient air and at 8 hr after the
beginning of a 6-hr O3 exposure (250 nL L⫺1). A complete list of the
genes used can be seen at http://www.biocenter.helsinki.fi/bi/koivu/
Arabidopsis/arrays.html.
b PR, pathogenesis-related protein; CHIb, basic chitinase; GST, glutathione S-transferase; ACO, ACC oxidase; SOD, superoxide dismutase; APX, ascorbate peroxidase; GR, glutathione reductase;
GPX, glutathione peroxidase; AOS, allene oxide synthase; PDF, plant
defensin.
c The values depict mRNA abundance relative to the mean of mRNA
abundance of actin genes ACT2 (EST accession number 179M16T7)
and ACT8 (accession number 179P1T7), which were shown to be
constitutively expressed by RNA gel blots.
d ↑, induction two- to fourfold; ⇑, induction fivefold and above. Similar results have been seen using mRNA from three independent experiments.
looked at other possible signaling pathways that could be
involved in lesion containment. Kangasjärvi et al. (1994) proposed that JA could act as signaling molecule in plant–O 3
interactions. Furthermore, Table 2 shows that fewer transcripts of defensin (PDF1.2), a gene that is coregulated by
ethylene and JA (Penninckx et al., 1998), were present in
both clean air–grown and O3-exposed rcd1. This suggests
that JA biosynthesis or signaling, or interaction of JA and
ethylene signaling, could be affected in rcd1, because the
ethylene biosynthesis and signaling pathways in rcd1 are
functional.
Results of the experiments addressing the role of JA in lesion containment are shown in Figure 8. rcd1 was first
treated with methyl jasmonate (MeJA) after 4 hr of O 3 exposure in an experiment similar to that used with the ethylene
antagonist NBD (cf. with Figure 6A). Exposure of rcd1 to 1.4
␮M MeJA after O3 exposure inhibited lesion propagation
from 4 to 8 hr (Figure 8A). This suggests that JA treatment is
Figure 5. O3 Induction of Ethylene Biosynthesis.
(A) Time course of O3 induction of the ACC synthase gene ATACS6. Transcript levels shown are relative to the hybridization signal
of control samples taken at the same times.
(B) Time course of concentrations of the ethylene precursor ACC
(⫾SE).
(C) Ethylene evolution (⫾SE) in clean air–grown and O3-exposed
Col-0 and rcd1. O3 exposure (250 nL L⫺1 for 6 hr) was followed by
postcultivation in pollutant-free air.
Black bars indicate the duration of exposure. FW, fresh weight.
1856
The Plant Cell
root elongation, rcd1 cannot be regarded as JA insensitive.
rcd1 is not deficient in JA biosynthesis either, because exposure of rcd1 to O3 led to much greater concentrations of
JA than in Col-0 (not shown).
To elucidate the role of JA in lesion propagation and containment, we further utilized the JA-insensitive mutant jar1
(Staswick and Howell, 1992). O3-exposed jar1 developed
spreading lesions similar to those of rcd1 (Figure 8B). Furthermore, in the margins of the spreading lesion, jar1 exhibited O2•⫺ accumulation (Figure 8D) in a manner similar to
rcd1. The same initial O2•⫺ accumulation as in Col-0 at 1 to
2 hr seemed sufficient to initiate lesion propagation in jar1.
However, the O3-induced lesion propagation in jar1 was not
accompanied by similar high amounts of ethylene as were
seen in rcd1 (Figure 8E). This suggests that jar1 is less efficient in lesion containment, in contrast to rcd1, which
shows increased lesion propagation. Thus, these results
suggest that JA signaling is required in processes that prevent the O2•⫺-dependent lesion propagation or that contain
lesion spread.
DISCUSSION
Figure 6. Role of Ethylene Perception and Signaling in the Formation of O3-Induced Leaf Damage.
(A) Inhibition of ethylene perception. rcd1 plants were exposed to O3
(250 nL L⫺1) for 4 hr (indicated by the black bar) and subsequently
exposed to clean air or to the ethylene antagonist NBD (30 ␮L L⫺1).
Cell death was measured as relative ion leakage (percentage of total
ions ⫾SE) after 0, 4, and 8 hr.
(B) O3 exposure and external ethylene. Col-0, rcd1, and the ethylene-insensitive mutant ein2 were exposed to clean air, 4 hr of O3
(250 or 400 nL L⫺1), or 4 hr of O3 followed by 4 hr of ethylene (12 ␮L
L⫺1). Cell death was measured as ion leakage (⫾SE) after 8 hr. Insert
shows ethylene evolution from Col-0 and ein2 in clean air (C) and after 4 hr of 400 nL L⫺1 O3.
(C) O2•⫺ and exogenous ACC. Leaves of Col-0, rcd1, and ein2 were
infiltrated with the O2•⫺-generating system X/XO, with the ethylene
precursor ACC (50 ␮M), or with an infiltration buffer. Cell death was
measured as relative ion leakage (percentage of total ions ⫾SE) after
20 hr.
(⫹), reagents included; (⫺), reagents not included.
The novel rcd1 mutant described here is sensitive to extracellular oxygen radicals and shows extensive formation of
HR-like lesions and activation of several pathogen defense–
related processes in response to ROS. Phenotypically, rcd1
is in part similar to the acd (Greenberg and Ausubel, 1993)
and lsd (Dietrich et al., 1994) mutants that have been utilized
to identify the components regulating hypersensitive cell
death downstream of pathogen infection. In particular, the
“runaway” spread of cell death in rcd1 during the first 8 to
12 hr is reminiscent of that in lsd1, in that a front of O2•⫺ accumulation has been found at lesion boundaries in living tissues in both mutants. However, in addition to avirulent
Pseudomonas, which induces an oxidative burst in the affected cells, O2•⫺ and O3 are the only treatments known
thus far to trigger the spreading lesion formation in rcd1,
whereas various other factors, such as SA, BTH, ethylene,
or growth conditions, trigger lesion formation in lsd or acd
mutants. Thus, rcd1 seems to define a component of the HR
related to the role of ROS in cell death regulation. The chromosomal location of rcd1, in addition, has not been described for any acd, lsd, antioxidant, or disease-related
mutants. Therefore, rcd1 describes a novel class of ROSresponsive lesion-mimic mutants.
rcd1 Appears to Be Deficient in the Control of
Lesion Propagation
The phenotypes described in various lesion-mimic mutants
suggest that three separate processes—initiation, propagation,
Ethylene and Jasmonate in Oxidative Cell Death
Figure 7. Progression of Cell Death in O3-Exposed Col-0, rcd1,
ein2, and rcd1 ein2 Double Mutant.
Three-week-old plants were exposed to O3 (250 nL L⫺1) for 4 hr (indicated by the black bar) and then kept in clean air. Whole rosettes
were collected at 0, 4, and 8 hr, and ion leakage from damaged cells
was determined as a measure of cell death. Cell death is expressed
as relative ion leakage (percentage of total ions ⫾SE).
and containment—are involved in lesion formation (Greenberg
and Ausubel, 1993; Dietrich et al., 1994; Greenberg et al., 1994;
Yu et al., 1998; Rate et al., 1999). For example, lsd1 and acd2
are defective in the control of lesion initiation; that is, several
inappropriate stimuli can trigger the HR-like lesions. In addition to quantitative differences in initiation, rcd1, reminiscent
of lsd1 and acd1, seems to be deficient in the control of lesion propagation. This deficiency, however, occurs only
transiently in rcd1, because after 24 hr, concomitant with the
cessation of O2•⫺ accumulation, the propagation of the lesion
is brought under control. In contrast, in the dnd1 (for defense,
no death1) mutant (Yu et al., 1998), initiation of cell death
takes place, but either propagation is deficient or containment is hyperactive. Similarly, lesion initiation but no
propagation takes place in O3-treated Col-0 (Figures 1 and 2),
tobacco (Schraudner et al., 1998), tomato (Tuomainen et al.,
1997), and birch (Pellinen et al., 1999). Related processes
can also be seen during systemic acquired resistance to
pathogens, in which the systemic leaves show microbursts
of H2O2 and subsequent cell death (micro-HR; Alvarez et al.,
1998) that resembles lesion initiation.
Increased O3 and O2•–-Induced Lesion Propagation in
rcd1 Is Not a Result of Deficient Antioxidative Protection
Propagation of the lesion is most likely to involve generation
at the initiation site of a signal that regulates death of the
neighboring cells. Both O2•⫺ and H2O2 have been shown to
act as signal molecules in various systems (Lamb and
Dixon, 1997). O3-exposed tobacco (Schraudner et al., 1998)
and birch (Pellinen et al., 1999) showed predominant H 2O2
but no O2•⫺ accumulation, whereas O3-exposed Arabidop-
1857
sis predominately accumulates O 2•⫺. Our results suggest
that in Arabidopsis, O2•⫺ can act as a positive regulator in
the amplification of a cellular oxidative burst and cell death
in the surrounding cells. O3, and O2•⫺ generated by X/XO,
induced O2•⫺ accumulation and cell death in rcd1 (Figure
1B), whereas H2O2 concentrations as great as 10 mM were
ineffective (Figure 3D). This raises the question of the functionality of O2•⫺ scavenging in rcd1. LSD1 is known to be involved in the induction of CuZnSOD in Arabidopsis
(Kliebenstein et al., 1999), and O3 sensitivity of vtc1 is a result of low ascorbate concentrations (Conklin et al., 1996).
Similarly, several O3-sensitive mutants that are deficient in
Fe and CuZnSOD gene expression have been isolated (D.
Kliebenstein and R. Last, unpublished data). However, transcript levels of several antioxidant genes were approximately
the same in rcd1 and Col-0 at 8 hr after the beginning of the
exposure (Table 2). Furthermore, ascorbate concentrations
and total catalase and SOD activities were similar in both clean
air–grown and O3-exposed Col-0 and rcd1 (K. Overmyer, H.
Tuominen, and J. Kangasjärvi, unpublished data). Thus, at an
overall level, deficient antioxidant capacity does not seem to
be responsible for lesion propagation in rcd1; other signals
and mechanisms must be involved.
Specific Activation of Ethylene Biosynthesis and
Functional Ethylene Signaling Are Required for
Amplification of O2•– Accumulation and Cell Death
Ethylene synthesis in the O3-exposed rcd1 was a result of
fast, specific activation of the ACC synthase gene AT-ACS6
(Figure 5A). A similar O3 responsiveness of only the ATACS6 gene has also been seen in Col-0 (Vahala et al., 1998).
In Col-0, both O3-induced ethylene evolution and O 2•⫺ accumulation lasted approximately the same time as the O 3
exposure, whereas in rcd1 they continued beyond the actual exposure period (Figures 1B and 5). Similarly, a correlation between ethylene evolution and O 2•⫺ accumulation was
evident in a survey of 11 Arabidopsis ecotypes differing in their
sensitivity to O3 (K. Mittelstrass, H. Wohlgemuth, and C.
Langebartels, unpublished data). A direct role for ethylene in
regulating ROS production was shown in further experiments. Both loss-of-function and gain-of-function experiments with O3-sensitive rcd1 and O3-tolerant Col-0 and ein2
revealed a requirement for ethylene synthesis and signaling
in cell death. This suggests that the O 3-induced ethylene
synthesis in Arabidopsis is not a mere indicator of damage
(Tingey et al., 1976; Heath and Taylor, 1997) but rather an
active component controlling the damage process itself. It
also implies that ethylene acts upstream of O 2•⫺ accumulation, which is consistent with its rapid induction by O 3. However, both 250 and 400 nL L⫺1 O3 induced ethylene
evolution and initial microbursts of O 2•⫺ in ein2 without lesion propagation. This suggests that the increase in ethylene synthesis is not under autocatalytic regulation and that
these very early responses related to lesion initiation are
1858
The Plant Cell
ethylene independent. The experiments performed with the
rcd1 ein2 double mutant also separated functionally O 3induced lesion initiation from propagation. The double
mutant showed an initial increase in cell death during O 3
exposure, a property of rcd1, but ceased further lesion
propagation in clean air because of the ethylene insensitivity
conferred by ein2 (Figure 7). Together, these results suggest
that ethylene-independent lesion initiation is followed by
ethylene-dependent amplification of O 2•⫺ accumulation,
which is responsible for the execution of spreading cell
death. We propose that ethylene primes the cells in the borders of the lesions for hypersensitivity to a message from
the O2•⫺-producing cells, which results in triggering O 2•⫺
production or in enhanced O2•⫺ production.
JAs and JA Signaling Are Involved in
Lesion Containment
Figure 8. Role of JAs and JA Signaling in Lesion Containment.
(A) Jasmonate protection of O3-induced cell death in the rcd1 mutant. Three-week-old rcd1 plants were exposed to O3 (250 nL L⫺1)
for 4 hr (indicated by the black bar under the x axis) and then exposed to clean air or MeJA (1.4 ␮M). Cell death was measured as relative ion leakage (percentage of total ions ⫾SE) after 0, 4, and 8 hr.
(B) O3 sensitivity of a JA-insensitive mutant jar1. Shown is the O3
damage in the leaves of jar1 at 24 hr after the onset of a 6-hr exposure to 250 nL L⫺1 O3.
(C) O3-induced accumulation of O2•⫺ in Col-0. Leaves of O3exposed Col-0 stained for O2•⫺ accumulation at 8 hr show no NBT
precipitation.
(D) O3-induced accumulation of O2•⫺ in a JA-signaling mutant. The
JA-insensitive mutant jar1 was exposed to 6 hr of O3 (250 nL L⫺1).
Plants were removed from the chamber at 8 hr, and detached leaves
were infiltrated with NBT. The presence of the purple formazan precipitate indicates the location and extent of O2•⫺ accumulation.
(E) Ethylene evolution from Col-0, rcd1, jar1, and ein2 in clean air and
after 2 hr of 250 nL L⫺1 O3. FW, fresh weight. Error bars indicate SE.
JA signaling appears to be involved in lesion containment.
Exposure of rcd1 to MeJA after 4 hr of O3 prevented further
lesion propagation (Figure 8A). This suggests that increased
JA can override the ethylene and O 2•⫺-dependent propagation of cell death that occurs in rcd1 at that time. Further evidence for this comes from the fact that the JA-insensitive
jar1 is O3 sensitive and displays a spreading-lesion phenotype similar to that of rcd1. Initially, at 1 to 2 hr after the beginning of O3 exposure, the pattern of O2•⫺ accumulation in
jar1 was similar to that of Col-0 and rcd1. Later, during lesion propagation, O2•⫺ accumulation continued in jar1, but
in contrast to rcd1, propagation did not involve ethylene
evolution to the same extent as in rcd1. This may seem at
first contradictory to the proposed requirement for ethylene
as an amplifier of O2•⫺ accumulation and cell death. Actually, however, similar lesion phenotypes would be expected
to result from an increase in lesion propagation caused by
increased ethylene-dependent O 2•⫺ accumulation, as in
rcd1, or from a reduced capacity for lesion containment,
without a large increase in ethylene, as in jar1.
Does Interaction and Cross-Talk among JA, SA, and
Ethylene Signaling Determine the Degree and Extent of
O2•–-Driven Lesion Formation?
Recently, an oxidative cell death cycle was postulated from
work with pathogen-infected plants undergoing HR (Van
Camp et al., 1998). In this model, ROS, SA, and cell death
were implicated in a self-amplifying cycle ultimately leading
to visible symptom development. Rao and Davis (1999)
have shown that NahG Arabidopsis plants failed to develop
HR-like lesions after a short exposure to O 3. They concluded that SA is involved in processes that contribute to
O3-induced cell death by potentiating ROS toxicity. We have
also verified this in our own experiments (H. Tuominen, K.
Ethylene and Jasmonate in Oxidative Cell Death
Overmyer, and J. Kangasjärvi, unpublished data). Thus,
three stress-related signaling pathways, for SA, JA, and ethylene, appear to interact with ROS in HR-like lesion development that takes place in O3-exposed Arabidopsis. Their
interactions and roles in ROS-dependent cell death are
summarized in Figure 9. During lesion initiation, initial cellular O2•⫺ accumulation in microbursts executes cell death in
separate, individual cells, resulting in micro-HR–like microscopic lesions. SA is required for the HR-like cell death, because O3 does not induce HR-like lesions in NahG plants
(Rao and Davis, 1999). Activation of the ACC synthase gene
AT-ACS6 in rcd1 results in increased ethylene evolution,
which promotes the ethylene signaling–dependent O 2•⫺ accumulation that drives lesion propagation. A similar role for
ethylene was also postulated for the activation of cell death
in the acd6 mutant (Rate et al., 1999) and for regulating
symptom development in Pseudomonas- and Xanthomonas-infected Arabidopsis (Bent et al., 1992) and in Xanthomonas-infected tomato (Ciardi et al., 2000).
Exogenously added JA promoted lesion containment in
rcd1, also overriding the positive effect of ethylene on cell
death. Furthermore, the JA-insensitive jar1 showed a lack of
lesion containment and thus formed visible lesions and O 2•⫺
accumulation without high ethylene synthesis. A similar role
for JA was proposed in an O3-sensitive poplar clone (Koch
et al., 1998).
Results from the experiments presented here give some
indication as to the possible location of RCD1 function with
respect to processes that regulate cell death (Figure 9). Ethylene biosynthesis was activated in rcd1 to a higher degree,
or by a lesser stimulus than in the parent ecotype Col-0.
RCD1 may be involved in processes related to regulation
of ethylene biosynthesis or to communication between the
ethylene, JA, and SA pathways, which are known to interact
with each other (Reymond and Farmer, 1998). Interaction
between these pathways appears to fine-tune the relative
contribution of lesion initiation, propagation, and containment, as reflected by different lesion sizes and formation kinetics. It therefore seems obvious to include both ethylene
and JA, as presented in Figure 9, as new regulating components in the oxidative cell death cycle.
Table 3. Inhibition of Root Elongation by MeJa in Col-0, rcd1, and jar1
MeJa (␮M)a
Col-0
0
10
34.4 ⫾ 1.6
11.6 ⫾ 1.1 a
cb
rcd1
jar1
26.6 ⫾ 0.7 b
10.2 ⫾ 0.4 a
28.3 ⫾ 1.0 b
26.4 ⫾ 0.7 b
a Seeds were germinated and grown on vertically placed agar plates
containing 0 or 10 ␮M MeJa.
b Root length (mm) was measured after 7 days. Means (⫾SE, n ⫽ 10
to 27) followed by the same letter are not significantly (P ⬍ 0.05) different according to Tukey’s multiple range test. Experiments were
repeated three times with similar results.
1859
Figure 9. Roles of Ethylene, JA, and SA in the O3-Induced Oxidative
Cell Death Cycle.
Locations of the components of ethylene and JA signaling, for which
mutants affecting the extent of cell death were used in this study,
are shown in green. Similarly, the sites of action for the inhibitors of
O2•⫺ accumulation (SOD and DPI), ethylene perception (NBD), and
SA action (NahG) that affect cell death are indicated with red bars.
Components addressed in this study are shown in green. The putative location of RCD1 function, which includes negative regulation of
ethylene synthesis, is shown.
METHODS
Mutant Screening, Crossing, and Genetic Mapping
Ethyl methanesulfonate–mutagenized Arabidopsis thaliana ecotype
Columbia (Col-0) seeds (M2E-1A-4; Lehle Seeds, Round Rock, TX)
were grown in a peat/sand mixture at a density of 3000 plants m⫺2
for 3 weeks in growth chambers (photon flux density 250 ␮mol m⫺2
sec⫺1, 12/12 hr day/night, 22/16⬚C, 50/75% relative humidity) and
exposed for 3 days to 250 nL L⫺1 ozone (O3) for 8 hr. Details of the
screening procedure for O3-induced foliar lesions have been described (Langebartels et al., 2000). Selected mutants were crossed
with each other as appropriate to determine allelism with Col-0 to determine the inheritance of the phenotype, and with Landsberg erecta
(Ler) for mapping. F2 populations of rcd1 ⫻ Ler with an O3-sensitive
phenotype were used for linkage analysis by using cleaved amplified
polymorphic sequence and microsatellite markers. O3 sensitivity of
the F2 lines used was verified in selfed F3 populations. The ethyleneinsensitive ein2-1 mutant and the jasmonic acid (JA)-insensitive jar1-1
were obtained from the Arabidopsis Biological Resource Center
(Ohio State University, Columbus). For double mutant analysis, ethylene-insensitive F2 individuals from the cross rcd1 ⫻ ein2 were
selected on plates containing 20 ␮M 1-aminocylcopropane-1-carboxylic acid (ACC) by screening for the lack of triple response. Double mutant F2 individuals homozygous for rcd1 were identified and
1860
The Plant Cell
confirmed with backcrosses of ethylene-insensitive F2 and F3 plants
to rcd1.
Treatments and Biochemical Analyses
O3 exposures were a single 6-hr pulse at 250 nL L⫺1, except where
otherwise indicated. Times of measurement refer to hours after the
onset of exposure. Superoxide (O2•⫺) accumulation in leaves was visualized with 0.1% nitroblue tetrazolium (NBT; Boehringer Mannheim), as described (Jabs et al., 1996), with a 20-min incubation
period. Plants were removed from the chambers at least 15 min before staining. Ethylene exposures (12 ⫾ 2 ␮L L⫺1) were performed
under conditions similar to those described above for O3 by mixing
2% ethylene with the air in the growth chamber. The concentration of
ethylene was measured with a photoionizer and regulated with a
computer-controlled system. Exposures to high-intensity light (photon flux density 1000 ␮mol m⫺2 sec⫺1, supplied by Philips multimetal
lamps; Oy Philips Ab, Espoo, Finland) were for 8 hr in growth chambers at 10⬚C. UV-B exposure (0.3 kJ m⫺2 hr⫺1) was for 7 days and
was essentially as described (Landry et al., 1995). Treatments with
30 ␮L L⫺1 ethylene antagonist norbornadiene (NBD; Sigma) and with
1.4 ␮M methyl jasmonate (MeJA; Sigma) were performed in a desiccator jar. Ethylene emission from plants was analyzed from two rosettes incubated for 1 hr at 20⬚C (Langebartels et al., 2000). ACC was
determined as described (Langebartels et al., 2000). Cellular O2•⫺
production was inhibited by diphenylene iodonium (DPI; Sigma), a
suicide inhibitor of NADPH oxidase and other flavin-containing oxidases (Cross and Jones, 1986). Leaf halves of rcd1 exposed to 250
nL L⫺1 O3 for 2 hr were infiltrated with DPI solutions at the indicated
concentrations. Leaf damage was measured after 24 hr in at least 10
replicates per treatment. Cell death was quantified by ion leakage
from rosette leaves into 5 mL of 18 Mohm water for 1 hr, measured
with a Mettler conductivity meter (Mettler Toledo GmbH, Switzerland).
rosettes that were maintained on wet filter paper during the course of
the treatment. Freshly grown Pseudomonas syringae pv tomato
DC3000 bacteria bearing the AvrB gene on the plasmid pPSG0002
(Bent et al., 1992) were resuspended, diluted at the indicated densities in 10 mM MgCl2, and delivered by vacuum infiltration. All experiments were repeated at least three times.
RNA Isolation, RNA Gel Blot, and cDNA
Macroarray Hybridizations
RNA isolation and hybridizations were essentially as described
(Carpenter and Simon, 1998). The expression of 92 selected genes
was studied with cDNA macroarrays using cDNA clones and with expressed sequence tag clones from the Arabidopsis Biological Resource Center. Inserts were amplified by polymerase chain reaction,
purified, and examined by agarose gel electrophoresis. Amplified
polymerase chain reaction products (100 ng) were denatured and
blotted onto a Hybond N⫹ membrane (Amersham) by using a 96well dot-blot device and then cross-linked with UV light. Used as
negative controls were 100 ng of pSPORT vector and 100 ng of
oligo-dT21. 32P-labeled cDNA probes were prepared from 1 ␮g of
mRNA by oligo-dT21⫺primed polymerization with M-MLV reverse
transcriptase (Promega) at 42⬚C for 1 hr, followed by 15 min at 70⬚C,
and then rapid cooling on ice. After addition of 1.5 units of RNase H
(Stratagene), the mixture was incubated at 37⬚C for 30 min and purified on G-50 columns (Amersham). Hybridizations were at 42⬚C in a
50% formamide buffer with stringent washes at 65⬚C (0.2 ⫻ SSC [1 ⫻
SSC is 0.15 M NaCl and 0.015 M sodium citrate] and 0.1% SDS), according to standard protocols. All hybridizations were quantified with
a phosphor imager (Bas-1500; Fujifilm, Tokyo, Japan) and an image
analysis program.
ACKNOWLEDGMENTS
In Vitro Treatments
The O2•⫺-generating system xanthine/xanthine oxidase (X/XO; 0.5
mM/0.05 unit mL⫺1; Sigma) or the H2O2-generating system glucose/
glucose oxidase (G/GO; 2.5 mM/2.5 to 250 units mL⫺1; Calbiochem)
in 10 mM sodium phosphate buffer, pH 7.0, was vacuum-infiltrated
into detached leaves (Jabs et al., 1996; Alvarez et al., 1998). When indicated, 440 units mL⫺1 MnSOD (Sigma) or 50 ␮M ACC (Sigma) was
included. All treatments were for 20 hr (or as indicated) at 22⬚C, and
only the first three fully expanded leaves were used. The lifetime of
O2•⫺ radical production by X/XO and of O2•⫺ production in leaves in
each treatment was assessed by NBT staining. H2O2 production was
monitored and calibrated against standard dilutions by luminol bioluminescence for concentrations in the micromolar range and by direct
spectrophotometric (A240) measurement for concentrations in the
millimolar range, as described (Chamnongpol et al., 1998). Direct
H2O2 treatments were performed in a large volume to prevent H2O2
clearance by detoxification systems (50 mL per 10 leaves); concentration determinations before and after treatments indicated a maximum of 20% decrease in H2O2 over the course of all experiments.
For JA root inhibition assays, surface-sterilized seed was grown on
vertically oriented plates of 0.5 ⫻ Murashige and Skoog (1962) medium in the presence or absence of 10 ␮M MeJA for 7 days, after
which root length was determined as described by Staswick and
Howell (1992). Pathogen treatments were performed with detached
We thank Mr. Timo Oksanen for help with the O3 exposures during
the mutant screening, Mr. Jorma Vahala for providing the AT-ACS6
clone, Professor Sheng Yang He for providing the Pseudomonas
strain used, and Professor Tapio Palva, Dr. Nigel Kilby, and Dr. Jörg
Durner for their comments on the manuscript. This work was supported by Academy of Finland Grant Nos. 43671 and 37995 to J.K.
and postdoctoral fellowship No. 41615 to H.T., by the Finnish Centre
of Excellence Program (2000-2005), and by grants from EU-FAIR
(Brussels, Belgium) and Bundesministerium für Bildung und Forschung
(BMBF) and Deutsche Forschungsgemeinschaft (DFG) (Bonn, Germany).
Received May 30, 2000; accepted July 10, 2000.
REFERENCES
Alvarez, M.E., Pennell, R.I., Meijer, P.-J., Ishikawa, A., Dixon,
R.A., and Lamb, C. (1998). Reactive oxygen intermediates mediate a systemic signal network in the establishment of plant immunity. Cell 92, 773–784.
Bent, A.F., Innes, R.W., Ecker, J.R., and Staskawicz, B.J. (1992).
Disease development in ethylene-insensitive Arabidopsis thaliana
Ethylene and Jasmonate in Oxidative Cell Death
infected with virulent and avirulent Pseudomonas and Xanthomonas pathogens. Mol. Plant-Microbe Interact. 5, 372–378.
Carpenter, C.D., and Simon, A.E. (1998). Preparation of RNA.
Methods Mol. Biol. 82, 85–90.
Chamnongpol, S., Willekens, H., Moeder, W., Langebartels, C.,
Sandermann, H., Jr., Van Montagu, M., Inzé, D., and van
Camp, W. (1998). Defense activation and enhanced pathogen tolerance induced by H2O2 in transgenic tobacco. Proc. Natl. Acad.
Sci. USA 95, 5818–5823.
Ciardi, J.A., Tieman, D.M., Lund, S.T., Jones, J.B., Stall, R.E., and
Klee, H.J. (2000). Response to Xanthomonas campestris pv. vesicatoria in tomato involves regulation of ethylene receptor gene
expression. Plant Physiol. 123, 81–92.
Conklin, P.L., Williams, E.H., and Last, R.L. (1996). Environmental
stress sensitivity of an ascorbic acid–deficient Arabidopsis
mutant. Proc. Natl. Acad. Sci. USA 93, 9970–9974.
Cross, A.R., and Jones, O.T.G. (1986). The effect of the inhibitor
diphenylene iodonium on the superoxide-generating system of
neutrophils. Biochem. J. 237, 111–116.
Dietrich, R.A., Delaney, T.P., Uknes, S.J., Ward, E.R., Ryals, J.A.,
and Dangl, J.L. (1994). Arabidopsis mutants simulating disease
resistance response. Cell 77, 565–577.
Drew, M.C., He, C.-J., and Morgan, P.W. (2000). Programmed cell
death and aerenchyma formation in roots. Trends Plant Sci. 5,
123–127.
Elstner, E.F., Osswald, W., and Youngman, R.J. (1985). Basic
mechanisms of pigment bleaching and loss of structural resistance in spruce (Picea abies) needles: Advances in phytomedical
diagnostics. Experientia 41, 591–597.
Greenberg, J.T., and Ausubel, F.M. (1993). Arabidopsis mutants
compromised for the control of cellular damage during pathogenesis and aging. Plant J. 4, 327–341.
Greenberg, J.T., Guo, A., Klessig, D.F., and Ausubel, F.M. (1994).
Programmed cell death in plants: A pathogen-triggered response
activated coordinately with multiple defense functions. Cell 77,
551–563.
Guzman, P., and Ecker, J.R. (1990). Exploiting the triple response
of Arabidopsis to identify ethylene-related mutants. Plant Cell 2,
513–523.
Heath, R.L., and Taylor, G.E., Jr. (1997). Physiological processes
and plant responses to ozone exposure. Ecol. Stud. 127, 317–368.
Jabs, T., Dietrich, R.A., and Dangl, J.L. (1996). Initiation of runaway cell death in an Arabidopsis mutant by extracellular superoxide. Science 273, 1853–1856.
Kangasjärvi, J., Talvinen, J., Utriainen, M., and Karjalainen, R.
(1994). Plant defense systems induced by ozone. Plant Cell Environ. 17, 783–794.
Kley, D., Kleinman, M., Sandermann, H., Jr., and Krupa, S. (1999).
Photochemical oxidants: State of the science. Environ. Pollut.
100, 19–24.
Kliebenstein, D.J., Dietrich, R.A., Martin, A.C., Last, R.L., and
Dangl, J.L. (1999). LSD1 regulates salicylic acid induction of copper zinc superoxide dismutase in Arabidopsis thaliana. Mol. PlantMicrobe Interact. 12, 1022–1026.
1861
Koch, J.R., Scherzer, A.J., Eshita, S.M., and Davis, K.R. (1998).
Ozone sensitivity in hybrid poplar is correlated with a lack of
defense-gene activation. Plant Physiol. 118, 1243–1252.
Lamb, C., and Dixon, R.A. (1997). The oxidative burst in plant disease resistance. Annu. Rev. Plant Physiol. Plant Mol. Biol. 48,
251–275.
Landry, L.G., Chapple, C.C.S., and Last, R.L. (1995). Arabidopsis
mutants lacking phenolic sunscreens exhibit enhanced ultravioletB injury and oxidative damage. Plant Physiol. 109, 1159–1166.
Langebartels, C., Ernst, D., Kangasjärvi, J., and Sandermann, H.,
Jr. (2000). Ozone effects on plant defense. Methods Enzymol.
319, 520–535.
Mehlhorn, H., and Wellburn, A.R. (1987). Stress ethylene formation
determines plant sensitivity to ozone. Nature 327, 417–418.
Murashige, T., and Skoog, F. (1962). A revised medium for rapid
growth and bioassays with tobacco tissue culture. Physiol. Plant.
15, 473–497.
Norman-Setterblad, C., Vidal, S., and Palva, E.T. (2000). Interacting signal pathways control defense gene expression in Arabidopsis in response to cell wall–degrading enzymes from Erwinia
carotovora. Mol. Plant-Microbe Interact. 13, 430–438.
O’Donnell, P.J., Calvert, C., Atzorn, R., Wasternack, C., Leyser,
H.M.O., and Bowles, D.J. (1996). Ethylene as a signal mediating
the wound response of tomato plants. Science 274, 1914–1917.
Örvar, B.L., McPherson, J., and Ellis, B.E. (1997). Pre-activating
wounding response in tobacco prior to high-level ozone exposure
prevents necrotic injury. Plant J. 11, 203–212.
Orzáez, D., and Granell, A. (1997). DNA fragmentation is regulated
by ethylene during carpel senescence in Pisum sativum. Plant J.
11, 137–144.
Pellinen, R., Palva, T., and Kangasjärvi, J. (1999). Subcellular
localization of ozone-induced hydrogen peroxide production in
birch (Betula pendula) leaf cells. Plant J. 20, 349–356.
Penninckx, I.A.M.A., Thomma, B.P.H.J., Buchala, A., Métraux, J.-P.,
and Broekaert, W.F. (1998). Concomitant activation of jasmonate
and ethylene response pathways is required for induction of a
plant defensin gene in Arabidopsis. Plant Cell 10, 2103–2113.
Pieterse, C.M.J., van Wees, S.C.M., van Pelt, J.A., Knoester, M.,
Laan, R., Gerrits, H., Weisbeek, P.J., and van Loon, L.C. (1998).
A novel signaling pathway controlling induced systemic resistance in Arabidopsis. Plant Cell 10, 1571–1580.
Rao, M.V., and Davis, K.R. (1999). Ozone-induced cell death
occurs via two distinct mechanisms in Arabidopsis: The role of
salicylic acid. Plant J. 17, 603–614.
Rate, D.N., Cuenca, J.V., Bowman, G.R., Guttman, D.S., and
Greenberg, J.T. (1999). The gain-of-function Arabidopsis acd6
mutant reveals novel regulation and function of the salicylic acid
signaling pathway in controlling cell death, defenses, and cell
growth. Plant Cell 11, 1695–1708.
Reymond, P., and Farmer, E.E. (1998). Jasmonate and salicylate
as global signals for defense gene expression. Curr. Opin. Plant
Biol. 1, 404–411.
Runeckles, V.C., and Vaartnou, M. (1997). EPR evidence for superoxide anion formation in leaves during exposure to low levels of
ozone. Plant Cell Environ. 20, 306–314.
1862
The Plant Cell
Sandermann, H., Jr., Ernst, D., Heller, W., and Langebartels, C.
(1998). Ozone: An abiotic elicitor of plant defence reactions.
Trends Plant Sci. 3, 47–50.
Schraudner, M., Moeder, W., Wiese, C., van Camp, W., Inzé, D.,
Langebartels, C., and Sandermann, H., Jr. (1998). Ozoneinduced oxidative burst in the ozone biomonitor plant, tobacco
Bel W3. Plant J. 16, 235–245.
Sharma, Y.K., and Davis, K.R. (1997). The effects of ozone on antioxidant responses in plants. Free Radical Biol. Med. 23, 480–488.
Sharma, Y.K., León, J., Raskin, I., and Davis, K.R. (1996). Ozoneinduced responses in Arabidopsis thaliana: The role of salicylic
acid in the accumulation of defense-related transcripts and
induced resistance. Proc. Natl. Acad. Sci. USA 93, 5099–5104.
Staswick, P.E., and Howell, S.H. (1992). Methyl jasmonate inhibition of root growth and induction of a leaf protein are decreased in
an Arabidopsis thaliana mutant. Proc. Natl. Acad. Sci. USA 89,
6837–6840.
Tingey, D.T., Standley, C., and Field, R.W. (1976). Stress ethylene
evolution: A measure of ozone effects on plants. Atmos. Environ.
10, 969–974.
Tuomainen, J., Betz, C., Kangasjärvi, J., Ernst, D., Yin, Z.H.,
Langebartels, C., and Sandermann, H., Jr. (1997). Ozone induction of ethylene emission in tomato plants: Regulation by differential transcript accumulation for the biosynthetic enzymes. Plant J.
12, 1151–1162.
Vahala, J., Schlagnhaufer, C.D., and Pell, E.J. (1998). Induction of
an ACC synthase cDNA by ozone in light-grown Arabidopsis
thaliana leaves. Physiol. Plant. 103, 45–50.
Van Camp, W., Van Montagu, M., and Inzé, D. (1998). H2O2 and
NO: Redox signals in disease resistance. Trends Plant Sci. 3,
330–334.
Wellburn, F.A.M., and Wellburn, A.R. (1996). Variable patterns of
antioxidant protection but similar ethene emission differences in
several ozone-sensitive and ozone-tolerant plant selections. Plant
Cell Environ. 19, 754–760.
Young, T.E., Gallie, D.R., and DeMason, D.A. (1997). Ethylenemediated programmed cell death during maize endosperm development of wild-type and shrunken2 genotypes. Plant Physiol.
115, 737–751.
Yu, I.-C., Parker, J., and Bent, A.F. (1998). Gene-for-gene disease
resistance without the hypersensitive response in Arabidopsis
dnd1 mutant. Proc. Natl. Acad. Sci. USA 95, 7819–7824.
Ozone-Sensitive Arabidopsis rcd1 Mutant Reveals Opposite Roles for Ethylene and Jasmonate
Signaling Pathways in Regulating Superoxide-Dependent Cell Death
Kirk Overmyer, Hannele Tuominen, Reetta Kettunen, Christian Betz, Christian Langebartels, Heinrich
Sandermann, Jr. and Jaakko Kangasjärvi
Plant Cell 2000;12;1849-1862
DOI 10.1105/tpc.12.10.1849
This information is current as of July 31, 2017
References
This article cites 43 articles, 16 of which can be accessed free at:
/content/12/10/1849.full.html#ref-list-1
Permissions
https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X
eTOCs
Sign up for eTOCs at:
http://www.plantcell.org/cgi/alerts/ctmain
CiteTrack Alerts
Sign up for CiteTrack Alerts at:
http://www.plantcell.org/cgi/alerts/ctmain
Subscription Information
Subscription Information for The Plant Cell and Plant Physiology is available at:
http://www.aspb.org/publications/subscriptions.cfm
© American Society of Plant Biologists
ADVANCING THE SCIENCE OF PLANT BIOLOGY