Introduction

Tropospheric ozone (O3) is one of the most important phytotoxic air pollutants in many parts of the world (Ashmore 2005; Van Dingenen et al. 2009). Crop losses are estimated to be $14–26 billion per year worldwide (Van Dingenen et al. 2009) and $17–18 billion in the year 2030 (Avnery et al. 2011). Even with the implementation of legislation to control emissions of its precursors, the primary air pollutants, ozone ambient concentrations are rising in much of the world during this century. The Royal Society predicted that ambient concentrations may increase by 20 to 25 % between 2015 and 2050, and again from 40 to 60 % in 2100. Accordingly, it is expected that ozone pollution may represent a relevant threat to global food security and climate changes in 2030 (Royal Society 2008).

The toxic effects of O3 on vegetation have been studied for over 50 years (Elägoz and Manning 2005), and it is well established that chronic exposure to relatively high concentrations can (i) cause oxidative stress and cellular alterations (Iriti and Faoro 2008; Faoro and Iriti 2009); (ii) produce leaf chlorotic, necrotic, and bronzing symptoms (Manning et al. 2002; Faoro and Iriti 2005); (iii) decrease photosynthetic efficiency, plant growth, and yield (Hayes et al. 2007; Mills et al. 2007; Booker et al. 2009; Guidi et al. 2009; Singh et al. 2009; Cascio et al. 2010); and (iv) induce premature senescence (Tonneijck et al. 2004) in a wide range of cultures and species in the (semi)natural communities.

The severity of O3 injury during the vegetative growth is affected both by extrinsic factors, such as the ozone concentration, duration and timing of exposure, environmental conditions that influence the stomatal uptake of the pollutant, and the intrinsic sensitivity of individual species or genotypes. Furthermore, the impact of O3 is greatly influenced by a series of acclimation and tolerance mechanisms, including antioxidant defenses and secondary metabolites (Ashmore et al. 2004; Matyssek et al. 2004; Fiscus et al. 2005; Black et al. 2007; Iriti and Faoro 2008, 2009; Betzelberger et al. 2010).

The effects of ozone pollution on vegetation have been reported on different plant species such as tomato (Calatayud and Barreno 2001), tobacco, bean, clover (Manes et al. 2003; Crous et al. 2006), watermelon (Gimeno et al. 1999; Benton et al. 2000; Fumagalli et al. 2001), and spinach (Calatayud et al. 2004). Due to ozone exposure, the leaves may present a typical bronzing pigmentation on adaxial (upper) leaf surface (EPA 1996; Pleijel 2000; Iriti et al. 2006), and they can also age prematurely, as in the case of wheat and other cereal species (Machler et al. 1995; Pleijel 2000). Detrimental physiological effects, such as reduced photosynthetic rate, may also occur without any visible symptom (Pleijel 2000). A decrease of yield has also been documented in a number of crops including carrots (Bennett and Oshima 1976) and tomatoes (Aguayo et al. 2006).

The use of bioindicator plant species is considered an inexpensive and reliable method in ozone monitoring. Bioindicators have the advantage to indicate a specific symptomatology that reflects the absorbed concentration of ozone, which is not necessarily correlated with O3 levels present in ambient air, because favorable conditions for ozone injury to occur are dependent on plant stomatal conductance and susceptibility to the pollutant, its concentration, and duration of exposure (Smith 2012; Pellegrini et al. 2014).

In the present work, carried out in fumigation chamber, we aimed to study the effects of three doses of ozone on two tomato (Solanum lycopersicum L.) widely cultivated varieties: Rechaiga II still largely used by farmers as a local variety, and De Colgar, the most commercialized tomato cultivar in Algeria. Visible leaf symptoms were recorded and some physiological parameters were measured, such as stomatal conductance, membrane integrity, potassium and sodium contents, amount of photosynthetic pigments, and soluble sugars, in order to assess the level of sensitivity/tolerance of both varieties. Indeed, as previously reported, these traits are relevant factors determining ozone sensitivity of selected plant cultivars (Faoro and Iriti 2005; Brosché et al. 2010; Tiwari and Agrawal 2011). Information on the effects of ozone on Algerian tomato varieties is still scarce and, to the best of our knowledge, this issue was not previously investigated in Algeria.

Materials and methods

Plant material

Two tomato (S. lycopersicum L.) cultivars were used in this study, De Colgar and Rechaiga II, the latter a local variety, both commercially available. Plants were grown in a greenhouse, in pots of 12 × 21 cm (2373 cm3) with a mixture of sand/ground/organic matter (1:1:1), at 24 ± 2 °C, 60 ± 5 % relative humidity (RH), and 16-h light/dark cycle. Plants were manually irrigated 3 days per week. After 6 to 9 weeks of planting up to six-leaf stage, the plants were transferred to the growth chamber for fumigation experiments.

Ozone fumigation

Ozone fumigation was carried out in a controlled chamber (~13.5 m3), at 26 ± 1 °C, 20 ± 5 % RH, and 500 μmol m−2 s−1 photon flux density at plant height. Ozone is generated by an electric discharge, passing pure oxygen through an ozone generator Ecobox (DBG Investments Group, LLC, Bristol, VA, USA). Ozone concentration in the fumigation chamber was continuously monitored with a photometric O3 analyzer (A-21ZX, Eco Sensors, Inc, Boudry, Switzerland), operating on UV absorption and interfaced with a personal computer. Control plants were kept in a growth chamber with filtered air, under the same conditions of temperature and RH. Greenhouse plants were pre-adapted to the conditions of the chamber for 48 h. Twelve plants for each cultivar were fumigated with 50 ppb of ozone, 4 h/day for 7 days, which are considered urban conditions (Heath 1994), simulating episodes of elevated ozone concentrations (Saitanis et al. 2014; Thwe et al. 2014a). Similar experiments were conducted with higher ozone concentrations, 80 and 100 ppb, while untreated plants were used as controls.

Calculation of necrotic areas

The leaves were examined at 0, 2, and 24 h after beginning of fumigation and, then, daily until the end of experiments (7 days). The percentage of necrotic symptoms on leaves was calculated from pictures recorded by high-resolution camera; the photos were processed by the MesurimPro® software that allows area calculation by assigning specific colors. The percentage of necrosis was calculated as NP (%) = NA / TA × 100, where NA = necrotic area and TA = total area.

Physiological parameters

Physiological parameters were determined at the end of each one of the three fumigations, i.e., after 7 days of ozone exposure.

Stomatal conductance

This parameter was measured using a porometer (AP4DELTA-T Devices, Cambridge, UK), which includes a portable unit for monitoring and analysis of stomatal conductance. The measurements were performed both on fumigated and control leaves, at 8 to 10 a.m., in fumigation chamber at 26 ± 1 °C, 20 ± 5 % RH, and 500 μmol m−2 s−1 photon flux density at plant height.

Membrane integrity

The membrane integrity was assessed by the method of Campos et al. (2003), to verify the transmembrane flow. Disks of freshly cut leaves (0.5 cm2) were rinsed three times (2–3 min) with deionized water; they were then floated on 10 mL of deionized water in Petri dishes. The leakage of electrolytes in the solution was measured after 22 h at ambient temperature, using a conductivity meter (pH-LF 3001-2). The total conductivity was obtained after maintaining the plates in a 90 °C oven for 2 h. The results are expressed in percentage of the total conductivity. The percentage of membrane integrity was calculated as MI (%) = (1-CC / TC) × 100, where CC = clear (free) conductivity and TC = total conductivity.

Photosynthetic pigments

The method of Lichtenthaler and Wellburn (1983), slightly modified by (Porra 2002), was used for determination of photosynthetic pigments. Leaf tissues (100 mg) were ground in a mortar and diluted in 8 mL of 80 % acetone. The mash was filtered in a test tube, using a Whatman paper 22. The volume of the tube was filled to 10 mL, by adding acetone. The absorbance was recorded with a flame spectrophotometer at wavelengths of 470, 645, and 663 nm. The results on the contents of chlorophyll a (chl a), chlorophyll b (chl b), total chlorophylls (chl T), and carotenoids are expressed as mg g−1 fresh weight (FW). Quantification of chlorophylls and calculation of the extinction coefficient were performed according to the Lambert-Beer law.

Soluble sugars

Soluble sugars were determined by the method of Dubois et al. (1956). Two milliliters of 80 % ethanol was added to leaf tissues (100 mg) in a test tube at room temperature in the dark. After 48-h extraction, ethanol was evaporated in a water bath at 70 °C. After cooling, each tube was filled with 20 mL distilled water. Then, 1 mL of this solution was transferred into another tube, adding 1 mL of 5 % phenol and 5 mL of concentrated sulfuric acid and vortexing to homogenize the solution. After 10 min, tubes were placed again in the water bath for 15 min at 30 °C. The absorbance of the solution was recorded at 490 nm and the results expressed as μg g−1 FW.

Determination of K+ and Na+

The collected leaf samples were washed in distilled water to remove salt and any external debris and dried at 80 °C for 48 h. The dry samples were ground to a fine powder using a mortar and pestle. The samples (1 g) were ashed at 600 °C in an electric furnace for 4 h. Five milliliters of 2 N HCl was added to cooled ash samples, which were then dissolved in boiling deionized water, filtered, and adjusted to a final volume of 50 mL. Na+ and K+ ions were measured using the standard procedure flame photometer (Vogel 1955), and the results were expressed as ppm.

Statistical analysis

The statistical treatment of the results was performed using the STATISTICA software. A description of the data is performed using the box plots to compare the percentage of necrotic leaves or physiological parameters of tomato (S. lycopersicum L.) cv. Rechaiga II depending on the dose of O3. ANOVA used to study the effect of O3 exposure on physiological parameters. Finally, a correlation was conducted with physiological parameters, necrotic lesions, and the ozone concentration; only important data has been shown.

Results

Leaf symptoms

The O3 exposure (50 ppb for 4 h) caused the appearance of visible symptoms on Rechaiga II cultivar, but not in De Colgar one (Fig. 1a). Shortly after the end of fumigation, Rechaiga II plants showed interveinal small chlorotic spots on the adaxial leaf surface, less than 1 mm in diameter (Fig. 1a), which widened and flowed together to form chlorotic and ivory lesions in the next 24 h (Fig. 1b). Necrosis appeared after further 24 h (Fig. 1c, d). After 7 days of fumigation, no symptom was observed in cv. De Colgar (Fig. 1e, f). The concentration of 80 ppb caused more intense symptoms, with reddish brown necrosis reaching a diameter of 2 mm (data not shown). Noteworthy, foliar injuries increased with increasing O3 concentration (Fig. 2).

Fig. 1
figure 1

Visible symptoms induced by a pulse of O3 (50 ppb for 4 h) on tomato (Solanum lycopersicum L.) cv. Rechaiga II leaves: a small chlorotic spots, b chlorotic lesions, c, d necrosis, e tomato cv. De Colgar leaves at the beginning of experiment, f no symptom was observed in cv. De Colgar

Full size image
Fig. 2
figure 2

Box plot of the percentage of necrotic lesions on tomato (Solanum lycopersicum L.) cv. Rechaiga II leaves as a function of the O3 dose

Full size image

Physiological parameters of tomato Rechaiga II as a function of the O3 dose

Both the concentrations of 50 and 80 ppb ozone modified some physiological parameters in fumigated plants (Fig. 3). In particular, the pollutant reduced stomatal conductance compared to control plants, as well as the levels of K+, Na+, chl a, chl b, chl T, and carotenoids (Fig. 3a–g). In contrast, membrane integrity and soluble sugars were higher in fumigated plants than in the control ones (Fig. 3h, i). At 100 ppb, similar trends were observed for all the assessed physiological parameters, apart from K+, which increased after O3 fumigation.

Fig. 3
figure 3

Box plots of the effects of O3 exposure on physiological parameters of tomato (Solanum lycopersicum L.) cv. Rechaiga II: a stomatal conductance (S cm−1), b membrane integrity (%), c Na+ ions (ppm), d K+ ions (ppm), e chlorophyll a (mg g−1 fresh weight, FW), f chlorophyll b (mg g−1 FW), g total chlorophylls (mg g−1 FW), h total carotenoids (mg g−1 FW), i soluble sugars (mg g−1 FW)

Full size image

Effect of ozone dose on physiological parameters

The “dose effect” was assessed by ANOVA analysis (Table 1). A highly significant effect of ozone concentration was observed on the content of K+ (p ≤ 0.01), soluble sugars (p ≤ 0.01), and carotenoids (p ≤ 0.001).

Table 1 ANOVA analysis on the effect of ozone dose on physiological parameters of tomato (Solanum lycopersicum L.) cv. Rechaiga II (N = 12)
Full size table

Correlation analysis

Results showed a significant positive correlation between the ozone dose and K+ (p ≤ 0.05), as well as between the ozone concentration and the levels of soluble sugars (p ≤ 0.01) and carotenoids (p ≤ 0.001) (Fig. 4). A significant negative correlation was reported between the percentage of necrosis and both chl a and chl T (p ≤ 0.01) (Fig. 4).

Fig. 4
figure 4

Correlation analysis among ozone concentration, physiological parameters, and necrotic lesions of tomato (Solanum lycopersicum L.) cv. Rechaiga II; ozone concentration versus a total carotenoids; b soluble sugars; c K+ ions; necrosis versus d total chlorophylls; e chlorophyll a

Full size image

Discussion

Early visible symptoms on Rechaiga II tomato leaves appeared 24 h after fumigation with 50 ppb of ozone, the lowest dose used in our experiments. In these conditions, stomatal conductance decreased, compared with control plants, possibly reducing the stomatal uptake of the pollutant. Nonetheless, stomatal closure represents a main defense mechanism in tolerant plants (Gerosa et al. 2003). With 80 ppb ozone, stomatal conductance further decreased, even if this mechanism was saturated with the highest ozone concentration (100 ppb), possibly due to the damage of guard cells. These results are in accordance with our previous data on currant tomato (Solanum pimpinellifolium) (Iriti et al. 2006), a bioindicator species of the pollutant, showing that, in controlled fumigation experiments, leaf symptoms increased with ozone concentrations (50, 80, and 100 ppb).

After the appearance of chlorotic spots, an obvious increase in ion leakage due to damaged cell membrane was observed in fumigated plants. As reported in other species, a short exposure to ozone can induce a deleterious effect on this parameter (Płazek et al. 2000; Guidi et al. 2001; Calatayud et al. 2003; Francini et al. 2007). Accordingly, the mineral balance was affected by the pollutant, with lower levels of both Na+ and K+ in fumigated plants than in the control ones. These changes may be explained by the alteration of the cell membrane integrity which can, in turn, generate an ion leakage (Marre et al. 1998; Rossard et al. 2006). It was shown that ozone altered plasmalemma permeability by the oxidation of membrane lipids and protein sulfhydryl groups (Guidi et al. 1999).

The decrease of chlorophylls documented in our study was in accordance with previous data (Della Torre et al. 1998; Saitanis et al. 2001). The chlorophyll degradation is one of the main and more rapid detrimental effect of ozone on the photosynthetic apparatus, as indicated by the decrease in chlorophyll content in other plant species (Broadmeadow and Jackson 2000; Bussotti et al. 2007). Many types of abiotic stress, closely associated with lipid peroxidation, can cause damage to chlorophyll-protein complexes located in thylakoids, particularly in the PSII reaction center. At the end of the fumigation and during the recovery, our plants showed a significant reduction in the chlorophyll and carotenoid contents, and, possibly, in the photosynthesis rate. There is some debate regarding the primary mechanism(s) involved in the decrease of photosynthesis, with evidence of direct effects of ozone exposure on light and/or dark reactions of photosynthesis (Power and Ashmore 2002) or indirect effects on stomatal closure, as previously introduced (Noormets et al. 2001). The loss of cholorphylls and carotenoids may impair the light absorbing capacity of the light harvesting complexes, thus affecting the capacity of thermal dissipation under O3 exposure. Furthermore, carotenoids are pivotal antioxidant molecules able to scavenge the harmful singlet oxygen and other reactive oxygen species (ROS) (Mikkelsen et al. 1995; Telfer et al. 1994). Therefore, the reduction in the levels and biosynthesis rate of carbohydrates are the result of the decrease of photosynthesis under elevated O3 conditions (Sun et al. 2014).

Wellburn and Wellburn (1994) fumigated the Aleppo pine (Pinus halepensis) with ozone and found that, during the summer, the plants showed significant accumulation of starch (especially in the endoderm), with a simultaneous crushing of phloem cells. Starch accumulation along the ribs of the sheet was also observed in other species, such as in birch (Betula pendula) (Landolt et al. 1997). The authors of both studies concluded that such starch accumulation was probably due to an altered phloem loading, an interpretation that may also be appropriate to explain our results. The response of tomato plants to acute ozone stress varied depending on the phenological stage. Leaf injuries were higher in the younger plants than in the older ones, whereas recovery processes were more efficient in the younger plants compared with the older ones (Thwe et al. 2013). Even if, in similar exposure conditions, tomato fruit yield was not significantly affected, fruit quality was influenced. In agreement with our results, total soluble sugars increased in fruits harvested from ozone-fumigated tomato plants, as well as organic acids and ascorbic acid, with a lower sugar/acid ratio mostly due to increased content of malic acid (Thwe et al. 2014b). Finally, in accordance with our data, the same authors reported a decrease of stomatal conductance and photosystem II efficiency in tomato plants after acute ozone fumigation (Thwe et al. 2014a).

Conclusions

This study aimed to investigate the physiological response of two tomato cultivars, De Colgar and Rechaiga II, exposed to different ozone concentrations (50, 80, and 100 ppb) 4 h per day over a period of 7 days. Our results showed a higher sensitivity of cv. Rechaiga II to even the lowest dose of the pollutant, compared with the other variety which remained asymptomatic. Ozone-sensitive species have been tested to be used as bioindicators for tropical conditions, as recently emphasized (Alves et al. 2011; Ferreira et al. 2012; Moura et al. 2014). Therefore, we suggest that, in the future, high-sensitivity species, as the tomato cv. Rechaiga II, may be also used as a potential and an effective bioindicator species of ozone pollution. Furthermore, some physiological parameters may be useful in detecting earlier sensitivity to ozone. However, our experiments were carried out in fumigation chamber, where climatic parameters are controlled, and for a short period of time. Hence, new surveys in open field conditions are necessary to correctly evaluate the potential of tomato cv. Rechaiga II as bioindicator to be used in Algeria, as well as the predictive potential of some physiological parameters as biomarkers of tolerance/sensitivity to the pollutant.