Water availability for irrigation is among the most critical factors that affect fruit tree growth in commercial orchards and tree production in nurseries, especially considering the long-term and costly investments of planting fruit tree orchards. Water stress promotes stomatal closure and reduction in CO₂ fixation that may cause over-reduction of the photosynthetic electron chain. If water stress is prolonged and/or severe, part of the energy supplied by incident photons may be redirected into processes that favor the generation of reactive oxygen species (ROS) such as hydrogen peroxide (H₂O₂), leading to oxidative damage to plant tissues (Jiang and Zhang, 2002). Nevertheless, plants can activate ROS-scavenging enzymes such as superoxide dismutase, catalase, ascorbate peroxidase or glutathione reductase, and non-enzymatic systems including secondary metabolites such as phenolic compounds, alkaloids, isoprenoids, phenylpropanoids, and other antioxidants such as glutathione and ascorbic acid (AsA) to reduce oxidative damage (Gill and Tuteja, 2010; Tattini et al., 2015). Furthermore, these systems can also play very important roles in the protection of cell membrane integrity (Khan et al., 2011; Patade et al., 2012).

Ascorbic acid and the ascorbate-glutathione pathway play an important role in detoxification of ROS and modulating other fundamental functions in plants under stress conditions (Gill and Tuteja, 2010; Foyer and Noctor, 2011; Akram et al., 2017). Ascorbic acid is also the major non-enzymatic antioxidant in the apoplast (Pignocchi and Foyer, 2003) where it also has a key role in environmental stress perception and stress signaling (Wolucka et al., 2005; Shapiguzov et al., 2012). Enhanced antioxidant systems induced by water stress have been described in herbaceous plants, for instance elevated AsA and increased activities of antioxidant enzymes in maize (Dolatabadian et al., 2009), or tomato (Murshed et al., 2013), and in fruit trees such as mandarin (Sarkar et al., 2016).

Plant tolerance to environmental stresses can be enhanced by the exogenous application of beneficial molecules such as proline, amino acids, humic acid, and other antioxidants (Hasanuzzaman et al., 2015). The physiological responses of herbaceous plants under water stress to exogenous AsA have been thoroughly studied (Athar et al., 2008; Dolatabadian et al., 2009; Malik and Ashraf, 2012; Xu et al., 2015; Mukhtar et al., 2016). Nevertheless, the effects of exogenous applications of AsA in fruit tree species under water stress have been scarcely researched [only in olive trees (Ibrahim, 2013) and grapevines (Zonouri et al., 2014)], and as of now, there are no studies on the effects of exogenous AsA on water-stressed deciduous fruit trees and their responses after rewatering.

Water stress can impair growth and entry into production of young fruit trees, and reduce growth, yield and fruit quality of mature trees. Water stress can be detrimental for young trees during the first year after planting in the field, especially in areas where the current practice is to start irrigation during the second year, as is commonly done in the Southeastern United States. In a previous study, young peach trees subjected to water stress for a prolonged period of time did not improve physiological indicators such as tree water status and sap flow rates even 1 week after irrigation is resumed (Remorini and Massai, 2003). The response to rewatering is especially relevant in young trees, for instance in nursery trees where water stress or even temporary irrigation problems could affect budding success and reduce maximum development rates and consequently nursery productivity.

Since exogenous AsA has been reported to positively impact physiological parameters in herbaceous plants under water stress, it was hypothesized that AsA applications may also improve responses of deciduous fruit trees to suboptimal water conditions. Thus, the goal of this research was to evaluate the efficacy of exogenous AsA applications in improving physiological responses (gas exchange and water relations, stomatal density and size, proline concentration, catalase activity and lipid peroxidation) of peach trees during water stress and upon rewatering.

Ascorbic acid is the most abundant antioxidant in plants (Buettner and Jurkiewicz, 2006), and it has importance in photoprotection and regulation of photosynthesis though stomatal or non-stomatal factors (Foyer and Harbinson, 1994; Forti and Elli, 1995). In this study we demonstrated that foliar application of AsA in young peach trees could be a useful practice to overcome short periods of water shortage.

In regards to gas exchange, exogenous applications of AsA to young water-stressed peach trees significantly increased A_CO2 upon rewatering in both cultivars to control levels. Biosynthesis of AsA occurs on the inner mitochondrial membrane through the oxidation of L-galactono-1,4-lactone (L-GalL). Recent research proved that exogenous applications of L-GalL to AsA-deficient Arabidopsis mutants increased A_CO2, photosynthetic electron transport rate, and leaf stomatal conductance (Senn et al., 2016). Also, AsA has a role in photosynthesis and donates electrons to photosystems I and II when the primary electron donor system is damaged (Gallie, 2013). Net photosynthesis, growth rate and chlorophyll concentration also increased in wheat seedlings grown under water stress and supplemented with AsA in comparison with non-treated seedlings (Malik and Ashraf, 2012). The results observed in our experiment are in line with these studies since trees that received exogenous AsA had increased A_CO2 values (after rewatering). This is of remarkable importance for alleviating the negative effects of water stress on photosynthesis reduction in young trees in commercial orchards undergoing a period of water stress (especially in areas where the current practice is to start irrigation after the second year on the ground), and in container-grown and field-grown young trees in tree nurseries, not only during periods of drought but also when field-grown nursery trees are dug and many fine roots are broken, which causes temporary water stress to trees until roots can reestablish.

In parallel with this increase in A_CO2, exogenous AsA applications also increased g_s after rewatering. This view is supported by the roles of AsA in stomatal regulation (through the control of H₂O₂ levels that otherwise could trigger stomatal closure; Chen and Gallie, 2004) and in maintaining water homeostasis (Athar et al., 2008). These roles are especially important in trees under water stress. In our experiment, one application of AsA significantly increased g_s in water-stressed ‘Scarletprince’ trees during the experiment, and two applications increased g_s in water-stressed trees of both cultivars upon rewatering. Interestingly, water-stressed ‘Scarletprince’ trees with a single application of AsA showed a less negative Ψ_w before rewatering than water-stressed trees that had not received AsA, and water-stressed ‘Scarletprince’ trees with either one or two AsA applications showed increased proline concentration before rewatering (twofold values compared to control trees), which correspond with Ψ_w values in AsA-treated water-stressed trees similar to control trees. However, proline concentration did not increase in water-stressed ‘CaroTiger’ trees, and only water-stressed trees with two applications of AsA had a 0.5-fold increase in proline concentration, which did not improve tree Ψ_w. The buildup of compatible solutes such as proline in water-stressed plants may contribute to the lowering of osmotic potential after rewatering and allow water to move into the cells (Tyree and Jarvis, 1982), which would partly explain the different responses of the two cultivars on gas exchange upon rewatering.

Furthermore, differences in ostiole area and WUE_leaf between cultivars were also found. Even though ostiole area of water-stressed trees with AsA applications was similar to that of control trees in both cultivars, water-stressed ‘Scarletprince’ trees with AsA applications had smaller ostiole area values than control (well-watered, no AsA) trees, whereas ‘CaroTiger’ water-stressed trees had greater area than control trees. This was probably the reason why ‘Scarletprince’ trees showed higher WUE_leaf than ‘CaroTiger’ trees (28% higher). Thus, both cultivars seem to have different response mechanisms when subjected to water stress, and these mechanisms seem to be controlled at the leaf level since both cultivars were grafted onto the same rootstock: accumulation of proline in ‘Scarletprince’ trees and stomatal control in ‘CaroTiger’. Different responses to AsA applications in cultivars of barley (Salama et al., 1994) and wheat (Athar et al., 2008) have also been reported.

The role of exogenous AsA on tree water status must be assessed because its ability to retain water under water stress conditions could improve tissue tolerance by protecting carboxylation, and inactivation and denaturation of enzymes (Anjum et al., 2012). For instance, our results show that even though each cultivar had different protective mechanisms, A_CO2 values in water-stressed trees with two AsA applications was similar in both cultivars.

Photosynthesis reduction caused by water stress-induced reduction in g_s in trees without AsA exposes chloroplasts to excessive excitation energy and induces oxidative stress by increasing the production of ROS such as H₂O₂. H₂O₂ is an important signaling molecule that promotes closing of stomata (Pei et al., 2000; Murata et al., 2001). Plants with a high level of ascorbate are less responsive to H₂O₂ signaling (Chen and Gallie, 2004) and, because AsA scavenges H₂O₂, exogenous applications of AsA have been found to reverse H₂O₂-induced stomatal closing (Zhang et al., 2001). Studies performed in several Prunus hybrids showed that water stress caused H₂O₂-induced oxidative stress and increased activity of the ascorbate-glutathione cycle-associated enzymes and their antioxidant substrates; nevertheless, this was reversed upon rewatering (Sofo et al., 2005), which is in agreement with our results.

In addition, lipid peroxidation causes cell membrane damage including decreased fluidity and increased permeability. Malondialdehyde is a product of lipid peroxidation and, thus, it is commonly used to determine cell membrane damage caused by oxidative processes (Calatayud et al., 2002; Ozkur et al., 2009). Our results show that AsA applications could have modified the antioxidant capacity of the trees since AsA-treated trees showed lower MDA levels: AsA “per se” scavenges ROS and prevents protein oxidation and degradation (Noctor and Foyer, 1998) and, additionally, enzymatic antioxidant systems in AsA-treated trees were stimulated. In this sense, proline has been described to act as an inhibitor of lipid peroxidation in drought-stressed rice (Guo et al., 2005) and wheat (Farooq et al., 2013), and as a scavenger of ROS under different abiotic stresses (Gill and Tuteja, 2010). Thus, proline accumulation in water-stressed trees treated with AsA in our study may have also participated in reducing lipid peroxidation as shown by the low levels of MDA in sprayed trees.

A simultaneous activity of antioxidants is assumed to be necessary for a fast response when plants face water stress (Kwak et al., 2009). In our experiment, the activity of catalase, which is involved in removing and/or scavenging ROS, was significantly stimulated in AsA-treated trees of both cultivars. A similar increase in catalase activity was reported in wheat treated with AsA and grown under salt stress (Athar et al., 2008). Specifically in our experiment, one application of AsA only increased catalase activity in ‘Scarletprince’ trees whereas ‘CaroTiger’ trees needed two applications to reach a similar value. Although other enzymatic antioxidant systems that have not been measured in our experiment such as peroxidases may have accounted for elimination of hydrogen peroxide, non-sprayed water-stressed trees of both cultivars had increased MDA and decreased photosynthesis even after re-watering, which indicates that antioxidants did not efficiently maintain ROS scavenging in water-stressed trees that had not received AsA. This is supported by the fact that major detoxification of ROS produced during photosynthesis are mediated by catalase and by reductive processes involving major redox buffers such as ascorbate (Noctor and Foyer, 1998; Foyer and Noctor, 2003). Thus, our results imply an accumulation of ROS as a consequence of water stress and a reduction due to the increased catalase activity caused by the application of AsA. By contrast, the untreated stressed trees showed the lowest stem water potential values, which was probably linked to a minor accumulation of proline, less catalase activity, and increased lipid peroxidation.

We can conclude that exogenous application of AsA could improve tree physiological responses to suboptimal water conditions and upon rewatering. Nevertheless, the parameters measured on water-stressed ‘Scarletprince’ and ‘CaroTiger’ trees revealed that each cultivar had a different response mechanism (the former using solute accumulation and the latter relying mainly on stomatal control), which can have implications on the potential use of exogenous biochemicals to improve water stress tolerance. Further applied research studies should explore the effect of applications of AsA at different concentrations on stress tolerance and recovery.

CP carried out the experiment under the supervision of ÁC (dissertation adviser) and JM (supervisor at Clemson University). Measurements were carried out by CP and JM. The paper was written and edited by the three authors.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.