One katal represents the conversion of one mole of substrate into one mole of product per second.
Author for correspondence: Jeremy Barnes Tel: +191 2227374 Fax: +191 2225229 Email: J.D.Barnes@ncl.ac.uk
• The potential limitations on net leaf carbon assimilation imposed by stomatal conductance, carboxylation velocity, capacity for ribulose 1,5-bisphosphate regeneration and triose phosphate ultilization rate were derived from steady-state gas exchange measurements made over the life-span of two leaves on plants of an ‘O3-sensitive’ population of Plantago major grown at contrasting atmospheric O3 concentrations.
• Parallel measurements of chlorophyll fluorescence were used to monitor changes in the quantum efficiency of PSII photochemistry, and in vitro measurements of Rubisco activity were made to corroborate modelled gas exchange data.
• Data indicated that a loss of Rubisco was predominantly responsible for the decline in CO2 assimilation observed in O3-treated leaves. The quantum efficiency of PSII was unchanged by O3 exposure.
• Stomatal aperture declined in parallel with CO2 assimilation in O3-treated plants, but this did not account for the observed decline in photosynthesis. Findings suggested that O3-induced shifts in stomatal conductance result from ‘direct’ effects on the stomatal complex as well as ‘indirect effects’ mediated through changes in intercellular CO2 concentration. Leaves on the same plant exposed to equivalent levels of O3 showed striking differences in their response to the pollutant.
Photosynthesis has long been known to be reduced by O3 and the effects play a key role in determining the adverse effects of the pollutant on the productivity of sensitive taxa (Heath, 1994; Pell et al., 1997; Farage & Long, 1999). However, many aspects related to the depression of photosynthesis by O3 remain poorly understood (Pell et al., 1994a).
It has been established that O3 commonly causes a concomitant decline in stomatal conductance (gs) and the rate of CO2 assimilation (A). However, few studies conducted under environmentally relevant conditions have questioned whether the shifts in gs in plants exposed to O3 are the product of effects of the pollutant on the stomatal complex or effects on A (Heath, 1994). Even fewer studies have examined stomatal vs non-stomatal limitations of photosynthesis (Farage & Long, 1995) and the timing of effects in relation to plant age/prior O3 history (Krupa & Manning, 1988). Efforts to understand the physiological basis of the decline in CO2 assimilation induced by O3 have focused on changes in the amount and activity of ribulose 1,5-bisphosphate carboxylase/oxygenase (Rubisco EC 22.214.171.124) (Farage & Long, 1992; Pell et al., 1994b). Indeed, O3 exposure has been shown to decrease both the amount and activity of Rubisco, effects that are accompanied by a decrease in mRNA transcript abundance for both the large (LSU) and small (SSU) subunits of the enzyme (Pell et al., 1994b; Pell et al., 1997). Rather less attention has been paid to the impacts of O3 on photosynthetic limitations imposed by (i) the capacity to regenerate Ribulose-1,5-bisphosphate (RuBP), dependent on the activity of the enzymes associated with the interconversion of 3-, 4-, 5- and 6-carbon intermediates in the Calvin cycle and the potential of the thylakoid reactions to supply ATP and NADPH (Sharkey, 1985), and (ii) end-product feedback inhibition which may arise as a consequence of the direct repression of photosynthetic genes by soluble carbohydrates (Sheen, 1994; Krapp & Stitt, 1995) and/or indirectly via limitations imposed by the utilization of triose phosphates, and subsequent regeneration of Pi, in the synthesis of starch and sucrose (Herold, 1980).
Ozone can directly affect stomatal aperture and impair stomatal performance (Robinson et al. 1998) – effects that may lead to reductions in the dose of the pollutant absorbed by plant tissues and afford a mechanism to restrict pollutant uptake. The extent to which O3-induced shifts in stomatal conductance contribute to the decline in CO2 assimilation remain uncertain. Some studies indicate that the shift in gs induced by O3 is the cause of the reduction in CO2 assimilation (Moldau et al., 1990, 1993; Kull et al., 1996), while others have shown that the shift in gs is a downstream consequence of the increase in the intercellular CO2 concentration (ci) resulting from the effects of O3 on photosynthetic metabolism (Farage et al., 1991; McKee et al. 1995). The reasons for these discrepancies are unclear, and it remains to be established whether there are genuine differences in responses between taxa or, whether these incongruities arise as a result of the diverse range of experimental methods and approaches that have been employed.
The present study was undertaken on an ‘O3-sensitive’ population of Plantago major. Previous work has revealed considerable variation in the O3 resistance of discrete populations of this species across Europe (Davison & Barnes, 1998) – a phenomenon related to the extent of the reduction in stomatal conductance induced by O3 (Reiling & Davison, 1995) and which is influenced by plant age/prior O3 history (Lyons & Barnes, 1998). It is not clear at this stage whether the observed differences in stomatal response contribute to the differences in O3 sensitivity between populations or whether the stomata respond to an increase in ci as a result of a decline in the capacity for CO2 fixation. The objective of the present study was therefore to: probe the relative limitations on A imposed by changes in stomatal conductance, Rubisco activity, RuBP regeneration capacity and triose phosphate utilization; test Reiling and Davison’s (Reiling & Davison, 1995) theory that the decrease in gs induced by O3 in Plantago major is caused by direct effects on the stomatal complex, rather than a down-stream consequence of the changes in ci that accompanies the inhibition of photosynthetic metabolism; and examine the effects of O3 on leaf gas exchange on plants at different developmental stages with contrasting O3 histories.
Materials and Methods
Plant culture and fumigation
Seed of Plantago major L. ‘Valsain’ was germinated in a propagator containing a standard potting compost (John Innes no. 2) in a controlled environment chamber ventilated with charcoal/Purafil®-filtered air (‘clean air’ < 5 nmol O3 mol−1 dry air). The propagator lid was removed following germination, and 7-d-old-seedlings (i.e. at the three-leaf stage) transplanted individually into 2.5 cm2 plugs of 10 × 7 modules containing the same compost. Seedlings were allowed 24 h to recover after transplantation, then transferred to duplicate controlled environment chambers ventilated with either ‘clean air’ (CFA) or ‘ozone’ (O3) (CFA plus 15 nmol mol−1 O3 overnight rising to a maximum between 12:00–16:00 hours of 75 nmol mol−1) (Zheng et al., 1998). Plants were transplanted into progressively larger pots (0.45 and 1.74 dm3) containing the same standard compost after 14 and 28 d, respectively, watered daily and fertilized 7, 21 and 35 d after transplantation with a medium-strength commercial nutrient solution (Phostrogen, Corwen, Clwyd, UK).
Leaf gas exchange
After 19 and 33 d exposure to CFA or O3, the newly emerging leaf borne on 2–3 plants per chamber was labelled (leaves 7 and 10, respectively) and leaf gas exchange determined at intervals following the attainment of full expansion (9 d from leaf emergence). Measurements were made using a spherical leaf cuvette (manufactured by PP Systems, Hitchin, UK), incorporated within an open gas exchange system, housed in an additional controlled environment cabinet constructed from 3 mm perspex, ventilated with ‘clean air’ (charcoal/Purafil®-filtered air) from the same air handling system supplying the fumigation chambers and illuminated by metal halide lamps (Connect Lighting Systems 250 W unit fitted with Sylvania HQI-TD 250 W/NDL lamp) providing a PPFD of 300 µmol m−2 s−1 at plant canopy height. Temperature, photoperiod, relative humidity and air velocity in the cabinet housing the leaf cuvette were maintained the same as that in the fumigation chambers. The leaf cuvette used was similar to that developed by Ireland & Long (1989), and described in detail by Long & Drake (1991). Atmospheric composition was modulated by mixing predetermined levels of compressed gases (CO2, N2 and dry CO2-free air) with the aid of mass flow controllers (Brooks Instruments B. V., Veenendaal, The Netherlands, model 5850TR MFCs/5878 controller). The gas flow to and from the cuvette was monitored with the aid of flow meters (Platon Flow Control Ltd, Basingstoke, UK). Before entering the leaf cuvette, the air was purified by passing through a charcoal/Purafil®/potassium permanganate-filter and humidified over distilled water maintained in a temperature controlled water bath to achieve a leaf-atmosphere vapour pressure deficit (VPD) of 1.3 ± 0.04 kPa. The air leaving the leaf cuvette was dried over anhydrous magnesium perchlorate before entering an infrared gas analyzer (IRGA). Absolute CO2 concentration was continuously monitored by an ADC IRGA (ADC Ltd, Hoddesdon, UK, model LCA-2), with the change in CO2 across the cuvette measured by a twin-channel LICOR IRGA (Li-Cor Inc., Lincoln, NB, USA, model 6252). Both IRGAs were calibrated against compressed mixtures of CO2 in air, cross-calibrated against gravimetrically prepared CO2 standards (Distillers MG, Reigate, Surrey; accuracy ± 2 µmol CO2 mol−1 dry air). The dewpoint of the air entering and leaving the cuvette was measured with the aid of cooled-mirrors (Protimeter Plc, Marlow, UK, model DP989M). Air and leaf surface temperatures were measured with the aid of cross-calibrated thermocouples. Measurements were made in ‘clean air’ at a leaf temperature of 24 ± 1°C, under a PPFD of 1000 µmol m−2 s−1 supplied by an integral fan-cooled 15 V/150 W Xenon lamp (Osram Xenophot HLX 64634, St Helens, Merseyside, UK). Prior experimentation revealed (i) no significant recovery in the light-saturated rate of CO2 assimilation (Asat) and stomatal conductance (gs) within 6 h of transferring plants from O3 to CFA and (ii) the level of irradiance employed was high enough to achieve Asat in these chamber-grown plants of Plantago major.
The attached leaf was sealed in the cuvette and allowed to equilibrate to the cuvette environment (30–40 min). Then, the CO2 concentration and dewpoint of the air entering and leaving the cuvette were measured at a reference CO2 concentration (Ca) of 350 µmol mol−1. Subsequent measurements were made on the same leaf following step-wise changes in the reference CO2 concentration, allowing 30–40 min for leaves to attain steady-state rates of gas exchange at each CO2 level. Reference CO2 concentrations used for the construction of A/ci response curves were: 80, 130, 165, 200, 250, 350, 700, 1000 µmol mol−1. The projected area of the leaf enclosed within the cuvette was traced and subsequently determined using a Delta-T Devices area meter (Cambridge, UK). Light-saturated rates of net CO2 assimilation (Asat), stomatal conductance (gs) and intercellular space CO2 concentration (ci) were calculated according to Von Caemmerer & Farquhar, 1981).
The light- and CO2-saturated rate of CO2 assimilation (Amax) was derived from asymptotic curves (y = a + bekx) fitted to A/ci response data according to Delgado et al. (1993); where y equals the rate of net CO2 assimilation; x equals the intercellular CO2 concentration and a represents Amax.
The relative stomatal limitation of photosynthesis (RSL) (i.e. the proportionate decrease in Asat attributable to stomata) was calculated from A/ci curves using the method of Farquhar & Sharkey (1982):
Where A0 represents the rate of CO2 assimilation when ci is 350 µmol mol−1 (i.e. the potential rate of CO2 assimilation in the absence of stomatal limitation at 350 µmol CO2 mol−1 dry air), and A represents the rate of CO2 assimilation at an ambient CO2 concentration of 350 µmol mol−1 (i.e. the actual rate of CO2 assimilation in the presence of stomatal limitation at 350 µmol CO2 mol−1 dry air).
In vivo estimates of the maximum rate of Rubisco carboxylation (Vc,max), maximum RuBP regeneration capacity mediated by light harvesting and electron transport (Jmax,RuBP) and the potential capacity of starch and sucrose synthesis to utilise triose phosphates and subsequently regenerate inorganic phosphate (Pi) for photophosphorylation (triose phosphate utilization (TPU)) were calculated by iteratively fitting curves (using nonlinear least square regression methods) to A/ci response data according to Harley & Sharkey (1991); under the assumption that (i) the transport of 4 electrons generates sufficient energy (ATP) and reductant (NADPH) for the regeneration of RuBP in the Calvin cycle (Farquhar & Von Caemmerer, 1982) (ii) CO2 assimilation is limited solely by the amount, activity and kinetic properties of Rubisco at a ci below 200 µmol mol−1 under light-saturating conditions (iii) inorganic phosphate (Pi) limits carboxylation at high ci, and (iv) the Michaelis–Menten constants for carboxylation and oxygenation (Kc and Ko, respectively), the specificity factor of Rubisco (τ) are similar in all C3 species, and unaffected by exposure to O3. Jordan & Ogren’s (1984) measured values for Kc, Ko and τ were employed in calculations, correcting for temperature- and pressure-dependencies according to Harley et al. (1992). The efficiency of light energy conversion (α) was assumed to be constant at 0.24 mol electrons mol−1 photons – based on an average quantum use efficiency of 0.073 mol CO2 mol−1 photons absorbed (Ehleringer & Bjorkman, 1977) and a leaf absorptance of 83% (Ehleringer & Pearcy, 1983). The rate of nonphotorespiratory CO2 evolution in the light (Rd) and the CO2 compensation point in the absence of Rd (Γ*) were derived according to Brooks & Farquhar (1985), from measurements made on three independent CFA and O3-treated plants of Plantago major‘Valsain’ after 32 d exposure in duplicate controlled environment chambers. Measurements were made employing the same gas exchange system used to undertake A/ci analyses. Rates of CO2 assimilation were measured over a range of ci’s close to the compensation point (15, 30, 60, 80 µmol mol−1) at contrasting PPFDs (150, 400 and 600 µmol quanta m−2 s−1) at a leaf temperature of 24°C. The response of A was found to be linear over this range of ci and regressions were fitted to data obtained at each level of irradiance. The co-ordinates at which the regression lines intersected each other were used to estimate both Rd and Γ*. This procedure provided values for CFA and O3-exposed plants of −0.033 and −0.17 µmol m−2 s−1 for Rd and 42.6 and 44.6 µmol mol−1 for Γ*, respectively (see Fig. 1). Further details are given by Zheng (1998) regarding the simplifying assumptions that were necessary, the estimation of Rd and Γ* and the manner in which the latter values were employed in the analysis of A/ci response curves.
Measurement of in vitro Rubisco activity
Measurements of in vitro Rubisco activity were undertaken on fully expanded leaves over the course of a single day, 23 d after the emergence of leaves 7 and 10. Measurements were performed using an NADH-linked coupled spectrophotometric assay (Ward & Keys, 1989), based on the original method developed by Lilley & Walker (1974). Optimized assays of this type have been shown to yield data comparable to radioisotopic measurements of Rubisco activity (Keys & Parry, 1990; Reid et al., 1997). Assays were performed on leaves detached from four independent plants per treatment (two plants per chamber). Leaves were harvested individually at regular periods during the day and their area/fresh weight recorded. Crude extracts were rapidly prepared by homogenising c. 200 mg fresh leaf tissue in 6 ml of ice-cold extraction buffer containing 100 mM HEPES-KOH [pH 7.5], 15 mM MgCl2, 5 mM EGTA, 15% (w/w) polyethylene glycol [PEG 20 000] and 14 mM mercaptoethanol; adding 100% (w/w) insoluble polyvinylpolypyrrolidone [PVPP] immediately before grinding. The extracts were centrifuged at 12 000 g for 1 min at 4°C, then the supernatant immediately decanted into fresh 1.5 ml Eppendorf tubes.
Initial Rubisco activity (i.e. the maximum activity measured as near as possible to the in vivo state of enzyme activation) was assayed at 25°C immediately following the preparation of crude extracts in a reaction mixture containing 150 mM Bicine (pH 8.0), 25 mM NaHCO3, 20 mM MgCl2, 3.5 mM ATP, 0.25 mM NADH, 5 mM phosphocreatine, 80 nkat1 glyceraldehyde-3-phosphate dehydrogenase (EC 126.96.36.199), 80 nkat 3-phosphoglyceric phosphokinase (EC 188.8.131.52), 80 nkat creatine phosphokinase (EC 184.108.40.206), plus 50 µl extract. NADH oxidation was initiated in the spectrophotometer by the addition of RuBP (to provide a final RuBP concentration of 0.5 mM). Total Rubisco activity (i.e. the maximum activity measured following the optimal achievable activation of the enzyme with Mg2+ and CO2) was assayed following 15 min incubation of the reaction mixture minus the three enzymes and RuBP at 25°C. Absorbance changes were recorded at 340 nm using an automated UV/Vis spectrophotometer (Unicam-Philips SP8700, Pye-Unicam Ltd, Cambridge, UK). Each assay was performed in duplicate and was corrected for independent controls run for every sample, to which RuBP was not added. The reaction was linear over 5 min and enzyme rates were found to be directly proportional to the amount of extract added to the reaction mixture. Rubisco activity was calculated from the change in absorbance over the first minute, based on a reaction stoichiometry of 2 : 1 (NADH : CO2). The activation state of Rubisco was estimated by expressing the initial activity as a percentage of the total activity. Data were expressed on a leaf area basis to enable comparisons with in vivo estimates of Vc,max.
Modulated chlorophyll fluorescence measurements were made 8 h into the photoperiod at weekly intervals over the first five weeks of the life-span of leaves 7 and 10. Measurements were made using a PAM-2000 fluorimeter (Walz, Effeltrich, Germany) with the minimum level of fluorescence (Fo) obtained under modulated red light (2 µmol m−2 s−1, frequency 20 kHz) and maximal fluorescence yields (Fm and Fm′) recorded following exposure to a saturating light pulse (0.8 s) of 8000 µmol m−2 s−1, provided by an 8-V/20 W halogen lamp (Bellaphot, Osram). Fluorescence signals were analysed according to Genty et al. (1989): the relative quantum efficiency of PSII photochemistry (ΦPSII=[Fm′– Fs]/Fm′) was measured in vivo under growth chamber conditions, while the maximum (or potential) quantum efficiency of PSII photochemistry (Fv/Fm = [Fm − Fo]/Fm) was measured after a 40-min period of dark-adaptation.
Leaf surface morphology
Cured and rapid-set dental wash material (Wright Health Group Ltd, Dundee, Scotland) was applied to the abaxial and adaxial surfaces of leaves 7 and 10 (9 d after leaf emergence) borne on 8 independent plants per treatment (4 plants per chamber). When set (c. 2 min), the wash material was peeled from the leaf surface, and cellulose acetate (CA) impressions made from the template. Randomly selected areas (7 × 0.1 mm2) were examined on each CA ‘film’ using a 10 × 40 light microscope. The number of stomata and epidermal cells and the aperture of three stomata in each field of view were recorded. This enabled the estimation of stomatal and epidermal cell densities (numbers per mm2 leaf surface); stomatal index (number of stomates expressed in terms of the number of epidermal cells per unit leaf area); stomatal aperture (pore size), and percentage pore area (the percentage of total leaf surface area occupied by stomatal aperture).
Statistical analyses were performed using SPSS (SPSS Inc., Chicago, Il, USA). Data were first checked for normal distribution and homogeneity of variation, then the influence of chamber on measured variables determined by multivariate analysis of variance (MANOVA). No significant chamber-to-chamber variation was found within treatments, so data were reanalysed using a reduced ANOVA model – under the assumption that plants in replicate chambers were as likely to be as similar, or as different from, plants within an individual chamber. Time-course data were subject to RM-ANOVA, then data at each of the individual measurement dates were reanalysed and significant differences between treatments were determined using one-way ANOVA.
Plants exposed to 15 nmol mol−1 O3 overnight rising to a maximum between 12 : 00 and 16 : 00 hours of 75 nmol mol−1 over a period of 88 d (cumulative AOT40 = 14960 nmol mol−1 h) developed no visible symptoms of foliar injury and no signs of O3-induced senescence during the measurement period.
Leaf gas exchange
Gas exchange measurements made 9, 23 and 45 d after leaf emergence at a c. of 350 µmol CO2 mol−1 dry air (i.e. the CO2 concentration at which plants were grown) revealed markedly different effects on leaves of the same plant exposed to equivalent O3 concentrations over their life-span (Fig. 2). Ozone reduced Amax (P < 0.001) and Asat,350 (P < 0.0001) in leaf 7, but there were no significant effects on leaf 10. Although the magnitude of the O3 effect on leaf 7 appeared to increase with leaf age, statistical analysis indicated that the O3 × leaf age interaction was not significant at the 5% level. Both leaves showed a significant (P < 0.0001) decline in Asat,350 with leaf age. Stomatal conductance, on the other hand, was found to decline significantly (P < 0.05) following exposure to O3 in both leaves; reductions in gs at growth CO2 concentrations averaging 15% and 10% for leaves 7 and 10, respectively. Effects of O3 on A and gs were associated with a significant (P < 0.004) increase in ci (+8 µmol mol−1 averaged across measurement dates) in leaf 7, but there was no significant change in ci in leaf 10.
Figure 3 shows A/ci response curves constructed 9, 23 and 45 d after the emergence of leaves 7 and 10 on plants exposed to CFA. Asat decreased in parallel with gs (Asat= 0.07gs– 4.2, r2 = 0.3607, P < 0.0001), but no significant shifts in RSL were observed in O3-treated plants over the first 5 wks in the life-span of leaves 7 and 10 (Fig. 4). These results imply therefore that the inhibition of CO2 assimilation by O3 was primarily caused by increased mesophyll limitations to photosynthesis. The relationship between A and ci was used to analyse those steps in the photosynthetic processes which are affected by O3 under light-saturating conditions. Vc,max (P < 0.0001), Jmax,RuBP (P < 0.001) and TPU (P < 0.001) were reduced significantly by O3 in leaf 7 (Fig. 4) and although there was some suggestion that the effects of O3 increased with leaf age, statistical analyses revealed no significant O3 × leaf age interaction (O3 × leaf age P > 0.05). No significant changes in Vc,max, Jmax,RuBP and TPU were observed in leaf 10 exposed to equivalent O3 concentrations at the same stages of leaf development (see Fig. 4). Regression of Vc,max against Jmax,RuBP revealed a strong linear relationship between these parameters across the entire dataset (Fig. 5).
Measurement of in vitro Rubisco activity
Figure 6 shows the effects of O3 on Rubisco activity determined immediately following extraction (initial) and after maximum achievable activation of the enzyme by Mg2+ and CO2 (total), 23 d after the emergence of leaves 7 and 10. No significant effects of O3 on initial and total Rubisco activity were evident 6 h into the photoperiod, following exposure to 15 nmol mol−1 O3 overnight. However, 3 h into the daily exposure to elevated O3 (a maximum between 12 : 00 and 16 : 00 hours of 75 nmol mol−1) initial Rubisco activity was reduced by 27% (P < 0.05) in leaf 7 and this effect persisted throughout the rest of the day, with no significant recovery in Rubisco activity when the concentration in the chambers returned to night-time levels late in the day (16 : 00–20 : 00). The reduction in Rubisco activity could not be overcome by fully activating the enzyme and changes in total activity mirrored the effects on initial activity (see Fig. 6). No significant effects of O3 on initial and total Rubisco activity were observed over the course of the day in leaf 10. Assays indicated an activation state of 49 ± 1.9% for Rubisco, and O3 exposure resulted in no significant change in the degree of enzyme activation.
In vitro measurements of initial activity (Fig. 6) showed relatively good agreement with in vivo estimates of Vc,max derived from A/ci response curves (Fig. 4) given the fact that data were recorded on different plants at the same stage of leaf development. However, in vitro estimates of initial Rubisco activity were consistently lower than estimates of Vc,max (20–30% lower on average). The reason for this is unclear. Light microscopy and low temperature scanning electron microscopy revealed no evidence to support an increase in stomatal heterogeneity or stomatal patchiness following the O3 treatment (Zheng, 1998). Thus, we believe the discrepancy between these datasets, which essentially measure the same parameter, to arise from differences in leaf development rates between experiments conducted under similar conditions, inhibition of enzyme activity by phenolic substances during the preparation of crude enzyme extracts and/or incomplete extraction of Rubisco during the homogenisation of material for in vitro assay.
Chlorophyll fluorescence measurements tracked over the first 5 wk in the life-span of leaves 7 and 10 revealed no significant effects of O3 on the relative quantum efficiency of photosystem II (PSII) photochemistry (ΦPSII) measured under growth conditions or the maximum quantum efficiency of PSII photochemistry (Fv/Fm) (see Fig. 7). Both parameters were observed to decline significantly (P < 0.001) as leaves aged, ΦPSII declining from 0.78 to 0.71 and Fv/Fm from 0.84 to 0.76, over a 5-wk period.
Leaf surface morphology
Measurements derived from epidermal impressions 9 d after the emergence of leaves 7 and 10 are shown in Table 1. Statistical analysis revealed that O3 significantly (P < 0.001) increased stomatal and epidermal cell densities on the adaxial and abaxial surface of leaf 7, though the effect was dominated by changes on the adaxial surface (O3* surface P < 0.01). Furthermore, O3-induced changes in stomatal density exceeded those in epidermal cell density resulting in a significant (P < 0.001) increase in the stomatal index on the adaxial surface of leaf 7. There was again evidence of contrasting effects dependent upon leaf surface (O3 × surface P < 0.05) and stage of plant development (O3 × plant age P < 0.001). Ozone decreased (P < 0.001) average stomatal aperture on the adaxial (29–51%) and abaxial (9–68%) leaf surfaces and reduced (P < 0.001) percentage pore area on both leaf surfaces.
Table 1. Impact of O3 on stomatal and epidermal cell densities, stomatal index (number of stomata relative to the number of epidermal cells on per unit leaf area), stomatal aperture and stomatal pore area as a percentage of total leaf area of the 7th and 10th leaves of Plantago major‘Valsain’
Epidermal cells mm−2
Stomatal aperture (nm2)
Values represent the mean of 56 measurements (8 leaves and 7 measurements per leaf) made on the youngest fully expanded leaf after 28 and 42 d exposure to either charcoal/Flurafil-filtered air (CFA, < 5 nmol mol−1 O3) or O3 (< 15 nmol mol−1 overnight rising to a maximum between 12 : 00–16 : 00 hours of 75 nmol mol−1). Main effects: *, P < 0.05; **, P < 0.01; ***, P < 0.001; ns, not significant. Within columns, means bearing the same superscript are not significantly different at the 5% level.
This study confirmed previous observations (Reiling & Davison, 1995; Zheng, Lyons & Barnes, 2000) that environmentally-relevant O3 concentrations inhibit photosynthesis and reduce stomatal conductance in sensitive lines of Plantago major in the absence of visible symptoms of foliar injury. Furthermore, substantial differences in response to similar levels of O3 were found between leaves on the same plant.
The maximum capacity for RuBP regeneration (Jmax,RuBP), modelled from A/ci response curves, was found to decline in parallel with the change in Vc,max in leaf 7 following exposure to O3 (Fig. 4). Moreover, there was a linear relationship between Vc,max and Jmax,RuBP (Fig. 5) across treatments suggesting that plants preserve a close functional balance in the allocation of resources between these processes. Similar observations of a tight coupling between Vc,max and Jmax,RuBP have been made in a range of species across a range of experimental conditions and have been interpreted to reflect adjustments within the photosynthetic apparatus facilitating the optimal utilization of available resources, especially nitrogen (Wullschleger, 1993). Since O3-induced reductions in Jmax,RuBP, were found in the absence of equivalent changes in ΦPSII and Fv/Fm (Fig. 7), it was concluded that the maximum rate of RuBP regeneration was not limited by effects of O3 on PSII photochemistry. This finding corroborates the conclusions drawn by previous authors who have investigated the effects of O3 on PSII activity (Farage et al., 1991; Baker et al. 1994; Farage & Long, 1995). Effects on Jmax,RuBP possibly reflected an inability to regenerate RuBP via the Calvin cycle following O3-treatment, as a result of the reduced activities of regenerative Calvin cycle enzymes. These enzymes are believed to be under a high degree of self-regulation (Geiger & Servaites, 1994) and a reduction in RuBP regeneration capacity might be expected in situations where there is a marked decline in Vc,max (Long & Drake, 1992). Effects on Jmax,RuBP similar to those induced by O3 in the present study have been reported following the exposure of leaves to damaging levels of ultraviolet-B radiation (Allen et al., 1997; Baker et al., 1997).
Modelled data suggested that O3 may also reduce triose phosphate utilization resulting in enhanced limitation of photosynthesis via the reduced availability of inorganic phosphate (Pi). A/ci measurements performed on leaf 7 in 21% and 2% oxygen revealed no significant increase in Asat on switching between O2 concentrations (data not shown) lending support to this conclusion (Sharkey et al., 1986). Moreover, O3 exposure resulted in the enhanced accumulation of nonstructural carbohydrates (mainly starch) in leaf 7, but not in leaf 10 (Zheng et al., 2000) – a situation that could result in top-down or bottom-up repression of Rubisco activity and other photosynthetic components in order to retain the functional balance between rates of CO2 fixation and carbohydrate utilization (Goldschmidt & Huber, 1992).
The O3-induced reduction in Asat in leaf 7 was accompanied by a parallel decline in conductance (15% lower on average). However, only minor changes in RSL were observed (Fig. 4). This indicates that the decline in gs was not responsible for the decline of Asat, an argument strongly supported by the absence of any change in Asat in leaf 10 despite significantly lower stomatal conductance in O3-treated plants (Fig. 2). The extent of the decrease in gs induced by O3 was greater in leaf 7 than leaf 10, consistent with the view that changes in gs were, at least in part, a response to the increase in ci induced via the negative effects of O3 on photosynthetic metabolism. However, similar effects could not explain the O3-induced decline in gs observed in leaf 10. Light microscopy revealed that the reduction in gs induced by O3 in both leaves was due to a decline in stomatal aperture, and not to a reduction in stomatal density (Table 1), so the data for leaf 10 suggest that ‘direct’ effects of O3 on stomatal aperture may have contributed to the O3-induced decline in gs observed in the present study. This conclusion is supported by other reports of O3-induced changes in gs independent of effects on the capacity for CO2 assimilation (Kleier et al. 1998; Torsethaugen et al. 1999). Indeed, recent studies indicate that gs may rise or fall in response to O3 damage, dependent on genotype, O3 concentration, and the sensitivity of stomatal guard cells to O3 relative to that of surrounding epidermal cells (Robinson et al., 1998). Improved understanding of the timing and mechanisms underlying stomatal closure in response to O3 is vital if the role that changes in stomatal aperture play in the avoidance of pollutant uptake is to be elucidated. Stomatal closure may afford little or no useful protection if, as it appears, the stomata close predominantly as a consequence of, rather than before, detrimental effects on leaf metabolism.
Marked differences were observed in the effects of O3 on photosynthetic metabolism between leaves on the same plant, despite the fact that leaves were exposed to equivalent O3 exposures at similar stages in their development. This finding is consistent with growth studies on Plantago major which have revealed a marked reduction in the impacts of O3 with plant age (Lyons & Barnes, 1998; Zheng, 1998), with similar shifts in O3 resistance reported for other species (Soja et al., 2000). The mechanisms underlying these shifts in resistance, and whether they are triggered by the prior exposure to O3 (i.e. ‘acclimation’) or are associated with plant developmental status, remain to be elucidated. Measurements of stomatal conductance (see Fig. 2) suggested that the contrasting effects of O3 on leaf 7 and 10 were not mediated by differences in O3 uptake – a finding consistent with the view that shifts in O3 resistance triggered by prior exposure to O3 and/or plant ontogeny are linked to changes in the tolerance of tissues following uptake (Schraudner et al., 1998). Indeed, the reduced impacts of O3 on leaf 10 vs leaf 7 are consistent with the systemic induction of cellular oxidative defence and repair systems under O3 stress (see Rao, Koch & Davis, 2000).
It is concluded that a loss of Rubisco protein constitutes the primary cause of the O3-induced decline in CO2 assimilation in Plantago major. The effect would appear to be accompanied by the down-regulation of the activity of other Calvin cycle enzymes and a possible increase in Pi limitation. Changes in the quantum efficiency of PSII were not involved with the O3-induced inhibition of CO2 assimilation. Findings relating to the effects of O3 on the photosynthetic physiology of other species show the same basic pattern of effects (Farage & Long, 1995). It remains to be established whether such shifts in photosynthetic metabolism are driven by ‘direct’ effects of O3 (or it’s reactive dissolution products) on Rubisco or whether they arise as a consequence of the enhanced accumulation of nonstructural carbohydrates in foliage. Although stomatal conductance declined in parallel with CO2 assimilation in O3-treated plants this did not account for the observed decline in photosynthesis. The change in stomatal conductance was found to be due to a reduction in stomatal aperture, rather than an effect on the number of stomates per unit leaf area, with data suggesting evidence of ‘direct’ effects of the pollutant on the stomatal complex as well as effects mediated through shifts in ci. Leaves on the same plant exposed to equivalent levels of O3 showed striking differences in the impacts of the pollutant on photosynthetic capacity with the pattern of effects consistent with the induction of defence and repair systems following O3-induced oxidative challenge.
We express our sincere thanks to Tom Lyons (Newcastle University), Kate Maxwell (Cambridge University), Ian McKee (Essex University), Chris Jeffrey (Edinburgh University) and Fred Last (retired) for assistance during the course of this study. The authors are indebted to The Royal Society, The Swales Trust and the British Overseas Development Administration for financing the study.