High atmospheric carbon dioxide-dependent alleviation of salt stress is linked to RESPIRATORY BURST OXIDASE 1 (RBOH1)-dependent H2O2 production in tomato (Solanum lycopersicum)

Highlight High CO2 concentrations can counteract the negative impact of salt stress in an apoplastic H2O2-dependent manner by regulating stomatal movement and Na+ delivery from the xylem to leaf cells.


Introduction
Crop productivity and food security are threatened by global climate change factors, such as projected temperature increases of more than 3.5 °C (Warren et al., 2011;Meehl et al., 2012). Moreover, atmospheric CO 2 levels are likely to double by the end of this century (Solomon et al., 2007). Crop yields are already limited in some areas by soil salinization, due in part to rising sea levels and agricultural practices such as irrigation and fertilization. Moreover, greenhouse crops, which are often grown with elevated levels of CO 2 , can experience salinity because of frequent irrigation and fertilization, which are common practices in greenhouse agriculture.
High atmospheric CO 2 levels are likely to have a profound effect on oxidative signalling in plants, particularly because of the suppression of photorespiration (Munne-Bosch et al., 2013). Atmospheric CO 2 enrichment increases photosynthetic efficiencies, at least in the short term, leading to increased growth and biomass production (Foyer et al., 2012). In contrast, salt stress decreases crop yields. Moreover, soil salinity is an important factor leading to the continuous loss of arable land (Shabala and Cuin, 2008). While many studies have focused either on plant responses to atmospheric CO 2 enrichment or salt stress, there is a dearth of literature on plant responses to the combined effects of salinity and high CO 2 . High CO 2 is known to induce salt tolerance  but the molecular and metabolic mechanisms that underpin this response are poorly understood.
High salt concentrations have a severe impact on plant growth and metabolism, decreasing water uptake and inhibiting key metabolic processes such as photosynthesis (Flowers and Yeo, 1995;Mäser et al., 2002;Deinlein et al., 2014). High soil salinity enhances the production of reactive oxygen species (ROS), a process that is accompanied by increased membrane lipid peroxidation (Mittler, 2002;Miller et al., 2010). Recent studies suggest that ROS are involved in the regulation salt tolerance (Bose et al., 2013;Schmidt et al., 2013). For example, Arabidopsis thaliana knockout mutants lacking the respiratory burst oxidase (Atrboh) F or D proteins, which are NADPH oxidases that catalyse the production of ROS in the apoplast, show increased salt sensitivity and altered Na + / K + homeostasis Jiang et al., 2012Jiang et al., , 2013.
Many plants have evolved protective mechanisms to cope with salinity and minimize salt toxicity (Zhu, 2002;Munns and Tester, 2008). For example, in Arabidopsis the salt overly sensitive (SOS) signalling pathway is considered to mediate salt stress responses that regulate ion homeostasis (Ji et al., 2013). The SOS1 transporter functions in long-distance transport of Na + through the xylem from roots to shoots during salt stress (Shi et al., 2002). The SOS2 gene encodes a putative serine-threonine type protein kinase that is required for intracellular Na + and K + homeostasis (Liu et al., 2000). Moreover, SOS3 encodes a myristoylated calcium-binding protein that functions as a primary calcium (Ca 2+ ) sensor, perceiving increases in cytosolic Ca 2+ that are triggered by excess Na + Halfter et al., 2000;Liu et al., 2000). As a result of SOS3-SOS2 interactions, SOS2 is recruited to the plasma membrane, a process that leads to the downstream activation of SOS1 and the extrusion of excess Na + from the cytosol (Qiu et al., 2002;Quintero et al., 2011). In addition to SOS, other components such as Na + /H + exchanger (NHX), high-affinity K + transporter (HKT) transporters and mitogen-activated protein kinase (MAPK) are also important in salt stress tolerance and in root-to-shoot Na + partitioning (Horie et al., 2009;Bassil and Blumwald, 2014;Li et al., 2014).
High atmospheric CO 2 levels increase the photosynthesis/ photorespiration ratio (Foyer et al., 2012). The enhanced photosynthesis rates triggered by high CO 2 levels are accompanied by decreased water loss through transpiration due to partial stomatal closure (Drake et al., 1997;Robredo et al., 2007). Elevated atmospheric CO 2 levels also lead to the activation of carbonic anhydrase (CA) proteins (Hu et al., 2010), a process that coincides with the activation of the open stomata 1 (OST1) protein kinase and the SLAC1 (slow anion channel 1) anion channel, which are involved in the regulation of stomatal movement (Tian et al., 2015).
In addition to increasing photosynthetic CO 2 assimilation rates, growth with high atmospheric CO 2 levels can mitigate against the negative impacts of abiotic stresses (Ameye et al., 2012;Bauweraerts et al., 2013;Zinta et al., 2014). For example, growth under high CO 2 led to enhanced tolerance to salinity, Fe deficiency and increased resistance to (hemi) biotrophic microbes such as tobacco mosaic virus and Pseudomonas syringae in tomato (Solanum lycopersicum; Jin et al., 2009;Takagi et al., 2009;Del Amor, 2013;Li et al., 2015). In contrast, growth under elevated CO 2 enhanced susceptibility to the necrotrophic pathogen, Botrytis cinerea (Zhang et al., 2015). Atmospheric CO 2 enrichment enhanced photosynthesis rates and increased the growth of several plant species under saline conditions (Bowman and Strain, 1987;Robredo et al., 2007;Del Amor, 2013). High CO 2 -induced salt tolerance is associated with reduced oxidative stress and transpiration rates, and with improved cellular hydration, intracellular Na + /K + homeostasis and water use efficiency (Bowman and Strain., 1987;Maggio et al., 2002;Yu et al., 2015). However, the mechanisms by which high atmospheric CO 2 levels suppress transpiration and hence decrease the delivery of Na + from roots to shoots remain to be characterized.
The tomato respiratory burst oxidase RBOH1 gene, SlRBOH1, is a homologue of Arabidopsis RBOHF. It has the highest transcript abundance within the RBOH family (Zhou et al., 2014b). SlRBOH1 is involved in the regulation of tolerance to oxidative stress and to high temperature stress. It also plays a key role in the regulation of stomatal movements mediated by the phytohormones, abscisic acid (ABA) and brassinosteroid (BR) (Nie et al., 2013;Xia et al., 2014;Zhou et al., 2014a). Apoplastic ROS are involved in the regulation of stomatal movement and Na + /K + homeostasis (Jiang et al., 2012). In order to test the hypothesis that elevated CO 2induced salt tolerance is linked to apoplastic H 2 O 2 accumulation, the impact of high atmospheric CO 2 concentrations on the responses to tomato plants to salt stress was evaluated, with a particular focus on the role of SlRBOH1.

Plant materials
Tomato (S. lycopersicum L cv. Ailsa Craig) seeds were sown in perlite and kept at 28 °C for 2 weeks. Seedlings were then transferred to plastic tanks (20 cm×30 cm×15 cm, 4 seedlings per tank) filled with Hoagland nutrient solution in growth chambers. The growth conditions used for subsequent growth of the plants were as follows: a 14/10 h (day/night) photoperiod, a temperature regime of 25/20 °C (day/night), a photosynthetic photo flux density (PPFD) of 600 µmol m -2 s -1 and a relative humidity of 85%. Plants at the 4-leaf stage were used for Experiment I.

Virus-induced gene silencing (VIGS) of RBOH1
The tobacco rattle virus (TRV) VIGS constructs used for silencing of the SlRBOH1 gene were generated by cloning a 311-bp RBOH1 cDNA fragment, which was amplified by PCR using the forward primer (5′-ATACGCGAGCTCAAGAATGGGGTTGATATTGT-3′) and reverse primer (5′-CGGAATTCAGACCCTCAACACT CAACCC-3′). The amplified fragment was digested with the restriction endonucleases, SacI and XhoI, and ligated into the same sites of the pTRV2 vector. The resulting plasmid was transformed into Agrobacterium tumefaciens strain GV3101, and VIGS was performed by infiltration of 15-d-old wild-type seedlings with a mix of A. tumefaciens strains harbouring pTRV1-or pTRV2 (Ekengren et al., 2003). Plants were then kept at 22 °C under a 14-h photoperiod for 30 d before they were used for Experiment II (Kandoth et al., 2007).

Salt and atmospheric CO 2 enrichment treatments
Both experiments involving salt and atmospheric CO 2 treatments were carried out in CO 2 -controlled growth chambers (ConvironE15; Conviron, Manitoba, Canada). Plants were kept at 25/20 °C, with a 14-h photoperiod under 600 μmol m -2 s -1 PPFD and 85% humidity conditions. In Experiment I, tomato plants at the 4-leaf stage were exposed to ambient CO 2 concentrations (380 μmol mol -1 ), ambient CO 2 concentration with 100 mM NaCl in the nutrient solution, elevated CO 2 (760 μmol mol -1 ), and elevated CO 2 with 100 mM NaCl in the nutrient solution. In Experiment II, pTRV and pTRV-RBOH1 plants at the 5-leaf stage were exposed to two levels of CO 2 (380 and 760 μmol mol -1 ) and a nutrient solution with or without NaCl (100 mM) for 11 d, giving a total of eight treatments for Experiment II. The nutrient solution was replaced every 3 d during the experiments and all measurements were performed with at least four replicates, with 20 plants per replicate. Leaf or root samples were frozen in liquid nitrogen and stored at -80 °C until used for the biochemical assays and gene transcript analyses. During the experiments, plants were harvested and oven-dried at 80 °C for 3 d before determination of dry weight and measurement of Na + and K + contents.

Physiological and biochemical measurements
The CO 2 assimilation rates (Pn), transpiration rates (Tr), and stomatal conductance (Gs) of the plants were determined with an infrared gas analyzer-based portable photosynthesis system (LI-6400; LI-COR, Lincoln, NE, USA). The air temperature, relative humidity, and PPFD were maintained at 25 °C, 85% and 1000 μmol m -2 s -1 , respectively, with variable CO 2 concentrations. The maximum quantum yield of photosystem II (PSII) (Fv/Fm) was measured using an imaging-PAM chlorophyll fluorimeter equipped with a computeroperated PAM-control unit (IMAG-MAXI; Heinz Walz, Effeltrich, Germany) as previously described (Zhou et al., 2014a). The seedlings were kept in the dark for at least 30 min before the measurements were taken. Fv/Fm values were calculated as Fv/Fm = (Fm−Fo)/Fm, where Fo is the minimal chlorophyll fluorescence measured during the weak measuring pulses and Fm is the maximum fluorescence measured by a 0.8 s pulse light at 4000 μmol mol −2 s −1 . Fv/Fm values were determined using the whole leaf.
Relative electrolyte leakage was measured in the leaves as previously described (Cao et al., 2007) using a conductivity detector (FE30K, Mettler-Toledo Instruments Co., Ltd., Switzerland). Briefly, leaf segments (0.3 g) were vacuum-infiltrated in 20 ml deionized water for 20 min and kept in the water for 2 h, and the conductivity (C1) of the resulting solutions were then determined. The leaf segments were then boiled for 15 min, cooled to room temperature, and the conductivity (C2) of the resulting solutions were determined. The C1:C2 ratios (C1/C2 × 100%) were calculated and used as a measure of the relative electrolyte leakage. The level of lipid peroxidation in leaves was assessed by measuring the malonyldialdehyde (MDA) content using 2-thiobarbituric acid as described by Hodges et al. (1999). Leaf water potential of intact excised leaves was measured using a Dew point Potential Meter (WP4; Decagon Device, Pullman, USA). Plant cell death was detected by Trypan Blue staining as previously described (Bai et al., 2012). Briefly, leaves were boiled for 5 min in a 1:1 mixture of ethanol and staining solution (10 ml lactic acid, 10 ml glycerol, 10 ml phenol, and 10 mg Trypan Blue dissolved in 10 ml distilled water) for staining. The leaves were then de-stained with chloral hydrate (2.5 g ml −1 distilled water), changing the solution every 12 h until the leaves were transparent. The stomatal apertures were measured as previously described (Xia et al., 2014) by peeling off the abaxial epidermis with forceps and floating it on a buffer containing 30 mM KCl and 10 mM 2-(N-morpholino)-ethanesulfonic acid, at a temperature of 25 °C. All images were captured using a light microscope equipped with a digital camera (Leica Microsystems, Wetzlar, Germany). NADPH oxidase activity was determined in isolated plasma membrane vesicles as previously described (Zhou et al., 2014b).

Visualization of cellular H 2 O 2 and Na + accumulation
Hydrogen peroxide (H 2 O 2 ) production in tissues was monitored using 2,7-dichlorofluorescein diacetate (H 2 DCF-DA), as previously described (Pei et al., 2000;Milling et al., 2011;Xia et al., 2011) with minor modifcations. Detached roots were washed with deionized water and incubated 15 min with 25 μM H 2 DCF-DA in 200 mM phosphate buffer (pH 7.4) and then washed five times with the same buffer without the dye. As negative controls, the root segments were incubated with 1 mM ascorbate (ASC) or 100 U ml -1 catalase (H 2 O 2 scavenger) for 30 min. The stem and petiole were excised using a scalpel and incubated for 15 min in 200 mM phosphate buffer (pH 7.4), then incubated for 15 min with 25 μM H 2 DCF-DA in 200 mM phosphate buffer (pH 7.4), and washed as described above. Fluorescence was observed using a Leica DM4000B microscope and images were captured using a Leica DFC425C camera and the Leica application suite V3.8 software (Leica Microsystems, Germany).
H 2 O 2 in the leaf apoplast was visualized using cytochemical CeCl 3 staining. Leaf fragments (3 mm 2 ) were excised from inoculated leaf panels and infiltrated with freshly prepared 5 mM CeCl 3 in 50 mM Mops at pH 7.2 for 1 h at 28 °C, and then fixed and embedded according to Bestwick et al. (1997). Leaf sections were examined using a transmission electron microscope (H7650, Hitachi, Tokyo, Japan) at an accelerating voltage of 75 kV, to reveal any electrondense CeCl 3 deposits that are formed in the presence of H 2 O 2 . H 2 O 2 was extracted and analysed as previously described (Willekens et al., 1997), in which plant material was first ground in liquid nitrogen and HClO 4 . After thawing and centrifugation, the supernatants were adjusted to pH 6.0 with KOH and passed through columns to remove lipid peroxides. H 2 O 2 was then quantified using a glutathione peroxidase/horseradish peroxidase assay system.
The detection of Na + in roots was carried out as described by Oh et al. (2009). Tomato plants were treated with 100 mM NaCl in Hoagland nutrition solution for 3 d, after which, whole root systems were incubated for 8 h in a Petri dish with the same media supplemented with 20 μM CoroNa Green-AM (Invitrogen). The tissue was washed with 200 mM phosphate buffer (pH 7.4) five times to remove excess dye and observed using a Leica DM4000B microscope (Leica Microsystems, Wetzlar, Germany). Images were captured using a Leica DFC425C camera and the Leica application suite V3.8 software.

Determination of ion content in tissues and xylem sap
The ion content in tissues and xylem sap was measured as previously described (Jiang et al., 2012). Dried root, stem, and leaf material (0.3 g) was digested in 5 ml concentrated HNO 3 (69%, v/v) for at least 12 h before extraction. The solution was then filtered, using quantitative filter paper and diluted with deionized water. Concentrations of Na + and K + in the diluted samples were determined in an airacetylene flame using an atomic absorption spectrometer (A6300; Shimadzu, Kyoto, Japan). During the experiment, the xylem sap exuding from the cut stem surfaces of 12 replicate de-topped plants was collected and pooled, then diluted with deionized water for ion content measurement, prior to quantification by atomic absorption spectrophotometry.
RNA extraction and quantitative RT-PCR analysis RNA was extracted from leaves and roots using the RNAprep pure Plant Kit (Tiangen biotech Co., Ltd. Beijing, China) and quantified using a NanoDrop ND-2000 Spectrophotometer (Thermo Scientific, Waltham, MA, USA), before quality assessment with a gel cartridge on a Bio-Rad platform (Bio-Rad, Hercules, CA, USA). RNA samples were used only if the ribosomal bands showed no degradation, and the 260/280 and 260/230 absorbance ratios were between 1.8 and 2.1. Total RNA (1 μg) was reverse transcribed using a ReverTra Ace quantitative (qPCR) RT Kit (Toyobo, Osaka, Japan), following the manufacturer's instructions. qRT-PCR was performed using the LightCycler 480 Real-Time PCR System (Roche Diagnostics, Germany). Each reaction (20 μl) consisted of 10 μl SYBR Green PCR Master Mix (Takara, Chiga, Japan), 8.6 μl sterile water, 1 μl cDNA, and 0.2 μl forward and reserve primers (10 μM). PCR was performed with 40 cycles of 30 s at 95 °C, 30 s at 58 °C, and 1 min at 72 °C. Genespecific primers are listed in Supplementary Table S1 (available at JXB online) and Actin2 was used as the reference gene. The relative gene expression was calculated according to Livak and Schmittgen (2001).

Statistical analysis
The experimental design was a completely randomized block design with four replicates. Each replicate contained 16 plants and at least four independent replicates were used for each determination. The data were subjected to analysis of variance, and the means were compared using Tukey's test at the 5% level.

Results
CO 2 enrichment attenuates salt-induced growth reduction, cellular membrane peroxidation, and Na accumulation but increases the levels of RBOH1 transcripts Biomass accumulation was significantly higher (~47%) in tomato plants grown with CO 2 enrichment than under ambient CO 2 conditions (Fig. 1A). While biomass was reduced by growth in the presence of salt, the salt-induced reduction in biomass was significantly lower in plants grown with CO 2 enrichment than under ambient CO 2 conditions (Fig. 1A). High CO 2 -grown plants had lower levels of electrolyte leakage and of MDA, which is an end product of lipid peroxidation. They also showed lower salt-induced decreases in photosynthetic CO 2 assimilation (Pn) and in the maximum quantum yield of photosystem II (PSII) (Fv/Fm, Fig. 1B-D, Supplementary Fig. S1 available at JXB online). However, stomatal conductance (Gs) values and transpiration rates (Tr) were lower in plants grown with CO 2 enrichment than under ambient CO 2 conditions, in the presence or the absence of salt stress (Supplementary Fig. S1).
Exposure to either salt stress or elevated CO 2 increased the levels of RBOH1 transcripts in leaves and roots (Fig. 1E). However, the effects of high salt on RBOH1 transcripts were greatest in plants grown under high CO 2 . The levels of H 2 O 2 were measured both qualitatively, using H 2 DCF-DA as a fluorescence probe, and quantitatively, using a spectrophotometric assay method. Both of these methods showed increased H 2 O 2 accumulation in the roots and petioles of salt-treated plants that were grown under high CO 2 compared with ambient CO 2 conditions (Supplementary Figs S2A,B and S3). Moreover, the salt-induced fluorescence signal was greatly decreased in the presence of ascorbate or catalase, confirming the specificity of H 2 DCF-DA for H 2 O 2 detection ( Supplementary Fig. S2C). Crucially, growth with CO 2 enrichment significantly decreased Na + accumulation in roots, stems, and leaves but it increased K + accumulation in roots, leading to a reduction in the Na + :K + ratio in these organs (Fig. 2).

Silencing of RBOH1 compromises the salt tolerance conferred by elevated CO 2 levels
RBOH1 transcript levels were reduced by 53.6% in tomato plants subjected to virus-induced gene silencing (pTRV-RBOH1) compared with empty vector (pTRV) controls ( Supplementary Fig. S4). However, no significant changes were found in the transcript levels of other RBOHs in the pTRV-RBOH1 plants. NADPH oxidase activity was decreased by 38.1 % in the leaves of the pTRV-RBOH1 plants compared with the pTRV controls. Both CO 2 enrichment and high salt conditions induced H 2 O 2 accumulation in roots, stems, leaf petioles, and leaves in the pTRV plants (Fig. 3). Salt-and CO 2 -induced H 2 O 2 accumulation was primarily localized within the parenchyma cells of the vascular tissue in the pericycle in the leaf petioles (Fig. 3C). Moreover, H 2 O 2 was predominately localized in the apoplast/cell wall compartment of the salt-treated leaf cells (Fig. 3D). Salt and high CO 2 -dependent H 2 O 2 accumulation was much lower in the pTRV-RBOH1 plants than the pTRV controls ( Fig. 3; Supplementary Fig. S5).
While the growth of the pTRV-RBOH1 plants was not significantly different from that of the pTRV controls in the absence of salt, the negative effects of high salt on biomass accumulation, electrolyte leakage, lipid peroxidation, and cell viability were increased in pTRV-RBOH1 plants relative to pTRV controls (Fig. 4). CO 2 enrichment attenuated the saltinduced decreases in biomass accumulation, tissue water potentials as well as reducing the salt-induced increases in electrolyte leakage and MDA contents in the pTRV controls, but not in the pTRV-RBOH1 plants (Fig. 4).

RBOH1-derived H 2 O 2 is associated with Na + transport from roots to leaves
Photosynthesis rates (Pn) were similar in the pTRV-RBOH1 plants and pTRV controls but the stomatal conductance (Gs) values, stomatal aperture sizes, and transpiration rates (Tr) were slightly higher in the pTRV-RBOH1 leaves than the pTRV controls (Fig. 5). The salt-induced decreases in Gs and Tr were significantly higher in plants grown under high CO 2 than under ambient CO 2 conditions (Fig. 5). The RBOH1 silenced tomato plants had consistently higher Gs, Tr, and stomatal aperture values (Fig. 5). High atmospheric CO 2 had little effect on the Gs and Tr values (Fig. 5).
The levels of Na + were higher in the pTRV-RBOH1 leaves than the pTRV controls under salt stress. In contrast, in the presence of salt, the leaf K + levels were lower in the pTRV-RBOH1 leaves than the pTRV controls (Fig. 6). When the pTRV controls were grown with high CO 2 , leaf Na + accumulation decreased and leaf K + levels increased. Hence, the Na + :K + ratios were decreased in the salt-treated pTRV controls grown under high CO 2 relative to plants grown under ambient CO 2 conditions (Fig. 6). In contrast, the pTRV-RBOH1 plants did not show an alleviation of salt-induced changes in Na + and K + accumulation in either leaves or roots when these plants were grown under elevated CO 2 concentrations. Moreover, the Na + :K + ratios of the salt-treated pTRV-RBOH1 roots and shoots were similar under ambient or high CO 2 .
Visualization of Na + accumulation in roots was performed on plants exposed to high salt using CoroNa Green, a greenfluorescent indicator, whose emission intensity increases upon Na + binding (Oh et al., 2009). CoroNa-Green fluorescence was barely detectable in the roots of pTRV controls and pTRV-RBOH1 plants grown under ambient CO 2 in the absence of salt stress. In contrast, a strong fluorescent signal was observed in the pericycle and vascular cells of the roots of both pTRV controls and pTRV-RBOH1 plants exposed to salt (Fig. 7A). When plants were grown under high CO 2 , fluorescence was only detected in one cell layer in the middle of the pTRV controls roots. In contrast, several cell layers in the high CO 2 -grown pTRV-RBOH1 roots showed high fluorescence.
The xylem-sap Na + concentration increased progressively over time in both pTRV-RBOH1 plants and pTRV controls following the onset of salt treatment (Fig. 7B). However, the increase in xylem-sap Na + concentrations were consistently higher in the pTRV-RBOH1 plants than the pTRV controls (Fig. 7B). Exposure to high CO 2 decreased xylem-sap Na + concentrations in the pTRV control plants 3 d after the onset of the salt stress treatment but not in the pTRV-RBOH1 plants. The pTRV controls showed the lowest xylem sap Na + :K + ratios under high CO 2 growth conditions. In contrast, the pTRV-RBOH1 plants had the highest xylem sap Na + :K + ratios under both ambient and high atmospheric CO 2 levels.

Apoplastic H 2 O 2 plays a role in CO 2 -induced Na + and K + homeostasis
When the pTRV-RBOH1 plants were grown under ambient CO 2 in the absence of salt, the levels of transcripts encoding salt response genes such as SOS1-3 and NHX1-3 (Ji et al., 2013;Bassil and Blumwald, 2014) and MAPK1-3 (Li et al., 2014) were similar to the pTRV controls (Fig. 8). Moreover, the abundance of transcripts encoding salt response genes was similar in plants grown under ambient or high CO 2 . However, SOS1, SOS3, and MAPK1 transcript levels in the leaves and SOS3, NHX1, NHX2, and MAPK2 transcript levels in the roots, were significantly higher in plants grown under CO 2 enrichment.
With the exception of SOS3, growth with high salt significantly increased the levels of transcripts encoding salt response genes in pTRV leaves and roots. In contrast salinity had negligible effects on the levels of most of measured transcripts, except for NHX3 in the leaves, and SOS2, NHX1, NHX2, and MAPK1 in the roots, which were significantly increased as a result of the salt treatment in pTRV-RBOH1 plants. Crucially, growth under high CO 2 had no effect on the levels of transcripts encoding the salt response genes in the pTRV-RBOH1 plants.

Discussion
Although plant responses to atmospheric CO 2 enrichment and soil salinity are well characterized, relatively little is known about the interactions between these environmental stresses. In general, high CO 2 levels promote plant growth due to increased photosynthesis, while high salinity inhibits growth by disrupting cellular Na + /K + homeostasis (Zhu, 2003;Shabala and Cuin, 2008). However, salt stress has milder effects on the metabolism and physiology of plants grown under high CO 2 conditions than those grown in air (Kanani et al., 2010). The data presented here shows that growth under high CO 2 alleviates the negative impacts of high NaCl on photosynthesis and biomass production (Figs 1 and 5; Supplementary Fig. S1). These findings agree with those of other studies concerning the relationships between elevated CO 2 levels and high salinity in tomato and other plant species Takagi et al., 2009;(Del Amor, 2013;Zaghdoud et al., 2013;Pinero et al., 2014;Yu et al., 2015). High CO 2 levels decreased leaf transpiration rates and stomatal conductance values, even in plants grown with high salt (Fig. 5A; Supplementary Fig. S1). The high CO 2 -dependent decreases in transpiration resulted in a decreased Na + accumulation and lower Na + : K + ratios in the tissues (Figs 2, 6, and 7B). These results suggest that the higher salt stress tolerance observed under elevated CO 2 is largely dependent on the suppression of transpiration. This conclusion is supported by other studies showing that high CO 2 alleviated the adverse effects of salinity by modulation of aquaporins leading to lower transpiration rates (Zaghdoud et al., 2013).
The excessive accumulation of Na + in leaves of plants grown under salt stress is dependent on both the transpiration rate and also the Na + concentration in the transpiration stream. In turn, the transpiration rate is controlled by the degree of stomatal closure, which is related to the apoplastic production of H 2 O 2 (Desikan et al., 2005;Bright et al., 2006;Desikan et al., 2006). RBOH1-mediated H 2 O 2 production not only plays a role in the control of stomatal movements but also in the acquisition of stress tolerance in tomato (Xia et al., 2014, Zhou et al., 2014a. The data presented here suggest that that CO 2 -induced stomatal movements are also linked to RBOH1-dependent H 2 O 2 generation in tomato (Figs 3 and 5).
The data presented here show that high CO 2 concentrations not only increase RBOH1 transcript levels in leaves but also result in higher H 2 O 2 accumulation in the roots and leaves, with particular effects in the vascular system (Figs 1 and 3; Supplementary Figs S2, S3, and 5). These high CO 2 -mediated responses were significantly lower in the pTRV-RBOH1 plants compared with the pTRV controls. The observations presented here show that RBOH1 is important in the regulation of CO 2 -induced stomatal movements. Moreover, the high CO 2 -dependent alleviation of Na + accumulation and salt-dependent growth inhibition were compromised in the pTRV-RBOH1 plants grown under high salt. These findings agree with previous results showing that the loss of AtRBOHF function in Arabidopsis enhanced salt sensitivity (Jiang et al., 2012). Taken together, these results provide evidence in support of the conclusion that growth under high CO 2 enhances salt stress tolerance by increasing H 2 O 2 -dependent stomatal closure.
Earlier studies have suggested that salt-induced H 2 O 2 accumulation in the vasculature is also involved in the regulation Fig. 6. Effects of NaCl treatments on accumulation of Na + and K + , and the Na + :K + ratio in pTRV empty vector control and RBOH1 silenced tomato plants grown under elevated (Ele: 760 µmol mol -1 ) or ambient (Amb: 380 µmol mol -1 ) CO 2 conditions. (A), Na + , K + content, and the Na + :K + ratio in leaves. (B), Na + and K + content, and the Na + :K + ratio in roots. The data are means ± SD of four replicates.
Na + transport to the shoot (Jiang et al., 2012(Jiang et al., , 2013. The data presented here show that high atmospheric CO 2 levels not only increased the H 2 O 2 in the vascular system but they also exacerbated salt-induced vascular H 2 O 2 accumulation ( Fig. 3; Supplementary Fig. S2B). Moreover, loss of RBOH1 function in the pTRV-RBOH1 plants led to lower H 2 O 2 accumulation in the cells of the vascular tissues under both ambient CO 2 and high CO 2 conditions (Fig. 3). The pTRV-RBOH1 plants exhibited higher Na + accumulation in the vascular cells and in the xylem sap and they showed lower high CO 2 -induced decreases in Na + accumulation in the vascular cells and in the xylem sap (Fig. 7). These results suggest that RBOH1-dependent H 2 O 2 production is critical for the regulation of Na + delivery to the leaves under high CO 2 . Although stomatal conductance values and transpiration rates were substantially decreased after 3 d of salt treatment, the Na + contents of the xylem sap and leaves increased. These findings are consistent with the concept that stomatal movement plays a significant role in CO 2 -induced salt tolerance during the early stages of salt stress, while the rate of Na + delivery via the xylem is more important at the later stages of the high salinity response.
The SOS and NHX families play a major role in maintaining cellular pH values, K + concentrations and Na + /K + ratios in order to prevent the perturbations in Na + /K + homeostasis caused by high salinity Bassil and Blumwald, 2014;Reguera et al., 2014). Salt-induced increases in the levels of transcripts involved in salt tolerance and in Na + homeostasis were higher in plants grown under high CO 2 compared with ambient CO 2 conditions (Fig. 8). Furthermore, the levels of the salt-induced and high CO 2 -induced transcripts were lower in the pTRV-RBOH1 plants than the pTRV controls, a finding that can be linked to the higher Na + :K + ratios in the former than the latter. Taken together, these results suggest that H 2 O 2 production is involved in salt-induced expression of the SOS and NHX genes under high CO 2 conditions. RBOH-mediated increases in ROS accumulation are thought to enhance SOS1 mRNA stability, which contributes to the maintenance of ion homeostasis under salt stress conditions (Chung et al., 2008). In addition to regulating Na + exclusion into the soil, SOS1 also functions in retrieving Na + from the xylem under high salt stress (Shi et al., 2002). This function may explain the reduced Na + levels observed in the xylem sap of plants grown under high CO 2 . The authors are unaware of any published evidence demonstrating an effect of H 2 O 2 on NHX expression or activity but it is possible that H 2 O 2 may regulate NHX expression indirectly via the SOS pathway. It has already been shown that over-expression of SlSOS2 in transgenic tomato plants confers salt tolerance by up-regulation of NHX genes (Huertas et al., 2012).
The mechanisms by which changes in apoplastic H 2 O 2 levels regulate the abundance of salt response transcripts are Fig. 7. Effects of NaCl treatments on the ion location and content in the vascular tissue of pTRV empty vector control and RBOH1 silenced tomato plants grown under elevated (Ele: 760 µmol mol -1 ) or ambient (Amb: 380 µmol mol -1 ) CO 2 conditions. (A) Cellular localization of Na + in roots. Confocal images of roots stained with CoroNa Green are shown. Samples were imaged 3 d after salt treatment. (B) Na + and K + content, and the Na + :K + ratio in xylem sap.
unknown. However, exposure to a range of abiotic stresses, including salt stress, generally induces ABA biosynthesis and accumulation (Tuteja, 2007), which regulates SOS2 gene expression (Ohta et al., 2003) and activates MAPKs such as MPK6 via RBOH-dependent H 2 O 2 production (Kovtun et al., 2000;Samuel et al., 2000;Lu et al., 2002;Zhang et al., 2006;Zhang et al., 2007;Xing et al., 2008). The C-terminal region of the Arabidopsis SOS1 gene is phosphorylated by MPK6 under high NaCl conditions (Yu et al., 2010). It has been previously shown that RBOH1-triggered H 2 O 2 accumulation activates MPK1/2 in tomato leaves and that silencing of RBOH1 impairs stress-induced ABA synthesis (Zhou Fig. 8. Effects of NaCl treatments on the transcript levels of genes involved in the SOS pathway, NHXs, and MAPKs in the pTRV empty vector control and RBOH1 silenced tomato plants grown under elevated (Ele: 760 µmol mol -1 ) or ambient (Amb: 380 µmol mol -1 ) CO 2 conditions. (A) Transcript levels in leaves. (B) Transcript levels in roots. Gene expression was evaluated 2 d after salt treatment. The data are means ± SD of four replicates. Bars denoted by the same letter did not differ significantly according to Tukey's test (P<0.05). et al., 2014b). MAPK3/6 are involved in salinity tolerance in rice (Li et al., 2014). Here it is shown that the levels of MAPK1/2 and MAPK3 transcripts, which are homologues of the Arabidopsis MAPK6 and MAPK3 genes, respectively, were higher in plants grown under conditions of salinity and high CO 2 (Fig. 8). These findings suggest that MAPK signalling cascades are involved in the increased expression of SOS genes under these conditions. High CO 2 -induced ROS accumulation may also increase free cytosolic Ca 2+ concentrations (Foreman et al., 2003), which may in turn lead to an improved cellular Na + /K + balance by enhancing Na + extrusion and maintaining K + influx (Demidchik et al., 2002;Zhu, 2002;Munns and Tester, 2008). These factors may also contribute to the higher tolerance to salt stress that was observed under high CO 2 conditions in this study. Moreover, an increase in RBOH-dependent H 2 O 2 production has been proposed as a mechanism to increase HKT-mediated Na + unloading from the xylem sap (Jiang et al., 2012). In turn, this process may trigger an antioxidant response to police ROS metabolism and prevent further increases in ROS accumulation, resulting in a further mitigation of the negative impacts of salt stress (Ben et al., 2014).
In summary, the findings reported here demonstrate that high CO 2 concentrations can counteract the negative impact of salt stress on photosynthesis and biomass production in an apoplastic H 2 O 2 -dependent manner in tomato plants. The regulation of apoplastic H 2 O 2 levels also makes a major contribution to the regulation of Na + transport from roots to shoots, a process that is influenced by stomatal movement and by Na + delivery from the xylem to leaf cells. Regulated changes in apoplastic H 2 O 2 levels may therefore play an important role in triggering the mechanisms that underpin salt tolerance and associated detoxification pathways.

Supplementary data
Supplementary data are available at JXB online. Table S1. Gene-specific primers designed for qRT-PCR. Fig. S1. Gas exchange in tomato plants in response to salt treatments and elevated CO 2 levels. Fig. S2. H 2 O 2 accumulation in response to salt treatments and elevated CO 2 levels. Fig. S3. Quantification of H 2 O 2 in salt-treated plants under ambient and elevated CO 2 levels. Fig. S4. Relative RBOHs transcript abundance and NADPH oxidase activity in the leaves from RBOH1-silenced tomato plants. Fig. S5. Quantification of H 2 O 2 in salt-treated pTRV control and RBOH1 silenced (pTRV-RBOH1) tomato plants under ambient and elevated CO 2 level.