This study examined whether two genotypes of hybrid poplar (Populus deltoides × Populus trichocarpa), previously classified as ozone tolerant and ozone sensitive, had differing physiological and biochemical responses when fumigated with 120 nL L−1 ozone for 6 h per day for eight consecutive days. Isoprene emission rate, ozone uptake and a number of physiological and biochemical parameters were investigated before, during and after fumigation with ozone. Previous studies have shown that isoprene protects plants against oxidative stress. Therefore, it was hypothesized that these two genotypes would differ in either their basal isoprene emission rates or in the response of isoprene to fumigation by ozone.
Our results showed that the basal emission rates of isoprene, physiological responses and ozone uptake rates were all similar. However, significant differences were found in visible damage, carotenoids, hydrogen peroxide (H2O2), thiobarbituric acid reactions (TBARS) and post-fumigation isoprene emission rates. It is shown that, although the classification of ozone tolerance or sensitivity had been previously clearly and carefully defined using one particular set of parameters, assessment of other key variables does not necessarily lead to the same conclusions. Thus, it may be necessary to reconsider the way in which plants are classified as ozone tolerant or sensitive.
Episodes of high tropospheric ozone concentration are predicted to increase in both duration and frequency in future climate scenarios, and so it is vitally important that plant responses to this particular stress are fully understood (Meehl et al. 2007). Ozone is known to be phytotoxic, causing cellular damage inside leaves. It enters a leaf primarily through the stomata and from there invades the intracellular leaf spaces (Musselman et al. 2006). Once inside the leaf, ozone induces oxidative stress by forming reactive oxygen species (ROS) such as O2, H2O2 and OH. Oksanen & Holopainen (2001) found that the dry shoot mass and stem heights of an ozone-sensitive clone of European white birch (Betula pendula Roth) were reduced by up to 17 and 46%, respectively, and Woo and Hinckley (2005) found decreases in a number of growth parameters of hybrid poplar (Populus deltoides × Populus trichocarpa) including total biomass and root/shoot ratio. Photosynthesis may decrease (Reich & Lassoie 1985; Wittig, Ainsworth & Long 2007), and disruption to the electron transport chain can occur (Ranieri et al. 2001). In addition, stomata may be induced to close to prevent further uptake of ozone (Nali et al. 1998; Guidi et al. 2001; Pasqualini et al. 2002), although this is not always the case (Pearson & Mansfield 1993), and increased resource allocation is required to detoxify and repair leaves (Ashmore 2005).
There have been numerous studies of the responses of plants to environmental stress and the mechanisms that underlie these responses. For example, the closure of stomata in ozone-sensitive clover in response to drought or ozone (Fuhrer & Booker 2003; Bermejo, Irigoyen & Santamaria 2006); the ability of the xanthophyll cycle to protect photosynthetic pigments against excess light either in isolation (Holt et al. 2005) or as part of ozone-induced injury (Alonso et al. 1999, 2001); and the role of antioxidants to prevent unregulated cell death, quench ROS or initiate early leaf senescence through signalling compounds (Pell, Schlagnhaufer & Arteca 1997).
Many of these plant defences are well understood, while the mode of action for others remains elusive. One such defence is the important, reactive biogenic volatile organic compound (BVOC) isoprene (C5H8, 2-methyl-1,3-butadiene). Isoprene plays a major role in the oxidative chemistry of the troposphere (Wiedinmyer et al. 2006; Sharkey, Wiberley & Donohue 2008). It is a precursor for the formation of tropospheric ozone, can react with the hydroxy radical (OH), thereby increasing the residence time of methane (CH4), and contribute to the formation of secondary organic aerosols (SOAs). Isoprene has been previously shown to protect against stresses such as high temperature and ozone; however, the mechanism by which this occurs is still unclear (Loreto & Velikova 2001; Sharkey et al. 2008). The reaction between isoprene and ozone is too slow to allow direct scavenging of ozone (Salter & Hewitt 1992; Loreto et al. 2001); however, isoprene may act indirectly through scavenging ROS species or strengthening membranes (Sharkey & Singsaas 1995; Velikova, Edreva & Loreto 2004). Recent studies have suggested that drought may override a plant's ability to emit isoprene, thereby impairing this protective capability (Brilli et al. 2007; Fortunati et al. 2008).
Although the ecophysiological role of isoprene has been studied in great detail, the results gathered have not always been consistent. Hewitt, Kok & Fall (1990) found that organic hydroperoxides, which are products of ozone–alkene reactions and are damaging to plants, were present in the leaves of isoprene-emitting plants after they had been exposed to ozone, but were absent from the leaves of non-emitting control plants. In contrast, Loreto et al. (2001) showed that endogenous isoprene-emitting plants and those fumigated with exogenous isoprene suffered less damage when exposed to ozone. Similarly, Sharkey, Chen & Yeh (2001) demonstrated that isoprene conferred thermotolerance to excised leaves when they were subjected to a transient high-temperature episode, whereas Logan & Monson (1999) showed that exogenous isoprene did not protect leaf discs from four isoprene-emitting species from the effects of high temperature.
The majority of the studies investigating the effect and role of isoprene used plants that were classified as either ozone tolerant or sensitive. The parameters on which these classifications are based often vary (Table 1) and may be one reason for the disharmony of results regarding isoprene's role. The intention of this study was to determine whether two genotypes of isoprene-emitting hybrid poplar (P. deltoides × P. trichocarpa), previously classified as ozone tolerant and ozone sensitive by Woo and Hinckley (2005) and Taylor (personal communication) (Table 1), had differing responses when fumigated with environmentally realistic concentrations of ozone.
Table 1. Parameters used to classify plant species as ozone tolerant or sensitive
Given the protective role of isoprene under oxidative stress, it was hypothesized that one mechanism for the contrasting ozone sensitivity would be because of differing, non-stressed basal isoprene emission rates or to the response of isoprene to fumigation by ozone. To this end, both genotypes were fumigated with 120 nL L−1 ozone for 6 h per day for 8 consecutive days. Fifteen separate parameters were examined covering biochemical, physiological, phenotypical and gas exchange responses (Table 1). Based on the results collected, the system for classifying plants as ozone tolerant or sensitive was investigated.
MATERIALS AND METHODS
The experiments were conducted on fully developed attached leaves of two poplar genotypes (P. deltoides × P. trichocarpa) previously determined to be ozone tolerant and sensitive. Woo and Hinckley (2005) used the responses of 10 different growth and productivity parameters to fumigation by 85–128 nL L−1 ozone for 6–9 h per day for a duration of 89 or 65 d dependent on screening period. Taylor (personal communication) used observations of necrotic flecking in response to both chronic exposure to ozone over several weeks, and to acute exposure of 150 nL L−1 of ozone applied over a few hours (Table 1). Plants were obtained from cuttings (vegetative propagation) grown at Southampton University, UK, and were initially planted in 4 L plastic pots containing a 1:1 (v/v) mix of John Innes No. 2 compost (JI2; John Innes Manufacturers Association, Harrogate, UK) and Perlite (Styroperl; LBS Horticulture, Lancashire, UK). The plants were grown in glasshouses at Lancaster University, UK, with a day length of 16 h and an irradiance of 1000 µmol photons m−2 s−1 photosynthetic photon flux density (PPFD). After 4 months, the plants were re-potted into 7.5 L plastic pots filled with the same compost mixture to avoid them from becoming pot bound.
Prior to the experiment, the plants were transferred into Teflon chambers and left to acclimatize to chamber growth conditions for 7 d. The poplar trees were watered every 2 d with tap water. Before any measurements were carried out, one fully expanded 3-week-old leaf from each tree was selected and tagged with polytetrafluoroethylene (PTFE) tape. The same leaf from each tree was used throughout the experiment for all non-destructive measurements. One leaf of a similar age and height from each tree was removed for biochemical analysis 1 d before fumigation with ozone (control), on the third and last day of the fumigation period (treatments) and 2 and 4 d post-fumigation (recovery).
Ozone fumigation was performed in a climate-controlled laboratory in two dynamic, PTFE lined, 1 m3 fumigation chambers, previously described by Stokes, Lucas & Hewitt (1993). Three or four ozone-sensitive and ozone-tolerant plants were placed in each of the two chambers approximately 7 d prior to commencing measurements. The chamber air temperature was maintained at 25 ± 1 °C with a relative humidity (RH) of 40 ± 10%. A PPFD of approximately 500 µmol photons m−2 s−1 at sample leaf height was provided for 14 h a day by four lamps (Powerstar HQI-BT, 400W/D daylight; OSRAM, Munich, Germany). Both chambers were fumigated with ambient air from an air-conditioning unit with a mean chamber inlet flow rate of 0.022 m3 s−1, equal to 1.3 complete air changes min−1. After 7 d acclimation, non-fumigation control measurements were taken over a 4 d period. Following this, ozone, produced by UV dissociation (Opsis, Furulund, Sweden) of pure oxygen (BOC Gases, Surrey, UK), was mixed with the ambient air at the inlet of one of the chambers to produce the desired volume-mixing ratio (VMR) of 120 ± 20 nL L−1 ozone inside the chamber. The VMR of ozone was monitored continuously during the fumigation period by a photometric ozone analyzer (Teledyne Instruments, San Diego, CA, USA). Ozone exposure consisted of a square-wave fumigation period of 6 h a day from 0930 to 1530 h UTC, for 8 consecutive days. This corresponds to an accumulated ozone exposure above 40 ppb (AOT40) of 3840 ppb h (NEGTAP 2001). While square-wave fumigation is unrealistic, this level of ozone exposure is not unrealistically high based on future climate scenarios and the results of observations taken during the 2003 European heat wave. At this time, a number of European countries saw daytime ozone concentrations during August 2003 exceed 120 nL L−1, and these occurrences are projected to increase in frequency and duration in the future (Lee et al. 2006; Meleux, Solmon & Giorgi 2007). The ozone-free control plants were exposed to ambient air only, in which the concentration of ozone never exceeded 15 ± 5 nL L−1. Following ozone fumigation, the plants were allowed to recover for 4 d during which time both chambers were supplied with ambient air.
Leaf gas exchange and chlorophyll a fluorescence measurements
Leaf gas-exchange measurements were coupled with chlorophyll a fluorescence measurements using an open gas-exchange system LI-6400 (Li-Cor, Lincoln, NE, USA) and taken on alternate days for ozone-tolerant and ozone-sensitive genotypes to coincide with measurements of the isoprene emission rates. All measurements were made on attached, pre-selected leaves as described earlier. Measurements of each physiological parameter were repeated 10 times on one leaf from each tree. The 10 points were averaged to give one measurement point for each parameter on each leaf. The CO2 concentration inside the cuvette was maintained at a constant 390 µL L−1 (approximate growth CO2 conditions) by the internal controls of the LI-6400. The airflow to the cuvette was kept at 300 µmol s−1 to maintain positive pressure within the cuvette.
A-Ci curves [assimilation rate versus intercellular (CO2)] were measured concurrently with chlorophyll a fluorescence on attached, pre-selected leaves using the LI-6400 at a leaf temperature of 25 °C and the given experimental irradiance. Cuvette [CO2] was altered to produce intercellular [CO2] ranges of 30–900 µL L−1. A and Ci values were corrected for diffusion of CO2 and water across the gaskets of the LI-6400 fluorometer head as described by Rodeghiero, Niinemets & Cescatti (2007). Mesophyll conductances (gm) for each A-Ci curve were estimated using the variable electron transport (J) method of Harley et al. (1992). The calculated values of gm were used to convert A-Ci curves to A-Cc (chloroplastic CO2 concentration) curves. Vcmax[the maximum potential velocity of ribulose bisphosphate carboxylase/oxygenase (Rubisco) for carboxylation] and Jmax[the maximum potential rate of electron transport contributing to ribulose bisphosphate (RuBP) regeneration] were calculated from A-Cc curves using the photosynthesis equations of Farquhar, Caemmerer & Berry (1980), von Caemmerer & Farquhar (1981) and Harley & Sharkey (1991). The Michaelis–Menten constant for CO2 (Kc) and O2 (Ko), and the CO2 compensation point in the absence of dark respiration (Γ*) were taken from Bernacchi et al. (2001). A non-linear curve-fitting routine (Microsoft Excel 2003; Microsoft Corporation, Redmond, WA, USA) was used to minimize the difference between the predicted values calculated by the models and the observed values, and solve for mitochondrial respiration in the light (Rd), Vcmax and Jmax.
All measurements were conducted at a constant temperature of 25 °C and irradiance of 500 µmol photons m−2 s−1 PPFD and were performed immediately post-ozone fumigation. At the given irradiance, the fluorescence yield at steady-state photosynthesis (Fs) and the maximum fluorescence yield produced by a 0.5 s saturating flash (PPFD of 7000 µmol m−2 s−1) were determined. The photosystem II (PSII) operating efficiency was calculated as F′q/F′m = F′v/F′m × F′q/F′v where F′m is the maximal fluorescence from a light-adapted leaf, F′q is the difference in fluorescence between F′m and the fluorescence emission from a leaf adapted to actinic light, F′v is the variable fluorescence from a light-adapted leaf, F′v/F′m is the maximum efficiency of PSII under the given light conditions, and F′q/F′v is the PSII efficiency factor (Baker & Rosenqvist 2004).
Net O3 uptake was calculated each day for both genotypes. Stomatal conductance values with respect to water vapour were converted into an O3 conductance by dividing by the coefficient of molecular diffusivity of ozone in water vapour (1.68) (Laisk, Kull & Moldau 1989). This assumes that the concentration of O3 inside the intracellular spaces of the leaf was approximately zero. The dose each fumigated plant received was defined as the total amount of pollutant that was actually absorbed into the plant through the stomata over a period of time (Fowler & Cape 1982). This was calculated as the O3 concentration multiplied by the duration of each fumigation episode per day. This was then converted into a daily O3 uptake rate (nmol m−2 d−1) for each genotype as a product of the conductance of O3 and daily uptake dose. Uptake through the cuticular layer was ignored because the cuticle is considered to be highly impermeable to O3 when compared to open stomata (Kerstiens & Lendzian 1989). In addition, boundary layer resistance was considered negligible because of the high velocity of the airflow through the chambers.
For biochemical analysis, one leaf from each tree was cut and removed at each time point. Three technical replicates were taken from each leaf, and the average of these was taken to give one data point per leaf.
Determination of carotenoid and chlorophyll content
Chlorophyll a and b, and total carotenoid in leaf samples were determined by extraction into 25 mL glass vials in ice, using acetone as the solvent. Then, 20 mg of frozen leaf material was homogenized with 10 mL of acetone, double distilled water solution (80:20) v/v in low light. This was then centrifuged for 13 min at 4600 rpm before the supernatant was decanted and the remaining pellet was resuspended in 10 mL of 100% acetone and centrifuged a second time. The resulting supernatant mixture from the two consecutive extractions was used for the determination of chlorophyll and carotenoid content by measuring the absorbance at 470, 646 and 663 nm according to the calculations of Lichtenthaler (1987) (Ultrospec 2100; GE Healthcare Bio-Sciences AB, Uppsala, Sweden).
Biochemical assay of H2O2
Hydrogen peroxide (H2O2) levels were determined according to the method of Velikova, Yordanov & Edreva (2000). Frozen leaf material (100 mg) was homogenized in an ice bath with 1 mL 0.1% (w/v) trichloroacetic acid (TCA). The homogenate was centrifuged at 12 000 g for 30 min, and 0.4 mL of the supernatant was added to 0.4 mL 10 mm potassium phosphate buffer (pH 7.0) and 0.8 mL 1 m potassium iodide (KI). The absorbance of supernatant was read at 360 nm. The coloured reaction product of H2O2 with KI develops within 25 min and is stable for at least 2 h. The content of H2O2 was calculated against a calibration curve using H2O2 standards.
Estimation of lipid peroxidation
The degree of lipid peroxidation of the leaf tissue was determined by malondialdehyde (MDA) accumulation, assayed using the thiobarbituric acid reactions (TBARS) method (Heath & Packer 1968). Approximately 40 mg of frozen leaf material was homogenized in 0.5 mL 0.1% (w/v) TCA. The homogenate was then centrifuged at 4 °C at 12 000 g for 30 min. Then, 0.5 mL of the supernatant was mixed with 1 mL of 20% TCA containing 0.5% (w/v) thiobarbituate acid (TBA), and the mixture heated in a water bath for 30 min at 95 °C. The reaction was stopped in an ice bath, and the mixture was then centrifuged at 10 000 g for a further 5 min. The absorbance of the supernatant was read at 532 and 600 nm to correct for non-specific turbidity. To calculate the content of lipid peroxides, an absorption coefficient of 155 mm cm−1 was used.
Total and reduced ascorbate determination
The levels of total and reduced ascorbate were determined according to the method of Gillespie & Ainsworth (2007). Approximately 40 mg of frozen leaf material was homogenized in 1 mL of 6% aqueous TCA (w/v). This was transferred to a 2 mL Eppendorf tube, and the mortar and pestle washed with another 1 mL of 6% TCA. This was collected in the same tube. Samples were centrifuged at 13 000 g for 5 min at 4 °C. The supernatant was transferred to a new 2 mL Eppendorf tube and then immediately placed in ice. Then, 100 µL of 75 mm phosphate buffer and 200 µL of either 6% TCA (blank), or sample supernatants were added to a 2 mL Eppendorf tube. Then, 100 µL of 10 mm dithiothreitol (DTT) was added to the total ascorbate tubes and incubated at room temperature for 10 min. Then, 100 µL of 0.5% N-ethylmaleimide (NEM) was then added to the total ascorbate tubes and incubated for a further 30 s; 200 µL of double distilled water was added to the reduced ascorbate tubes. To all tubes, 500 µL 10% TCA, 400 µL 43% phosphoric acid (H3PO4), 400 µL 4% α-α′-bipyridyl and 200 µL 3% iron chloride (FeCl3) was added. All tubes were then incubated in a water bath for 1 h at 37 °C before the absorbance of each tube was read at 525 nm. The content of total and reduced ascorbate was calculated from a standard curve.
Isoprene emission measurements
During the 4 control days, the isoprene emission rate of each genotype in unstressed conditions was measured by gas chromatography mass spectroscopy (GC–MS), gas chromatography flame ionization detection (GC-FID) and proton transfer reaction mass spectroscopy (PTR-MS; Ionicon GmbH, Innsbruck, Austria) as described in Hayward et al. (2001) and Possell et al. (2005). Measurements of the isoprene emission rates were taken on alternate days for ozone-tolerant and ozone-sensitive genotypes. The pre-selected leaf from each plant was clamped in a leaf cuvette attached to the LI-6400. A PTFE sampling line from the cuvette head was connected to the PTR-MS and sampled with a flow rate of approximately 100 mL min−1. The isoprene emission rate from one leaf on each tree was monitored continuously until it became stable, and then 10 further measurements were taken. The 10 points were averaged to give one measurement point for each parameter on each leaf. Total leaf area was determined at the end of the experiment by a leaf area meter (LCA4-AM100; ADC BioScientific Ltd., Hertfordshire, UK).
A repeated-measures analysis of variance (anova) was used to determine significant differences between treatments (ozone fumigated or non-fumigated controls) at each measurement point (±S.E. unless otherwise stated) were separated by paired T-test. Significant statistical differences are displayed at the P ≤ 0.05, P ≤ 0.01 or P ≤ 0.001 levels. Where necessary, values have been arcsine transformed so that assumptions of homogeneity of variance and normality are not violated. Biological replication consisted of three (ozone-sensitive) or four (ozone-tolerant) trees of each genotype for each treatment. Statistical analysis was performed with SPSS for Windows (release 14; SPSS, Chicago, IL, USA).
The ozone-sensitive genotype of hybrid poplar developed both necrotic and chlorotic lesions and spots on the adaxial surface of several of the mature leaves, approximately 72 h after the initiation of ozone fumigation (Fig. 1a). These symptoms were absent from the ozone-fumigated tolerant genotype. However, a number of the mature leaves of this genotype appeared to begin exhibiting signs of early foliar senescence (Fig. 1b). Indeed, 3 weeks after completion of the experiment, the majority of the mature leaves on the ozone-tolerant genotype had senesced. By comparison, the mature leaves of the ozone-sensitive genotype were still attached.
Effect of ozone exposure on leaf gas exchange
Prior to ozone fumigation, there were no significant differences in the net photosynthetic rate (Pn), stomatal conductance (gs), maximum potential rate of carboxylation (Vcmax) or the maximum potential rate of electron transport contributing to RuBP regeneration (Jmax) of either genotype (Fig. 2).
There continued to be no difference in Pn or gs of either genotype during or after fumigation when compared to their respective non-fumigated control plants (Fig. 2a–d). The only exception to this occurred on day 6 of ozone fumigation when Pn of the ozone-sensitive fumigated plants was significantly lower than the corresponding controls (P = 0.049) (Fig. 2b). In addition, there was no significant difference in Pn when comparing the two genotypes after fumigation had ceased (P = 0.68).
Vcmax of the ozone-tolerant plants was significantly lower than the corresponding controls only after 7 d of fumigation (t = 3.24, P = 0.02; Fig. 2e). In comparison, Vcmax of the fumigated ozone-sensitive genotype was significantly lower than controls after only the fourth day of fumigation (t = 3.29, P = 0.02; Fig. 2f). However, there were no differences for either genotype after fumigation.
Ozone fumigation had no effect on Jmax of the ozone-tolerant genotype (Fig. 2g), whereas a significant reduction was observed only after the final day of fumigation for the ozone-fumigated sensitive plants (control Jmax = 98.3 ± 4.8 µmol m−2 s−1, fumigated Jmax = 72.0 ± 10.6 µmol m−2 s−1; t = 3.93, P = 0.01) (Fig. 2h). There were no differences for either genotype during the 4 recovery days.
Chlorophyll a fluorescence
Measurements of chlorophyll a fluorescence parameters enabled non-invasive monitoring of the efficiency of PSII. Using this approach, the operating efficiency of PSII (F′q/F′m) was measured as a function of the maximum potential efficiency of PSII (F′v/F′m) and the PSII efficiency factor (F′q/F′v) (Genty, Briantais & Baker 1989; Baker & Rosenqvist 2004).
Prior to ozone exposure, there were no differences in the arcsine-transformed values of PSII operating efficiency (F′q/F′m), PSII maximum efficiency (F′v/F′m) or PSII efficiency factor (F′q/F′v) (Fig. 3).
Ozone fumigation only affected the arcsine transformed F′v/F′m of the ozone-tolerant genotype when compared to controls on the second recovery day (t = 3.22, P = 0.03; Fig. 3c). Fumigation also caused a decrease in F′q/F′m and F′v/F′m on day 8 of fumigation of the ozone-sensitive genotype (t = 2.62, P = 0.04; t = 3.40, P = 0.02, respectively; Fig. 3b,d). There were no significant differences in any efficiency parameter after 4 d of recovery when comparing the two previously fumigated genotypes.
Overall, no differences were observed in the daily cumulative O3 uptake rates of either genotype during the fumigation period (Fig. 4a). There was a significant correlation between isoprene emission rate and ozone uptake for the ozone-tolerant genotype, but not for the sensitive genotype (F = 150.24, P < 0.01, R2 = 0.987; F = 2.16, P = 0.28, R2 = 0.520, respectively; Fig. 4b). Those plants with higher rates of isoprene emission showed greater uptake of ozone.
Prior to fumigation, data for both control and treatment trees for each genotype were pooled as both sets of plants were held under the same conditions. Results showed that all measured biochemical parameters were statistically similar in both genotypes. The results from each biochemical assay are given in Table 2.
Table 2. Biochemical data ± standard deviation collected from non-fumigated tolerant controls (TC), ozone-fumigated-tolerant plants (TF), non-fumigated sensitive controls (SC) and ozone fumigated-sensitive plants (SF)
Treatment and genotype
Fumigation day 3
Fumigation day 8
Recovery day 2
Recovery day 4
Fumigation consisted of exposure to 120 nL L−1 ozone for 6 h per day for 8 consecutive days. Superscript letters indicate significant differences between control and fumigated plants of the same genotype. Significance is shown at P = 0.05 (a), P = 0.01 (b) and P < 0.001 (c). Superscript * indicates significant differences between control plants at P = 0.05 (*), P = 0.01 (**) and P < 0.001 (***). Superscript numbers indicate significant difference between fumigated plants at P = 0.05 (1), P = 0.01 (2) and P < 0.001 (3).
Exposure to ozone decreased the chl a/b ratio of the ozone-fumigated-tolerant genotype when compared to non-fumigated controls only on the second recovery day (t = 3.60, P = 0.02; Table 2). By comparison, ozone fumigation caused a decline in chl a/b of the ozone-sensitive genotype when compared to controls only on the eighth day of fumigation (t = 4.02, P = 0.02). No differences were observed during the recovery period.
Comparison of the non-fumigated control plants indicated the chl a/b ratio of the ozone-sensitive genotype decreased more than the tolerant plants over the duration of the experiment (F = 16.15, P = 0.02), with a specific reduction on the fourth recovery day (t = 6.43, P < 0.01; Table 2). The ratios for the fumigated ozone-sensitive plants were significantly lower than the fumigated tolerant plants on both measured days of fumigation (t = 7.26, P < 0.01; t = 2.95, P = 0.04 respectively).
Fumigation with ozone had no observed effect on the total chlorophyll content of the ozone-tolerant genotype. In contrast, fumigation restricted total chlorophyll in the sensitive plants when compared to controls (F = 10.11, P = 0.03), promoting a recovery after fumigation ceased (F = 14.00, P = 0.03; Table 2). Total chlorophyll in the fumigated ozone-tolerant genotype was generally higher than the fumigated ozone-sensitive genotype, although this difference was not significant. Total chlorophyll was significantly higher in the control ozone-tolerant genotype when compared to the control ozone-sensitive plants on the fourth day of the experiment (fumigation day 3) (t = 3.68, P = 0.02).
No difference in total carotenoid was observed in the ozone-fumigated tolerant genotype when compared to their corresponding non-fumigated controls (Table 2). In contrast, fumigation caused a significant decline in total carotenoid of the ozone-sensitive genotype when compared to control plants on the last day of fumigation (F = 4.08, P = 0.02) and both recovery points (F = 4.24, P = 0.02; F = 4.02, P = 0.03 respectively; Table 2). It was also noted that total carotenoid content of the ozone-tolerant control plants was generally higher than the sensitive controls (F = 10.01, P = 0.03). Similarly, the carotenoid content of the ozone-fumigated-tolerant genotype was significantly higher than the fumigated sensitive plants (F = 48.56, P < 0.01). Furthermore, total carotenoid in the fumigated tolerant plants increased during recovery, whereas those in the sensitive plants did not (F = 20.62, P = 0.02).
There was a significant increase in H2O2 content of the ozone-fumigated-tolerant genotype relative to its non-fumigated controls on the third fumigation day (1.89 ± 0.04 µmol g−1 FW; 1.24 ± 0.09 µmol g−1 FW, respectively, P < 0.01; Table 2). Following this, the H2O2 content slowly decreased resulting in a significant difference between the two treatments (F = 49.78, P = 0.002). During the recovery period, the H2O2 content increased once again relative to the non-fumigated control plants, and specific differences were observed on both measured days of the recovery period (t = 3.42, P = 0.03; t = 11.01, P < 0.001 respectively).
In contrast, the H2O2 content of the fumigated ozone-sensitive genotype increased for the whole duration of ozone fumigation (F = 101.04, P < 0.001), resulting in a difference between fumigated and control treatments by the final day of fumigation (2.3 ± 0.08 µmol g−1 FW, 0.97 ± 0.08 µmol g−1 FW; t = 20.17, P < 0.001; Table 2). There was no difference between the two treatments during the recovery period.
While there was no difference between ozone-tolerant and sensitive controls, the H2O2 content of the fumigated sensitive genotype was significantly lower than that of the tolerant plants on the third fumigation day (1.54 ± 0.07 µmol g−1 FW; 1.89 ± 0.04 µmol g−1 FW, respectively; t = 4.55, P = 0.01), but significantly higher on the final day of ozone fumigation (2.34 ± 0.05 µmol g−1 FW; 1.52 ± 0.09 µmol g−1 FW, respectively, t = 7.68, P < 0.01; Table 2).
In comparison to the non-fumigated control plants, ozone fumigation had no effect on the level of TBARS of the ozone-tolerant genotype (Table 2). In contrast, TBARS concentration increased in the fumigated ozone-sensitive genotype on the last day of ozone fumigation (22.99 ± 0.74 µmol g−1 FW, ozone fumigated, 16.54 ± 1.45 µmol g−1 FW, non-fumigated controls, t = 3.95, P = 0.02), as well as on the second day of recovery (26.99 ± 1.32 µmol g−1 FW, ozone fumigated, 16.26 ± 1.02 µmol g−1 FW, non-fumigated controls, t = 16.66, P < 0.01). Furthermore, there were significant differences between the fumigated and control treatments at both times (F = 12.78, P = 0.02 during fumigation, F = 16.66, P = 0.02 during recovery).
While the levels of TBARS in the control genotypes were similar throughout the experiment, the ozone-fumigated-sensitive plants showed greater lipid peroxidation than the tolerant plants during fumigation (F = 13.28, P = 0.02; Table 2). This continued through the recovery period (F = 16.78, P = 0.03; Table 2).
Total and reduced ascorbate
There were no differences in total ascorbate content when comparing the ozone-fumigated-tolerant genotype with its control plants during fumigation (Table 2). Reduced ascorbate was higher in the fumigated plants when compared to controls on the third day of fumigation (t = 3.10, P = 0.03). On the final recovery day, both total and reduced ascorbate increased with respect to controls (t = 3.37, P = 0.02; t = 3.72, P = 0.01, respectively). Fumigation had no effect on either total or reduced ascorbate of the ozone-sensitive genotype.
While the control plants of both genotypes had similar levels of total and reduced ascorbate, reduced ascorbate was elevated in the fumigated tolerant genotype when compared to the sensitive plants (F = 8.15, P = 0.04). Similarly, total ascorbate was higher in the tolerant plants when compared to the sensitive on the final day of recovery (t = 3.61, P = 0.04).
Isoprene emission rate
Prior to ozone fumigation, the isoprene emission rates of the ozone-tolerant (12.3 ± 1.3 nmol m−2 s−1) and sensitive (12.8 ± 1.0 nmol m−2 s−1) genotypes of hybrid poplar were statistically similar (Fig. 5a,b). Throughout and subsequent to ozone fumigation, the isoprene emission rate of the tolerant genotype remained statistically similar to the non-ozone-fumigated control plants (Fig. 5a). In contrast, fumigation caused a decline in the isoprene emission rate of the sensitive genotype when compared to controls (F = 37.27, P < 0.01), with specific reductions on days 4 and 6 of fumigation (t = 7.37, P < 0.01; t = 4.35, P = 0.01, respectively). There continued to be a difference between the two treatments during recovery (F = 9.99, P = 0.03), with the rate of isoprene emission gradually increasing from the fumigated plants. By the final day of the recovery period, the rate of isoprene emission of the fumigated ozone-sensitive plants (7.9 ± 1.2 nmol m−2 s−1) was significantly lower than that of the fumigated ozone-tolerant genotype (12.8 ± 3.0 nmol m−2 s−1, t = 3.01, P = 0.04).
Despite the prior classification of our experimental material as either ozone tolerant or ozone sensitive (Table 1), only five of the 15 parameters measured demonstrated significant differences when comparing responses to fumigation with ozone.
In this study, we see no effect of ozone fumigation on Pn of the ozone-tolerant genotype and only minimal effect on Pn of the ozone-sensitive genotype. It is interesting to note that Vcmax and Jmax along with the potential operating efficiency and potential maximum efficiency of PSII were also reduced when comparing the fumigated ozone-sensitive genotype to its corresponding control plants. However, these effects did not occur simultaneously. There were no differences in any of the parameters when a comparison of the two fumigated genotypes was made after fumigation had ceased. This contrasts with prior studies that have shown that exposure to either acute, short-term or chronic, long-term concentrations of ozone causes, in general, significant reductions in either Pn and/or gs (Nali et al. 1998; Guidi et al. 2001; Loreto & Velikova 2001; Pasqualini et al. 2002; Calfapietra et al. 2007). However, Wittig et al. (2007) found that exposure to mean concentrations of 79 nL L−1 ozone and 104 nL L−1 ozone reduced net assimilation and stomatal conductance in species of poplar by an average of 21 and 19%, respectively. Furthermore, they noted that increasing ozone concentration reduced these parameters further. Although the reductions in Pn and gs found in this study were not statistically significant because of the error terms, the average reductions were 19 and 28%, respectively, for the ozone-sensitive poplar. This is in keeping with the findings of Wittig et al. (2007) and indicates the activation of stomatal closure, an importance ozone uptake avoidance mechanism (Guidi et al. 2001).
Ozone enters a plant mainly through the stomata, and therefore, any reduction in stomatal conductance also reduces ozone uptake (Emberson, Wieser & Ashmore 2000). Given the lack of difference when comparing the stomatal conductance of these two genotypes of poplar, it is not surprising that there is also a corresponding similarity in ozone uptake. Our data indicate that cumulative ozone uptake or dose does have a detrimental effect on photosynthesis, as well as encouraging necrotic spotting and early leaf senescence. However, despite similar doses, the responses of these two genotypes of poplar were different. As such, cumulative uptake by itself cannot be used to quantify the expected level of response to ozone (e.g. Pell et al. 1999; Massman, Musselman & Lefohn 2000; Fuhrer & Booker 2003; Massman 2004).
Previous studies using hybrid poplar clones have found decreases in either Vcmax or Jmax. Ranieri et al. (2001) found that the actual rate of electron transport was significantly reduced in ozone-sensitive poplar clones but not in tolerant clones, while Guidi et al. (2001) found that Vcmax was reduced in ozone-tolerant clones but not in sensitive plants. It has been suggested that a reduced electron transport rate may be a consequence of a reduction in Rubisco content and activity (Miller, Arteca & Pell 1999). Although we do observe some reductions in Vcmax, no measurement of actual Rubisco activity or content was made; therefore, it is not possible to ascertain whether this is the case here.
The visible symptoms that appeared on the surface of the ozone-sensitive leaves suggest that ROS accumulation caused cell death (Fig. 1a). Pell et al. (1997) described how, when stomata are open, as was the case here, there was a rapid loss of semipermeability of the plasma membrane, followed by plasmolysis and, if the stress is maintained, cell death. Given the received dose of ozone (Fig. 4a), this is likely to be a regulated induction of programmed cell death in order to destroy the most vulnerable cells and signal to adjacent cells to increase protection against oxidative stress (Pell et al. 1997). Indeed, the increase in chlorotic and necrotic patches on the surface of the leaves of the ozone-sensitive genotype was reflected in a decline during fumigation of both the ratio of Chl a/b as well as the content of total chlorophyll (Reich 1983; Ranieri et al. 2001; Iglesias et al. 2006) (Table 2). However, a recovery after fumigation was also observed as the ratio of Chl a/b stabilized, and the content of total chlorophyll increased. This lends further weight to the explanation of programmed cell death during ozone fumigation in order to protect the photosynthetic apparatus from damage, rather than uncontrolled cell death.
Unlike the ozone-sensitive genotype, minimal cell death appears to have occurred in the leaves of the ozone-tolerant genotype until some time after the completion of the experiment when the majority of the older, mature leaves senesced (Fig. 1b). This is reflected in the ratio of Chl a/b, which only declined after fumigation had ended although ozone fumigation had no effect on the total chlorophyll content (Table 2).
These results are in keeping with previous findings by Woo and Hinckley (2005) who demonstrated that although this genotype was classified as ozone tolerant, it had a higher rate of leaf senescence than some of the more sensitive genotypes. It has previously been suggested that the observed leaf yellowing prior to senescence may imply that nutrients were being relocated from the older leaves to the newer, still expanding leaves (Frost, Taylor & Davies 1991; Gardner et al. 2005; Woo & Hinckley 2005). In this way, these trees are able to maintain productivity and biomass under elevated ozone. In contrast, the ozone-sensitive species are unable to reassign the nutrients and as such are unable to carry out foliar repair leading to programmed cell death (Amthor 1988; Pell et al. 1997).
One clear difference in responses of the two genotypes was the total carotenoid content. It has previously been suggested that ozone causes light-dependent inhibition of photosynthesis (Leipner, Oxborough & Baker 2001). This would produce a decrease in chlorophyll and total carotenoid content, alongside down-regulation of PSII (Ljudmila, Rech & Jahns 2007). Our results suggest this may have occurred in the fumigated ozone-sensitive genotype, but that this was not the case in the ozone-tolerant plants. These data indicate that no modifications of the photosynthetic and accessory pigments were required in order to maintain the functionality of the photosynthetic apparatus in the tolerant genotype. Furthermore, these plants were able to maintain a supply of protective carotenoids under ozone fumigation (Table 2). We propose this is a fundamental difference in the ability of the two genotypes to withstand the effects of exposure to ozone based on the previous classifications by Woo and Hinckley (2005) and Taylor (personal communication). Calfapietra et al. (2008) also found that total carotenoid content was higher in aspen clones that were resistant to ozone fumigation. They suggest that tolerance to ozone is related to an up-regulation of the metabolic pathway that controls the production of carotenoids, as well as other volatile and non-volatile antioxidant compounds including isoprene.
Prior to fumigation with ozone, the isoprene emission rates of the ozone-tolerant and sensitive genotypes were statistically similar (Fig. 5). While the isoprene emission rate of the ozone-tolerant plants was unaffected by fumigation with ozone, the rate of emission of isoprene of the ozone-sensitive plants was significantly reduced. Isoprene is highly hydrophobic (Rudzinski 2006) and is therefore likely to partition into the lipid phases of membranes. Previous studies have suggested two ways in which isoprene may act to protect against oxidative stress, either through directly quenching ROS species (Velikova et al. 2004) or strengthening membranes (Sharkey et al. 2001; Velikova et al. 2004).
It has also been shown that plants that are able to emit isoprene are better protected against oxidative stress than those that are unable or inhibited from emitting isoprene (e.g. Loreto & Velikova 2001; Loreto et al. 2001; Velikova et al. 2004, 2005; Calfapietra et al. 2007). Calfapietra et al. (2007) found that expressions of the isoprene synthase gene and isoprene synthase protein levels were both reduced under elevated ozone, contributing to a reduction in isoprene emission rate from aspen trees (Populus tremuloides Michx). More recent work by Calfapietra et al. (2008) has also found that plants that are resistant to ozone fumigation are better able to maintain an isoprene emission rate similar to that of non-fumigated control plants. Our data support the hypothesis that resistance to ozone is a combination of several factors that may include the ability to maintain an isoprene emission rate.
Once ozone has entered a leaf through the stomata, it dissolves into the apoplastic fluid where it reacts with H2O and solutes to form ROS (Byvoet et al. 1995). This initiates the onset of various protective scavenging mechanisms, as well as increasing the potential for premature leaf senescence (Tingey & Taylor 1982; Reich & Amundson 1985; Scandalios 1997; Pell et al. 1999; Ranieri et al. 2001). Combined with the similarity in ozone uptake of the two genotypes, the results from the biochemical assays suggest the ozone-tolerant genotype has a higher potential to detoxify and/or protect against the effects of ROS.
This suggestion is further supported by the differences in the generation and subsequent effect of hydrogen peroxide (H2O2) when comparing the two genotypes (Table 2). While H2O2 in the ozone-tolerant genotype increased compared to controls, there was no associated increase in lipid peroxidation as indicated by the level of TBARS. In contrast, increases in H2O2 in the fumigated ozone-sensitive plants were accompanied by simultaneous significant increases in lipid peroxidation. In addition, while the H2O2 content returned to control values during the recovery period, lipid peroxidation increased further, although this returned to values comparable to non-ozone-fumigated controls by the end of the recovery period (Table 2). This suggests that the increase in ROS was sufficient to overwhelm the antioxidant system and attack the membranes causing lipid peroxidation and denaturation.
Furthermore, there was no effect of fumigation on either total or reduced ascorbate in the ozone-sensitive plants. Ascorbate normally acts as an antioxidant by being available for energetically favourable oxidation. ROS oxidize ascorbate to form monodehydroascorbate and then dehydroascorbate (DHA). The ROS are reduced to H2O while the oxidized forms of ascorbate are then themselves reduced (Asada 2006). The lack of change in both total and reduced ascorbate and subsequent increase in the level of lipid peroxidation and H2O2 in the ozone-sensitive genotype, suggests there was minimal scavenging of ROS by ascorbate.
In summary, we suggest that the ability of the ozone-tolerant genotype to maintain growth and productivity under elevated ozone is caused by an increased antioxidant potential when compared to the ozone-sensitive plants. This is at least in part because of elevated total carotenoids and the ability to maintain an isoprene emission rate comparable to the level of control plants. From our data, it is clear that the classification of ozone sensitivity is highly dependent upon the variables chosen for assessment. This is evidenced by the obvious differences in growth and productivity of the subject genotypes clearly shown by Woo and Hinckley (2005) versus the few significant physiological and biochemical differences detected in this study. While net gas exchange and plant photosynthetic efficiency are useful indicators of plant response, they are not always reliable surrogates for plant or crop productivity or stress tolerance (Flowers et al. 2007). In conclusion, we suggest that plant productivity, yield, carotenoid content and isoprene emission rate are fundamental parameters that should be measured when assessing plant sensitivity to ozone.
This work was funded by the NERC (studentship award NER/S/A/2005/13680), BBSRC (grant number BBS/B/12172) and the European Commission Marie Curie Research Training Network ‘ISONET’, and the European Science Foundation ‘VOCBAS’ programme. The authors thank Gail Taylor, Southampton University, for supplying the hybrid poplars used in this work, and Lancaster University Mathematics and Statistics Department for advice on data analysis. We also thank Francesco Loreto, Sue Owen and Peter Harley for their helpful comments and discussions.