Analysis of knockout mutants suggests that Arabidopsis NADP-MALIC ENZYME2 does not play an essential role in responses to oxidative stress of intracellular or extracellular origin

Abstract

NADPH is a pivotal molecule in oxidative stress, during which it is potentially produced by several cytosolic NADP-linked
dehydrogenases. This study investigated the response and functional importance of the major leaf cytosolic NADP-malic enzyme
in Arabidopsis (NADP-ME2) during oxidative stress. Data from both microarray and targeted quantitative PCR analyses showed that NADP-ME2 transcripts accumulated in response to ozone or in mutants undergoing intracellular oxidative stress. To test the functional
importance of this response, loss-of-function nadp-me2 mutants were obtained and the effects of oxidative stress of intracellular and extracellular origin were tested. Despite
much decreased leaf NADP-ME activity, nadp-me2 showed a wild-type phenotype when exposed to ozone. Introduction of the nadp-me2 mutations into the catalase-deficient cat2 background did not alter growth inhibition or lesions triggered by intracellular oxidative stress. Similarly, loss of NADP-ME2 function had little effect on cat2-triggered changes in glutathione or NADPH. While single nadp-me2 mutations produced slight effects on basal resistance to one type of bacteria, they did not affect resistance induced by
the cat2 mutation. Taken together, the results suggest that, although NADP-ME2 induction is part of the response to oxidative stress, the enzyme is not an essential determinant of the outcome of such
stress.

Full text

Journal of Experimental Botany, Vol. 64, No. 12, pp. 3605–3614, 2013
doi:10.1093/jxb/ert194 Advance Access publication 12 July, 2013
© The Author [2013]. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved.
For permissions, please email: journals.permissions@oup.com
Abbreviations: CAT, catalase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; G6PDH, glucose-6-phosphate dehydrogenase; GR, glutathione reductase;
GSH, glutathione; GSSG, glutathione disulphide; ICDH, isocitrate dehydrogenase; ME, malic enzyme.
ReseaRch papeR
Analysis of knockout mutants suggests that Arabidopsis
NADP-MALIC ENZYME2 does not play an essential
role in responses to oxidative stress of intracellular or
extracellularorigin
ShengchunLi
1,
*, AmnaMhamdi
1,
*, CyndieClement
2,3,4
, YvesJolivet
2,3,4
and GrahamNoctor
1,†
1
Institut de Biologie des Plantes, Université de Paris sud, UMR CNRS 8618, 91405 Orsay cedex, France
2
Université de Lorraine, UMR1137 EEF, F-54500 Vandoeuvre-lès-Nancy, Cedex, France
3
INRA, UMR1137 EEF, F-54280 Champenoux, France
4
IFR110 EFABA, F-54500 Vandoeuvre-lès-Nancy, Cedex, France
*
These authors contributed equally to this work.
†
Author for correspondence: graham.noctor@u-psud.fr
Received 8 April 2013; Revised 20 May 2013; Accepted 28 May 2013
Abstract
NADPH is a pivotal molecule in oxidative stress, during which it is potentially produced by several cytosolic NADP-
linked dehydrogenases. This study investigated the response and functional importance of the major leaf cytosolic
NADP-malic enzyme in Arabidopsis (NADP-ME2) during oxidative stress. Data from both microarray and targeted
quantitative PCR analyses showed that NADP-ME2 transcripts accumulated in response to ozone or in mutants
undergoing intracellular oxidative stress. To test the functional importance of this response, loss-of-function nadp-
me2 mutants were obtained and the effects of oxidative stress of intracellular and extracellular origin were tested.
Despite much decreased leaf NADP-ME activity, nadp-me2 showed a wild-type phenotype when exposed to ozone.
Introduction of the nadp-me2 mutations into the catalase-deficient cat2 background did not alter growth inhibition
or lesions triggered by intracellular oxidative stress. Similarly, loss of NADP-ME2 function had little effect on cat2-
triggered changes in glutathione or NADPH. While single nadp-me2 mutations produced slight effects on basal resist-
ance to one type of bacteria, they did not affect resistance induced by the cat2 mutation. Taken together, the results
suggest that, although NADP-ME2 induction is part of the response to oxidative stress, the enzyme is not an essential
determinant of the outcome of such stress.
Key words: Glutathione, H
2
O
2
, NADP(H), ozone, photorespiration, redox homeostasis.
Introduction
NADP(H) is a key player both in assimilatory metabolism
and in cellular redox homeostasis. In the chloroplast in the
light, ferredoxin-NADP
+
reductase generates the reduced
form, NADPH, which then mainly powers the reduction of
1,3-bis-phosphoglycerate in the reaction catalysed by the stro-
mal NADPH-dependent glyceraldehyde-3-phosphate dehy-
drogenase (GAPDH). In the dark, or in non-photosynthetic
tissues, enzymes such as glucose-6-phosphate dehydroge-
nase (G6PDH) play important roles in converting plastidial
NADP
+
to NADPH (Anderson and Duggan, 1976; Von
Schaewen et al., 1995; Wakao and Benning, 2005). While
attention has been paid to the roles of NADP-linked enzymes
in other highly redox-active organelles, such as the mitochon-
dria and peroxisomes (Møller and Rasmusson, 1998; Meyer
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from

3606
|
Li etal.
etal., 2011), less is known about the production and turno-
ver of NADP(H) in the cytosol. As many signalling processes
occur in the cytosol, and because cytosolic and nuclear redox
states are likely to be closely linked, identifying the fac-
tors determining NADP(H) turnover in this compartment
remains key to understanding plant stress responses (Foyer
and Noctor, 2009). In particular, an important gap in our
knowledge is uncertainty over which enzymes are the most
important in producing reducing power to meet the increased
demand during stress (Valderrama etal., 2006, Dizengremel
etal., 2009). As well as being required to maintain the pools of
antioxidants such as ascorbate and glutathione in the reduced
form (Noctor, 2006), NADPH is considered to be the main
source of electrons for respiratory burst oxidase homologues,
also known as NADPH oxidases, which among other things
play important roles in biotic stress (Torres etal., 2006).
There are at least four major types of cytosolic NADP-
linked dehydrogenase that can oxidize various carbon sub-
strates to generate reducing power in the form of NADPH.
As well as cytosolic G6PDH, the major form of leaf NADP-
linked isocitrate dehydrogenase (cICDH) is found in this com-
partment (Hodges etal., 2003; Mhamdi etal., 2010a). Plants
also have a cytosolic non-phosphorylating NADP-GAPDH,
in addition to the classical NAD-linked enzyme (Kelly and
Gibbs, 1973; Rius et al., 2006). A fourth type of cytosolic
NAPDH-producing dehydrogenase is NADP-linked malic
enzyme (NADP-ME), which oxidatively decarboxylates
malate to pyruvate (Gerrard Wheeler etal., 2005, 2008, 2009).
While NAD- and NADP-dependent malic enzymes have long
been known to play key roles in C
4
photosynthetic metabo-
lism (Furbank and Foyer, 1988; Dever etal., 1995; Langdale,
2011; Maier etal., 2011), the roles of NADP-ME in C
3
plants
are much lessclear.
One possible function of C
3
-type NADP-ME is in plant
defence (Casati et al., 1999). Roles for NADP-ME in stress
responses have received support from studies that have shown
that the enzyme is upregulated by various environmental
challenges, including pathogen attack (Voll et al., 2012, and
references therein). However, such data can only provide cor-
relative evidence that any given enzyme activity is important,
and the redox network may be composed of a complex matrix
of functionally interacting components that show complete
or partial redundancy. In olive plants subject to salt stress, for
example, multiple NADPH-generating activities are induced
in concert (Valderrama et al., 2006). Within this complex
redox network, knockout mutants are useful tools to identify
the most important players or to establish functional redun-
dancy. Studies using this approach suggest that cICDH plays a
non-replaceable role in response to biotic and oxidative stress.
Loss-of-function icdh mutations in Arabidopsis cause activa-
tion of pathogenesis-related responses and bacterial resistance,
and alter the glutathione redox state during oxidative stress
(Mhamdi etal., 2010a; Dghim etal., 2013). In a similar vein,
a recent study reported altered responses to fungal infection
in Arabidopsis nadp-me2 knockout mutants (Voll etal., 2012).
However, it remains unclear whether NADP-ME2 plays an
important role in inuencing the outcomes of oxidative stress,
a key factor in unfavourable environmental conditions.
The aim of the present study was to address this specic
question. To establish whether or not the major Arabidopsis
leaf malic enzyme (NADP-ME2) plays a major role during
biotic and oxidative stress, knockout mutants were subjected
to challenge with three bacterial strains as well as to two
independent types of stress that are triggered by increased
cellular oxidation: ozone exposure and a catalase-decient
genetic background (cat2). The data showed that, while the
NADP-ME2 gene is induced both by exposure to ozone and
in cat2, loss of its function had only slight effects on resist-
ance to bacteria and had little effect on the cellular redox
state or phenotypes determined by oxidative stress. Thus,
while induction of NADP-ME2 is part of the oxidative stress
responses, the enzyme does not appear to be an indispensable
player in these conditions.
Materials and methods
Plant material and growth conditions
All genotypes were in the Arabidopsis Columbia (Col-0) ecotype.
The cat2 mutant was cat2-2 (Queval et al., 2009), while two inde-
pendent T-DNA lines for NADP-ME2 were obtained from the Salk
collection (SALK_073818 and SALK_020607). Double cat2 nadp-
me2 mutants were produced by crossing. After verication of the
double heterozygotes in F1 plants by PCR, double homozygotes
were identied similarly in the F2 generation (Supplementary Fig.
S1 at JXB online) and allowed to produce F3 seeds, which were
used for experiments. All seeds were sown on soil, incubated for 2
d in the dark at 4°C, and then transferred to a controlled-environ-
ment growth chamber with long days (16 h light/8 h dark) and an
irradiance of 200μmol m
–2
s
–1
at the leaf level, 20/18°C, and 65%
humidity. Plants were supplied with nutrient solution twice weekly.
Following snap freezing in liquid nitrogen, samples were stored at
–80°C until analysis. All data are means ±standard error (SE) of at
least three independent samples from different plants.
Ozone treatment
Ozone treatments were performed in phytotron chambers constantly
ventilated with charcoal-ltered air. The treatment (350 ppb) was
begun following a 7-d acclimation period. Ozone was produced from
pure O
2
with two ozone generators (OZ500, Fischer, Bonn, German;
CMG3-3; Innovatec II, Rheinbach, Germany) and injected directly
along with the ltered air entering the chambers. Control plants
were exposed to ambient ltered air. A set of automated systems
and analysers (O341M; Environment S.A. Paris, France) were used
to monitor the concentrations and the length of ozone exposure.
Fumigation started 1 h after the beginning of the photoperiod (16 h
light/8 h dark). Ozone concentration was maintained at 350 ± 10 ppb
of ozone for 7 h (8 h into the photoperiod). Two 7 h fumigations
were performed on consecutive days. Prior to sampling, plants were
allowed to recover overnight following the second fumigation.
Quantification of lesions
The percentage of total rosette area displaying lesions during ozone
stress or in the cat2 background was quantied by imaging affected
areas of at least ten plants of each type using IQmaterials software.
Transcript analysis
Whole rosettes of four replicate plants per treatment were harvested
and frozen immediately in liquid nitrogen. Total RNA extraction
and quantitative reverse transcriptase (RT)-PCR were conducted as
described by Queval etal. (2007). The primer sequences are shown in
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from

NADP-malic enzyme and oxidative stress
|
3607
Supplementary Table S1 at JXB online. Microarray data were ana-
lysed in a dataset of which the other features were described previ-
ously by Mhamdi etal. (2010b).
Pathogentests
Three strains of Pseudomonas syringae were used in this study. To
test resistance to an avirulent bacterium, P. syringae pv. tomato
(Pst) strain DC3000 avrRpm1 was used. Resistance to a virulent
bacterium was tested using Pst strain DC3000, while P.syringae pv.
maculicola ES4326 (Psm) was employed as a second, less virulent
strain than P.syringae DC3000. Bacteria were selected on medium
containing 50 mg l
–1
of rifampicin (Psm) or 100 mg l
–1
of rifampicin
and 25 mg l
–1
of kanamycin (Pst). Using a 1 ml syringe with no nee-
dle, the central leaves of ve to seven plants of each genotype were
inoculated with bacteria. Leaf discs were taken for analysis either
immediately after inoculation (0 h, control for possible differences in
bacterial entry) and 24 or 48 h later to quantify bacterial prolifera-
tion in planta.
Enzyme and metaboliteassays
Soluble protein and extractable NADP-malic enzyme activity was
measured as described by Dghim etal. (2013). Pyridine nucleotides,
ascorbate, and glutathione were assayed spectrophotometrically
using a plate-reader protocol as described in detail by Queval and
Noctor (2007).
Results
Of the four NADP-ME genes found in Arabidopsis,
NADP-ME4 encodes a plastidial enzyme while the other
three enzymes are predicted to be localized in the cytosol
(Gerrard Wheeler etal., 2005). Of these three, NADP-ME2
encodes the major enzyme in leaf tissues (Gerrard Wheeler
et al., 2005; Voll et al., 2012). Consistent with this,
Genevestigator analysis (Hruz et al., 2008) showed that
NADP-ME2 was the most highly expressed in most tissues
and also during stress conditions (Supplementary Fig. S2A,
B, at JXB online). RT-PCR analysis in the growth condi-
tions used in this study showed that, while NADP-ME2 tran-
scripts were readily detected, NADP-ME1 and NADP-ME3
were expressed at very low levels in leaves (Supplementary
Fig. S2C). Knockout mutants for NADP-ME2 show less
than 10% wild-type extractable leaf activity (Voll et al.,
2012). The absence of detectable transcripts was conrmed
in the nadp-me2 mutants by RT-PCR (Supplementary Fig.
S3 at JXB online).
The knockout mutants were used to analyse the potential
role of NADP-ME2 in response to biotic and oxidative stress.
First, the impact on the response to bacterial challenge was
analysed. Two experiments, each based on quadruplicate
samples, were performed for three bacterial strains. Control
samples were taken immediately following bacterial inocula-
tion (0 h post-inoculation; Fig.1, left panels) and bacterial
proliferation was analysed in samples taken 24 or 48 h later
(Fig.1, right panels). For avirulent and virulent Pst, no signif-
icant difference between Col-0 and the nadp-me2 mutant lines
was observed in either experiment. For Psm, one experiment
revealed no signicant difference, while the second showed
slightly but signicantly increased bacterial growth in the
nadp-me2 mutants (Fig.1).
Ozone is an important pollutant whose stressful effects on
plants involve initial oxidation in the apoplast followed by
adjustments in intracellular metabolism, notably involving
respiratory pathways (Kangasjärvi etal., 2005; Dizengremel
et al., 2009). Exposure of Arabidopsis to ozone caused vis-
ible lesions to appear on the leaves, but there was no signif-
icant difference in the extent of lesions between Col-0 and
the nadp-me2 mutant (Fig. 2A, B). Total extractable leaf
4
5
6
7
8
9
4
5
6
7
8
9
4
5
6
7
8
9
4
5
6
7
8
9
Col-0
nadp-me2-1
nadp-me2-2
Col-0
nadp-me2-1
nadp-me2-2
Colony forming units (log
10
cm
-2
leaf area)
Pst-DC3000
Pst-avrRpm1
Psm-ES4326
(1)
*
*
Psm-ES4326
(2)
Fig.1. Effect of nadp-me2 mutations on challenge with three
different bacterial strains. Shaded bars, Col-0; filled bars,
nadp-me2 mutants. Left, leaves sampled at 0 h post-inoculation.
Right, leaves sampled at 24 (Pst-DC3000) or 48 (others) h post-
inoculation. *Significant difference from Col-0 at P <0.05. For
Pst-DC3000 and Pst-avrRpm1, each graph shows the results of
one of two repeat experiments in which no significant difference
was observed. For Psm-ES4326, the two individual experiments
gave slightly different results, and so both are shown.
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from

3608
|
Li etal.
NADP-ME activity was enhanced by ozone exposure in Col-
0, while remaining at very low levels in nadp-me2 (Fig.2C).
Quantitative (q)RT-PCR analysis showed that the ozone-
induced increase in activity in Col-0 was accompanied by
enhanced NADP-ME2 transcript abundance (Fig.2D).
Given its induction by ozone, the response of NADP-ME2
was examined under conditions where oxidative stress is ini-
tiated by intracellular H
2
O
2
production. This was done by
exploiting the Arabidopsis cat2 mutant, a stress-mimic model
that is a useful system for uncovering the functions of enzymes
involved in producing reductant during oxidative stress
(Mhamdi etal., 2010a,b,c). In microarray analyses of cat2,
NADP-ME2 showed some induction, although the increase
was not statistically signicant at the P <0.05 level compared
with the Col-0 control (Fig.3, left). However, a statistically
signicant induction of NADP-ME2 was observed in a dou-
ble cat2 gr1 mutant, which showed exacerbated intracellu-
lar oxidative stress caused by additional loss of glutathione
reductase 1 activity (Fig.3, left). Amore sensitive quantica-
tion of responses in the single cat2 mutant using qRT-PCR
showed that NADP-ME2 was signicantly induced, to more
than twofold the wild-type values (Fig.3, right).
To establish whether NADP-ME2 is functionally impor-
tant in response to intracellular oxidative stress, the two
allelic nadp-me2 mutants were each crossed into the cat2
background, and the effects on phenotype and redox state
were examined. Oxidative stress in the cat2 mutant induces
visible phenotypes of decreased growth and character-
istic lesions that appear on the leaves (Queval et al., 2007;
Chaouch etal., 2010). These responses are modulated by the
introduction of secondary mutations for several NADP(H)-
linked enzymes located in the cytosol or at the plasmalemma
(Mhamdi etal., 2010a,b; Chaouch etal., 2012). However, nei-
ther the decreased rosette size nor lesion spread observed in
cat2 was affected by the introduction of the secondary nadp-
me2 mutations (Fig.4).
0
5
10
15
20
Col-0
nadp-me2-1
Col-0 + O
3
*
*
B
A
Col-0 nadp-me2-1
Control+ Ozone
nadp-me2-1 + O
3
0,0
0,2
0,4
0,6
Transcript abundance
rel. to ACTIN2
Col-0
Col-0 + O
3
*
D
Control
Col-0 nadp-me2-1
+ Ozone
0
3
6
Lesion area (%)
Col-0
nadp-me2-1
Col-0 + O
3
nadp-me2-1 + O
3
nd nd
C
NADP-ME activity
nkat.g
-1
FW
Fig.2. Ozone responses of NADP-ME in Col-0 and nadp-me2-1 plants. (A) Representative photographs of plants before and
after ozone treatments. (B) Quantification of ozone-induced leaf damage. (C) Extractable NADP-ME enzyme activities in Col-0 and
nadp-me2-1 plants before and after ozone treatment. (D) NADP-ME2 transcript abundance in Col-0 plants. *Significant difference from
Col-0 at P <0.05.
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from

NADP-malic enzyme and oxidative stress
|
3609
The potential impact of the nadp-me2 mutations on oxi-
dative stress-induced changes in cellular redox state was
explored. In agreement with other studies in barley and
tobacco (Smith et al., 1984; Willekens et al., 1997), the
clearest and most reproducible biochemical effect of leaf
catalase deciency in Arabidopsis was on glutathione. In
cat2, this key redox marker of intracellular redox state
becomes more oxidized and the disulphide form (GSSG)
typically accumulates to 10- to 20-fold higher levels than
in Col-0 (Queval et al., 2007; Fig. 5). In contrast to glu-
tathione, changes in H
2
O
2
are much less apparent or unde-
tectable in cat2, reecting the difculty of quantifying this
reactive molecule and/or its rapid removal by reducing
enzymes in the absence of catalase-dependent dismuta-
tion (Chaouch etal., 2010; Han etal., 2013). In the Col-0
background, one of the nadp-me2 mutants had a decreased
content of glutathione (Fig.5), but neither showed a sig-
nicant change in glutathione redox state. In the cat2 oxi-
dative stress background, neither allelic nadp-me2 mutation
affected the accumulation of GSSG, and the glutathione
redox state was similar in all three cat2 backgrounds, at
about 60% glutathione (Fig. 5). No signicant effects of
any of the mutations on ascorbate or NAD(H) pools were
observed (Fig. 5). While neither single nadp-me2 mutant
showed a signicant change in leaf NADP(H), the cat2
mutant showed a tendency towards an increase in the total
pool, and this was associated with signicantly increased
NADPH compared with Col-0. A similar effect was
observed in cat2 nadp-me2-1 but was less evident in cat2
nadp-me2-2 (Fig.5).
Oxidative stress in cat2 leads to induction of resistance to
virulent bacteria above basal levels, and metabolite proling
suggests that this response shows many of the hallmarks of
induced resistance triggered by biotic challenge (Chaouch
etal., 2010, 2012). To analyse whether the nadp-me2 muta-
tions affected this response, bacterial growth was analysed
in cat2 and the double mutants. While cat2 showed lower
growth of both Pst-DC3000 and Psm relative to Col-0, this
enhanced resistance was not affected by the secondary nadp-
me2 mutations (Fig.6).
Discussion
Both detoxication and signalling processes linked to oxi-
dative stress depend on NADPH production, but the rela-
tive importance of different NADPH-generating enzymes
remains unclear (Foyer and Noctor, 2009). Some conclusions
have been drawn based on the relative extractable activities
of NADP-generating enzymes (Valderrama et al., 2006;
Dizengremel et al., 2009). While such data may be useful
pointers, for several reasons it is not possible to reach deni-
tive conclusions based solely on them. Measurable in vitro
enzyme activities are often a composite of several isoforms
Rel. ACTIN2
Log
2
transcript
(rel. Col-0)
cat2
cat2 gr1
0
0.5
1.0
1.5
0
0.2
0.4
0.6
cat2
Col-0
Microarray
qRT-PCR
*
*
Fig.3. Response of NADP-ME2 transcript abundance to
intracellular oxidative stress. Left, microarray analysis of
transcripts in cat2 and cat2 gr1 plants. Right, qRT-PCR analysis
of NADP-ME2 transcripts in Col-0 and cat2 plants. *Significant
difference from Col-0 at P <0.05.
0
200
400
600
ACTIN2
NADP-ME2
Col-0 cat2
nadp-me2-1
cat2 nadp-me2-2nadp-me2-2
cat2 nadp-me2-1
A
B
C
Rosette FW
(mg)
Lesion area
(%)
0
4
8
12
123456
1 2 34 5 61 2 34 5 6
nd nd nd
Fig.4. Characterization of cat2 nadp-me2 double mutants.
(A) NADP-ME2 transcript levels in Col-0 and single and
double mutants. Lanes: 1, Col-0; 2, cat2; 3, nadp-me2-1;
4, cat2 nadp-me2-1; 5, nadp-me2-2; 6, cat2 nadp-me2-2.
(B)Representative photographs of plants. (C) Rosette fresh weight
(left) and H
2
O
2
-triggered lesions (right) in cat2 and the double
mutants. No significant differences in either rosette size or lesion
extent were observed (n≥10 plants). Genotypes are indicated by
the same numbers as in (A).
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from

3610
|
Li etal.
located in different subcellular compartments. Secondly, such
assays are usually carried out at optimal substrate concentra-
tions, whereas these may be less optimal or variable in vivo.
Post-translational regulation, well known for the chloroplast
G6PDH (Anderson and Duggan, 1976; Von Schaewen etal.,
1995; Wenderoth etal., 1997), may also modify in vivo activi-
ties compared with those measured in vitro. For these reasons,
a genetic approach able to target specic isoforms can pro-
vide important information on the functional importance of
a specic enzyme. Using this strategy, evidence was reported
that cytosolic NADP-ICDH plays a non-redundant role in
oxidative stress responses and signalling (Mhamdi et al.,
2010a). The present genetically based study of NADP-ME2
allows the following conclusions to bedrawn.
Cytosolic NADP-ME is induced by oxidative stress of
intracellular and extracellularorigin
Previous work has established that NADP-ME2 is induced
by various stresses (Voll etal., 2012, and references therein).
The present work focused on inducibility by oxidative stress,
because this is a component common to many environmen-
tally induced challenges. Both in response to ozone, whose
primary action is at the cell surface, and in the cat2 mutant,
where the initial oxidative trigger is peroxisomal, NADP-ME2
was signicantly induced, by several fold. In the case of
ozone, this induction at the transcript level appeared to be of
signicance for enzyme capacity, because it was accompanied
Fig. 5. Redox profiling of Col-0, nadp-me2, cat2, and cat2
nadp-me2 double mutants. Open bars, reduced forms; filled bars,
oxidized forms. Significant differences are indicated at P<0.05
by black symbols on the white bars (reduced forms) and white
symbols on the black bars (oxidized forms). *Comparison of
mutants with Col-0.
+
Comparison of double mutants with cat2.
4
5
6
7
8
9
4
5
6
7
8
9
Colony forming units (log
10
cm
-2
leaf area)
Pst-DC3000
Psm-ES4326
*
*
*
*
*
*
cat2
cat2 nadp-me2-1
cat2 nadp-me2-2
cat2
cat2 nadp-me2-1
cat2 nadp-me2-2
Fig.6. Impact of nadp-me mutations on cat2-induced resistance
to Pst-DC3000 and Psm-ES4326. Open bars, cat2; filled bars,
cat2 nadp-me double mutants. Left, leaves sampled at 0 hours
post-inoculation; right, leaves sampled at 24 (Pst-DC3000) or
48 (Psm-ES4326) h post-inoculation. *Significant difference from
Col-0 at P <0.05 (for both bacteria, growth was significantly lower
in cat2 backgrounds than in Col-0). Dotted lines indicate the Col-0
value.
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from

NADP-malic enzyme and oxidative stress
|
3611
by an increase in extractable NADP-ME activity, an effect
that was not observed in the nadp-me2 knockout (Fig. 2).
Thus, increased NADP-ME activity seems to be part of the
oxidative stress responses, and this increase is linked to induc-
tion of NADP-ME2 at the transcriptionallevel.
A search of available microarray datasets for the cat2
mutant showed that NADP-ME2 was the only ME-encoding
gene that was signicantly induced. Indeed, of all the candi-
date NADPH-generating enzymes in the cytosol, NADP-ME2
showed the clearest response. Despite this, it seems that the
responses of Arabidopsis to oxidative stress are little affected
by the loss of its function.
NADP-ME does not play an irreplaceable role in
responses to oxidativestress
Ectopic expression of NADP-ME2 has been reported to
enhance stress tolerance (Laporte etal., 2002; Liu etal., 2007).
However, such studies cannot provide information on gene-
specic functions. Using a specic loss-of-function approach,
NADP-ME2 was shown to be required for responses of
Arabidopsis to fungal infection (Voll et al., 2012). The pre-
sent data on resistance to different bacteria also suggest
some role for the enzyme in biotic stress responses, although
statistically signicant increases in pathogen growth were
observed in nadp-me2 compared with Col-0 only in one of
two experiments in which responses to Psm were analysed
(Fig. 1). A key point may be timing, as the effects on fun-
gal responses were observed during the very early signalling
events that occur in the rst few minutes after infection (Voll
et al., 2012). However, plants can be exposed to prolonged
oxidative stress in the natural environment, and this may be
a major determinant of plant performance and yield (Foyer
and Noctor, 2009). Such conditions can occur, for example,
during chronic exposure to ozone. Despite its induction by
ozone treatment, loss of NADP-ME2 function produced no
effect on the visible symptoms produced by exposure to ele-
vated doses of this oxidizing pollutant.
The cat2 mutant is a useful system for evaluating potential
functions of other antioxidative enzymes. When CAT2 is func-
tional, it prevents photorespiration-linked oxidative stress in
optimal conditions (Vandenabeele etal., 2004; Queval etal.,
2007). As the other two catalase genes (CAT1 and CAT3) are
not thought to be expressed at appreciable levels in photo-
synthetic cells (Mhamdi etal., 2010c), loss of CAT2 function
places a greater load on pathways that supply reductants to
peroxidases, and this is notably reected in the shift of cellular
thiol-disulde status towards an oxidized condition (Fig. 5).
This increased oxidative load provides a useful context for
evaluating the potential functions of reductant-generating
enzymes. For example, loss of glutathione reductase 1 func-
tion in gr1 knockout mutants has in itself no obvious effect on
plant phenotype, but it dramatically impacts on phenotypes
and redox state in the cat2 background (Mhamdi etal., 2010b).
With respect to NADPH-producing enzymes, loss of cytosolic
NADP-ICDH function also modulates responses in cat2:
both pathogenesis responses and glutathione oxidation are
reinforced in double cat2 icdh mutants (Mhamdi etal., 2010a).
The gr1 and icdh mutations also modulate ozone responses in
Arabidopsis (Dghim et al., 2013), showing that their impact
during oxidative stress is not limited to the cat2 background.
Another NADPH-linked enzyme, the NADPH oxidase
encoded by AtRbohF, also produces specic effects on cat2-
triggered redox and pathogenesis responses (Chaouch etal.,
2012). In contrast to these previous observations, and despite
its clear inducibility by oxidative stress, loss of NADP-ME2
function produced little or no effect on cat2 characteristics.
Introduction of this secondary mutation did not alter cat2
phenotypes (Fig.4) or affect cat2-induced resistance to bac-
teria (Fig.6). Aprevious study of nadp-me2 mutants reported
increases in the carbon substrate, malate, and decreases in the
carbon product, pyruvate (Voll etal., 2012), suggesting that the
enzyme contributes to respiratory pathways in vivo. However,
no data were presented on NADP(H) or related redox pools.
Similar to the lack of effect on cat2 phenotypes, introduction
of the secondary nadp-me2 mutations had little effect on redox
proles. Although one double mutant showed a tendency
towards lower NADPH, in the other double mutant the status
of NADP(H) was very similar to cat2 (Fig.5).
Accumulation of GSSG is a well-known response in cat-
alase-decient plants but also occurs in response to many
environmental stresses (Vanacker et al., 2000; Bick et al.,
2001; Gomez etal., 2004; Koornneef etal., 2008). Recent data
suggest that modulation of glutathione status may play an
important role in linking oxidative stress to downstream phy-
tohormone-linked pathways (Han etal., 2013; Mhamdi etal.,
2013). Previous studies of the cat2 mutant strongly suggest
that a large part of the accumulated GSSG is found in com-
partments other than the cytosol, notably the chloroplast and
vacuole (Queval etal., 2011; Han etal., 2013). Nevertheless,
the extent of GSSG accumulation in cat2 is secondarily
affected by loss of function of other NADPH-linked enzymes
found in the cytosol, even though these mutations produce
much less obvious effects on NADP(H) contents measured in
whole leaf extracts (Mhamdi etal., 2010a,b). Thus, irrespec-
tive of the subcellular compartments in which GSSG most
strongly accumulates, glutathione status can be considered
a useful marker that provides a more sensitive indication of
changes in cytosolic redox processes than NADP(H) itself
in oxidative stress conditions. However, nadp-me2 mutations
produced little or no effect on the cat2-dependent modulation
of glutathione status. This contrasts with the effect of knock-
ing out the cytosol-located GR1 activity, which dramatically
exacerbates cat2-triggered accumulation of GSSG (Mhamdi
etal., 2010b). In view of this observation, a requirement for
NADP-ME to contribute NADPH to maintain GR activity
would predict more severe accumulation of GSSG in cat2
nadp-me2 mutants than in cat2. This effect is indeed observed
in cat2 icdh mutants, implicating cICDH as a non-redundant
player in providing NADPH to GR (Mhamdi etal., 2010a).
If, on the other hand, NADP-ME activity were essential to
provide reducing power for reactive oxygen species-producing
NADPH oxidases, cat2-triggered GSSG accumulation might
be expected to be weakened in cat2 nadp-me2 double mutants,
as observed in cat2 atrbohF (Chaouch etal., 2012). The present
results do not allow us to discount that NADP-ME is required
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from

3612
|
Li etal.
to generate NADPH for both antioxidant and pro-oxidant
processes, and that the similar glutathione status in cat2
and cat2 nadp-me2 lines is the result of opposing effects that
exactly cancel each other. Nevertheless, the simpler conclu-
sion, based on the observations of Figs 4 and 5 taken together,
is that NADP-ME does not play an important, irreplaceable
role in producing NADPH for oxidative stress responses.
Concluding remarks
A recent report implicated NADP-ME2 in early events during
plant–pathogen interactions (Voll etal., 2012), but the present
data suggest that this enzyme is not functionally required for
longer-term redox homeostatic or phenotypic responses to
oxidative stress. This contrasts with the evident response of
the gene to oxidative stress at the transcriptional level. While
the slight effects on basal resistance to one bacterial strain
(Fig.1) provide further evidence that NADP-ME2 plays a role
during biotic stress responses, such effects may not be redox-
linked: emerging evidence suggests that specic organic acids
could have a signalling role in stress responses (Finkemeier
etal., 2013). Given that NADP-ME2 encodes most of the leaf
NADP-linked ME activity, the apparent dispensability of its
expression could reect biochemical redundancy due to the
presence in plant cells of other enzymes that can link malate
to pyruvate. Similarly, in terms of NADPH, several NADP-
linked dehydrogenases may cooperate to generate reductant
for redox reactions during oxidative stress. Interestingly, the
dispensability of NADP-ME in such conditions contrasts with
our previous data for cICDH, even though the latter is much
less obviously induced by oxidative stress than NADP-ME2.
Together, these observations emphasize the importance of
establishing the impact of loss of function, as well as the dan-
ger of drawing conclusions on a given enzyme’s importance
from its activity or inducibility. Future studies will aim at
identifying the importance or redundancy of other NADPH-
generating enzymes in the response to oxidative stress.
Supplementarydata
Supplementary data are available at JXB online.
Supplementary Fig. S1. Genotyping of cat2 nadp-me dou-
ble mutants. Primer sequences are listed in Supplementary
Table S1.
Supplementary Fig. S2. Expression analysis of genes
encoding cytosolic NADP-ME in Arabidopsis.
Supplementary Fig. S3. Characterization of nadp-me2
mutants.
Supplementary Table 1. Oligonucleotide sequences used
in this study for genotyping (PCR) or analysis of transcript
abundance (RT-PCR, qRT-PCR).
Acknowledegments
We thank the Salk Institute Genomic Analysis Laboratory,
CA, USA, for providing the sequence-indexed Arabidopsis
T-DNA insertion mutants and the Nottingham Arabidopsis
Stock Centre, UK, for supply of seed stocks. This work
was partly funded by the French Agence Nationale de la
Recherche projects ‘Vulnoz’ project no. ANR-08-VULN-012
and ‘Cynthiol’ project no. ANR-12-BSV6-0011. The authors
thank Stéphane Martin for technical support. S.L.thanks the
Université de Paris Sud and Michel Dron (IBP, Orsay) for
help with funding towards his PhD studies.
References
Anderson LE, Duggan JX. 1976. Light modulation of glucose
6-phosphate dehydrogenase. Partial characterisation of the
light inactivation system and its effects on the properties of the
chloroplastic and cytoplasmic forms of the enzyme. Plant Physiology
58, 135–139.
Bick JA, Setterdahl AT, Knaff DB, Chen Y, Pitcher LH, Zilinskas
BA, Leustek T. 2001. Regulation of the plant-type 5′-adenylyl sulfate
reductase by oxidative stress. Biochemistry 40, 9040–9048.
Casati P, Drincovich MF, Edwards GE, Andreo CS. 1999. Malate
metabolism by NADP-malic enzyme in plant defence. Photosynthesis
Research 61, 99–105.
Chaouch S, Queval G, Noctor G. 2012. AtrbohF is a crucial
modulator of defence-associated metabolism and a key actor in the
interplay between intracellular oxidative stress and pathogenesis
responses in Arabidopsis. The Plant Journal 69, 613–627.
Chaouch S, Queval G, Vanderauwera S, Mhamdi A, Vandorpe
M, Langlois-Meurinne M, Van Breusegem F, Saindrenan P,
Noctor G. 2010. Peroxisomal hydrogen peroxide is coupled to biotic
defense responses by ISOCHORISMATE SYNTHASE1 in a daylength-
related manner. Plant Physiology 153, 1692–1705.
Dever LV, Blackwell RD, Fullwood NJ, Lacuesta M, Leegood
RC, Onek LA, Pearson M, Lea PJ. 1995. The isolation and
characterization of mutants of the C4 photosynthetic pathway. Journal
of Experimental Botany 46, 1363–1376.
Dghim AA, Mhamdi A, Vaultier MN, Hasenfratz-Sauder MP,
Le Thiec D, Dizengremel P, Noctor G, Jolivet Y. 2013. Analysis
of cytosolic isocitrate dehydrogenase and glutathione reductase 1
in photoperiod-influenced responses to ozone using Arabidopsis
knockout mutants. Plant, Cell & Environment doi: 10.1111/pce.12104
(in press).
Dizengremel P, Le Thiec D, Hasenfratz-Sauder MP, Vaultier MN,
Bagard M, Jolivet Y. 2009. Metabolic-dependent changes in plant
cell redox power after ozone exposure. Plant Biology 11, 35–42.
Finkemeier I, König AC, Heard W, Nunes-Nesi A, Pham PA,
Leister D, Fernie AR, Sweetlove LJ. 2013. Transcriptomic analysis
of the role of carboxylic acids in metabolite signaling in Arabidopsis
leaves. Plant Physiology 162, 239–253.
Foyer CH, Noctor G. 2009. Redox regulation in photosynthetic
organisms: signaling, acclimation, and practical implications.
Antioxidants & Redox Signaling 11, 861–905.
Furbank RT, Foyer CH. 1988. C
4
plants as valuable model
experimental systems for the study of photosynthesis. New
Phytologist 109, 265–77.
Gerrard Wheeler MC, Arias CL, Maurino VG, Andreo CS,
Drincovich MF. 2009. Identification of domains involved in the
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from

NADP-malic enzyme and oxidative stress
|
3613
allosteric regulation of cytosolic Arabidopsis thaliana NADP-malic
enzymes. FEBS Journal 276, 5665–5677.
Gerrard Wheeler MC, Arias CL, Tronconi MA, Maurino VG,
Andreo CS, Drincovich MF. 2008. Arabidopsis thaliana NADP-malic
enzyme isoforms: high degree of identity but clearly distinct properties.
Plant Molecular Biology 67, 231–242.
Gerrard Wheeler MC, Tronconi MA, Drincovich MF, Andreo CS,
Flugge UI, Maurino VG. 2005. A comprehensive analysis of the
NADP-malic enzyme gene family of Arabidopsis. Plant Physiology
139, 39–51.
Gomez LD, Vanacker H, Buchner P, Noctor G, Foyer CH. 2004.
Intercellular distribution of glutathione synthesis and its response to
chilling in maize. Plant Physiology 134, 1662–1671.
Han Y, Chaouch S, Mhamdi A, Queval G, Zechmann B, Noctor
G. 2013. Functional analysis of Arabidopsis mutants points to novel
roles for glutathione in coupling H
2
O
2
to activation of salicylic acid
accumulation and signaling. Antioxidants & Redox Signaling 18,
2106–2121.
Hodges M, Flesch V, Gálvez S, Bismuth E. 2003. Higher plant
NADP-dependent isocitrate dehydrogenases, ammonium assimilation
and NADPH production. Plant Physiology and Biochemistry 41,
577–585.
Hruz T, Laule O, Szabo G, Wessendorp F, Bleuler S, Oertle L,
Widmayer P, Gruissem W, Zimmermann P. 2008. Genevestigator
V3: a reference expression database for the meta-analysis of
transcriptomes. Advances in Bioinformatics 2008, 420747.
Kangasjärvi J, Jaspers P, Kollist H. 2005. Signalling and cell death
in ozone-exposed plants. Plant, Cell & Environment 28, 1021–1036.
Kelly GJ, Gibbs M. 1973. Nonreversible
d-glyceraldehyde
3-phosphate dehydrogenase of plant tissues. Plant Physiology 52,
111–118.
Koornneef A, Leon-Reyes A, Ritsema T, Verhage A, Den Otter
FC, Van Loon LC, Pieterse CMJ. 2008. Kinetics of salicylate-
mediated suppression of jasmonate signaling reveal a role for redox
modulation. Plant Physiology 147, 1358–1368.
Langdale JA. 2011. C
4
Cycles: Past, present, and future research on
C
4
photosynthesis. Plant Cell 23, 3879–3892.
Laporte MM, Shen B, Tarczynski MC. 2002. Engineering for
drought avoidance-expression of maize NADP-malic enzyme in
tobacco results in altered stomatal function. Journal of Experimental
Botany 53, 699–705.
Liu S, Cheng Y, Zhang X, Guan Q, Nishiuchi S, Hase K, Takano
T. 2007. Expression of an NADP-malic enzyme gene in rice (Oryza
sativa. L) is induced by environmental stresses; over-expression of the
gene in Arabidopsis confers salt and osmotic stress tolerance. Plant
Molecular Biology 64, 49–58.
Maier A, Zell MB, Maurino VG. 2011. Malate decarboxylases:
evolution and roles of NAD(P)-ME isoforms in species performing
C
4
and C
3
photosynthesis. Journal of Experimental Botany 62,
3061–3069.
Meyer T, lscher CH, Schwöppe C, Schaewen AV. 2011.
Alternative targeting of Arabidopsis plastidic glucose-6-phosphate
dehydrogenase G6PD1 involves cysteine-dependent interaction with
G6PD4 in the cytosol. The Plant Journal 66, 745–758.
Mhamdi A, Hager J, Chaouch S, etal. 2010b. Arabidopsis
GLUTATHIONE REDUCTASE1 plays a crucial role in leaf responses
to intracellular hydrogen peroxide and in ensuring appropriate gene
expression through both salicylic acid and jasmonic acid signaling
pathways. Plant Physiology 153, 1144–1160.
Mhamdi A, Han Y, Noctor G. 2013. Glutathione in phytohormone-
dependent responses: teasing apart signaling and antioxidant
functions. Plant Signaling & Behavior 8, e24181.
Mhamdi A, Mauve C, Gouia H, Saindrenan P, Hodges M, Noctor
G. 2010a. Cytosolic NADP-dependent isocitrate dehydrogenase
contributes to redox homeostasis and the regulation of pathogen
responses in Arabidopsis leaves. Plant, Cell & Environment 33,
1112–1123.
Mhamdi A, Queval G, Chaouch S, Vanderauwera S, Van
Breusegem F, Noctor G. 2010c. Catalase in plants: a focus on
Arabidopsis mutants as stress-mimic models. Journal of Experimental
Botany 61, 4197–4220.
Møller IM, Rasmusson AG. 1998. The role of NADP in the
mitochondrial matrix. Trends in Plant Science 3, 21–27.
Noctor G. 2006. Metabolic signalling in defence and stress: the
central roles of soluble redox couples. Plant, Cell & Environment 29,
409–425.
Queval G, Issakidis-Bourguet E, Hoeberichts FA, Vandorpe M,
Gakière B, Vanacker H, Miginiac-Maslow M, Van Breusegem
F, Noctor G. 2007. Conditional oxidative stress responses in the
Arabidopsis photorespiratory mutant cat2 demonstrate that redox
state is a key modulator of daylength-dependent gene expression,
and define photoperiod as a crucial factor in the regulation of H
2
O
2
-
induced cell death. The Plant Journal 52, 640–657.
Queval G, Jaillard D, Zechmann B, Noctor G. 2011. Increased
intracellular H
2
O
2
availability preferentially drives glutathione
accumulation in vacuoles and chloroplasts. Plant Cell & Environment
34, 21–32.
Queval G, Noctor G. 2007. A plate-reader method for the
measurement of NAD, NADP, glutathione and ascorbate in tissue
extracts. Application to redox profiling during Arabidopsis rosette
development. Analytical Biochemistry 363, 58–69.
Queval G, Thominet D, Vanacker H, Miginiac-Maslow M,
Gakière B, Noctor G. 2009. H
2
O
2
-activated up-regulation of
glutathione in Arabidopsis involves induction of genes encoding
enzymes involved in cysteine synthesis in the chloroplast. Molecular
Plant 2, 344–356.
Rius SP, Casati P, Iglesias AA, Gomez-Casati DF. 2006.
Characterization of an Arabidopsis thaliana mutant lacking a cytosolic
non-phosphorylating glyceraldehyde-3-phosphate dehydrogenase.
Plant Molecular Biology 61, 945–957.
Smith IK, Kendall AC, Keys AJ, Turner JC, Lea PJ. 1984.
Increased levels of glutathione in a catalase-deficient mutant of barley
(Hordeum vulgare L.). Plant Science Letters 37, 29–33.
Torres MA, Jones JDG, Dangl JL. 2006. Reactive oxygen species
signaling in response to pathogens. Plant Physiology 141, 373–378.
Valderrama R, Corpas FJ, Carreras A, Gómez-Rodríguez MV,
Chaki M, Pedrajas JR, Fernández-Ocaña A, Del Río LA, Barroso
JB. 2006. The dehydrogenase-mediated recycling of NADPH is a key
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from

3614
|
Li etal.
antioxidant system against salt-induced oxidative stress in olive plants.
Plant, Cell & Environment 29, 1449–1459.
Vanacker H, Carver TL, Foyer CH. 2000. Early H
2
O
2
accumulation
in mesophyll cells leads to induction of glutathione during the hyper-
sensitive response in the barley-powdery mildew interaction. Plant
Physiology 123, 1289–1300.
Vandenabeele S, Vanderauwera S, Vuylstecke M, Rombauts S,
Langebartels C, Seidlitz HK, Zabeau M, Van Montagu M, Inzé D,
Van Breusegem F. 2004. Catalase deficiency drastically affects gene
expression induced by high light in Arabidopsis thaliana. The Plant
Journal 39, 45–58.
Voll LM, Zell MB, Engelsdorf T, Saur A, Wheeler MG, Drincovich MF,
Weber APM, Maurino VG. 2012. Loss of cytosolic NADP-malic enzyme
2 in Arabidopsis thaliana is associated with enhanced susceptibility to
Colletotrichum higginsianum. New Phytologist 195, 189–202.
Von Schaewen A, Langenkämper G, Graeve K, Scheibe R.
1995. Molecular characterization of the plastidic glucose-6-phosphate
dehydrogenase from potato in comparison to its cytosolic counterpart.
Plant Physiology 109, 1327–1335.
Wakao S, Benning C. 2005. Genome-wide analysis of glucose-6-
phosphate dehydrogenases in Arabidopsis. The Plant Journal 41,
243–256.
Wenderoth I, Scheibe R, Von Schaewen A. 1997. Identification of
the cysteine residues involved in redox modification of plant plastidic
glucose-6-phosphate dehydrogenase. Journal of Biological Chemistry
272, 26985–26990.
Willekens H, Chamnongpol S, Davey M, Schraudner M,
Langebartels C, Van Montagu M, Inzé D, Van Camp W. 1997.
Catalase is a sink for H
2
O
2
and is indispensable for stress defense in
C3 plants. EMBO Journal 16, 4806–4816.
at Biomedische Bibliotheek on April 7, 2016http://jxb.oxfordjournals.org/Downloaded from