Author for correspondence: L. Guidi Tel: +39 50 9719320 Fax: +39 50 598614 Email: firstname.lastname@example.org
• Bean seedlings (Phaseolus vulgaris cv. Pinto) were grown in the greenhouse at a light intensity of 400 µmol m−2 s−1. When the primary leaf was fully expanded, plants were divided into four groups and subjected to one of the following treatments: light intensity of 400 µmol m−2 s−1 and filtered air (control); light intensity of 400 µmol m−2 s−1 and ozone (O3) (150 nl l−1 for 5 h) (ozonated); light intensity of 1000 µmol m−2 s−1 for 5 h and filtered air (HL); and light intensity of 1000 µmol m−2 s−1 and O3 (150 nl l−1) for 5 h (HL + O3).
• At the end of the treatments (HL and/or O3) a strong decrease in CO2 assimilation rate as well a decrease in stomatal conductance were observed, while no changes in intercellular CO2 concentration were recorded. In addition the Fv : Fm ratio (maximal quantum yield for PSII photochemistry) decreased in the stressed leaves (HL and/or O3), indicating photoinhibition, and they showed a corresponding increase in minimal fluorescence (F0), indicating a higher number of deactivating photosystem II (PSII) centres.
• The maximum catalytic activity of the Benson–Calvin cycle enzymes, fructose-1,6-bisphosphate phosphatase (FBPase) and Rubisco, decreased following HL + O3 stress but activation was enhanced. A linear relation was found between activation state of NADP-malate dehydrogenase (MDH) and the flux of electrons through PSII and in HL + O3-treated plants NADP-MDH activity decreased at high irradiance levels, indicating a limitation in linear electron flux.
Tropospheric ozone (O3) is considered one of the most potent phytotoxic air pollutants. A decrease in the rate of light-saturated net photosynthesis as a result of exposure to elevated O3 concentration has been reported (Heath, 1994; Farage, 1996; Farage & Long, 1999). However, the site of O3 damage in the cell bringing about the decrease in light-saturated photosynthesis has not been clearly identified. Some authors have correlated the observed decrease in photosynthesis under O3 exposure with a reduction in stomatal conductance (Reich, 1987; Mansfield & Pearson, 1996). In other cases it was assumed that the decline in photosynthetic rate was the result of the interaction of O3 with photosynthetically active mesophyll cells (Farage et al., 1991). These latter authors found a decrease in carboxylation efficiency, but no changes in the photochemical capacity (Fv : Fm) and apparent quantum yield of CO2 uptake, and concluded that an impairment of carboxylation efficiency was the initial effect of O3 on photosynthesis. Reichenauer et al. (1997) found in Populus nigra that O3 interacts primarily with components of the Calvin cycle, which results in a decrease of photosynthetic rate with subsequent feedback on the current photochemical capacity.
Photoinhibition of photosynthesis has been shown to be of relevance under natural conditions in the absence of any stress other than high light (Ögren, 1988). However, suboptimal environmental conditions, such as the presence of O3, promote photoinhibition even when the photon flux density (PFD) is moderate (Guidi et al., 1999). When carbon assimilation is limited by the decrease in stomatal conductance during O3 exposure, chloroplasts may be subjected to an excess of energy, resulting in the downregulation of photosynthesis or in photoinhibition (Demmig-Adams & Adams, 1996).
Some recent investigations suggest that O3 and light environment may display a strong interaction (Fredericksen et al., 1996; Günthardt-Goerg et al., 1997; Topa et al., 2001), but these were cases in which the effect of O3 was more pronounced in low light or shaded environments. In a natural environment, O3 is often associated with high light and it has been suggested that light stress may cause additional damage to plants suffering from O3 stress (Guidi et al., 2000). However, in field conditions stomatal closure occurs, diminishing the O3 flux into the leaves. Therefore, laboratory experiments and field chamber experiments under low light and low vapour pressure deficit (VPD) always overestimate the detrimental effect of ozone (Grünhage & Jäger, 1994).
The light energy absorbed by an ozonated leaf may largely exceed the photon requirement for photosynthetic electron transport (because net CO2 uptake by leaves under O3 is reduced) and overload the mechanisms that protect photosystem II (PSII) activity from photoinhibition. The factor that determines what level of light intensity is excessive has not yet been established, but it appears to be linked with the redox state of QA, the primary acceptor of PSII (Ögren, 1991). Therefore, environmental conditions, such as high levels of O3, that reduce the rate of chloroplastic electron transport through reduced CO2 assimilation, increase the excitation pressure at any given irradiance and, consequently, the sensitivity of a leaf to photoinhibition (Hurry et al., 1992)
A knowledge of the relationship between the activities of PSII, the redox state of the stroma and the activation states of the light-saturated enzymes of the Calvin–Benson cycle is necessary, if not fundamental, for the understanding of photosynthetic electron transport and photosynthetic control in stressed plants. It is clear that electron transport can operate efficiently only when the supply of NADP, ADP and inorganic phosphate (Pi) is nonlimiting and the parameter of the redox and phosphorylation potentials, termed assimilatory power or assimilatory force, is low (Heber et al., 1986).
The present study was undertaken to evaluate the effects of ozone and/or high light on photosynthetic apparatus in bean leaves in vivo (via chlorophyll (Chl) fluorescence, CO2 assimilation and the activities of component enzymes) and to determine the primary site of damage to the photosynthetic apparatus. The activity of NADP-malate dehydrogenase (MDH) can be used as a physiological indicator of the redox state of the stroma of the mesophyll chloroplasts, where it is localized, because of the dependence of the activation state of this enzyme on the electron flow through the ferredoxin-thioredoxin system and the status of the NADPH/NADP pool (Foyer et al., 1992; Foyer, 1993). Similarly, FBPase, which is located in the stroma, is a hysteretic enzyme regulated by a supply of reducing equivalent from the electron transport system and by the availability of its substrate, fructose-1,6-bisphosphate (FBP) (Harbinson et al., 1989; Foyer, 1993). The enzyme measured in the present work gives information on the availability of reducing equivalents in the stroma (via the activation state of NADP-MDH) and the capacity for catalysis by the Calvin cycle (FBPase and Rubisco). Evaluations of enzyme activities were made simultaneously with those of PSII efficiency to provide new insight into the coordinate control of thylakoid function and carbon metabolism.
Materials and Methods
All measurements were made on the fully expanded primary leaf of bean (Phaseolus vulgaris L., cultivar Pinto) plants (15 d old). Bean seeds were germinated at 25°C on washed vermiculite. Seedlings (5 d old) were transplanted into pots containing a steam-sterilized soil–peat–perlite (1 : 1 : 1, by volume) mixture and placed in a greenhouse (20–25°C and 55–80% relative humidity (r.h.) under a 14-h photoperiod at a PFD of about 400 µmol m−2 s−1).
When the primary leaves were completely expanded, the bean plants were homogeneously divided into four groups. One group was placed in a controlled environment chamber (25 ± 4°C and 85 ± 2% r.h.) in filtered air and subjected to 1000 µmol m−2 s−1 PFD for 5 h (HL). Another group was maintained in the same controlled conditions but at an irradiance of about 400 µmol m−2 s−1 PFD (control). A third group of plants was exposed to a single pulse of O3 (150 nl l−1 for 5 h) (ozonated) and the fourth was exposed to a single pulse of O3 and high light treatment (HL + O3). The O3 treatment was carried out under the same controlled conditions as used for the high light treatment.
The O3 treatment was performed in a controlled environment chamber (0.483 m3) (Cavallo, Milan, Italy). Air supplied to the chamber was passed through a charcoal filter. Ozone was generated from pure oxygen by electric discharge (ozone generator Fisher 500; Fisher, Meckenheim, Germany) and added to the filtered air. The flow of ozone-enriched air to the controlled environment chamber was regulated by a flow controller. The ozone concentration inside the chamber was monitored continuously with an analyser connected to a PC (Model 8810; Monitor Laboratories, Englewood, USA). For this experiment, the target O3 concentration was 150 nl l−1 for 5 h. Control plants were maintained under the same experimental conditions as O3-treated plants but exposed to charcoal-filtered air.
Measurements of gas exchange were made on the primary leaf at the end of each treatment (high light and/or O3) using an open system (CMS-400; Walz, Effeltrich, Germany). The experimental procedures were as described by Soldatini & Guidi (1992). During gas exchange measurements in an assimilation chamber (GK-022; Walz), the temperature was 25 ± 2.8°C, r.h. 65 ± 12%, CO2 concentration was 350 µmol mol−1 and O2 was 21%. Responses to irradiance (0–1000 µmol m−2 s−1 PFD) of photosynthetic CO2 assimilation in the leaf were calculated using the Smith equation:
( is the quantum yield for CO2 uptake; Amax is CO2 assimilation rate at light saturation level; I is the irradiance; Tenhunen et al., 1976). Stomatal conductance (gs), transpiration rate (E) and intercelluar CO2 concentration were measured at a CO2 mole fraction of 350 µmol mol−1 and at saturating light. The quantum yield for CO2 uptake () was calculated from the initial slope of the light-response curve using least-squares linear regression analysis.
Chlorophyll fluorescence measurements
Modulated chlorophyll a fluorescence measurements were made with a PAM-2000 fluorometer (Walz) on the primary bean leaves dark-adapted for 40 min The fluorometer was connected to a leaf-clip holder (2030-B; Walz) with a trifurcated fibre optic (2010-F, Walz) and to a computer with data acquisition software (DA-2000; Walz). The minimal fluorescence, F0, was determined using the measuring modulated light which was sufficiently low (< 1 µmol m−2 s−1) not to induce any significant variable fluorescence. The maximal fluorescence level, Fm, was determined by a 0.8-s saturating pulse at 8000 µmol m−2 s−1 in dark-adapted leaves. The saturation pulse method was used for the analysis of quenching components (Schreiber et al., 1986). Intermittent, abrupt illumination by sufficiently strong light causes a transient, but complete removal of photochemical quenching, causing a corresponding increase in variable fluorescence, Fv to Fv′; any residual quenching is assumed to be nonphotochemical. The intensity of the actinic light was maintained at about 400 µmol m−2 s−1 and saturating flashes of white light 15 000 µmol m−2 s−1 and 800 ms duration were given every 20 s. After the saturating pulse, the maximal fluorescence reached the Fm′ value and the actinic light allowed steady-state photosynthesis and modulated fluorescence yield at this steady-state (Fs) to be reached. Quenching components qP and qNP were calculated as defined by Schreiber et al. (1986). The values of Φexc. (Fv′ : Fm′), and the apparent electron transport rate (PSII = qP × Φexc. × PFD × 0.5 × 0.84) were computed (Schreiber et al., 1986), where Fm′ is the maximal fluorescence, F0′ is the minimal fluorescence and Fv′ is the difference between Fm′ and F0′ in the light-adapted state.
To measure the fluorescence-PFD response, primary bean leaves were held horizontally in the leaf-clip holder. The PFD on the leaves, provided by a halogen lamp (2050-H, Walz), was adjusted from darkness to 1000 µmol m−2 s−1 in steps of 50–200 µmol m−2 s−1. The halogen lamp was equipped with a heat-reflecting filter to reduce heat generation by the lamp. The PFD on the leaf was monitored with a microquantum sensor installed on the leaf-clip holder next to the spot where fluorescence was measured. After the leaf was exposed to the desired PFD for 20 min, the chlorophyll a fluorescence of PSII was measured using the PAM-2000.
For the calculation of the parameter (1/F0 − 1/Fm), the chlorophyll fluorescence yields after dark incubation (F0 and Fm) were all normalized to the F0 value of the leaf control. Because of this normalization, 1/F0 − 1/Fm = Fv/Fm for the control sample and represents 100% of the functional PSII complexes present before any photoinhibitory treatment. This parameter has been suggested to be an indicator of PSII reaction centre functionality (Havaux et al., 1991; Walters & Horton, 1993), and empirically correlated linearly with the oxygen yield per single-turnover flash (Park et al., 1995; Lee et al., 1999).
Total nonphotochemical quenching, qN, is heterogeneous and has three distinct phases of relaxation (Horton & Hague, 1988; Krause & Weis, 1991). Energy quenching, qNE = 1 − ((Fm′ − F0′)/(Fm″ − F0)) is also referred to in this study as qF, the fast-relaxing component of qN, while the quenching remaining after 3 min darkness provides an estimate of the slowly relaxing component of qN, referred to in this study as qS; the latter can be attributed, at least in part, to the state transitions (Krause & Weis, 1991) and is defined as 1 − ((Fm″ − F0)/(Fm − F0)). For the analysis of qN, the relaxation method reported by Keiller et al. (1994) was adopted. After the induction of qN using the saturation pulse method, 3 min of darkness was applied. Following these 3 min, another saturating pulse was given and a final level of light-saturated fluorescence, Fm″, was measured.
Before leaf sampling, plants were kept at increasing photosynthetic photon flux densities (PPFDs) for 20 min in the leaf clip in the PAM-2000. Immediately after the chlorophyll fluorescence measurements, sampling was carried out on the irradiated portion of the leaf, which was removed by forcing a brass cutter, previously chilled with liquid nitrogen, through the leaf, such that irradiation was uninterrupted as reported by Foyer et al. (1992). The leaf discs were removed from the apparatus while still frozen and stored at −80°C until needed.
Concentrations of NADP-MDH and FBPase were measured following the method described by Harbinson et al. (1990). Frozen samples were ground in liquid nitrogen and resuspended in a 0.1 m Tricine-KOH buffer (pH 8.0) containing 10 mm MgCl2 1 mm ethylenediaminetetraacetic acid (EDTA) and 0.1% Triton X-100 and 1 mm dithiothreitol (DTT). The extract was centrifuged at 12 000 g for 10 min at 4°C and the supernatant was immediately used to determine initial enzyme activity. The NADP-MDH was measured following the oxidation of NADPH in a 1-ml reaction mixture containing 0.1 mm Tricine-KOH (pH 8.0), 10 mm MgCl2, 1 mm EDTA, 0.2 mm NADPH, 0.5 mm oxaloacetate and 20 µl of leaf extract. Maximum activation was achieved by incubation of the extract with 100 mm DTT for 10 min at 25°C before assay. The FBPase activity was measured by the increase in the absorbance at 340 nm of a 1-ml reaction mixture containing 100 mm Tricine-KOH (pH 8.0), 10 mm DTT, 10 mm MgCl2, 0.2 mm NADP+, 0.5 mm FBP, 2.5 units of Glc-6-P-dehydrogenase and 10 units of phosphoglucose isomerase. Maximum activation was measured following incubation of the leaf extract with 2 mm FBP and 0.1 m Tricine buffer (pH 8.0) for 10 min at 25°C before assay.
The activity of Rubisco was determined according to the method of Sawada et al. (1990) and Holaday et al. (1992) with minor modifications, using an extraction buffer containing 250 mm Tris-HCl (pH 7.8), 50 mm MgCl2, 2.5 mM EDTA and 37.5 mg DTT. The extract was centrifuged at 10.000 g for 10 min at 4°C and the supernatant was immediately used to determine initial Rubisco activity. The total Rubisco activity was determined after the incubation of leaf extract for 10 min with 20 mm MgCl2 and 10 mm NaHCO3 at 25°C. The carbamylation ratio of Rubisco was determined as initial Rubisco activity/total Rubisco activity. Rubisco activity was determined in an assay buffer containing 50 mm Tris-HCl (pH 7.8), 10 mm MgCl2, 30 mm NaHCO3, 5 mmβ-mercaptoethanol (β-MeSH) and 5 mm ATP. The following were also added: 5 units of 3-phosphoglyceric phosphokinase, 5 units of glyceraldehyde-3-phosphate dehydrogenase, 1 unit of glycero-3-phosphate dehydrogenase, 6 units of triose phosphate isomerase and 0.15 mm NADH. The reaction was initiated by the addition of 5 mm of ribulose-1,5-bisphosphate and the oxidation of NADH was monitored at 340 nm.
Data derived from three independent experimental trials. Data were subjected to a two-way analysis of variance (anova) with ozone and high light as factors. The LSD0.05 test was used for comparison of the means.
The quantum yield for CO2 assimilation was determined from the slope of the regression of photosynthetic CO2 assimilation vs light intensity (30–80 µmol m−2 s−1).
Polynomial second-order curves were fitted for the response of enzymatic activity to increasing irradiance levels. For the relationship between rates of photosynthetic electron transport through PSII (JPSII) and CO2 assimilation rate or NADP-MDH activity, the linear regression was applied and the correlation coefficients were analysed.
Preliminary experiments were conducted for the choice of intensity and duration of the exposure to high light. Only at a light intensity of 1000 µmol m−2 s−1 for 5 h was a sustained photoinhibition recorded, and this by the strong decrease of the Fv : Fm ratio and an increase in F0 value. Plants exposed to this light intensity recovered from photoinhibition 24 h after they were exposed to the grown light intensity (data not shown).
Fixation of CO2
The fixation of CO2 by control bean plants increased with increasing irradiance up to the maximum of 600 µmol m−2 s−1. At this irradiance, CO2 assimilation approached light saturation (Fig. 1; note that units are required on the axes of Fig. 1) with values of maximal photosynthetic CO2 fixation of about 8 µmol CO2 m−2 s−1 (Table 1). Following O3 treatment, there was also an effect on this relationship below 200 µmol m−2 s−1 (Fig. 1), suggesting an effect on the quantum yield of CO2 fixation under light-limiting conditions and implying the presence of photoinhibition. At irradiance higher than 200 µmol m−2 s−1, light saturation of CO2 fixation in ozonated leaves was observed with a maximal rate of photosynthetic CO2 assimilation in air between 2 µmol CO2 m−2 s−1 and 3 µmol CO2 m−2 s−1 (Table 1). High light treatment also reduced CO2 assimilation at low and high irradiance and a substantially similar response occurred when bean plants were treated with O3 and high light at the same time (Fig. 1).
Table 1. Gas exchange parameters determined on the primary leaves of Phaseolus vulgaris cultivar Pinto
Plants were exposed to high light (1000 µmol m−2 s−1 for 5 h) (HL) and/or a single pulse of ozone (O3, 150 nl l−1 for 5 h) (HL + O3 and ozonated, respectively). Controls were plants maintained in charcoal-filtered air and at a light intensity of about 400 µmol m−2 s−1 (Control). is the quantum yield of CO2 uptake and is expressed as µmol CO2 µmol−1 photons. Photosynthetic activity (Amax, µmol CO2 m−2 s−1), stomatal conductance to water vapour (gs, mmol H2O m−2 s−1), transpiration rate (E, mmol H2O m−2 s−1) and intercellular CO2 mole fraction (Ci, µmol mol−1) were determined at light saturation (about 600 µmol m−2 s−1), 345 µmol mol−1 CO2, 21% O2, 25°C and 65% relative humidity. Each value represents the mean of six replicates. Means followed by the same letter are not different at P = 0.05 according to the least significant difference (LSD) test determined by two-way anova.
HL + O3
Table 1 shows the parameters related to gas exchange. At a light intensity of about 600 µmol m−2 s−1, maximum photosynthetic CO2 assimilation (Amax) was significantly reduced by the stresses (O3 and/or high light), as were stomatal conductance and transpiration rate. Intercellular CO2 concentration was not influenced by the stresses, while a strong decrease in the value of quantum yield for CO2 assimilation () was observed following the different treatment. Therefore it appears that ozone and/or high light induced a decrease in CO2 fixation attributable both to strong stomatal closure but also to the decreased capacity of the mesophyll to carry out CO2 fixation.
Sensitivity to photoinhibition
The ratio between variable and maximal fluorescence, Fv : Fm, is widely used as an estimate of the maximum quantum yield for PSII electron transport. Control bean plants showed an Fv : Fm value of 0.811, which is typical for healthy plants (Björkman & Demmig, 1987) (Table 2). Following the O3 treatment, a significant decrease of this ratio was observed (−11%). HL- and HL + O3-treated plants exhibited a 19% and 31% reduction in Fv : Fm, respectively, compared with the photochemical efficiency measured in the control. Measurements carried out 24 h after removing the stresses indicated a complete recovery of the value of Fv : Fm in HL-treated plants, while for O3- and/or HL-treated plants, the Fv : Fm ratio returned to values similar to the controls 4 days after the end of the treatments (data not shown).
Table 2. Chlorophyll a fluorescence parameters determined on the primary leaves of Phaseolus vulgaris cultivar Pinto
(1/F0 – 1/Fm)
Plants were exposed to high light (1000 µmol m−2 s−1 for 5 h) (HL) and/or a single pulse of ozone (O3, 150 nl l−1 for 5 h) (HL + O3 and ozonated, respectively). Controls were plants maintained in charcoal-filtered air and at a light intensity of about 400 µmol m−2 s−1 (Control). Measurements were made at room temperature. For the (1/F0 − 1/Fm) parameter, all fluorescence yields were normalized to the F0 of control leaf. Each value represents the mean of nine replicates. Means followed by the same letter are not different at P = 0.05 according to the least significant difference (LSD) test determined by two-way anova. F0, minimal fluorescence; Fm, maximal fluorescence; Fv/Fm, maximal quantum yield for PSII photochemistry.
HL + O3
The values of initial fluorescence, F0, significantly changed in all treatments (Table 2). In the HL + O3 stressed plants, however, the F0 value increased dramatically and this indicates that PSII reaction centres in the stressed plants were severely damaged. The Fm values decreased significantly by 10, 24 and 20% in the ozonated, HL and HL + O3 plants, respectively, compared with the controls. The sustained decrease in dark-adapted Fv : Fm and the increase in F0 indicate the occurrence of photoinhibitory damage in response to O3 and/or high light. However, changes in Fv : Fm and F0 are still accepted and widely used as reliable diagnostic indicators of photoinhibition (He et al., 1996; Valladares & Pearcy, 1997).
The parameter (1/F0 − 1/Fm) obtained from measurements of F0 and Fm after dark adaptation represents an alternative indicator of PSII functionality (Havaux et al., 1991; Walters & Horton, 1993) and is reported in Table 2. As expected, this parameter was similar to Fv : Fm in controls and declined significantly in O3- or HL-treated leaves. The greater reduction was observed in leaves subjected to the combined O3 and HL treatments.
The relationship between linear photosynthetic electron transport and CO2 fixation
The product of ΦPSII and irradiance gives the JPSII, a value that is proportional to the rate of photosynthetic electron transport through PSII, although it is not a measure of the latter. Accurate determination of PSII electron transport cannot be made unless the absorbance of light by the photosynthetic pigments and the distribution of excitation energy between PSI and PSII are known. A nonlinear relationship, with respect to the increasing yield of CO2 fixation, was evident as JPSII increased in control bean plants (Fig. 2) (y = 0.35 + 0.35x − 0.0002x2, r = 0.973***). A very different picture appeared in stressed plants. Very low values of CO2 fixation and JPSII were recorded following the O3 treatment (Fig. 2) and the relationship between CO2 assimilation rate and JPSII disappeared (y = 0.27 + 0.01x, r = 0.444 ns); in these leaves the heterogeneity of the response was greater than in control plants. High light reduced the slope of the regression line calculated between CO2 fixation and JPSII (y = 0.73 + 0.02x, r = 0.974***) as well as in high light and ozonated plants (y = 0.14 + 0.02x, r = 0.964***) (Fig. 2).
Effect of stresses on PSII excitation pressure
The reduction state of Primary acceptor of PSII (QA) (i.e. (QA)red/((QA)red + (QA)ox) can be estimated by steady state chlorophyll a fluorescence as 1 − qP at growth irradiance, reflecting PSII excitation pressure (Dietz et al., 1985; Demmig-Adams et al., 1990). The results in Table 3 indicate that HL treatment resulted in a fourfold increase in PSII excitation pressure compared with the controls. A strong and significant increase in the 1 − qP parameter was also observed in O3- and O3 + HL-treated plants.
Table 3. Steady-state fluorescence quenching characteristics determined on the primary leaves of Phaseolus vulgaris cultivar Pinto
1 − qP
Plants were exposed to high light (1000 µmol m−2 s−1 for 5 h) (HL) and/or a single pulse of ozone (O3, 150 nl l−1 for 5 h) (HL + O3 and ozonated, respectively). Controls were plants maintained in charcoal-filtered air and at a light intensity of about 400 µmol m−2 s−1 (Control). Measurements were made at room temperature at a photon flux density (PFD) of 400 µmol m−2 s−1. Each value is the mean of nine replicates. Means followed by the same letter are not different at P = 0.05 according to the least significant difference (LSD) test determined by two-way anova.
HL + O3
Chlorophyll fluorescence parameters
In control plants ΦPSII reached a value of 0.573 and decreased significantly in ozonated leaves (Table 3). The greater decrease in ΦPSII was recorded in HL- and O3 + HL-treated leaves, with a reduction of about 70% in comparison with the controls.
The parameter Φexc. is an estimate of the efficiency of excitation energy transfer to open PSII traps. This parameter decreased significantly in ozonated plants (23%); a greater reduction was observed in HL- and HL + O3-treated plants, in which the reduction was of about 40% compared with the controls (Table 3).
The higher reduction state of QA in stressed plants also corresponded to the increase in qNP (Table 3). In O3- and/or HL-stressed plants a strong and significant increase of nonphotochemical quenching was observed.
The partitioning of total nonphotochemical quenching, qNP, into its fast (qNPf) and slow relaxation-time components (qNPs) are reported in Table 4. The qNPf component increased significantly in ozonated plants (+119%), but more pronounced was the increment in HL- and HL + O3-stressed plants (> 200%).
Table 4. Fast (qNPf) and slow (qNPs) relaxing components of nonphotochemical quenching qNP determined on the primary leaves of Phaseolus vulgaris cultivar Pinto
Plants were exposed to high light (1000 µmol m−2 s−1 for 5 h) (HL) and/or a single pulse of ozone (O3, 150 nl l−1 for 5 h) (HL + O3 and ozonated, respectively). Controls were represented by plants maintained in charcoal-filtered air and at a light intensity of about 400 µmol m−2 s−1 (Control). Measurements were made at room temperature. Each value represents the mean of nine replicates. Means followed by the same letter are not different at P = 0.05 according to the least significant difference (LSD) test determined by two-way anova.
HL + O3
An increase in qNPs was only observed in the HL + O3-stressed plants, while in O3- and HL-stressed plants a decrease and no change, respectively, were found.
Effect of stresses on activities of the carboxylation enzyme
Since CO2 assimilation was significantly depressed in stressed plants, the Rubisco enzyme involved in CO2 photoassimilation was examined (Fig. 3). The activity of Rubisco was measured immediately after extraction (initial activity) and after incubation with Mg2+ and CO2 (total activity). In dark-adapted bean leaves the initial activity was 1.4 (± 0.8) mmol CO2 min−1 mg−1 protein for controls, 2.5 (± 0.2) mmol CO2 min−1 mg−1 protein for O3-treated leaves, 5.0 (± 1.2) mmol CO2 min−1 mg−1 protein for HL-treated leaves and 0.7 (± 0.0) mmol CO2 min−1 mg−1 protein for HL + O3-treated leaves. In vitro activation increased the activity to 54 (± 2.4) mmol CO2 min−1 mg−1 protein (controls), 27 (± 16) mmol CO2 min−1 mg−1 protein (O3), 23 (± 17) mmol CO2 min−1 mg−1 protein (HL) and 0.9 (± 0.0) mmol CO2 min−1 mg−1 protein (HL + O3). During the light period, maximal activated Rubisco activity increased to about 19 mmol CO2 min−1 mg−1 protein in controls (Fig. 3) and the light-induced increase in the maximal activated Rubisco activity following the different treatments was 50 mmol CO2 min−1 mg−1 protein (O3), 70 mmol CO2 min−1 mg−1 protein (HL) and 7 mmol CO2 min−1 mg−1 protein (HL + O3). A pronounced effect of HL + O3 treatment was the increase in activation state of Rubisco from 2% in controls to 22% and 78%, respectively, in HL- and HL + O3-treated leaves.
There was a different response by Rubisco to increasing irradiance levels in HL-treated plants: these plants showed an increase in total Rubisco activity for values of irradiance between 400 µmol m−2 s−1 and 800 µmol m−2 s−1, but a decrease was observed at higher irradiance. A different picture for HL + O3-treated leaves was seen, where the total Rubisco activity was lower compared with controls in response to increasing irradiance levels.
Effect of stresses on thiol-regulated stromal enzymes
For determination of NADP-MDH and FBPase activities, plants were exposed to various light intensities for 20 min after the treatments (O3 and/or HL), before the leaves were rapidly extracted and tested for enzymes activities.
In dark-adapted plants the maximal catalytic activities (activated by incubation with DTT) of FBPase were 77 (± 90) mmol NADPH min−1 mg−1 protein for controls, 159 (± 60.5) mmol NADPH min−1 mg−1 protein for O3-treated leaves, 72 (± 0.0) mmol NADPH min−1 mg−1 protein for HL-treated leaves and 36 (± 0.28) mmol NADPH min−1 mg−1 protein for HL + O3-treated leaves. The activation state of FBPase in controls was about 95% but strongly decreased in stressed plants, which showed values of FBPase activation state of 4% (O3), 5% (HL) and 19% (HL + O3).
Total FBPase activity in leaves of the control was almost constant at increasing irradiance while an enhancement in its activity was observed in ozonated leaves at irradiance of up to 500 µmol m−2 s−1 (Fig. 4). In HL-treated leaves, the total activity of FBPase was higher than the controls at all irradiance levels, while in plants subjected to the combined treatment of HL + O3, FBPase decreased at increasing irradiance after an initial increase (Fig. 4).
In dark-adapted leaves, maximal catalytic NADP-MDH activity (activated by incubation with DTT) was 30 (± 0.0) mmol NADPH min−1 mg−1 protein for controls, 391 (± 204) mmol NADPH min−1 mg−1 protein for O3-treated leaves, 17.7 (± 5.77) mmol NADPH min−1 mg−1 protein for HL-treated leaves and 88 (± 91) mmol NADPH min−1 mg−1 protein for HL + O3-treated leaves. NADP-MDH activity increased markedly with increasing illumination, up to a maximal value at 750 µmol m−2 s−1 (Fig. 5). At very low irradiance, the NADP-MADH activity of leaves from the ozonated plants was strongly increased compared with the controls; the increased variability following O3 treatment did not allow a more precise interpretation. A similar pattern was also shown by values of NADP-MDH activity of HL-treated leaves (Fig. 5). In leaves subjected to the combined treatments HL and O3, after an initial increase in NADP-MDH activity at low irradiance, the catalytic activity of this enzyme was lower than the controls at high irradiance levels (Fig. 5).
The NADP-MDH activity increased markedly with increasing JPSII in control and HL-treated leaves (y = 50 + 2.37x, r = 0.661**, y = 145 + 1.83x, r = 0.687**, respectively) (Fig. 6). In ozonated and HL + O3-treated leaves, NADP-MDH activity strongly decreased as JPSII increased (y = 428 − 1.56x, r = 0.261 ns, y = 144 – 0.52x, r = 0.196 ns, respectively); the increased variability following these treatments did not allow a more precise interpretation.
The present results showed some impairment in the CO2 assimilation rate at increasing irradiance, owing to the presence of O3. This impairment depended on stomatal limitations, as shown by the lower gs, but also by the altered mesophyll activity, as indicated by the maintenance of intracellular CO2 concentration (Ci) after the decrease of Amax. In addition, the quantum yield for CO2 uptake significantly diminished in ozonated leaves. A similar picture was also seen for HL- and HL + O3-treated leaves. It has often been suggested that stomatal closure in the presence of a pollutant, such as O3, could constitute an important mechanism for avoidance on the part of internal tissues. This cannot be disputed, but if stomatal closure occurs in response to intercellular CO2 after the mesophyll has been damaged to some degree, the protection offered may be of only limited benefit (Mansfield & Pearson, 1996). Other authors have proposed that the elevation of CO2 concentration in the internal gas spaces of the leaf is the reason for closure of stomata (Farage et al., 1991).
It is known that high light intensity also induces impairment of CO2 assimilation rate (Krause, 1988), as can be seen in bean leaves. The reduction induced by high light and/or ozone leads not only to a decrease in the capacity for CO2 fixation related to a decrease in stomatal conductance, but also to an alteration in mesophyll capacity for photochemistry as indicated by the unchanged Ci. These results are in agreement with the previous work (Guidi et al., 2000). Noormet et al. (2001) reported a decrease in CO2 assimilation rate in response to O3 induced primarily by altered mesophyll processes, as confirmed by changes in Rubisco content and the stability of intercellular CO2 concentration.
The relationships between JPSII and CO2 fixation suggest that the capacity of electron transport in driving CO2 assimilation increases with the rate of the electron transport until a certain level, after which no increase in CO2 assimilation rate is observed. In stressed plants this relationship is qualitatively modified and quantitatively reduced; a fraction of reducing equivalents produced by electron transport was not utilized by CO2 fixation. Ozone in particular had a detrimental effect on the relationship between CO2 assimilation and JPSII, which lost its linearity.
The Fv : Fm ratio, the maximum quantum efficiency of PSII photochemistry, decreased significantly in O3- and/or HL-stressed leaves and the decrease in the ratio is attributable to a strong increase in F0 and a decrease in Fm. Sustained increases in F0 are associated with photoinhibitory damage (Cleland et al., 1986) and may be explained as a less efficient trapping and exciton conversion to heat. Generally, the increase in F0 would indicate damage to PSII reaction centres, which is not readily reversible. However, this is not always the case since photoinhibition of leaves showing strong increases in F0 may well exhibit fast and complete recovery. This was observed in cv. Pinto leaves with recovery time of 1 d and 4 d in HL and O3 + HL treatments, respectively. In the same species subjected to photoinhibition, other authors also found a total recovery of F0 (Greer et al., 1986).
The reduction in Fv : Fm and the associated increase in F0 may be due to the formation of nonfunctional PSII centres (PSIINF), as reported by Krause (1988). Havaux et al. (1991) and Walters & Horton (1993) have suggested that the chlorophyll fluorescence parameter (1/F0 − 1/Fm) may reflect PSII functionality. Indeed (1/F0 − 1/Fm) declines linearly with the loss of functionality of PSII as reported by Lee et al. (1999). The value of (1/F0 − 1/Fm) recorded in O3- or HL-treated leaves declined significantly, with a more pronounced reduction in HL + O3-treated leaves. In leaves subjected to HL + O3 treatment, the strong reduction in (1/F0 − 1/Fm) indicates extremely reduced PSII functionality, and was accompanied by a strong reduction in the PSII photochemical efficiency (Fv : Fm) and a strong increase in F0. This suggests that the combination of ozone and high light cause an increase in PSII damage.
The actual photochemical efficiency of PSII, ΦPSII, decreased in stressed plants with a more pronounced reduction in HL- and HL + O3-treated leaves and a significant inhibition was found in JPSII in O3- and/or HL-stressed leaves.
It is known that photoinhibition is related to the redox state of PSII (Ögren, 1991; Öquist et al., 1992a,b). Ögren & Rosenqvist (1992) have reported that photoinhibition in different taxonomic groups was dependent on PSII excitation pressure measured as 1 − qP. The PSII excitation pressure increased in stressed bean leaves, with the highest reduction observed in HL-treated plants. The HL treatment seems to enhance the photoinhibitory effect induced by ozone but also the excitation pressure on PSII. The higher reduction state of QA found in stressed plants may be linked also to the higher internal CO2 concentration (Ci) present in leaves of O3- and/or HL-stressed plants. In a recent review Melis (1999) reported that the probability of PSII photodamage depends on the redox state of QA. Photodamage occurs with less probability when QA is oxidized and excitation energy is used in electron transport. When QA is reduced, excitation energy is dissipated by nonassimilatory charge recombination. Ozone and high light determined a reduction in CO2 assimilation rate and therefore a lower utilization of reducing power, resulting in an increased portion of closed PSII reaction centres (reduced QA) in thylakoid membranes. This condition leads to greater photodamage, which is more evident in O3 + HL-treated leaves.
The PSII excitation pressure increased in stressed plants and this indicates that the reoxidation of the electron acceptor QA is less effective and implies that a fraction of the PSII traps are closed during actinic illumination. These closed traps, which are unable to undergo charge separation and to take part in linear photosynthetic electron transport, should lead to a decrease in actual quantum yield (ΦPSII). The decrease in ΦPSII observed in stressed plants with a more pronounced effect in HL- and O3 + HL-treated leaves, was accompanied by decreases in Φexc.. The decline in this parameter indicates an increase in the probability of a photon absorbed by PSII antennae being dissipated as heat, and demonstrates the occurrence of a stress-induced down-regulation of photosynthesis. In summary, O3 and HL induced a decrease in Φexc., and a large increase in nonphotochemical quenching, as also demonstrated by the significantly higher qNP, with concomitant decreases in ΦPSII and . The decrease in ΦPSII must also be attributed to an increase in nonphotochemical quenching.
Photoinhibition also results in an increase in thermal energy dissipation, which constitutes a photoprotective mechanism and does not represent damage (Demmig-Adams & Adams, 1992. In bean leaves subjected to O3 and/or HL, a strong increase in qNP was observed. Partitioning total nonphotochemical quenching qNP into its fast (qNf) and slow relaxation-time components (qNs) indicated that, despite a relatively high constant level of qNP in stressed leaves (O3 and/or HL), there were changes in qNf and qNs. Both ozone and high light determined an increase in qNf components. Demmig & Winter (1988) described qNf as a sensitive indicator of the point at which light starts to become excessive. This fast-relaxing component of qNP is thought to be largely linked to the trans-thylakoid pH gradient (Krause & Weis, 1991). Thus, the increased values of qNf are consistent with the decreased capacity to assimilate CO2 and also to the need to reduce PSII activity.
Demmig & Winter (1988) suggested that increased levels of qNs in spinach leaves predominantly occur when the leaves are subjected to potentially photoinhibitory conditions. This also seems to be the case for leaves subjected to O3 + HL treatment, where qNs increased in comparison with the controls and was accompanied by the greatest decrease in Fv : Fm ratio. However, the role of this component of qNP is not completely clear, since it diminished in O3- or HL-treated leaves. The variations in qNs are related to heat dissipation in the xanthophyll cycle (Demmig-Adams & Adams, 1992) but also to photoinhibition of photosynthesis. Furthermore, photoinhibition may be caused by an increase in xanthophyll cycle-dependent heat dissipation and/or reversible inactivation or irreversible damage of some PSII reaction centres (Long et al., 1994). An increase in xanthophyll cycle-dependent heat dissipation leads to a decrease in F0, while irreversible damage or reversible inactivation of PSII reaction centres leads to an increase in F0 (Demmig et al., 1987; Krause & Weis, 1991). Therefore, from our results, it seems that irreversible damage or reversible inactivation of PSII reaction centres occurred without the involvement of the xanthophyll cycle-dependent heat dissipation.
In optimal conditions, the over-reduction of the stroma is prevented by coordination of ATP and NADPH synthesis with rates of use in carbon metabolism (Foyer et al., 1990, 1992). The evidence presented here shows that the enzymes of the Calvin cycle in dark-adapted leaves are inhibited by O3 and/or HL, since total Rubisco activity was decreased. In the dark, maximal extractable FBPase activity was also decreased following exposure to the two stresses combined, while an increase and no effects were observed in ozonated and HL-treated leaves. The loss of these enzymes after O3 and/or HL stress, with the exception of total FBPase activity in ozonated leaves, indicates that stresses induced changes in the enzyme proteins. In the light, total Rubisco activity in ozonated leaves was higher than in the controls and showed a similar light response; the increase in total Rubisco activity was accompanied by a reduction in activation state (initial to total activity ratio) of the enzyme. This suggests that in ozonated leaves a lower proportion of the Rubisco sites exits in the carbamylate form since this ratio is used as a measure of carbamylation of Rubisco (Butz & Sharkey, 1989). Different behaviour was observed in the HL + O3-treated leaves in which Rubisco activity was lower at increasing irradiance compared with the controls. In these leaves, a higher activation state of Rubisco was found. This would suggest that after the combined treatment of O3 and HL the stroma is more reduced and the increased activation of Rubisco suggests that the system senses a decreased capacity for CO2 assimilation and attempts to redress the balance via an increased activation of the component enzyme.
The NADP-MDH enzyme is modulated by reversible thiol reduction via the Fd-thioredoxin system. At low irradiance, NADP-MDH activity in stressed leaves is higher than in the controls and this indicates that the mesophyll stroma is relatively reduced. The NADP-MDH activity increases as the flux of reducing equivalents increases with increasing irradiance. A linear relationship has been demonstrated to exist between activation states of NADP-MDH and the flux of electrons through photosystems (Harbinson et al., 1990) and this also was found to occur in bean leaves. In O3- and HL + O3-treated leaves, NADP-MDH activity decreased at high irradiance and may indicate a limitation of linear electron flux. However, in these leaves, a strong and significant reduction in JPSII was observed, which confirms this hypothesis. There was a different response in HL-treated plants in which, at high irradiance, NADP-MDH activity tended to increase: the relationship between the NADP-MDH activity and JPSII was similar to the controls, as was the relationship between CO2 assimilation and JPSII. This indicates that, in these plants, the mechanism to prevent the overreduction of the stroma was effective even if not completely efficient, since the 1 − qP parameter strongly increased.
It is clear that because of scatter encountered in enzymatic activity results, it is difficult to draw any conclusions. It is also true that these types of enzymatic assays frequently result in very wide scattering, as other authors have also reported. Kingston-Smith et al. (1997) reported that the large variability of data was probably intrinsic to maize. Harbinson et al. (1990), who worked on pea found the same large degree of variability.
In conclusion, plants treated with a single pulse of O3 showed a decreased photosynthetic CO2 assimilation associated with a reduced stomatal conductance and altered mesophyll activity. The O3-treated leaves showed clear symptoms of photoinhibition, as confirmed by the reduction in Fv : Fm ratio. The increase in F0 value is linked to a reduced PSII functionality, which also determines an increase in excitation pressure at PSII level (increase in 1 − qP). The increase in nonphotochemical quenching did not prevent damage at PSII level. Total Rubisco activity increased in O3-treated leaves, as well as the FBPase activity, while total NADP-MDH activity was higher than the control at low irradiance but lower at higher irradiance levels.
The HL-treated leaves showed an alteration in CO2 assimilation, which was accompanied in this case both by reduced stomatal conductance and altered mesophyll activity. Rubisco activity was lower than the control while a strong increase in FBPase activity was observed.
In the combined treatments of ozone and high light, the picture was substantially similar to those recorded for O3- or HL-treated leaves, but the response was more noticeable: a substantial increase in F0 and a strong decrease in Fv : Fm were found compared with the other two treatments (HL or O3). Total Rubisco activity was lower than in the controls, even though the carbamylation ratio was stimulated.
To summarize, sustained reductions in the efficiency of photosynthetic energy conversion were observed in plants exposed to high light that were also subjected to an additional stress factor such as ozone. These sustained reductions were associated with sustained increases in thermal energy dissipation but also to deactivation of PSII reaction centres. It is concluded that high light stress can modify the reaction of the leaves to ozone stress by inactivation of the PSII reaction centres, also assisted by a strong reduction state of the electron chain.
This study was financially supported by the MIUR (National Projects), Rome (Italy).