Differential impact of chronic ozone exposure on expanding and fully expanded poplar leaves.

Abstract

Populus tremula L. × Populus alba L. (Populus ×c anescens (Aiton) Smith) - clone INRA 717-1-B4 saplings (50 cm apex to base and carrying 19 leaves on average) - were followed for 28 days. Half of the trees were grown in charcoal-filtered air while the other half were exposed to 120 ppb ozone for 11 h a day during the light period. The expanding leaf number 4 was tagged at the beginning of the experiment and finished expansion between 7 and 14 days. These leaves were harvested weekly for biochemical and proteome analyses using quantitative bidimensional electrophoresis (DiGE). Independent of the ozone treatment, all the analyses allowed a distinction between expanding and adult leaves. The results indicate that during the expansion phase (Days 0-7) the enzymatic machinery of the leaves is set up, and remains dynamically stable in the adult leaves (Days 14-28). Although ozone had no apparent effect on expanding leaves, the metabolic stability in fully expanded leaves observed in ozone-free plants was disturbed after 2 weeks of exposure and a stress-induced response became apparent.

Full text

Page 1

Differential impact of chronic ozone exposure on expanding and fully
expanded poplar leaves
SACHA BOHLER,1,2KJELL SERGEANT,1ISABELLE LEFÈVRE,1YVES JOLIVET,2
LUCIEN HOFFMANN,1JENNY RENAUT,1PIERRE DIZENGREMEL2
and JEAN-FRANCOIS HAUSMAN1,3
1
Department of Environment and Agro-biotechnologies, CRP-Gabriel Lippmann, 41 rue du Brill, L-4422 Belvaux, GD Luxembourg
2
Nancy-Université, UMR 1137 Ecologie et Ecophysiologie Forestières, F-54506 Vandoeuvre-lès-Nancy, France
3
Corresponding author (hausman@lippmann.lu)
Received April 9, 2010; accepted August 16, 2010; handling Editor Marc Abrams
Summary
L. (Populus×canescens (Aiton) Smith) — clone INRA 717-
1-B4 saplings (50 cm apex to base and carrying 19 leaves
on average) — were followed for 28 days. Half of the trees
were grown in charcoal-filtered air while the other half were
exposed to 120 ppb ozone for 11 h a day during the light
period. The expanding leaf number 4 was tagged at the
beginning of the experiment and finished expansion
between 7 and 14 days. These leaves were harvested weekly
for biochemical and proteome analyses using quantitative
bidimensional electrophoresis (DiGE). Independent of the
ozone treatment, all the analyses allowed a distinction
between expanding and adult leaves. The results indicate
that during the expansion phase (Days 0–7) the enzymatic
machinery of the leaves is set up, and remains dynamically
stable in the adult leaves (Days 14–28). Although ozone
had no apparent effect on expanding leaves, the metabolic
stability in fully expanded leaves observed in ozone-free
plants was disturbed after 2 weeks of exposure and a stress-
induced response became apparent.
PopulustremulaL.×
Populusalba
Keywords: air pollution, biochemistry, difference in gel
electrophoresis (DiGE), photosynthesis, primary carbon
metabolism, proteomics.
Introduction
The formation of leaves, the site of photosynthesis and thus
the source from which the plant draws the energy for its life
cycle, is an important event in plant development. Leaf for-
mation can be divided into three phases: (i) leaf initiation at
the site of the shoot apical meristem, which can be followed
by bud dormancy, (ii) development to the state of a fully
expanded adult leaf and finally (iii) senescence and abscis-
sion. Among these three phases, leaf initiation and
senescence are well studied in model plants such as
Arabidopsis (initiation and senescence) (Grbic and Bleecker
1995, Buchanan-Wollaston et al. 2005) and poplar (mainly
senescence) (Keskitalo et al. 2005); however, fewer studies
focus on leaf expansion.
The process of leaf initiation is tightly controlled
(Fleming 2005) by multiple genes (Schoof et al. 2000,
Saiga et al. 2008) and auxin fluxes (Benkova et al. 2003,
Reinhardt et al. 2003). In the process of senescence
initiation, several growth regulators (e.g. ethylene, salicylic
acid and jasmonic acid) can be involved (Grbic and
Bleecker 1995, Buchanan-Wollaston et al. 2005, Reinbothe
et al. 2009). Like leaf initiation, senescence is tightly regu-
lated (Hopkins et al. 2007), allowing the plant to recycle
carbon and nitrogen (Andersson et al. 2004) before leaf
abscission.
Information about the physiological and biochemical vari-
ation of a leaf between initiation and senescence is only
poorly represented in the literature. It was shown that the
photosynthetic rate increases in expanding leaves, that chlor-
ophyll accumulates and that respiration is highest during the
expansion phase (Kennedy and Johnson 1981, Reich 1983).
There are some reports about the control underlying leaf
shape development (Eshed et al. 2001), but only very few
describe the metabolic changes in leaves during expansion
and adult leaf life (Maayan et al. 2008).
Tropospheric ozone is a secondary pollutant formed by
the decomposition of nitric oxides under solar UV radiation,
a decomposition catalyzed by a number of other atmos-
pheric components such as volatile organic compounds and
hydroxy radicals (Oke 1987). During the industrialization
age, background concentrations and the number and inten-
sity of peak episodes of ozone have increased. As recently
reviewed, ozone has a negative effect on plants (Tkacz et al.
2008, Fuhrer 2009, Renaut et al. 2009). Ozone enters the
leaves through the stomata and, upon contact with the apo-
plast, the molecule degrades into reactive oxygen species
Tree Physiology 30, 1415–1432
doi:10.1093/treephys/tpq082
© The Author 2010. Published by Oxford University Press. All rights reserved.
For Permissions, please email: journals.permissions@oxfordjournals.org
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 2

(ROS) (Rao et al. 2000). These highly reactive ROS attack
cell components such as lipids and proteins, and induce a
series of signaling pathways involving several plant growth
regulators (Kangasjärvi et al. 2005, Castagna and Ranieri
2009). Reduction in biomass yield, symptoms such as
chloroses and necroses on leaves and alterations of metab-
olism are typical plant responses to ozone exposure (Guidi
et al. 2001, Degl’Innocenti et al. 2002). It is well known
that the impact that ozone may have on plants depends on
the effective ozone flux, which is a balance between the
degree of stomatal opening and the level of antioxidant
defense (Wieser and Matyssek 2007, Dizengremel et al.
2008).
In a previous study, we have shown that exposure of
poplar to 120 ppb ozone for 1 month leads to the formation
of necroses in adult leaves, but not in expanding ones
(Bohler et al. 2007). Variations in carbon metabolism and
othercellular functionswere
14 days of treatment, but nearly absent after only a few
days, except for ribulose-1,5-bisphosphate carboxylase oxy-
genase activase (RuBisCO activase), which decreased as
early as after 3 days (Bohler et al. 2007). It was confirmed
that physiological parameters of photosynthesis and photo-
respiration do not change in leaves under expansion under
ozone stress, and that variations appear only after full
expansion of the leaves (Bagard et al. 2008).
The present study attempts to give an overview of the
proteome changes observed during the development of
leaves upon ozone exposure. These results will be compared
with the observed changes during normal leaf development.
The discussion of the results of this proteomic study will be
supported by morphological measurements and biochemical
analyses of pigments and carbohydrates.
clearly visibleafter
Materials and methods
Plant material and ozone treatment
Plant material was obtained and treated as previously
described (Bohler et al. 2007). Briefly, poplar clones
(Populus tremula L. × Populus alba L. (Populus×canescens
(Aiton) Smith)—clone INRA 717-1-B4) were multiplied in
vitro, transferred to a mixture of peat/perlite (Gramoflor SP1
Universel), transplanted into 5-l pots containing peat/perlite
fertilized with 15 g of slow-release nutritive granules
(Nutricote T-100, N/P/K/MgO 13/13/13/2, Fertil, Boulogne-
Billancourt, France), and grown at 75/85% relative humidity
(day/night) over a 14-h light period (Sun T Agro, Philips,
Eindhoven, The Netherlands; intensity: 250–300 µmol m−2
s−1) and at 22/18 ± 2 °C (day/night) for a period of 5 weeks.
For the ozone treatment, plants were transposed into eight
fumigation chambers with 10 plants per chamber. Four
chambers were randomly used as controls and four for
ozone treatment. Control plants were grown in charcoal-
filtered air, whereas treated plants were exposed to charcoal-
filtered air supplemented with 120 ppb ozone for 13 h per
day during the light period or to charcoal-filtered air for
controls. All other conditions (humidity, light and tempera-
ture) were as described above. Leaf number 4, counted from
the last leaf emerged at the apex, was tagged at the begin-
ning of the treatment on each plant. They are referred to as
leaves of interest. Two leaves from each chamber were har-
vested for analysis and immediately frozen in liquid nitro-
gen after 0, 7, 14, 21 and 28 days of treatment.
Morphological measurements
Growth (height and diameter) and the number of newly
formed leaves were determined for each plant after 0, 7, 14,
21 and 28 days of exposure. Height measurement was based
on the length of the stems from the collar to the apex, and
diameter expansion was estimated via measurements of the
stem 1.5 cm above the collar with a caliper. The number of
abscised leaves was recorded daily. Concomitantly, visible
symptoms of ozone treatment (chlorosis and necrosis) were
recorded.
The lobe length (petiole excluded) and width of the
leaves of interest were measured on each sampling date
between 0 and 21 days of treatment. Leaf area was calcu-
lated using a formula developed specifically for this poplar
clone (Bagard et al. 2008).
Chlorophyll fluorescence
Chlorophyll fluorescence measurements were performed on
the leaf just below the leaf of interest (leaf number 5) with a
Fluorescence Monitoring System (FMS 1; Hansatech,
Norfolk, UK). Leaf clips were attached to target leaves 20
min prior to measurements for dark acclimation. The fol-
lowing programwas created
Monitoring Software 3.11 based on the default program
from the manufacturer: gain setting 50, modulating light
intensity setting 1, script logging on. Fv/Fmwas measured
for 2.5 s at an intensity of 90 units with a light pulse of 0.8 s.
After 10 s, actinic light was switched on to 30 units, and
after an additional 300 s PhiPSII was measured for 2.5 s at
an intensity of 90 units with a light pulse of 0.8 s. Actinic
light was then switched off, and after 10 s Fo0was measured
for 2 s. Three plants per chamber were measured at each
sampling date, amounting to 12 measurements per treatment
per sampling date.
ontheFluorescence
Soluble protein extraction and labeling
Proteins were extracted from one leaf per chamber per
sampling date, i.e., four leaves per treatment and per date,
amounting to four biological replicates. Soluble protein
extraction was carried out as described previously (Kieffer
et al. 2009), with small modifications. Briefly, 300 mg of
fresh leaf material were ground in liquid nitrogen, and pro-
teins were suspended in precipitation buffer (20% trichloro-
acetic acid (TCA) and 0.1% w/v dithiothreitol (DTT) in
BOHLER ET AL.1416
TREE PHYSIOLOGY VOLUME 30, 2010
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 3

ice-cold acetone) and washed three times with ice-cold
acetone containing 0.1% w/v DTT. Precipitated protein
pellets were freeze dried. Dried samples were re-solubilized
in labeling buffer (7 M urea, 2 M thiourea, 4% w/v 3-[(3-
Cholamidopropyl)dimethylammonio]-1-propanesulfonate
(CHAPS), 30 mM Tris) and the pH of the supernatant was
adjusted to 8.5. The protein concentration was determined
using the 2D Quant Kit (GE Healthcare, Little Chalfont,
UK) according to the manufacturer’s instructions. After
extraction, proteins were used for a multiplex analysis by
fluorescence difference in
(Skynner et al. 2002).
Protein extracts and a pooled internal standard were
labeled with CyDyes™ (GE Healthcare) prior to electro-
phoresis as previously described (Kieffer et al. 2009).
Ninety micrograms of proteins (two samples of 30 µg each
and 30 µg of internal standard) were loaded on each gel and
separated by 2D electrophoresis.
gelelectrophoresis(DiGE)
Bidimensional electrophoresis
To each sample mix consisting of two samples and the
internal standard and made up to 450 µl with lysis buffer
(labeling buffer without Tris), 2.7 µl of Destreak reagent (GE
Healthcare), 9 µl of IPG buffer 3–10 NL (GE Healthcare) and
5 µl of a saturated bromophenol blue solution were added.
Sample loading on ReadyStrip IPG strips pH 3–10 NL, 24
cm (Biorad, Hercules, CA, USA), was carried out overnight
by passive rehydration. Isoelectric focusing (IEF) was carried
out on an Ettan IPGphor Manifold (GE Healthcare) in an
IPGphor IEF unit (GE Healthcare). The following protocol
was used: 150 V for 2 h, 300 V for 2 h, gradient to 1000 V
over 6 h, 1000 V for 1 h, gradient to 8000 V over 4 h, 8000
V until ?70,000 V h were reached at 20 °C with a maximum
current setting of 50 µA per strip. The second dimension was
run as previously described (Kieffer et al. 2009).
Additional preparative gels were run as described above
with 300 µg of an unlabeled protein sample. Glass plates
were pretreated with Bind Silane (GE Healthcare) to guaran-
tee adherence of the gels to the glass plates during gel
picking. These gels were poststained with Lava Purple
(Fluorotechnics, Sydney, Australia) according to the manu-
facturer’s instructions.
Image capture and analysis
The gel images of the different samples were acquired using
a Typhoon Variable Mode Imager 9400 (GE Healthcare) at
a resolution of 100 µm according to the instructions pro-
vided for each dye. Images were analyzed using the
Decyder v6.5.14.1 software (GE Healthcare). Selected spots
of interest (absolute abundance variation of at least 1.5-fold,
ANOVA, P< 0.01) were located and a picking list was gen-
erated. Spots of interest were excised and digested as pre-
viously described (Bohler et al. 2007). MS and MS/MS
spectra were acquired using an Applied Biosystems 4800
Proteomics Analyzer (Applied Biosystems, Sunnyvale, CA,
USA). Calibration was carried out with the peptide mass
calibration kit for 4700 (Applied Biosystems). Proteins were
identified by searching against the NCBI poplar protein
databases (downloaded 11 June 2009, 99,629 sequences)
and poplar EST (downloaded 11 June 2009, 419,944
sequences) using MASCOT (Matrix Science, http://www.
matrixscience.com, London, UK). All searches were carried
out allowing for a mass window of 150 ppm for the precur-
sor mass and 0.75 Da for fragment ion masses. The search
parameters allowedfor carboxyamidomethylation
cysteine as fixed modification. Oxidation of methionine,
oxidation of tryptophan (single oxidation, double oxidation
and kynurenin) and pyrrolidone carboxylic acid (N-terminal
glutamine and glutamic acid) were set as variable modifi-
cations. Homology identification was retained with prob-
ability set at 95%.
The sub-cellular localization of the proteins was estimated
with the following tools: ChloroP 1.1 Server (http://www.cbs.
dtu.dk/services/ChloroP/), MITOPROT (http://ihg2.helmholtz-
muenchen.de/ihg/mitoprot.html) and PTS1 predictor (http://
mendel.imp.ac.at/mendeljsp/sat/pts1/PTS1predictor.jsp).
of
Chlorophyll and carotenoid analysis
Pigments were extracted from one leaf per chamber per
sampling date, i.e. four leaves per treatment and per date,
amounting to four biological replicates. Two extracts were
made per sample, amounting to two technical replicates.
Pigment extraction was carried out as previously described
(Bohler et al. 2007). In order to assess the recovery of the
method, β-apo-80-carotenal was added as the internal stan-
dard to the extraction mixture to reach a final concentration
of 20 µg ml-1. Identification and quantification of chloro-
phylls and carotenoids were carried out as previously
described (Andre et al. 2007), with some modifications. A
gradient system was used involving four mobile phases:
eluent A was 100% water, eluent B was 100% acetonitrile,
eluent C was 80% methanol v/v with 0.5 M ammonium
acetate at pH 7.2 and eluent D was 100% ethyl acetate. The
flow rate was 1 ml min−1and the column temperature was
26 °C. The gradient elution profile was as previously
described (Kieffer et al. 2009).
Carbohydrate analysis
Carbohydrates were extracted from one leaf per chamber per
sampling date, i.e. four leaves per treatment and per date,
amounting to four biological replicates. Carbohydrate extrac-
tion was carried out as previously described (Kieffer et al.
2009). Carbohydrates were separated by high-performance
anion exchange chromatography coupled with pulsed ampero-
metric detection as previously described (Oufir et al. 2008).
Starch analysis
Extraction
(Gaucher
was
et
carried
al.
outas previously
some
described
modifications. 2003), with
LEAF DEVELOPMENT AND OZONE EXPOSURE 1417
TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 4

Approximately 100 mg of fresh weight poplar leaves were
ground in liquid nitrogen, mixed thoroughly with 750 µl of
70% v/v methanol and centrifuged for 10 min at 17,000g at
4 °C. The pellet was re-suspended in 750 µl of 70 % v/v
methanol and centrifuged once more. The pellet containing
the starch was then dried for 1 h at 30 °C. Starch was sub-
sequently degraded to glucose by adding 500 µl of 2% v/v
HCl to the dried pellet and incubated at 99 °C for 2 h. After
centrifugation for 10 min at 17,000g and 4 °C, the super-
natant was collected.
Starch was quantified by a spectrophotometric method
using anthrone reagent. A glucose solution of 1 mg ml−1
was used as a standard. Lysate (5 µl) was mixed with 45 µl
of water, 50 µl of 2% v/v HCl and 1.5 ml of anthrone
reagent (0.15% [w/v] anthrone, 71% v/v sulfuric acid),
incubated at 99 °C for 20 min and cooled on ice. The absor-
bance was read at 620 nm.
Statistical analysis
Statistical analysis was carried out with the following
number of replicates: 4 biological replicates for proteomics,
carbohydrates and starch, 4 biological and 2 technical repli-
cates for pigment analysis, 12 biological replicates for fluor-
escence and 40 biological repetitions for growth, for each
condition at each sampling date. The proteomic effects of
leaf development were tested with a one-way ANOVA on
control samples only. The average ratio was calculated for
each time point versus Day 0. The effects of ozone were
tested with a two-way ANOVA, with treatment as factor one
and time as factor two. The average ratio was calculated for
each time point versus its control counterpart. Differentially
abundant proteins with an absolute ratio of at least 1.5-fold
for minimum one time point and a P-value of <0.01 were
selected. A principal component analysis (PCA) was run for
qualitative appreciation of the proteomic results. ANOVA
and PCA were performed on Decyder Extended Data
Analysis (v.1.0.14.1). For morphological measurements,
carbohydrates, starch and pigments, significance was calcu-
lated via a one-way ANOVA, with control/treated as factors
and growth and concentration, respectively, as the response.
These analyses were performed on MINITAB Statistical
Software v.15.1.1.0.
Results
Morphological measurements
Stem height growth of ozone-treated and untreated plants
(Table 1) was strongest during the first 7 days of the exper-
iment, but continued until the end of the experiment. There
were no significant effects of ozone. Similarly, stem diam-
eter (Table 1) increased all through the fumigation period in
both growth conditions. Significantly lower diameter growth
was observed for treated plants between Days 7 and 21.
Exposed and control plants shed their lower leaves during
the experiment. For treated plants, leaf loss became signifi-
cantly higher starting at Day 24 (Figure 1). The rate of new
leaf formation in fumigated plants also increased signifi-
cantly starting at Day 14 (Figure 1).
The leaves of interest (leaf number 4 counted from the
top and tagged at the beginning of the treatment) had strong
growth between Days 0 and 7 that continued briefly
between Days 7 and 14. Leaf length significantly increased
between Days 0 and 7, and leaf width between Days 0 and
14 (Figure 2a; 8 ×5 to 20× 15 cm). A statistically signifi-
cant difference in leaf area between control and ozone con-
ditions was detected only at Day 14 (Figure 2b). Leaves
probably reached maximum expansion shortly after 7 days
into the experiment.
Pinpoint-sizednecroses
ozone-exposed plants after ?14 days. They first developed
on mature leaves, and started to show up only very rarely
on young leaves later in the experiment. The leaves of inter-
est developed necroses when they were mature. The
necroses spread over mature leaves and were clearly visible
at the end of the treatment on all exposed plants. The young
and mature leaves of exposed plants also showed a marked
chlorosis starting around Day 21, with a loss of glossiness
of the cuticle.
startedappearing on
Chlorophyll fluorescence
Only two chlorophyll fluorescence parameters changed
during the ozone treatment, maximum quantum yield of
photosystem II (Fv/Fm) and nonphotochemical quenching
(NPQ; Figure 3). Fv/Fmincreased until Day 14 in ozone-
free leaves and then remained stable until the end of the
Table 1. Per cent increment of stem height and stem diameter compared with Day 0±standard deviation, in control and treated plants.
7 days14 days 21 days 28 days
Height (cm)
Control
Ozone
Diameter (cm)
Control
Ozone
34.60±3.65
34.31±8.03
68.88±7.88
71.00±10.34
98.77±12.84
102.45±13.91
126.68±20.13
133.71±18.58
22.98±6.65
19.96±7.62
44.39±9.52
34.61±7.09
60.58±11.40
37.83±6.52
74.32±14.93
45.15±6.54
Statistical significance was tested by one-way ANOVA between controls and treated. Significance is shown by bold numbers at P<0.001,
n=40.
BOHLER ET AL.1418
TREE PHYSIOLOGY VOLUME 30, 2010
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 5

experiment, while in treated leaves it increased only until
Day 7 and decreased again from Day 14 on. From Day 14
to the end of the treatment, Fv/Fm was distinctly lower in
treated leaves (Figure 3a). NPQ decreased gradually over
the time course of the experiment in both growth con-
ditions, but from Day 14 on remained significantly higher
in ozone-treated leaves (Figure 3b). The quantum yield of
photosystem II (PhiPSII) and photochemical quenching (qP)
decreased gradually over the time course of the experiment
in treatment and control conditions and showed no signifi-
cant variation due to ozone exposure (data are not shown).
Pigments
Content of chlorophyll a and b (Figure 4a), lutein and neox-
anthin (Figure 4b) increased in control leaves until Day 21
and then remained at the same level until Day 28. In
ozone-exposed plants, the highest concentration for these
pigments was already reached at 14 days, except for neox-
anthin where it was reached at 21 days. For each of these
pigments, the content was significantly lower in ozone-
treated plants from Day 14 on. Violaxanthin and zeaxanthin
are represented as a ratio of their content to that of chloro-
phyll a as previously described (Niyogi et al. 1997)
(Figure 4c and d). Both ratios clearly decreased over time in
both growth conditions, but remained higher in exposed
plants at the later time points, starting at Days 14 (violax-
anthin) and 21 (zeaxanthin). The violaxanthin/zeaxanthin
ratio was higher in treated plants, although significant only
on Day 14 (Figure 4e). Beta-carotene (Figure 4f) behaved
similarly in the two conditions, increasing until Day 21.
Carbohydrates
Measured carbohydrate levels were stable (glucose, sucrose
and starch) or slightly increasing (fructose) during the first
7 days in control plants, and then their content decreased
more or less strongly at Day 14 (Figure 5a–d). Sucrose
content then immediately increased again until Day 21 to
remain stable until Day 28, whereas glucose and fructose
remained at a low level at Day 21 and rose again at Day 28.
Starch levels rose slowly in control leaves from Days 14 to
28. Carbohydrate content in ozone-exposed plants roughly
followed the pattern of control leaves, with some significant
changes. Fructose levels were significantly higher in treated
plants at Days 14 and 21, glucose and sucrose only at
Day 14. Starch, on the other hand, was higher in treated
plants at Days 21 and 28.
Proteomics
The proteomic effects of normal leaf development were
tested with a one-way ANOVA on control spot maps only.
The average ratio was calculated for each time point versus
Day 0. The effects of ozone were tested with a two-way
ANOVA, with treatment as factor one and time as factor
two. The average ratio was also calculated for each ozone
treatmenttime pointversus
Differentially abundant proteins with an absolute ratio of at
least 1.5-fold for minimum one time point and a P-value
<0.01 were selected. When considering leaf development
and ozone treatment, the intensity of 470 protein spots was
significantly different. Of these, 404 proteins could be
matched and picked on preparative gels. Finally, 216 pro-
teins could be significantly identified. The proteins dis-
cussed below are presented in Table 2. Detailed tables are
available as Supplementary data at Tree Physiology Online.
This made a final total of 162 identified spots for leaf devel-
opment, 123 for ozone exposure, of which 69 were
common to both conditions (Figure 6a). The distribution of
the spot maps on the PCA plots shows that the most impor-
tant differences were between the samples of Day 0, corre-
sponding to small immature leaves, and the other samples
corresponding to mature leaves (Figure 6b), irrespective of
the growth conditions. When the Day 0 gels are removed
from the PCA to focus on the effects ozone has during leaf
itscontrol counterpart.
Figure 1. (a) Leaf loss count since Day 0 and (b) new leaf formation since Day 0. Statistical significance was tested by one-way ANOVA
between control and treated and is represented as follows: *P<0.05; **P<0.01; ***P <0.001. n =40. Error bars represent standard
deviation.
LEAF DEVELOPMENT AND OZONE EXPOSURE1419
TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 6

development, it becomes visible that the organization along
the axes separates the spot maps according to time and con-
dition (Figure 6c). It also illustrates that the results of the
different biological repetitions for each time point and con-
dition are grouped together, even though not perfectly. In
the cluster analysis, the Day 0 gels cluster together, as do
the Day 7 gels, independently of the treatment (Figure 6d).
The effects of ozone treatment become visible with Day 14
when the experimental points become clustered according
to the treatment. The variations in abundance of six promi-
nent proteins in ozone-free and ozone-treated samples are
illustrated in Figure 7, and the abundance profiles of all pro-
teins discussed below can be found in Supplementary data
S3 at Tree Physiology Online.
Discussion
Leaf development in control plants
In plants, leaves are the site of photosynthesis and hence the
major site of carbon metabolism. They are morphologically
and physiologically specialized for light-energy absorption,
CO2uptake and assimilation, and energy production. Only
little information can be found about expanding leaves, but it
is considered that photosynthesis and chlorophyll content
increase during leaf expansion (Kennedy and Johnson 1981,
Nogués et al. 2008).In the present study, important proteomic
changes in primary carbon metabolism were observed during
leaf expansion. The changes in carbon metabolism during leaf
Figure 2. (a) Representation of leaf size from measured lobe length and width (b) Calculated leaf area. Statistical significance in (a) is for the
values of leaf width only. Statistical significance was tested by one-way ANOVA between control and treated (b) and between each time
point of one condition against Day 0 of the same condition (a and b). Letters represent statistical significance for P<0.05, n=16 for Days 0
and 7, n=12 for Day 14 and n =8 for Day 21. Error bars represent standard deviation.
BOHLER ET AL. 1420
TREE PHYSIOLOGY VOLUME 30, 2010
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 7

development between Days 0 and 7 are summarized in
Figure 8. An important number of the identified proteins
involved in carbon metabolism belong to the Calvin cycle,
and are present as more than one isoform (Table 2). Most of
these increased in relative abundance during the first 7 days of
development, in parallel with the leaf expansion phase
(Figure 2). One of the proteins that increased in abundance is
RuBisCO LSU. This increase in RuBisCO LSU is one of the
few observations concerning leaf development that has pre-
viously been reported (Maayan et al. 2008). In parallel with
the accumulation of Calvin cycle proteins during the first 7
days (e.g., chloroplastic glyceraldehyde-3-phosphate dehydro-
genase, transketolase and sedoheptulose-1,7-bisphosphatase)
was the increase in abundance of photosystem proteins, as
was already observed previously (Maayan et al. 2008), and
associated pigments (Figure 4). Both phenomena seem logical
for expanding leaves, where chloroplasts are slowly develop-
ing and photosynthesis is activated, as is illustrated by the
chlorophyll fluorescence parameter Fv/Fm (Figure 3). During
the first 7 days of leaf expansion, a drop in abundance of
some glycolytic proteins was observed (e.g., cytosolic fruc-
tose bisphosphate aldolase and triose phosphate isomerase).
These enzymes are involved in the first half of glycolysis
where glucose is transformed into triose phosphate. It is
known that glycolysis and mitochondrial respiration are inhib-
ited by light in adult leaves (Tcherkez et al. 2005), but are
probably very active in expanding leaves (Reich 1983). Since
photosynthesis is only increasing slowly during leaf expan-
sion, it is most probable that carbohydrates for catabolism are
imported into the young leaves in a source–sink relationship
with older leaves (Dickson 1989). This is also illustrated by
the high concentration of glucose and fructose at the early
time points (Figure 5). As the chloroplastic metabolism is
built up, more triose phosphate is produced by the Calvin
cycle and the leaf starts sustaining itself. Although most of
these molecules are diverted into storage carbon molecules
during the day, a part of it starts fueling glycolysis. After
7 days, most carbon metabolism proteins observed seemed to
have reached a more or less stable level, and it is probably
between 7 and 14 days of expansion that the leaf becomes
mature and switches its role from sink to source leaf. It is also
during this time that glucose and fructose levels were lowest
as carbohydrates were exported to younger leaves in a now
inverted source–sink relationship. Mitochondrial malate dehy-
drogenase (MDH) levels dropped during the first 7 days of the
experiment, indicating a switch in the substrate for the Krebs
cycle from pyruvate to malate.
Between 7 and 28 days of the experiment, Calvin cycle
and glycolysis enzymes remained at relatively stable levels.
Chloroplast electron transport chain proteins, on the other
hand, continued to increase in abundance. As the plants
grew, younger leaves shaded the older ones. To maintain a
sufficient level of photosynthesis, it is likely that the
machinery was adapted to a more efficient use of available
light (Ishida et al. 1999).
Cytosolic glutamine synthetase (GS) increased gradually
overtheexperiment. This
for ammonium fixation from primary nitrogen uptake, and
also from protein recycling and photorespiration (Bernard
and Habash 2009). Since primary nitrogen assimilation
mostly happens in roots, it can be assumed that the
increase in GS observed in leaf development was due to an
increase in ammonium recycling from proteins, possibly
simply due to their turnover, and the increase in ammonium
productionin photorespiration.
and S-adenosylmethionine synthase were observed to
decrease over the entire time course of the experiment, the
strongest decrease clearly being during the first 7 days.
S-adenosylmethionine is the precursor for several pathways,
like biotin, polyamines and ethylene, and acts as a general
donor for methyl groups (Amir et al. 2002). Similar behav-
ior could be observed for alanine aminotransferase, which
catalyzes the transfer of amino groups to different substrates
(Orzechowski et al. 1999). Translational elongation factors
and ribosomal proteins decreased in abundance in different
patterns over the time of the experiment. This shows that
enzyme isresponsible
Methionine synthase
Figure 3. (a) Fv/Fm, maximum PhiPSII, and (b) NPQ, calculated from chlorophyll fluorescence. Statistical significance was tested by
one-way ANOVA between control and treated and between each time point of one condition against Day 0 of the same condition. Letters
represent statistical significance for ANOVA P<0.05. n=12. Error bars represent standard deviation.
LEAF DEVELOPMENT AND OZONE EXPOSURE 1421
TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 8

protein synthesis was lower in the mature leaf. This is
logical, because during development the leaf is ‘built’ with
and by proteins. Only elongation factor G increased in
abundance between Days 0 and 7, and a ribosome recycling
factor and elongation factor G are increased at Day 28.
They are both involved in the disassembly of ribosomal
subunits and RNA at the end of protein translation
(Fujiwara et al. 2004). Chaperones such as heat shock
protein 70 (HSP70) and disulfide isomerase, which are
responsible for correct protein folding, also show a down-
ward trend, confirming reduced protein synthesis in mature
leaves. Some isoforms of HSP are at relatively low levels
from the beginning, showing that specific isoforms may be
specifically needed during the developmental phase, while
others are needed to maintain basic metabolism. All these
changes indicate that the general metabolic activity in adult
leaves is lower than that in developing leaves.
Up and down variations among isoforms of the identified
detoxification proteins are not always consistent, which
could be due to different isoforms for each identified spot
or to unidentified posttranslational modifications. Catalase
showed an upward trend in abundance at the beginning of
the experiment, and a downward trend toward the end.
Catalase is one of the major enzymes of peroxisomes
(Reumann and Weber 2006). They are, among other things,
responsible for the detoxification of H2O2 from photore-
spiration. The increase in abundance of catalase would
therefore be linked to an increase in RuBisCO, which is
responsible for O2fixation in photorespiration. The levels of
some peroxidase isoforms were stable throughout the 28
Figure 4. Quantification of the pigments (a) chlorophyll a and b, (b) lutein and neoxanthin, (c) violaxanthin and (d) zeaxanthin, expressed as
their ratio to chlorophyll a, (e) the ratio between violaxanthin and zeaxanthin; (f) beta-carotene. Statistical significance was tested by
one-way ANOVA between control and treated and between each time point of one condition against Day 0 of the same condition. Letters
represent statistical significance for ANOVA P <0.05. n =8. Error bars represent standard deviation.
BOHLER ET AL. 1422
TREE PHYSIOLOGY VOLUME 30, 2010
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 9

days, while some were higher at Day 0 and reached the
level of the others at Day 7. It has been shown before that
peroxidase activity in leaves can be associated with leaf
expansion and is highest shortly before the end of leaf
growth (Macadam et al. 1992). A large part of peroxidase
activity can be associated with the cell wall (Bacon et al.
1997), but in this study it is not possible to distinguish
between cell wall and alternative locations of the identified
enzymes. It is possible that peroxidase activity protects the
cells from the ROS formed during leaf expansion and cell
wall development, while catalase is responsible for those
formed during photorespiration.
Changes due to ozone treatment
It was shown in a previous publication from our group that
there were barely any changes in the proteome profiles of
poplar leaves after 3 days of ozone treatment, but that
changes appeared after 14 and 35 days (Bohler et al. 2007).
In the present study, the time points were adjusted to those
results, to get a better understanding of the variations on a
smaller time-scale (28 vs 35 days) but with more obser-
vation points (every 7 days). The observations tend to
confirm the previous results.
No effects could be observed in the current experiment
before 14 days of treatment at a proteomic or biochemical
level, probably due to small experimental or technical
differences between the two experiments (e.g., size of the
plants at the beginning of the treatment, small differences in
conditions due to technical fluctuations and variation of
precise fumigation concentrations due to nonautomated
ozone adjustment). From Day 14 on, effects were visible at
the proteomic (Table 2 and Figure 9), biochemical
(Figures 4 and 5), physiological (Figure 3) and morphologi-
cal levels (Table 1 and Figure 1).
For carbon metabolism, four enzymes of different function
presented a decreased abundance in the Calvin cycle due to
ozone, including RuBisCO, which was previously shown
clearly to be reduced by ozone exposure (Pell et al. 1992,
Brendley and Pell 1998, Miller et al. 1999, Pelloux et al.
2001). Similarly, carbonic anhydrase, RuBisCO subunit
binding protein and RuBisCO activase, all involved in the
optimal functioning of RuBisCO, were present at lower
levels than in controls. This undoubtedly resulted in reduced
carbon fixation and triose phosphate production. Since part
of the triose phosphate is exported to the cytosol as a sub-
strate for sucrose synthesis or glycolysis, this would have had
an impact on these pathways. The analyses showed that three
proteins of glycolysis upstream of triose phosphate were
reduced by the treatment, but no changes could be detected
for enzymes of the lower part of glycolysis, which could be
expected to respond to the variations of the Calvin cycle. It
Figure 5. Quantification of (a) glucose, (b) fructose, (c) sucrose and (d) starch. Statistical significance was tested by one-way ANOVA
between control and treated and between each time point of one condition against Day 0 of the same condition. Letters represent statistical
significance for ANOVA P<0.05. n=4. Error bars represent standard deviation.
LEAF DEVELOPMENT AND OZONE EXPOSURE1423
TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 10

Table 2. Summary of identified proteins in carbon metabolism, protein metabolism, protein folding, protein synthesis, detoxification and
pathogenesis related.
Protein nameNumber of spots identifiedSubcel. loc.
AgingOzone In common
Up Down UpDown
Carbon metabolism
Carbonic anhydrase
RuBisCO subunit binding-protein alpha subunit
RuBisCO subunit binding-protein beta subunit
RuBisCO large subunit
RuBisCO activase 1
RuBisCO activase 2
Transketolase 1 (chloroplastic)
Transketolase 1 (cytosolic)
Glyceraldehyde-3-phosphate dehydrogenase
(chloroplastic)
Glyceraldehyde-3-phosphate dehydrogenase (cytosolic)
Phosphoglycerate kinase, putative
Triose phosphate isomerase (cytosolic)
Fructose-1,6-bisphosphatase
Fructose bisphosphate aldolase (chloroplastic)
Fructose bisphosphate aldolase (cytosolic)
Sedoheptulose-1,7-bisphosphatase
NADP-dependent malic protein
NAD-dependent MDH
MDH
Glycolate oxidase
Enoyl-ACP reductase
Quinone oxidoreductase-like protein
Mitochondrial lipoamide dehydrogenase
Lactoylglutathione lyase
ATP synthase beta subunit
Light-harvesting complex I protein Lhca1
Light-harvesting complex I protein Lhca3
Light-harvesting complex II protein Lhcb1
Oxygen-evolving enhancer protein 1
Oxygen-evolving enhancer protein 2
Chloroplast oxygen-evolving enhancer protein
Photosystem I reaction center subunit II
Chloroplast photosystem I protein
PSI reaction center subunit III
Ferredoxin-NADP oxidoreductase
Polyphenol oxidase
Protein metabolism
Glutamine synthetase
Vitamin-B12independent methionine synthase,
5-methyltetrahydropteroyltriglutamate-homocysteine
S-adenosylmethionine synthetase 4
Alanine aminotransferase 2
Serine carboxypeptidase (carboxypeptidase D)
Glycyl–tRNA synthetase glycine–tRNA ligase
Mitochondrial serine hydroxymethyltransferase
Protein folding
HSP
Chloroplast HSP70
HSP90
HSP70
HSP, putative
Protein disulfide isomerase, putative
1
1
0
8
5
1
4
1
3
0
0
1
0
6
1
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
0
0
2
9
2
3
1
0
chl
chl
chl
chl
chl
chl
chl
cyt
chl
10
9
2
7
3
0
3
0
1
0
0
0
3
3
0
0
0
0
1
2
1
1
2
3
2
7
2
1
1
1
1
4
0
1
1
3
0
2
1
1
2
1
2
0
1
1
0
0
0
0
0
0
2
0
0
0
0
0
1
0
0
0
0
0
0
0
0
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
3
0
7
3
1
0
1
0
0
0
1
1
0
1
1
0
0
2
2
0
2
1
1
0
0
0
3
0
0
1
2
0
0
0
0
2
0
0
0
0
1
0
0
0
2
2
0
2
1
1
0
0
0
2
0
cyt
chl
cyt
cyt
chl
cyt
chl
cyt
cyt
mit
per
n.a.
cyt
mit
cyt
chl
chl
chl
chl
chl
chl
chl
chl
chl
chl
chl
chl
1
0
0
3
1
0
0
1
1
1
cyt
cyt
0
0
0
1
0
3
2
0
0
1
0
0
1
0
0
0
3
0
0
0
0
2
0
0
0
cyt
per
per
cyt
mit
0
1
0
0
2
1
1
1
0
2
3
1
0
0
1
2
2
2
0
0
0
0
0
0
0
0
0
0
2
2
chl
chl
cyt
cyt
cyt
cyt
Continued
BOHLER ET AL.1424
TREE PHYSIOLOGY VOLUME 30, 2010
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 11

was notable that starch levels were higher in treated leaves.
Starch accumulates during the day in chloroplasts, to be used
during the night by carbon catabolism to produce energy and
sustain growth (Smith and Stitt 2007). Possibly, the lower
levels of carbon that still were produced during the day were
preferentially stored in the chloroplast as starch, which could
then have been used in the night for respiration and repair. It
should also be noted that an import of carbon molecules
could come from senescing leaves, which were more abun-
dant in ozone-treated trees (Figure 1). Furthermore, phloem
translocation could be impaired by ozone.
The cytosolic NADP-dependent malic enzyme (NAD-ME)
downstream of glycolysis was higher in abundance in
treated plants. This enzyme catalyzes the conversion of
malate to pyruvate. Mitochondrial MDH, on the other hand,
decreased in abundance. It has been shown that under
ozone exposure, phosphoenolpyruvate carboxylase (PEPc)
is typically induced (Fontaine et al. 2003). The high mol-
ecular weight of this enzyme made it impossible to observe
it in this proteomic study. PEPc leads to malate production
through the anapleurotic pathway. This malate can then be
imported into the mitochondrion and converted to pyruvate
by NAD-ME, which has been shown to be induced by
ozone (Dizengremel et al. 1994). The NADH produced by
this conversion could then be used in the respiratory chain
(Dizengremel 2001). Although no variation in NAD-ME
could be observed in the present study, the decreased MDH
could shift the balance to the conversion of malate to pyru-
vate rather than oxaloacetate. Cytosolic NADP-ME was
observed to increase in abundance, which would also point
to a preferential use of pyruvate as a precursor for the mito-
chondrion, instead of malate. The reaction also leads to the
production of NADPH in the cytosol, needed for the regen-
eration of, e.g. ascorbate (Dizengremel et al. 2009).
Although there is no evidence for this here, it has been pre-
viously shown that the second half of glycolysis is induced
under ozone exposure, as is the Krebs cycle. Increased
activity of phosphofructokinase and fumarase was previously
revealed (Dizengremel et al. 1994; Sehmer et al. 1998).
During ozone exposure, we also observed changes in two
enzymes involved in photorespiration, mitochondrial dihy-
drolipoamide dehydrogenase and peroxisomal glycolate
oxidase, which both showed a late downward trend.
Although these are only two proteins of the pathway, the
results are in agreement with previously published obser-
vations (Bagard et al. 2008). It is probable that a decrease
in photorespiration is directly linked to the decrease in
RuBisCO abundance, hence CO2and O2fixation.
Table 2. Continued
Protein nameNumber of spots identified Subcel. loc.
Aging Ozone In common
UpDown UpDown
Isomerase peptidyl-prolyl cis–trans isomerase
Protein grpE, putative
Protein synthesis
60S acidic ribosomal protein P0
Ribonucleoprotein
Ribosome recycling factor
Chloroplast translational elongation factor Tu
Eukaryotic translation initiation factor 5A isoform I
Translation elongation factor G
Detoxification
Monodehydroascorbate reductase
Cytosolic ascorbate peroxidase
Glutathione-S-transferase theta
Glutathione peroxidase
Catalase
Peroxidase
Pathogenesis related
Class I chitinase
Class IV chitinase
Acidic endochitinase SE2
Beta-1,3-glucanase
Kunitz trypsin inhibitor TI3
0
0
1
1
0
0
0
0
0
0
chl
chl
1
0
1
0
0
1
2
1
0
2
1
0
0
0
0
0
0
0
0
1
1
2
1
1
0
1
1
2
1
1
cyt
chl
n.a.
chl
cyt
cyt
0
1
1
1
2
1
1
2
0
0
1
1
0
0
0
0
0
4
0
0
3
0
5
0
0
0
1
0
3
0
cyt
cyt
cyt
n.a.
per
cyt
1
0
0
1
0
0
0
0
0
0
2
1
2
4
1
0
0
0
0
0
1
0
0
1
0
cyt
cyt
cyt
cyt
cyt
The table shows protein name, number of spots containing the protein increasing and decreasing in abundance in both conditions and
number of significant varying proteins that are common to both conditions, and the subcellular localization of the proteins as determined by
the tools described in the Materials and methods section. The variations up (increased abundance) and down (decreased abundance)
represent the most prominent trend of the protein abundance observed. Tables with detailed information are available as Supplementary data
S1 and S2 at Tree Physiology Online. Abbreviations: ACP, acyl carrier protein; chl, chloroplastic; cyt, cytosolic; mit, mitochondrial; n.a., not
available; LSU, large subunit.
LEAF DEVELOPMENT AND OZONE EXPOSURE 1425
TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 12

Figure 6. (a) Venn diagram of the number of identified proteins for each treatment and the overlap, (b) PCA of all gel images, (c) PCA analysis of
all gel images excluding Day 0 and (d) cluster analysis of all experimental groups. C, control; O, ozone. Numbers 0 to 28 indicate days of treatment.
BOHLER ET AL.1426
TREE PHYSIOLOGY VOLUME 30, 2010
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 13

The effects on the primary carbon metabolism discussed
above mostly became visible at Day 14. Changes in abun-
dance of chloroplast electron transport chain proteins, on
the other hand, mostly appeared at 21 days of exposure.
Effects on the biochemical part of photosynthesis seemed to
appear before effects on the photochemistry, as already
suggested (Dizengremel 2001, Heath 2008, Dizengremel
et al. 2009). If the biochemistry was affected first, and the
Calvin cycle was inhibited, ATP and NADPH could
accumulate in the chloroplast, and act on the regulation of
theelectrontransportchain.
dropped already after 14 days of treatment (Figure 4).
Fluorescence parameters also varied already significantly
after 14 days of treatment. The strong decrease in Fv/Fm
Pigment concentrations
(Figure 3a) showed that there was a loss in the maximum
efficiency of PSII. The relative increase in NPQ when com-
pared with control illustrates that electron flow was too high
for the photochemical processes and had to be quenched at
least partially by the xanthophyll cycle, as already shown
(Castagna et al. 2001). Indeed, the levels of violaxanthin
and zeaxanthin relative to chlorophyll a are higher in treated
leaves. Altogether this shows that there is an effect on
photochemistry before variations in protein abundance can
be detected. Polyphenol oxidase, a protein that is thought to
play a protective role in the chloroplast electron transport
chain (Sherman et al. 1991), was higher in abundance in
treated leaves starting at Day 14, showing that an oxidative
attack occurs at this level as early as 14 days.
Figure 7. Examples of variation in protein abundance of a selection of prominent proteins. Values are calculated from the exported normal abun-
dance from Decyder software. Multiple lines of the same color and symbol represent separate protein spots containing the same protein function.
Similar figures for all the discussed protein spots can be found in Supplementary data S3 at Tree Physiology Online.
LEAF DEVELOPMENT AND OZONE EXPOSURE1427
TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 14

Since ozone clearly leads to oxidative stress and to the
formation of ROS such as H2O2(Diara et al. 2005), it is
expected that detoxifying proteins would increase, most
notably ascorbate-related detoxification (Luwe et al. 1993).
In the present study, the abundance of the observed ascor-
bate peroxidase did not vary under ozone exposure,
although it has been shown to increase in activity (Sehmer
et al. 1998). Other peroxidases increased in abundance.
They could be responsible for some of the ozone-induced
ROS detoxification, because they are often associated
Figure 8. Representation of the changes in protein abundance of
primary carbon metabolism in a developing leaf at Day 7 compared
with Day 0 in ozone-free and ozone-treated conditions alike. Magenta
arrows indicate reduced abundance while cyan arrows or numbers
indicate increased abundance of the protein. A gray arrow indicates a
protein presenting a downward and upward variation in different
spots. (a) Chloroplast electron transport chain, (b) Calvin cycle, (c)
glycolysis and (d) Krebs cycle. TP, triose phosphates. 1, carbonic
anhydrase; 2, RuBisCO; 3, RuBisCO activase; 4, RuBisCO subunit
binding; 5, phosphoglycerate kinase; 6, glyceraldehyde-3-phosphate
dehydrogenase; 7, fructose-1,6-bisphosphate aldolase; 8, transketo-
lase; 9, sedoheptulose-1,7-bisphosphatase; 10, fructose-1,6-bisphos-
phatase; 11, fructose-1,6-bisphosphate aldolase; 12, triose phosphate
isomerase; 13, glyceraldehyde-3-phosphate dehydrogenase; 14,
NADP malic enzyme; 15, NADP MDH; 16, oxygen-evolving enhan-
cer protein; 17, oxygen-evolving enhancer protein 1; 18, oxygen-
evolving enhancer protein 2; 19, light-harvesting complex II protein
Lhcb1; 20, PSI reaction center subunit II; 21, PSI reaction center
subunit III; 22, light-harvesting complex I protein Lhca1; 23, light-
harvesting complex I protein Lhca3; 24, ferredoxin NADP+
reductase; 25, ATPase beta; 26, polyphenol oxidase.
Figure 9. Representation of the changes in protein abundance of
the primary carbon metabolism of an ozone-treated leaf compared
with a control leaf on Day 28. Magenta arrows or numbers indicate
reduced abundance while cyan arrows or numbers indicate
increased abundance of the protein. Numbered items correspond to
those presented in the legend of Fig. 8.
BOHLER ET AL.1428
TREE PHYSIOLOGY VOLUME 30, 2010
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from

Page 15

with cell wall detoxification (Bacon et al. 1997), where
ozone-inducedROS
first
other hand, were present at lower abundance in ozone-
treatedleaves. Becausecatalases
photorespiration-induced H2O2detoxification (Reumann and
Weber 2006), the reduction in RuBisCO abundance could
explain the decrease of this enzyme. Reduced fixation of O2
would decrease the rate of photorespiration.
Pathogenesis-related proteins are proteins that were initially
found to be involved in pathogen attack (Van Loon 1997),
but recently also associated with different types of abiotic
stress (Sandermann et al. 1998, Renaut et al. 2004). Here, we
observed an increase in abundance of multiple isoforms of
chitinase. It is known that chitinases are induced by many
stress conditions, ethylene, salicylic acid and jasmonic acid,
all play a role in ozone stress signaling (Kangasjärvi et al.
2005).It was also shown that acidic class III chitinases can be
secreted into the apoplast, which is the primary location of
ozone attack (Arie et al. 2000). Similar to chitinase, the abun-
dance of several beta-1,3-glucanases increased. The upward
variation of these two pathogenesis-related proteins was prob-
ably due to the ozone-induced oxidative burst (Wohlgemuth
et al. 2002) that is also typical of a pathogen attack.
appear.Catalases,onthe
are involvedin
Ozone and leaf expansion
If little is known about leaf expansion in general, even less is
known about it in combination with ozone exposure. It was
shown in previous observations that necroses only appear
once leaves are fully expanded (Bohler et al. 2007, Bagard
et al. 2008). In the present study, it was shown that leaves
expanding during the treatment grow to a slightly smaller
area than normally expanding leaves, but with barely any
statistical significance. As discussed above, leaf expansion is
complete after a little more than 7 days, which is also visible
at the proteomic level. The proteomic variations could
clearly be divided into changes between 0 and 7 days and
changes between 7 and 28 days (Figure 10). In the first
phase, ozone had strictly no effect on the protein level.
Ascorbate peroxidase and peroxidase levels were higher at
the beginning of leaf expansion than in the mature leaf,
which could have conferred a certain resistance to ROS on
the young leaves. Under ozone exposure, peroxidases were
induced only at later time points, which supports this hypoth-
esis. Similarly, the higher levels of chaperones in young
leaves could protect them from ozone-induced damage.
These considerations could explain why younger leaves are
more resistant to ozone, simply because cellular processes
are very active and the inherent antioxidant potential is stron-
ger (Nogués et al. 2008). Protective measures that are already
in place because they are required for the expansion process
can handle the ozone treatment, and the effective ozone flux
is not sufficient to induce a response. It has also been shown
that the parenchyma of developing leaves is denser than that
of mature leaves (Paoletti et al. 2009). In denser tissue, less
ozone reaches the inner leaf tissues which remain protected
from direct ozone exposure.
Furthermore, 162 identified protein spots varied during
leaf development, 123 for ozone exposure; only 69 of these
changed under both conditions. This does not mean that
proteins different in the two conditions have different func-
tions. Some proteins, such as those related to pathogenesis,
Figure 10. Summary of the proteomic variations in developing leaves and in ozone-treated leaves. The clear distinction between the expan-
sion phase and the fully expanded leaf is underlined. The effects of ozone only appear in the adult leaf.
LEAF DEVELOPMENT AND OZONE EXPOSURE1429
TREE PHYSIOLOGY ONLINE at http://www.treephys.oxfordjournals.org
by guest on November 2, 2010
treephys.oxfordjournals.org
Downloaded from