•The effects of zinc (Zn) toxicity on photosynthesis and respiration were investigated in sugar beet (Beta vulgaris) plants grown hydroponically with 1.2, 100 and 300 μM Zn.
•A photosynthesis limitation analysis was used to assess the stomatal, mesophyll, photochemical and biochemical contributions to the reduced photosynthesis observed under Zn toxicity.
•The main limitation to photosynthesis was attributable to stomata, with stomatal conductances decreasing by 76% under Zn excess and stomata being unable to respond to physiological and chemical stimuli. The effects of excess Zn on photochemistry were minor. Scanning electron microscopy showed morphological changes in stomata and mesophyll tissue. Stomatal size and density were smaller, and stomatal slits were sealed in plants grown under high Zn. Moreover, the mesophyll conductance to CO2 decreased by 48% under Zn excess, despite a marked increase in carbonic anhydrase activity. Respiration, including that through both cytochrome and alternative pathways, was doubled by high Zn.
•It can be concluded that, in sugar beet plants grown in the presence of excess Zn, photosynthesis is impaired due to a depletion of CO2 at the Rubisco carboxylation site, as a consequence of major decreases in stomatal and mesophyll conductances to CO2.
Respiratory metabolism also interacts with photosynthesis, and respiratory mutants (especially those with altered alternative oxidase (AOX) activity) show decreased photosynthesis, as well as decreased gs and gm (reviewed by Flexas et al., 2008). The effects of Zn excess on respiration are a matter of controversy. On the one hand, in vitro studies have suggested that Zn inhibits the activity of the mitochondrial cytochrome bc1 complex, probably by interfering with the enzyme’s reaction with ubiquinol at the UQH2 binding niche that oxidizes UQH2, the quinol oxidase (Qo) site (Link & von Jagow, 1995), as well as the activity of AOX (Affourtit & Moore, 2004). On the other hand, in vivo studies have shown that respiration is enhanced progressively with increasing Zn doses (Ismail & Azooz, 2005), and use of inhibitors of cytochrome bc1 and AOX has suggested that there is a specific increase in AOX activity (Webster & Gadd, 1999), although, to the best of our knowledge, no report on the in vivo activity of AOX under Zn excess has been published. Regardless of the mechanisms involved, Zn-induced alterations in respiration would result from metabolic effects that could potentially reduce net photosynthesis.
The effects of Zn excess on the photochemistry and biochemistry of photosynthesis are also a matter of controversy. In some studies, the quantum yield of photosystem II (PSII)-related electron transport, estimated at low photosynthetic photon flux density (PPFD), was not affected in Zn-treated plants (Van Assche & Clijsters, 1986a), in agreement with studies showing only small decreases (from 0.82 to 0.70) of the maximum potential PSII efficiency (Fv/Fm) under Zn excess (Schuerger et al., 2003; Dhir et al., 2008; Sagardoy et al., 2009). However, other studies have shown an inhibition of thylakoid electron transport rates (ETRs) by high Zn concentrations (Kim & Jung, 1993). Specific effects on PSII photochemistry, related to competitive substitution of manganese (Mn) by Zn at the site of water photolysis inhibiting photosynthetic electron transport and oxygen evolution, have also been reported (Van Assche & Clijsters, 1986a; Ralph & Burchett, 1998). Consistent with this, Zn excess was found to decrease PSII efficiency (ΦPSII) and noncyclic photophosphorylation (Van Assche & Clijsters, 1986a; Bonnet et al., 2000; Schuerger et al., 2003; Sagardoy et al., 2009), although, alternatively, ΦPSII could be limited by excess Zn-mediated reductions in dark photosynthetic processes. Other studies have shown no changes in the ADP : ATP and NADP+ : NADPH ratios with changes in Zn concentration (Dhir et al., 2008). In some studies, the decrease in Fv/Fm was associated with increased levels of minimal Chl fluorescence in the dark (Fo) (Vaillant et al., 2005), suggesting PSII photoinactivation or photooxidation, consistent with the observation that lipid peroxidation was found to be enhanced in plants treated with Zn excess (Chaoui et al., 1997). The activity of antioxidant enzymes such as Mn-superoxide dismutase (Mn-SOD) and ascorbate peroxidase (APX) increased in plants exposed to high concentrations of Zn (del Río et al., 1985; Chaoui et al., 1997; Bonnet et al., 2000). Catalase (CAT) activity was decreased in high-Zn plants (Chaoui et al., 1997; and references therein), which may suggest decreased photorespiratory activity. However, photosynthetic ETR : AN ratios increased under Zn excess (from c. 9 in controls to 31–77; recalculated from Monnet et al., 2001), which is probably indicative of increased electron consumption diverted to photorespiration or to alternative processes.
Sugar beet (Beta vulgaris L. ‘Orbis’) seeds were germinated and grown in vermiculite for 3 wk, in a growth chamber at a constant temperature (25°C) and a PPFD of 600 μmol m−2 s−1 with a 16 h : 8 h light/dark regime. Plants were then moved to a glasshouse, and transplanted into 20-l plastic buckets (four plants per bucket) containing half-strength Hoagland nutrient solution (Terry, 1980) with 45 μm Fe(III)-EDTA. Once plants were established, experiments with different concentrations of Zn were initiated. Zn (ZnSO4) concentrations were 1.20 μM (control), 100 μM and 300 μM. Treatments used were based on a previous study (Sagardoy et al., 2009). Half-strength Hoagland’s solution was added, when necessary, several times throughout the experiment. Plants were used for measurements 10–14 d after the treatments were imposed. Measurements were performed in young, fully expanded leaves, 3 to 4 h into the light period.
At the end of the experiment, 14 d after the start of the treatments, plants were collected and separated into roots and shoots. Fresh weights (FWs) and dry weights (DWs) were determined for each fraction, and water content (WC) was calculated as (FW − DW)/FW. Leaf areas were measured with an AM-100 leaf area meter (ADC, Herts, UK). Eight plants per treatment were used.
For nutrient analysis, plant tissues were washed with pure water and dried in an oven at 60°C for 76 h to constant weight. Samples were then dry-ashed and dissolved in HNO3 and HCl following the Association of Official Analytical Chemists (AOAC) procedure (Association of Official Analytical Chemists, Washington, DC, USA); Zn was determined by flame atomic absorption spectroscopy (FAAS) (Igartua et al., 2000).
Gas exchange and chlorophyll fluorescence measurements
Stomatal conductance (gs) was estimated nondestructively with a portable leaf porometer (SC-1; Decagon Devices, Pullman, WA, USA). Porometry was used for rapid assessment of stomatal status throughout the experimental period, in order to ensure all experiments were performed under similar initial experimental conditions.
Leaf gas exchange and Chl fluorescence were measured simultaneously using an open gas exchange system (Li-6400; Li-Cor, Inc., Lincoln, NE, USA) with an integrated Chl fluorescence chamber head (Li-6400-40 leaf chamber fluorometer; Li-Cor, Inc.) according to Flexas et al. (2007). These measurements were typically carried out on days 10–13 after the start of the treatments. All measurements were performed at 25°C and 1500 μmol photons m−2 s−1 (10% blue light). The reference CO2 concentration (Ca) was set at 400 μmol CO2 mol−1 air, and vapor pressure deficit (VPD) was kept at 2.0 ± 0.2 kPa.
In addition to net photosynthesis (AN) and gs, the substomatal CO2 concentration (Ci) was calculated. The possible occurrence of Ci overestimation was evaluated in preliminary experiments in all treatments as follows. Average gs values were found to be much higher (data not shown) than those typically causing cuticular-associated Ci overestimations (Boyer et al., 1997; Flexas et al., 2009). Also, leaf Chl fluorescence images (obtained using a FluorCAM; PSI Instruments, Brno, Czech Republic; data not shown) demonstrated that patchiness did not occur, indicating that heterogeneous stomata closure did not cause errors in the calculation of Ci (Terashima, 1992).
The incorporated fluorometer allowed determination of the actual PSII efficiency (ΦPSII = (F ′m − F ′)/F ′m, where F ′ and F ′m are the steady-state and maximum Chl fluorescences, respectively) (Genty et al., 1989). F ′m was determined with a light-saturating pulse of c. 8000 μmol photons m−2 s−1.
The ETR was then calculated as ΦPSII × PPFD × α. In this equation, α (a term that includes the product of leaf absorbance and the partitioning of absorbed quanta between PSI and PSII) was determined for each treatment as the slope of the relationship between ΦPSII and ΦCO2 (i.e. the quantum efficiency of gross CO2 fixation), obtained by varying the light intensity under nonphotorespiratory conditions in an atmosphere containing < 1% O2 (Valentini et al., 1995). Three light curves per Zn treatment were recorded to determine α.
Mesophyll conductance (gm) was estimated from combined gas exchange and Chl fluorescence measurements (Harley et al., 1992) as AN/(Ci − (Γ*(Jflu + 8(AN + Rd))/(Jflu − 4(AN + Rd))) by the variable J method. In this equation, AN and Ci were obtained from gas exchange measurements at saturating light, whereas Γ* was taken after Bernacchi et al. (2002), and dark respiration was used as a proxy for Rd (Pinelli & Loreto, 2003). Jflu is the electron transport rate calculated from fluorescence measurements, Rd is the mitochondrial respiration in the light, and Γ* is the CO2 compensation concentration in the absence of mitochondrial respiration.
Calculated gm values were used to convert AN–Ci into AN–CC curves (where CC is the chloroplastic CO2 concentration) using the equation CC = Ci − (AN/gm). The maximum carboxylation and ETR capacities (Vc,max and Jmax, respectively) were calculated from the AN–CC curves, using Rubisco constants and the temperature dependence of Rubisco kinetic parameters described on a CC basis by Bernacchi et al. (2002). The Farquhar model was fitted to the data by applying iterative curve-fitting (minimum least-square difference) using Microsoft Excel’s Solver tool.
Photosynthesis limitation analysis
Relative photosynthetic limitations were partitioned into their functional components, following the method of Wilson et al. (2000), modified by Grassi & Magnani (2005) to take into account gm. This approach requires measuring AN, gs, gm and Vc,max, and it allows for the partitioning of photosynthesis limitations into different components related to stomatal (SL) and mesophyll conductance (MCL) and leaf biochemical characteristics (BL), compared with a hypothetical reference state when gs, gm and Vc,max are at their maximum. This approach makes it possible to compare absolute or relative limitations to assimilation over any period of time, assuming that a reference maximum assimilation rate can be defined as a standard (see Grassi & Magnani (2005) for further details and formulae underlying this rationale). In our case, control plants were fixed as reference maximum values, and their limitations were set to 0 (see Table 3). Diffusional (DL) and nonstomatal (NSL) limitations are taken as the sum of SL and MCL, and of MCL and BL, respectively.
An isotope ratio mass spectrometer (Delta Plus XP; Thermo LCC, Bremen, Germany) with a dual-inlet system from the Serveis Científico Tècnis of the Universitat de les Illes Balears (UIB) was used for respiratory measurements. Analysis of respiration and oxygen-isotope fractionation was performed in 10-cm2 leaf discs at a controlled temperature of 25°C as in Florez-Sarasa et al. (2007). The electron partitioning through the alternative oxidase pathway (τa) was calculated as (Δn − Δc)/(Δa − Δc), where Δn, Δc, Δa are the oxygen isotope fractionation in the absence of inhibitors, in the presence of SHAM (hydroxy-salicylic acid) and in the presence of KCN, respectively. For inhibitor treatments, leaf discs were incubated in the presence of 10 mM KCN for 30 min and the Δa value obtained was 32‰. As cytochrome oxidase pathway discrimination has been shown to be remarkably constant in several species (Ribas-Carbóet al., 2005), a Δc value of 20.0‰ was used. The individual activities of the cytochrome oxidase (vcyt) and alternative oxidase (valt) pathways were obtained from the total oxygen uptake rate (Vt) and τa as Vt(1 − τa) and Vtτa, respectively.
Four discs per treatment were collected for respiratory measurements 10–12 d after Zn treatments were imposed.
Experiments on stomatal conductance responses and application of the Ball–Woodrow–Berry model
Stomatal conductance responses were studied with the Li-6400 gas exchange system using three different treatments: leaf desiccation (DES treatment); exogenous application of abscisic acid (ABA treatment); and changes in relative humidity (RH treatment). The DES treatment was carried out by cutting the leaf petiole in air and letting the leaf desiccate for ≥ 90 min under ambient conditions. Measurements were taken every 1 min, using four leaves per treatment. Exogenous application of ABA consisted of a single application of 100 μM ABA (ABA was previously dissolved in MeOH) to the plant nutrient solution. Measurements were taken in attached leaves for 4 d after the start of the treatment, using four replicates per treatment. In the RH treatment, 70%, 50% and 30% RH values were used. Measurements were carried out once gs had reached a steady-state level (typically 20–40 min after changing the RH value), using four leaves per treatment. DES and RH experiments were carried out in plants not treated with ABA.
To assess the effects of Zn stress on the stomatal conductance response to environmental conditions and its coupling to photosynthesis, the model proposed by Ball et al. (1987) was used, plotting gs vs (ANHS)/Cs, where AN is the net photosynthesis rate (in μmol CO2 m−2 s−1), Cs is the CO2 concentration at the leaf surface (μmol mol−1) and HS is the RH value (%). Data obtained from the AN–Ci curves and RH and ABA measurements (9–25 data points depending on the treatment) were used.
Xylem sap collection
In order to analyze ABA concentration, sugar beet xylem sap was obtained using leaf petioles as described elsewhere (López-Millán et al., 2000). Malate dehydrogenase (MDH, EC 18.104.22.168) was used as a cytosolic contamination marker by checking the activity in xylem sap against the corresponding activities in petiole total homogenates.
Xylem from six plants per treatment was analyzed for ABA following the procedures described in Albacete et al. (2008) on a high-performance liquid chromatography (HPLC)/mass spectrometry (MS) system (CEBAS-CSIC, Murcia, Spain).
Extracts used for the measurement of carbonic anhydrase (CA, EC 4.2.11) activity were prepared by grinding fresh leaf discs (0.95 cm2) in 1 ml of extraction buffer (50 mM HEPES-NaOH, pH 8.3, 0.5 mM EDTA, 10 mM dithiothreitol (DTT), 10% (v/v) glycerol and 1% (v/v) Triton X-100) at 4°C. Extracts were centrifuged at 2400 g for 10 min at 4°C. Supernatants were put in Eppendorf tubes and frozen at −20°C until assayed. CA activity was measured using a method adapted from that of Gillon & Yakir (2000). Assays were carried out in 7-ml flat-bottom glass vials placed on ice with continuous stirring. A volume of 4.5 ml of reaction buffer (Na-Barbital 20 mM, pH 8.3) was supplemented with 75 μl of extract, and 1.5 ml of CO2-saturated water (at 0°C) was added to start the reaction. CA activity was obtained from the reaction time of pH change from 8.3 to 7.3.
Scanning electron microscopy
Leaf samples were taken from sugar beet plants grown according to Sagardoy et al. (2009), and electron microscopy was used to study leaf surfaces (scanning electron microscopy (SEM); ICB-CSIC, Zaragoza, Spain) and cryo-fractured leaf pieces (low temperature-SEM (LT-SEM); Cryotrans CT-1500, Oxford, UK). A Hitachi S-3400 N microscope (Krefeld, Germany) was used to visualize hydrated leaf surfaces. Fresh pieces were trimmed to an adequate size, mounted on stubs and observed directly (uncoated) with an accelerating voltage of 1 kV. Also, fresh leaf sections were mounted on aluminum stubs with adhesive (Gurrs, O.C.T. B.D.H, UK; Gurr®, OCT compound, BDH, Poole, UK), cryofixed in slush nitrogen (−196°C), cryotransferred to a vacuum chamber at −180°C, and fractured using a stainless steel spike. Once inside the microscope, samples underwent superficial etching in a vacuum (at −90°C and 2 kV for 120 s), and were overlaid with gold (Au) for observation. Fractured samples were observed at low temperature using a Zeiss digital scanning microscope (DSM 960), employing secondary and back-scattered electrons (SEM-BSE).
Hydrated leaf surfaces were analyzed by measuring stomatal density and pore size on the adaxial and abaxial sides of the leaf. Stomatal density was calculated from SEM images by measuring the number of stomata in a given leaf surface. Stomatal density data are the mean ± SE of 10 (adaxial) and three (abaxial) images in each treatment. Pore size was measured manually using the Adobe Photoshop CS3 image analysis software. Pore size data are the mean ± SE of 50 (adaxial) and 25 (abaxial) stomata from the same images used in stomatal density measurements.
The spaces occupied by intercellular air spaces and substomatal cavities were quantified using the SEM images (nine images from control leaves and eight from 300 μM Zn-grown plants) and the ImageJ Image Processing and Analysis software from the Wright Cell Imaging Facility (http://www.uhnresearch.ca/facilities/wcif). Chloroplast size was measured manually in each SEM image by counting the number of pixels occupied by the axis of the chloroplast relative to the scale, using the Adobe Photoshop CS3 image analysis software (for 18 chloroplasts each in the control and high-Zn treatments, using nine and eight images, respectively).
Protoplast and chloroplast isolation and size measurements
Intact protoplasts and chloroplasts were isolated as described in González-Vallejo et al. (2000) and using the chloroplast isolation kit (Sigma), respectively, and measured using the Adobe Photoshop CS3 image analysis software. Data shown are for 40 protoplasts and 30 chloroplasts in each treatment (obtained using eight different images in each case).
High Zn concentrations in the nutrient solution reduced whole-plant DW and leaf area (Table 1), and also shoot and root DW (data not shown) in sugar beet plants grown in a glasshouse. Plants treated with high Zn contained less water than control plants (Table 1). These results agree with previously reported data for Zn-stressed sugar beet plants grown at lower PPFD in a growth chamber (Sagardoy et al., 2009).
Table 1. Growth parameters, water content (WC) and zinc (Zn) concentrations in sugar beet plants grown in a glasshouse in hydroponics with different Zn concentrations for 14 d
Data are the mean ± SE of eight replicates.
Different letters indicate significant differences (Duncan’s test) at P < 0.05.
DW (g per plant)
2.8 ± 0.4 a
1.3 ± 0.2 b
0.8 ± 0.2 b
Area per leaf (cm2)
136.7 ± 5.5 a
48.3 ± 5.2 b
29.8 ± 3.0 b
96.8 ± 0.2 a
91.1 ± 0.1 b
87.6 ± 0.3 c
Zn in roots (μg g−1 DW)
136.1 ± 15.0 a
218.4 ± 2.9 b
202.9 ± 0.1 c
Zn in shoots (μg g−1 DW)
129.7 ± 11.9 a
1223.7 ± 64.5 b
1184.3 ± 102.7 b
Zinc concentration increased significantly in tissues of plants grown with high Zn in the nutrient solution; 1.5- to 1.6-fold in roots and almost 10-fold in shoots (Table 1). Leaf Zn concentrations were markedly higher than those found in a previous growth chamber study (Sagardoy et al., 2009). This was probably a result of the reduced plant growth rates in the glasshouse (this study) as compared with growth chamber conditions (Sagardoy et al., 2009). However, total amounts of Zn extracted per plant (362, 1139 and 742 μg in the 1.2, 100 and 300 μM Zn treatments, respectively) were comparable to those found in Sagardoy et al. (2009). Zn allocation (as a percentage of total Zn, for shoot : roots) was 78 : 22%, 92 : 8% and 94 : 6% in the 1.2, 100 and 300 μM Zn treatments, respectively.
Photosynthetic parameters and respiratory measurements
Leaves of plants grown in high Zn showed 50% decreases in photosynthetic rate (AN), with no significant differences between the 100 and 300 μM Zn treatments (Table 2). However, although there was a tendency for some parameters to decrease with excess Zn, Duncan analysis showed no significant differences at P < 0.05 in Fv/Fm (P < 0.061), ETR (P < 0.056) and ΦPSII (P < 0.056) (Table 2). There were significant differences in Fv/Fm only when Tukey’s analysis (P < 0.027), a softer statistical test, was used. Stomatal conductance (gs) was reduced by 70% in excess Zn, whereas mesophyll conductance was reduced by 44%, and in both cases differences between the 100 and 300 μM Zn treatments were not significant. Similar gs changes were observed when using a leaf porometer (Fig. 1).
Table 2. Photosynthetic parameters measured with a Li-6400 gas exchange system at 400 μmol CO2 mol−1 air in sugar beet plants grown in hydroponics with different zinc (Zn) concentrations
Data are the mean ± SE of five replicates.
Different letters indicate significant differences (Duncan’s test at P < 0.05).
AN, net photosynthesis; Fv/Fm, maximum potential photosystem II efficiency; ETR, electron transport rate; ΦPSII, actual photosystem II efficiency; gs, stomatal conductance; gm, mesophyll conductance; Ci, substomatal CO2 concentration; CC, chloroplastic CO2 concentration; Vc,max, in vivo maximum rate of Rubisco carboxylation; Jmax, in vivo maximum rate of electron transport driving regeneration of RuBP.
AN (μmol CO2 m−2 s−1)
21.4 ± 1.2 a
12.4 ± 1.4 b
11.1 ± 1.8 b
0.821 ± 0.001 a
0.807 ± 0.005 a
0.802 ± 0.008 a
ETR (μmol e− m−2 s−1)
143.8 ± 3.1 a
114.0 ± 5.8 a
120.0 ± 17.3 a
0.218 ± 0.005 a
0.173 ± 0.009 a
0.182 ± 0.026 a
gs (mol CO2 m−2 s−1)
0.231 ± 0.033 a
0.070 ± 0.014 b
0.055 ± 0.010 b
gm (mol CO2 m−2 s−1)
0.389 ± 0.091 a
0.243 ± 0.055 ab
0.204 ± 0.048 b
Ci (μmol CO2 mol−1 air)
286 ± 8 a
200 ± 11 b
176 ± 17 b
CC (μmol CO2 mol−1 air)
221 ± 19 a
143 ± 14 b
115 ± 9 b
Vc,max (μmol CO2 m−2 s−1)
104.4 ± 4.5 a
102.8 ± 5.0 a
126.5 ± 19.1 a
Jmax (μmol CO2 m−2 s−1)
128.7 ± 6.7 a
122.6 ± 1.3 a
The slope of the relationship between ΦPSII and ΦCO2 (α) was found to be 0.44 in all Zn treatments (not shown). The intercept of the relationship, which indicates the amount of electrons channeled to alternative sinks (Laisk & Loreto, 1996; Long & Bernacchi, 2003), was very close to zero, being slightly negative in the controls (−0.024) and in plants grown at intermediate Zn concentrations (−0.031) and positive but of similar magnitude (0.025) in the 300 μM Zn-grown plants (not shown). As a negative ΦPSII is not possible, these differences are mere statistically nonsignificant deviations from the origin.
At the measuring CO2 concentration (400 μmol CO2 mol−1 air), Ci and CC were lower (30–38% and 35–48%, respectively) in plants grown in an excess of Zn, although no differences were found between the two high Zn concentrations (Table 2). The AN vs Ci and AN vs CC curves also showed that, in excess Zn, Ci reached values of 700–800 μmol CO2 mol−1 air, whereas CC never reached values of 400 μmol CO2 mol−1 air (Fig. 2). Analysis of the data shown in Fig. 2 indicated that there were no significant differences between treatments in slope (Vc,max); P < 0.198) and (Jmax), although the latter parameter could not be calculated in the 300 μM Zn treatment (Table 2).
The AN vs Ci curve for Zn-treated plants saturated at values below the saturation values for control plants, which would appear to reflect a biochemical limitation that is not associated with CO2 availability. However, when AN was expressed on a CC basis, CO2 availability was found to be clearly responsible for the decreased photosynthesis. Data from the limitation analysis also showed that biochemical limitations were negligible in excess Zn-grown plants (Table 3). In other words, the ‘apparent’ biochemical limitation observed in the AN vs Ci curves can be fully explained by the decreased mesophyll conductance in excess Zn-treated plants.
Table 3. Photosynthesis limitation parameters (%) in sugar beet plants grown in hydroponics with different zinc (Zn) concentrations
The control, 1.2 μM Zn treatment was taken as a reference, for which all limitations were set to 0 (see text for a full explanation). TL, total conductance; SL, stomatal conductance; MCL, mesophyll conductance; BL, biochemical limitation. Data are the mean ± SE of five replicates.
42 ± 6
48 ± 8
38 ± 5
42 ± 7
4 ± 2
5 ± 2
Similar to Grassi & Magnani, (2005) the control treatment can be defined as the actual reference where all three parameters (gs, gm and Vc,max) were at their maximum: gs and gm declined in response to Zn treatment, whereas Vc,max did not change significantly (although it tended to increase; Table 2). Therefore, the same Vc,max value was used for all treatments, so that all biochemical limitations (BL) were 0 (Table 3). Using the AN, gs, gm and Vc,max values found, stomatal conductance limitation (SL) was estimated to account for 79%–86% of the total limitation (TL), whereas nonstomatal limitations (NSL = MCL + BL) were only 14% and 21% in the 100 and 300 μM Zn-treated plants, respectively (Table 3). The implications of the fact that Vc,max values tended to increase with excess Zn merit some consideration. Although mathematically one could calculate a negative limitation, such a value would imply the alleviation of a pre-existing limitation in the control plants. But, if control plants had some limitation, then the net photosynthesis displayed by them would not be really maximal and therefore control plants could not be used as a reference. Establishing a hypothetical optimal state may be uncertain, as maximum values in all three parameters (gs, gm and Vc,max) would result in net photosynthesis values higher than any of those measured but still unknown; that is, the total limitation could not be properly stated and hence a limitation analysis would not be possible. Because of this, even when Vc,max values were (non-significantly) higher in treated than control plants, BL was considered null and not negative, as explained by Flexas et al. (2009). Nevertheless, we also performed the analysis allowing negative limitations to occur. The results were identical in 100 μM plants, whereas in 300 μM plants SL, MCL and BL become 49, 6 and −7, respectively, as compared with 42, 5 and 0 when BL was forced to 0. In other words, the main conclusion remains the same, that is, that SL was by far the most important limitation and BL did not occur in response to high Zn.
Stomatal conductance experiments
The response of stomatal conductance was studied by inducing stomatal closure, in a short-term (DES) and a long-term (ABA) treatment, and also by using different vapor pressure deficits (RH treatment). After detachment, gs increased transiently for a few minutes in control plants. This transient gs increase was smaller at 100 μM Zn and absent at 300 μM Zn (Fig. 3a). These data show the progressive inability of stomata in Zn-treated plants to respond to hydraulic stimuli. In the long-term experiments with ABA, gs decreased markedly in the controls whereas in plants grown in 300 μM Zn it was not affected (Fig. 3b). Also, when relative humidity was reduced, gs decreased in the controls but not in the two high-Zn treatments (Fig. 3c).
The Ball–Berry model (gs vs (AN.Hs)/Cs) distinguished clearly between control and 300 μM Zn-treated plants (Fig. 4a). However, when ABA was used, stomata closed in the controls and the data for controls and 300 μM Zn-treated plants had similar slopes (Fig. 4b).
Biochemical parameters and respiratory measurements
CA activity in leaves was similar in control and 100 μM Zn-treated plants, whereas in plants grown at 300 μM Zn it increased markedly (by 80%; Table 4). Xylem ABA concentrations were reduced by 70% at 100 μM Zn compared with the controls, whereas at 300 μM Zn changes were not statistically significant (Table 4).
Table 4. Carbonic anhydrase (CA) activity in leaf extracts and abscisic acid (ABA) concentration in xylem sap from sugar beet plants grown in hydroponics with different zinc (Zn) concentrations
Data are the mean ± SE of five and six replicates, respectively.
Different letters indicate significant differences (Duncan’s test) at P < 0.05.
CA (μmol CO2 m−2 s−1)
305.5 ± 12.3 a
328.8 ± 29.6 a
550.5 ± 44.7 b
ABA (ng ml−1)
111.4 ± 22.7 a
33.5 ± 5.0 b
63.6 ± 7.1 ab
Leaf respiration increased 2-fold in both high-Zn treatments when compared with the control. The increase was more marked for the cytochrome oxidase pathway (COP; 2.3-fold) than for the alternative oxidase pathway (AOP; 1.8-fold) respiration pathway (Table 5).
Table 5. Total leaf respiration (Vt) and the contribution of the cytochrome oxidase pathway (COP; vcyt) and alternative oxidase pathway (AOP; valt) in sugar beet plants grown in hydroponics with different zinc (Zn) concentrations
Data are the mean ± SE of four replicates.
Different letters indicate significant differences (Duncan’s test) at P < 0.05.
Vt (μmol O2 m−2 s−1)
0.37 ± 0.04 a
0.81 ± 0.02 b
0.80 ± 0.05 b
vcyt (μmol O2 m−2 s−1)
0.25 ± 0.03 a
0.59 ± 0.02 b
0.58 ± 0.04 b
valt (μmol O2 m−2 s−1)
0.12 ± 0.02 a
0.22 ± 0.01 b
0.23 ± 0.01 b
Scanning electron microscopy images
Leaf surface samples were scanned at ×300 and ×3000. Both the adaxial (Fig. 5c) and abaxial (Fig. 5d) epidermis of high-Zn plants had a smoother appearance than those of control plants (Fig. 5a,b). Moreover, excess Zn induced a decrease in both stomatal density and size (Table 6, Fig. 5). A closer look at stomata showed structural differences, and plants grown at 300 μM Zn (Fig. 6c,d; adaxial and abaxial sides, respectively) had rounder stomata and smaller stomatal slits than the control plants (Fig. 6a,b; adaxial and abaxial sides, respectively). In some cases, the stomatal slit of high-Zn leaves appeared to be completely sealed.
Table 6. Stomatal density and pore size in the abaxial and adaxial epidermis of hydroponically grown control and 300 μM zinc (Zn)-grown sugar beet plants
1.2 μM adaxial
1.2 μM abaxial
300 μM adaxial
300 μM abaxial
Stomatal density data are the mean ± SE of 10 (adaxial) and three (abaxial) images, and pore size data are the mean ± SE of 50 (adaxial) and 25 (abaxial) stomata from the same images used in stomatal density measurements (Fig. 5).
Different letters indicate significant differences (Duncan’s test) at P < 0.05.
Density (stomata mm−2)
218 ± 7 a
223 ± 3 a
172 ± 5 b
156 ± 12 b
Pore size (range in μm)
Pore size (mean in μm)
14.5 ± 0.3 a
14.0 ± 0.5 a
10.8 ± 0.2 b
10.1 ± 0.4 b
Mesophyll anatomical features and the arrangement of chloroplasts around the mesophyll cell plasma membrane were different in plants grown under high Zn concentrations than in the controls. LT-SEM showed a more compact mesophyll tissue in Zn-treated plants (Fig. 7c) than in the controls (Fig. 7a), with smaller cells. Isolated protoplasts were smaller in 300 μM high-Zn plants than in controls (33 ± 1 vs 46 ± 3 μm, respectively; n = 40). Also, intercellular spaces and abaxial and adaxial substomatal cavities were smaller in the high-Zn leaves. The space occupied by intercellular air spaces decreased by 31% in the highest Zn treatment, from 9.8% (controls) to 6.8% (300 μM Zn-grown plants) of the whole mesophyll, whereas that occupied by the substomatal cavities decreased by 27%, from 5.2% (controls) to 3.8% (300 μM Zn-grown plants) of the mesophyll.
Chloroplast length was reduced in high-Zn leaves (Fig. 7d) compared with controls (Fig. 7b); data obtained from the SEM images were 2.8 ± 0.2 vs 3.9 ± 0.1 μm (n = 18) in 300 μM high-Zn and control plants, respectively. Isolated chloroplasts were also smaller in high-Zn plants (3.0 ± 0.2 μm) than in the controls (4.5 ± 0.3 μm) (images not shown; n = 30). High Zn concentrations reduced the apparent adherence of chloroplasts to the mesophyll cell plasma membrane (Fig. 7d) compared with control plants (Fig. 7b).
The aim of the present study was to investigate the causes of the decrease in photosynthesis that occurs under Zn stress. Both 100 and 300 μM ZnSO4 resulted in high Zn concentrations in roots and shoots, far above the leaf critical toxicity concentrations (400–500 μg g−1 DW; Marschner, 1995), with shoot Zn concentrations being similar in the 100 and 300 μM Zn treatments. Both biomass and photosynthetic rates decreased markedly in Zn-treated plants compared with the controls. Decreased growth and photosynthesis under excess Zn have already been described for other species (Ismail & Azooz, 2005; Vaillant et al., 2005; Mateos-Naranjo et al., 2008). In a previous study, we described changes in gas exchange properties and photosynthetic pigments in sugar beet grown under high Zn at lower PPFDs than those used in the present study (Sagardoy et al., 2009).
The most important effect of high Zn on photosynthetic parameters was a 70% decrease in stomatal conductance, with mesophyll conductance decreasing by 44%, whereas other possible causes for photosynthetic rate decreases such as PSII photochemistry were not significantly affected. A photosynthesis limitation analysis (Grassi & Magnani, 2005) revealed that, of a total photosynthesis limitation of 42–48% under excess Zn, up to 38–42% could be accounted for by SLs and only 4–5% could be accounted for by MCLs, whereas significant BLs did not occur. Furthermore, the decrease in stomatal conductance was caused by physical and/or structural stomatal changes, whereas hydraulic and chemical signaling, which usually control stomatal closure (Christmann et al., 2005, 2007), were not involved. Stomata of high Zn-treated plants did not respond at all to either chemical or hydraulical signals, and the concentration of ABA in the xylem was decreased rather than increased under excess Zn, indicating that stomatal closure was not mediated by ABA signals. Also, gs in Zn-treated plants was unaffected by exogenous ABA and changes in VPD. In summary, Zn-treated plants showed a stomatal closure similar to that of control plants supplied with exogenous ABA, according to a Ball–Woodrow–Berry analysis.
Zn-stressed plants had a lower stomatal frequency and a smaller stomatal size than control plants, and similar characteristics were found in Phaseolus vulgaris grown in high Zn (Van Assche et al., 1980). Scanning electron microscopy showed that stomata of 300 μM Zn-grown plants were round in shape and had a shorter slit than the stomata of control plants, and in many cases the slit was apparently sealed with unidentified, wax-like substances. Preliminary experiments (using a 10-s wash with chloroform:MeOH 2 : 1, a 10-s wash with hexane or a sequential combination of the two procedures) confirmed the waxy nature of these substances. Further studies are needed to elucidate the nature of this stomatal seal, and the mechanism for its accumulation under excess Zn.
The causes of the decreased CO2 mesophyll conductance were also investigated using LT-SEM. High Zn-grown plants had a lower leaf porosity than control plants, with the surface of leaf mesophyll cells being less exposed to intercellular air spaces than those of control leaves. Also, chloroplasts were smaller and the interaction of chloroplasts with cell membranes was hampered as a result of the changes in shape, and both factors would increase the length of the CO2 diffusion pathway in the cytosol. All these factors have been shown to be determinants of gm under very different experimental conditions (Sharkey et al., 1991; Flexas et al., 2008; Evans et al., 2009; Li et al., 2009).
The photochemistry-related parameters Fv/Fm, ETR and ΦPSII were not decreased significantly under high Zn, consistent with previous observations in sugar beet grown in relatively low PPFDs (Sagardoy et al., 2009) as well as in other species (Van Assche & Clijsters, 1986a; Schuerger et al., 2003; Dhir et al., 2008). This indicates that decreased photosynthesis was not caused by impaired leaf photochemistry. Zn treatments did not induce the operation of alternative sinks for electrons. On the one hand, this evidence comes from the close to 0 intercept of the relationship between ΦPSII and ΦCO2 under nonphotorespiratory conditions (Laisk & Loreto, 1996; Long & Bernacchi, 2003). On the other hand, the ETR : AN ratios increased under Zn excess, from 6.7 in the controls to 10.8 in the 300 μM Zn treatment (calculated from Table 2); values for ETR : AN + respiration (values taken from Table 5) ratios were lower, at 6.6 (controls) and 10.1 (300 μM Zn), whereas those for the ETR : AN + respiration + photorespiration (not shown) ratios were lower still, at 5.2 (controls) and 6.9 (300 μM Zn), indicating photorespiration as the cause of the increased ETR : AN ratios in excess Zn-grown plants and ruling out the existence of alternative sinks for electrons under Zn excess. The slope of the relationship between ΦPSII and ΦCO2 was not affected by the Zn treatments, which suggests that there were no changes either in leaf absorbance or in energy partitioning between PSI and PSII. Photosynthetic biochemistry was also unaffected by Zn, based on the in vivo estimates of Vc,max and Jmax. Nevertheless, Zn excess increased the activity of CA, a metalloprotein with Zn in its active center, although only at 300 μM. Zn was previously reported to inhibit CA in vitro at high concentrations (Ivanov et al., 2007). Although it has been suggested that CA is involved in the regulation of mesophyll conductance to CO2 (Gillon & Yakir, 2000), gm did not increase under excess Zn, but instead decreased by 44% compared with the control. This decrease, moderate when compared with the much greater decrease in stomatal conductance, may be related to the increase in CA activity.
Dark respiration increased markedly in Zn-treated plants, as observed previously in other species (Ismail & Azooz, 2005). The increase in total respiration was associated with significant increases in the activity of both cytochrome (vcyt) and alternative (valt) pathways. Data obtained in this study are not consistent with previous data suggesting a Zn-induced preferential increase in AOX (Webster & Gadd, 1999), or an inhibitory effect of Zn on AOX (Affourtit & Moore, 2004). The cytochrome pathway is associated with the growth component of respiration and results in high ATP production, whereas AOX is associated with the maintenance component of respiration and results in lower ATP production (Florez-Sarasa et al., 2007). In the case of the sugar beet plants in our study, where growth of high-Zn plants was severely reduced, increased ATP synthesis through increased cytochrome respiration would probably be used to increase ion uptake, exchange and compartmentalization (Lambers et al., 2005), to minimize the impact of Zn toxicity. The direct effect of increased respiration on the decrease in net photosynthesis would have been very small (approx. 3%); assuming that the measured rates of dark respiration also applied during the light period, the increased respiration would have decreased the total reduction in photosynthesis induced by excess Zn from 42% to 39% and from 48% to 45% under 100 μM and 300 μM Zn, respectively. The actual effect was probably even smaller, as the rates of respiration in the light are often lower than in the dark (Priault et al., 2006; Juszczuk et al., 2007).
In conclusion, 100–300 μM Zn resulted in large reductions in sugar beet biomass (> 50%) and photosynthetic rates (40–50%), whereas leaf respiration rates doubled through increased activity of both the cytochrome and alternative pathways, probably resulting in increases in capacities for ion compartmentalization and Zn exclusion. Under excess Zn, stomatal conductance was reduced by 70%, and stomata became insensitive to environmental variables such as leaf water status, exogenously applied ABA and VPD. In high Zn-treated plants, stomata were round in shape and smaller than in control plants and, in many cases, were covered by a wax-like seal of unknown nature. Excess Zn, therefore, affected primarily stomatal conductance, apparently through alterations of guard cell development (lower stomatal density on the leaf surface) and guard cell function. Leaf photochemistry and photosynthetic biochemistry were not significantly affected by high Zn. Mesophyll conductance to CO2 also showed 44% decreases, despite concomitant 2-fold increases in CA, possibly as a result of changes in mesophyll ultrastructure and chloroplast size and arrangement with respect to the mesophyll cell plasma membrane.
This work was supported by Spanish Ministry of Science and Innovation grants AGL2007-61948 (to J.A.) and BFU2008-01072/BFI (to M.R-C.). R.S. was supported by an I3P-CSIC predoctoral fellowship. We thank P. Pons, M. Truyols and G. Cabot for assistance with growing plants, J. Bota for help with CA measurements, I. Tacchini and F. Pinto for help with SEM and LT-SEM, respectively, and J. M. Andrés for the use of the SEM apparatus. Finally, we would like to thank Dr Biel Martorell for his technical help with the IRMS and all the staff at the Serveis Cientifico-Tecnics of the Universitat de les Illes Balears for their help during the running of these experiments.