Ectopic overexpression of the cell wall invertase gene CIN1 leads to dehydration avoidance in tomato

Highlight Overexpression of a cell wall invertase gene promotes dehydration avoidance and WUE, and delays senescence in tomato through limiting stomatal conductance and inducing sink metabolism and photosynthesis in source leaves.


Introduction
The prospect of global warming coinciding with the continuous increase in the human population is projected to have a significant impact on the demands imposed on contemporary agricultural production. Abiotic stresses in general, and, more specifically, drought stress, have a strong negative impact on crop yield, because they create suboptimal growth conditions. Therefore, drought tolerance has attracted much attention and investment from private, public, academic, and philanthropic sectors (Marris, 2008;Lybbert and Bell, 2010). There is an urgent need to generate plants that are optimized in terms of tolerance to reduced water quantity and quality to support growth and development (Lybbert and Bell, 2010), which most probably cannot be achieved using only traditional methods of plant breeding. Drought stress leads to physiological modifications such as reduced photosynthesis, transcriptional and post-transcriptional regulation of various genes, and osmolyte biosynthesis (Seki et al., 2007;Shinozaki and Yamaguchi-Shinozaki, 2007). Classic genetic engineering approaches involved target genes that function in mechanisms used by plants to avoid and/or tolerate drought, such as stomatal conductance, ionic homeostasis, or osmolyte production (Hirayama and Shinozaki, 2010), which, however, typically resulted in a trade-off causing reduced fitness or yield penalties under normal growth conditions. Such genes, frequently identified through expression profiling, include signalling components and downstream effector genes (Skirycz et al., 2011). However, the advances provided by modern plant molecular biology and genetics could identify a new class of target genes (Webster et al., 2012) linked to useful traits such as resistance to drought, salt, and other abiotic stress conditions, without negative impacts on plant fitness.
Plant fitness is intrinsically linked to the ability to produce and consume carbohydrates in a tissue-specific manner. In higher plants, growth and metabolism of sink tissues is sustained by the carbohydrates synthesized in source leaves, which are transported mainly as sucrose through the phloem into the sink tissues (Koch, 2004;Roitsch and González, 2004). Source-sink relationships are dynamic and change during development and in response to different biotic and abiotic stresses (Roitsch, 1999;Balibrea et al., 2000Balibrea et al., 2003Roitsch and González, 2004;Albacete et al., 2014b). In general, under abiotic stress conditions, the competition between different physiological processes and sink organs for the limited carbon supplies leads to a reduction in the sink strength, affecting overall plant growth and crop yield (Cuartero and Fernández-Muñoz, 1998). The use of sucrose in the sink tissues requires cleavage of the glycosidic bond, catalysed by both sucrose synthase and invertases. Three types of invertase isoenzymes are distinguished based on solubility, subcellular localization, pH optima, and isoelectric point: vacuolar invertase (vacInv), cytoplasmic invertase (cytInv), and cell wall-bound invertase (cwInv) (Roitsch and González, 2004). CwInv has been shown to play a crucial role in plant development by regulating sink strength, ensuring the steady supply of photoassimilates to sink tissues (Tang et al., 1999;Goetz et al., 2001;Roitsch et al., 2003;Weschke et al., 2003). Under water stress, differences in storage carbohydrate accumulation in drought-sensitive and drought-tolerant wheat were correlated with differences in sugar profiles, expression of cwInv genes, and levels of fructan biosynthesis in the anther and ovary (Ji et al., 2010). Also, Koonjul et al. (2005) reported that transitory water deficit in wheat during male meiosis selectively down-regulated the transcription of two genes encoding a vacuolar (Ivr5) and a cell wall (Ivr1) invertase isoform in the anthers. It has been suggested that pollen sterility, or the concomitant inhibition of starch accumulation in water-stressed rice plants, is unlikely to be caused by carbohydrate starvation per se. Instead, an impairment of enzymes of sugar metabolism and starch synthesis may be among the potential causes of this failure (Sheoran and Saini, 1996). Nevertheless, the physiological mechanisms involved in altered assimilate partitioning and tolerance towards abiotic stress by invertases remain unclear (Albacete et al., 2011(Albacete et al., 2014bPérez-Alfocea et al., 2010).
The increase of sink strength through cwInv activity is a general response under stress conditions. In particular, expression of the CIN1 cwInv gene from C. rubrum has been reported to increase in response to a number of different stress-related stimuli (Roitsch and Ehness, 2000). In suspension-cultured cells, the inducing effect of the fungal elicitor chitosan could be mimicked by phosphatase inhibitors and benzoic acid (Ehness and Roitsch, 1997). Mechanical wounding of source leaves of Chenopodium rubrum plants also resulted in increased CIN1 mRNA levels, showing the wide range of stress-related stimuli affecting CIN1 expression (Roitsch et al., 1995). In addition, cwInv is regulated at transcriptional and post-translational levels by many different factors including sugars (Roitsch et al., 1995(Roitsch et al., 2003Roitsch, 1999), phytohormones (Roitsch et al., 2003;Balibrea et al., 2004;Roitsch and González, 2004), and proteinaceous inhibitors (Rausch and Greiner, 2004;Bonfig et al., 2010;McKenzie et al., 2013). Together, these mechanisms allow a fine-tuned regulation of the cwInv activity to control growth, development, and, therefore, plant adaptation under abiotic stress conditions. CwInv thus works as a pivotal enzyme at the integration point of metabolic, hormonal, and stress signals (Proels and Roitsch, 2009). In this study, it is shown that overexpression of the cwInv gene CIN1 in tomato dramatically increases whole-plant water use efficiency (WUE) under water stress conditions, through a strategy of dehydration avoidance, providing a novel approach to overcome droughtinduced limitations to crop productivity.

Materials and methods
InvLp6g::CIN1 construction and overexpression The full-length 1.7 kb CIN1 cDNA under the control of a 2.5 kb fragment of the promoter of vacuolar invertase pInvLp6g from Solanum pimpinellifolium (Elliott et al., 1993) (GenBank accession no. Z12028.1) was cloned into the vector pBI101. After transfer into the Agrobacterium tumefaciens strain LBA4404, cotyledons from the cv. P-73 of tomato (Solanum lycopersicum L.) were transformed with the CIN1 overexpression construct. T 2 plants from five different transgenic lines containing the InvLp6g::CIN1 construct were identified as transgenic (homozygous or heterozygous) or azygous for the T-DNA based on the presence of the marker gene NPTII that confers resistance to the antibiotics kanamycin and neomycin determined by PCR. Subsequently, the homozygous or heterozygous transgenic state of the (PCR positive) T 2 plants was determined in their respective T 3 progeny by PCR.

Plant growth conditions
Wild-type plants from the P-73 cultivar (WT) were used as controls. Tomato seeds were germinated at 28 °C and 90% relative humidity, and grown under 25/18 °C (day/night temperatures), 16 h light (245 μmol m -2 s -1 ) and a relative humidity of 60-70%. At 15 d after sowing, 12 plants of each transgenic line and the WT were transferred to 10 litre pots filled with peat and grown for 14 d. At this point, watering was withheld for a period of 9 d. Three plants per line were irrigated at field capacity during this period and used as controls.

CIN1 expression analysis
Fresh tissues from mature leaves, roots, seedlings, and fruits were used for total RNA isolation, and 1 μg of total RNA was used for first-strand cDNA synthesis according to standard methods, using oligo(dT) primers. Semi-quantitative real-time PCR (RT-PCR) using actin to normalize the obtained cDNA amounts was performed as described previously (Großkinsky et al., 2011). For CIN1 expression analyses, the primers CIN1-Forward (5'-CCTGGGAGTATAGTGGCTGAACC-3') and CIN1-Reverse (5'-AGGTCTTCTCTGAATCCG-3') were used.

Soil water potential and relative water content
Measurements of the soil water potential were done with a Watermark Soil Moisture Meter. Leaf relative water content (RWC) was determined as: RWC=(fresh weight-dry weight)/(turgid weightdry weight). To determine the turgid weight, leaves were kept in distilled water in darkness at 4 ºC to minimize respiration losses until they reached a constant weight (full turgor, typically after 24 h).

Sucrolytic and other carbon metabolism enzyme assays, and invertase inhibitor activity
Sucrolytic and other carbon metabolism enzyme activities were assayed by determining the NADH delivered in a coupled enzymatic reaction using specific substrates/enzymes depending on the target enzyme (Balibrea et al., 1999. The absorbance was monitored at 340 nm. The proteins were analysed with Bradford reagent using bovine serum albumin (BSA) as standard. The invertase inhibitor assay was performed as previously described (Bonfig et al., 2010).

Sugar determination
A 100 mg aliquot of leaf plant material was ground in liquid nitrogen and 0.9 ml of water was added. After homogenization with cationic and anionic exchange resins and centrifugation for 10 min at 20 000 g and 4 ºC, the supernatant was filtered and 10 μl were injected in a normal-phase liquid chromatography system (Shimadzu Corporation, Kyoto, Japan), using acetonitrile/water (85/15, v/v) as the mobile phase at a flow rate of 1 ml min -1 .

CO 2 exchange measurements
Gas exchange measurements were conducted in the fifth fully expanded leaf in each genotype with a gas exchange system (LI-6400; Li-Cor, Lincoln, NE, USA). Leaves were first equilibrated at a photon density flux of 500 μmol m -2 s -1 for at least 2 min. After this, photosynthesis was induced with a photon density flux of 1000 μmol m -2 s -1 and 400 μmol mol -1 CO 2 surrounding the leaf (C a ). Leaf temperature was maintained at 25 ºC, and the leaf to air vapour pressure deficit was kept between 1 kPa and 1.3 kPa for the determination of the photosynthetic rate (A).

Transmission electron microscopy
For ultrastructural studies, small pieces of mature leaves were fixed in 2.5% glutardialdehyde/2% paraformaldehyde in 0.06 M cacodylate buffer (pH 7.2) for 90 min, post-fixed, dehydrated, and embedded as previously described (Zechmann et al., 2007). Ultrathin sections (80 nm) were investigated after post-staining with uranyl aceate and lead citrate with a Philips CM10 transmission electron microscope.

Accumulated transpiration and water use efficiency
Transpired water was measured gravimetrically by daily weighing the potted plants during the experiment. A pot with the same amount of soil but without a plant was weighed daily and used as a reference to determine the amount of water evaporated. WUE was determined as the biomass generated during the drought period (in grams) divided by the accumulated transpiration during that period (in millilitres).

Chlorophyll fluorescence
Modulated chlorophyll fluorescence was measured in tagged and dark-adapted (30 min) leaves, using a chlorophyll fluorometer OS-30 (OptiSciences, Herts, UK) with an excitation source intensity of 3000 μmol m -2 s -1 . A special version of an Imaging-PAM Chlorophyll Fluorometer (Walz) was used to investigate spatio-temporal changes in photosynthetic parameters (Schreiber, 2004).

Antioxidant enzymes
The leaf apoplastic fraction was isolated by vacuum infiltration in the presence of 50 mM TRIS-acetate buffer pH 6.0. Samples were concentrated and pre-purified by chromatography on Sephadex G-25 NAP-10 columns (GE Healthcare) (Hernández et al., 2001). Leaf residues (2 g), which resulted from the apoplastic extraction, were homogenized using a mortar and pestle in 4 ml of ice-cold 50 mM TRIS-acetate buffer pH 6.0 containing 0.1 mM EDTA, 2 mM cysteine, and 0.2% (v/v) Triton X-100, and used as the symplastic fraction. Peroxidase (POX) and superoxide dismutase (SOD) activities were assayed as described previously (Hernández et al., 2001).

Glutathione and electrolyte leakage determination
Total and oxidized glutathione were extracted and analysed by highperformance liquid chromatography (HPLC) as described previously (Kranner and Grill, 1995). For electrolyte leakage measurements, tomato leaves (2 g) were cut into pieces (~2 cm 2 ) and incubated in 8 ml of MilliQ water in sealed tubes, for 2 h at room temperature. After incubation, the conductivity of the bathing solution was measured with a conductivity meter. This value was referred to as value A. The bathing solutions were returned to the sealed tubes, containing the pieces of leaves, which were then incubated in a water bath at 95 ºC for 25 min. After cooling to room temperature, the conductivity of the bathing solution was measured again. This is referred to as value B. For each measurement, electrolyte leakage was expressed as percentage leakage: [(value A/value B)×100].

Proteomic analysis
Two-dimensional DIGE minimal labelling (Alban et al., 2003) was used to identify leaf apoplast protein abundance differences. Gel image analysis was performed using Progenesis SameSpots v3.0 (Nonlinear Dynamics, Newcastle, UK) as described previously (Martínez-Esteso et al., 2011). Spots whose normalized volume (% total spot volume) increased or decreased according to the treatment across the experiment were selected based on analysis of variance (ANOVA; P<0.05). Selected spots were manually excised from the gels and processed for identification by matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF) and liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) in the PROTEORED © proteomic facility at the University of Alicante (Spain). Protein in-gel digestion and identification by MALDI-TOF or LC-MS/MS and database search were done as described previously (Martínez-Esteso et al., 2009).

Metabolite profiling
Metabolite profiling was performed using gas chromatography coupled to a LECO Pegasus IV time-of-flight (Leco Corp Inc., St. Joseph, MI, USA) mass analyser (GC-TOF-MS) as previously described (Scherling et al., 2009).

Hormone extraction and analysis
Hormones were analysed as described previously .

Statistics
All experiments were repeated three times, and the results of one representative experiment are presented in each case. Data were subjected to an ANOVA using the SPSS software (Version 19.0, SPSS Inc., Chicago, IL, USA). The statistical significance of the results was analysed by Student-Newman-Keuls test at the 5% level.

CIN1 overexpression increases water use efficiency and photosynthesis under drought stress
It has recently been shown that transgenic tomato plants overexpressing the cwInv gene CIN1 from C. rubrum under the control of the putative fruit-specific promoter InvLp6g recover sink strength and fruit growth under suboptimal conditions imposed by salinity (Albacete et al., 2014a). Although the InvLp6g promoter should predominantly confer expression in tomato fruits, semi-quantitative RT-PCR analyses for five selected independent homozygous T 3 lines revealed that the CIN1 transgene is also expressed in seedlings and in leaves from plants grown under control conditions (Fig. 1A). Therefore, the effect of CIN1 overexpression on whole-plant physiology was also investigated to distinguish between fruit-specific and systemic, general changes in physiology. Furthermore, overexpression of CIN1 dramatically increased cwInv activity in the leaves of transgenic lines compared with the WT under normal watering regimes (Fig. 1B). Therefore, invertase activity was used as a more direct and specific indicator of the 'performance' of the transgenics rather than CIN1 expression. Consequently, three groups of plant lines could be considered: (i) lines CIN1-12, CIN1-91, and CIN1-93 which all behave similarly to high cwInv activity lines; (ii) line CIN1-10 as an intermediate cwInv activity line; and (iii) line CIN1-8 as a low activity line, similar to the WT despite showing CIN1 expression.
Transgenic and WT plants were subjected to drought stress conditions by withholding water at the vegetative stage. Although drought stress resulted in a general decrease of the cwInv activity, transgenic plants showed significantly higher cwInv activity, similar to or even higher than that of the WT plants under control conditions (Fig. 1B). After 7 d of drought stress, CIN1 plants remained turgid whereas WT plants partially wilted (Fig. 1C, left), and the same effect was observed at the flowering stage 2 d after withholding water (Fig. 1C, right). This strong difference in drought tolerance was even more apparent in seedlings after 15 d of withholding water under uncontrolled conditions in the greenhouse (Fig. 1D, left) and re-watering for 1 d (Fig. 1D, right). Shoot dry weight and leaf area did not differ significantly between WT and transgenic tomato plants growing under drought stress (Fig. 1E, F). Measurements of the root zone water potential ( Fig. 2A) revealed that the WT and the CIN1-8 line dried the substrate faster than the other CIN1 lines. Significant differences were observed from day 5 without watering onwards until the end of the drought experiment. The CIN1-8 line behaved as WT plants in all subsequent experiments and served as an aphenotypic transgenic control line. Unlike other abiotic stresses, water availability is directly related to productivity through the maintenance of healthy leaves. Although the RWC was reduced during the drought period (Fig. 2B), transgenic plants were superior to WT plants in maintaining soil water potential and consequently leaf RWC, even though they partially wilted.
To gain additional insights into the effect of cwInv in plants expressing pInvLp6g::CIN1, accumulated wholeplant transpiration, WUE, stomatal conductance (g s ), chlorophyll fluorescence (F v /F m ), and photosynthetic rate (A) were measured and compared. Accumulated transpiration measured during the drought period was significantly reduced in most of the transgenic lines (by 30%) with respect to the WT (Fig. 2C). Plants expressing pInvLp6g::CIN1 exhibited a significantly increased WUE (30-50%), measured as the ratio of biomass produced to water used (Fig. 2D). Interestingly, under control conditions, g s was significantly lower in the transgenic plants (up to 50% in line CIN1-93) compared with the WT plants. In contrast, at the end of the stress period, g s was significantly higher in the CIN1 plants, due to increased water content in the substrate (Fig. 2E).
Furthermore, regarding photosynthetic parameters, analyses of the spatio-temporal changes in (Fig. 3A) and absolute values of (Fig. 3B) F v /F m revealed that chlorophyll fluorescence was higher in the CIN1 plants than in the WT throughout the drought period. In agreement with this, the photosynthetic rate (A) was less reduced in CIN1 plants (30%) than in the WT plants (75%) and the aphenotypic line CIN1-8 (60%) (Fig. 3C). Interestingly, ultrastructural analysis by transmission electron microscopy revealed an increase in sugar storage since CIN1-91 leaves presented distinct and larger starch grains than the WT under drought stress (Fig. 3D).

CIN1 expression affects sugar and antioxidant metabolism
In contrast to cwInv (Fig. 1B), the activity of the other two invertase isoenzymes, vacInv and cytInv, was not or only weakly affected (Fig. 4A, B); the sucrose synthase activity was significantly lower in the transgenic plants under control conditions (Fig. 4C). Despite the increased cwInv activity in the CIN1 plants, fructose and glucose contents in the leaf were lower than those of the WT and the aphenotypic line CIN1-8 under drought stress conditions (Fig. 4D). Invertase inhibitor activity was significantly higher in CIN1 plants than in the WT under normal watering regimes, especially in line CIN1-93 with the highest cwInv activity (Fig. 4E). Under  drought stress, a significant increase of the invertase inhibitor activity was observed in the WT and, to a lower extent, in the CIN1-8 and CIN1-10 lines, while in the other CIN1 lines, a decrease was detected (Fig. 4E). To determine the effect of the transgene expression on metabolism, in addition to the sucrolytic activities, a set of nine additional key enzymes of primary carbohydrate metabolism was tested (Fig. 5). In general, these enzyme activities were lower under drought conditions. With the exception of phosphoglucoisomerase, all other activities were reduced in the transgenic plants under control conditions compared with the WT, although to a different extent. Under drought, the transgenic plants were characterized by higher activities of the glycolytic enzymes aldolase, phosphofructokinase, phosphoglucoisomerase, phosphoglucomutase, and UDP-glucose-pyrophosphorylase (Fig. 5A-E). The activity of glucose-6-phosphate dehydrogenase was lower in the CIN1 plants under both control and drought stress conditions (Fig. 5F). Hexokinase, fructokinase, and ADP-glucosepyrophosphorylase activities did not show consistent changes under water stress conditions (Fig. 5G-I).
During stress, disruption of cellular homeostasis is accompanied by the generation of reactive oxygen species (ROS), and the extent of stress-induced damage can be attenuated by the action of the cell's antioxidant systems, including glutathione and enzymes capable of scavenging ROS. The oxidized state of glutathione increased during drought in the WT and the aphenotypic line CIN1-8, but decreased in the other transgenic CIN1 lines (Fig. 6A). Apoplastic POX and SOD activities were strongly reduced by drought stress in the WT but were maintained or induced in CIN1 plants (Table 1). This was related to the performance of the different CIN1 lines and the electrolyte leakage (membrane stability) during the drought period. In this sense, cell membrane damage under drought stress was more severe in the WT and the aphenotypic CIN1-8 line than in the other CIN1 lines analysed (Fig. 6B). Proteomic analysis ( Fig. 7; Table 2) confirmed a high expression for one POX isoform (spot 618) in CIN1 plants and other proteins related to plant stress defence responses: chitinase (spot 845) and proteases of the subtilisin-like clan (spots 393 and 414).

Changes in leaf hormonal balance
Leaf concentrations of the active cytokinin (CK) trans-zeatin (tZ) and the ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC) were analysed in control and waterstressed plants at the end of the experiment. Although tZ levels were reduced under drought stress, CIN1 lines that showed a strong phenotype also presented a significant increase in tZ compared with WT plants (Fig. 8A). Drought stress resulted in a general increase in the ACC concentrations in WT and CIN1 plants, but remained significantly lower (by 25%) in the CIN1 lines, except for the aphenotypic CIN1-8 line (Fig. 8B). While auxin levels were slightly increased in the transgenic plants (Fig. 8C), no clear trend was evident for the abscisic acid (ABA) levels (Fig. 8D).

CIN1 expression modifies the metabolite profile under drought stress
To characterize further the mechanism of drought stress tolerance in CIN1 plants, metabolite profiling was performed. This identified 100 compounds, of which 22 showed significant changes due to drought stress conditions (Table 3). CIN1 plants showed specific accumulation of phosphorylated sugar intermediates glucose-6-phosphate and fructose-6-phosphate, organic acids, and phenolic compounds. Five out of six mono-and disaccharides and a sugar-alcohol were reduced in the leaves of transgenic lines under drought conditions, whereas amino acids showed a differential response (Table 3). This set of metabolites was subjected to principal component analysis (PCA). Although the CIN1 plants clustered together with the WT plants under control conditions, at the end of the drought period CIN1 plants clearly clustered separately from WT plants (Fig. 9). The fact that the stressed CIN1 plants clustered in between plants under control conditions and stressed WT plants for the principal component 1 (PC1) indicates that CIN1 plants are less affected by drought stress conditions (Fig. 8).

Discussion
Although plant responses to drought stress have been intensively studied for many years, progress in the development of tolerant crops has been slow. A number of genes, including those involved in the production of osmotic adjustment, detoxification of ROS, and transcription factors, have been employed to increase drought tolerance in transgenic plants (Munns and Tester, 2008;Faize et al., 2011;Skirycz et al., 2011). However, the success rate is typically very low, probably due to the lack of knowledge about the mechanisms that are controlling growth under water stress conditions. In general, adaptation or engineered tolerance to drought is accompanied by adverse effects on development and yield. Therefore, maintaining growth, photosynthesis, and metabolism are the major goals to minimize the impact of abiotic stress on crop yield. Here it is shown that ectopic expression of the CIN1 gene encoding a cwInv from C. rubrum resulted in various metabolic changes that improved tomato adaptation to drought stress conditions by increasing WUE.
Many studies demonstrated that decreased stomatal conductance and transpiration under drought stress conditions is the main factor limiting photosynthesis (Chernyad'ev, 1997;Chaves et al., 2009;Galmés et al., 2011), and assimilate transport in phloem sets conditions for gas exchange (Nikinmaa et al., 2013). The present results show that CIN1 plants had limited whole-plant transpiration during drought stress, and importantly WUE increased (Fig. 2). In fact, under control conditions, stomatal conductance was reduced in the CIN1 plants (Fig. 2E), leading to reduced water consumption during the drought period, while photosynthetic activity was maintained ( Fig. 3A-C). In contrast, the increased stomatal conductance observed in the transgenic plants at the end of the drought period is explained by the higher water availability in the soil, compared with the WT plants. Therefore, it seems that CIN1 plants present a better control of stomatal closure explained by metabolic factors since it has been demonstrated that sugar levels and sucrolytic activities play important roles in guard cell function with impacts on WUE (Antunes et al., 2012). Indeed, it has been reported that vacuolar invertase activity was higher in guard cells than in other epidermal cells and it plays an important role in regulating stomatal aperture in Arabidopsis (Ni, 2012). Therefore, the increased cwInv activity (Fig. 1B) and hexose concentrations  Data are presented as means ±SE, *P<0.05, one-way ANOVA, n=3.

Table 1. Superoxide dismutase (SOD) and peroxidase (POX) activities (U g -1 FW), and total peroxidase activity (expressed as a percentage of the control) in tomato leaves of WT and CIN1 plants under normal and drought stress conditions
( Fig. 4D) observed in the leaves of CIN1 plants under normal watering regimes could provoke a reduction in the stomatal conductance retaining water for the drought period. Similarly, it has been reported that overexpression of the Arabidopsis trehalase AtTRE1 gene, the only enzyme known in this species specifically to hydrolyse trehalose into glucose, leads to increased drought stress tolerance through an induction of stomatal closure (Van Houtte et al., 2013). Other trehalose enzymes and intermediates, such as trehalose-6-phosphate, have also been linked to abiotic stress tolerance in tomato (Delorge et al., 2014). In contrast, the lower stomatal conductance observed in the WT plants at the end of the drought period can be explained by the higher level of stress due to reduced soil moisture compared with CIN1 plants (Fig. 2A).
After the initial assimilate accumulation under stress, photosynthesis may be inhibited by metabolic factors, stomatal-derived signals, and other regulatory mechanisms, while stress-induced leaf senescence limits whole-plant photosynthesis (Munns and Tester, 2008), as seems to be the case in the WT and the aphenotypic CIN1-8 line (Fig. 3A-C) due to the expected hexose accumulation (Fig. 4D). Avoiding feedback inhibition of photosynthesis by co-ordinating the assimilate transport between source and sink tissues by invertases and/ or partitioning of sugars towards starch accumulation may maintain photosynthetic activity under harmful environmental conditions (Stitt, 1991;Koch, 1996) and thus delay leaf senescence. Recently, it has been shown that engineered drought tolerance in tomato is reflected in chlorophyll fluorescence emission signatures (Mishra et al., 2012), and similar changes were also observed in the CIN1 plants (Fig. 3A,  B). An increase in the concentrations of the invertase products fructose and glucose would be expected, as has been recently shown in rice plants that overexpress the cwInv gen GIF1 related to induced pathogen defence (Sun et al., 2014). Strikingly, despite higher cwInv activities (Fig. 1B), fructose and glucose contents in droughted leaves of CIN1 plants were lower than those in the WT, but similar to those of control leaves (Fig. 4D). These results resemble the effect of ectopic expression of cwInv in delaying natural or light-induced senescence in transgenic tobacco plants (Balibrea et al., 2004). The unexpected finding that an increase in cwInv in CIN1 plants does not result in an increase in hexose steadystate concentrations would explain the maintenance of photosynthetic activity despite the activation of sink metabolism. Therefore, the delay of senescence induced by cwInv in CIN1 plants may be related to an activation of the metabolic carbohydrate flux. The resulting higher rate of sugar utilization causes a decrease of hexose levels in the transgenic plants (Balibrea et al., 2004). Thus, despite the activation of sink metabolism, it seems that the hexose concentration does not reach the threshold level that would result in the feedback inhibition of photosynthetic gene expression. This provides a mechanism to uncouple the usually observed inverse and co-ordinated regulation of source and sink metabolism. The apparent stimulation of carbohydrate fluxes is also reflected by the finding that the transgenic plants were able to maintain higher activities of additional key enzymes of primary carbohydrate metabolism under drought stress conditions, such as aldolase, phosphoglucomutase, phosphofructokinase, phosphoglucoisomerase, and UDP-glucose pyrophosphorylase (Fig. 5). Those enzymes are involved in sucrose and starch biosynthesis from triose-phosphate resulting from the higher photosynthetic activity, thus increasing the source activity. For example, it has been demonstrated that aldolase, a key  enzyme for energy production, plays important roles in the tolerance to drought and salt stress by activating glycolysis (Fan et al., 2009). Thus, CIN1 plants were able to maintain metabolic fluxes of primary carbohydrate metabolism in support of growth and energy production for stress tolerance compared with WT plants, also reflected by higher levels of the phosphorylated intermediates fructose-6-phosphate and glucose-6-phosphate (Table 3; Fig. 9). Evidence for carbon flux shortage under water stress in sink organs (nodules) has been shown in pea plants (Gálvez et al., 2005). Apparently, drought tolerance is related to a fine-tuned interaction and balance between extracellular hydrolysis by the cwInv and the metabolic flux of the sink cell to avoid the accumulation of carbohydrates under harmful conditions. Furthermore, the large starch grains in leaves from CIN1 plants (Fig. 3D) indicate that changing the cycle of starch synthesis and breakdown and thus metabolic channelling could be an additional mechanism to contribute to drought stress tolerance in CIN1 plants (Sulpice et al., 2009) in order to maximize carbon uptake and growth, while minimizing osmotic impacts on photo-inhibition and photo-oxidation. Photo-oxidative damage and premature leaf senescence are common effects of abiotic stress challenges, inhibiting growth (Munns, 2002). Photo-inhibition and photo-oxidation are caused by the impaired consumption of NADPH by the Calvin cycle, with the subsequent transfer of photosynthetic electrons from over-reduced ferredoxin to oxygen (Mehler reaction) producing ROS that damage cell structures (Stitt, 1991;Paul and Foyer, 2001). SOD and POX activities were maintained in CIN1 plants under drought stress (Table 1), especially apoplastic SOD, which agrees with the reported accumulation of superoxide radicals under salt stress conditions in this compartment (Hernández et al., 2001). It has been reported that glucose and CK agonistically regulate POX activity in Arabidopsis (Kushwah and Laxmi, 2014), thus explaining the higher and stable levels of this protein in the CIN1 plants (Table 2). Furthermore, many reports indicate that the ratio between reduced and oxidized glutathione (GSH/GSSG) is an effective marker of cellular redox homeostasis and may be involved in ROS activity perception by plants under drought conditions (Labudda and Azam, 2014). In this way, GSH/GSSG play a direct or indirect key role in regulating and signalling at the transcriptional and/or posttranslational level due to the interaction of these molecules with other cellular redox systems such as the aforementioned enzymes SOD and POX (Anjum et al., 2012), as well as with the hormonal balance (Miao et al., 2006). Maintaining the activity of these two ROS-scavenging enzymes, together with decreased concentrations of oxidized glutathione (Fig. 6A) and reduced electrolyte leakage (Fig. 6B), resulted in a better control of ROS levels and sustained membrane protection in drought-stressed CIN1 plants, such as reported in the senescence-regulated CK-overproducing (SARK2::IPT) tobacco plants in response to drought (Rivero et al., 2007). A possible link between antioxidants and carbohydrate metabolism could be the glucose-6-phosphate dehydrogenase, which is essential for maintaining the cellular redox balance and shown to be critically involved in salt and drought stress responses (Dal Santo et al., 2012). Possibly due to the posttranslational regulatory mechanism involved in the activation of this enzyme, the measured activities do not reflect the in vivo situation.
Sugar and hormones such as CKs are fundamental to plants and regulate a number of similar processes agonistically (Kushwah and Laxmi, 2014). It has been recently reported that glucose induces CK biosynthetic (IPT3) and perception (AHK4) genes, while it represses some CK-degrading enzymes (CKX5) in Arabidopsis (Kushwah and Laxmi, 2014). Ectopic expression of CIN1 under the control of the senescence-associated promoter SAG12 increased source strength and delayed developmentally regulated leaf senescence in tobacco, and identified cwInv as an essential component of the CK-mediated delay of senescence (Balibrea et al., 2004). More recently, the extreme drought tolerance observed in tobacco plants overexpressing the CK biosynthesis gene IPT was also related to delayed senescence due to increased CK levels, without affecting ABA levels (Rivero et al., 2007), as it occurred in the CIN1 plants (Fig. 8D). Furthermore, rice plants overexpressing the IPT gene showed increased water stress tolerance and grain yield due to CK-mediated source/sink modifications (Peleg et al., 2011). Therefore, delayed leaf senescence of CIN1 plants under drought stress could be explained by both higher cwInv activity (Fig. 1B) and higher  tZ levels in CIN1 leaves (Fig. 8A), thus mimicking the IPToverexpressing plants. Additionally, stressed CIN1 plants showed a significant decrease in the ethylene precursor ACC (Fig. 8B). Ethylene has been long considered a stress-related hormone that mediates leaf senescence under abiotic stress conditions (Munné-Bosch and Alegre, 2004), probably by inhibiting cwInv . Because cwInv appears to activate metabolic carbohydrate fluxes (Balibrea et al., 2004) that repress ethylene biosynthesis (Mayak and Borochov, 1984), probably through glucose suppression of ACC oxidase activity (Hong et al., 2004), leaf senescence could additionally be delayed by the reduction of ethylene levels. Together these data indicate that the physiological changes conferred by CIN1 overexpression during the vegetative stage are closely connected with hormonal (CKs and ethylene, but not ABA) and sugar metabolism, suggesting that a causative effect of cwInv on the regulation of plant hormonal balance and function cannot be ruled out, as has been recently demonstrated in fruits of tomato plants subjected to salt stress (Albacete et al., 2014a).
Post-translational relief of invertase from inhibition by a proteinacous inhibitor has been implicated in abiotic stress tolerance (Ruan et al., 2010). The in vivo functionality and the physiological significance of these inhibitors in plant growth, development and stress responses have been demonstrated only recently (Jin et al., 2009;Bonfig et al., 2010). Such posttranslational regulation is particularly relevant to cwInv because these proteins are intrinsically stable due to their glycosylation (Rausch and Greiner, 2004). Increased invertase inhibitor activity in the CIN1 plants under control conditions probably reflects feedback regulation to decrease the ectopic transgenic cwInv activities and minimize negative impacts on phloem loading (Fig. 4E). In contrast, drought stress resulted in a strong decrease of the invertase inhibition, thus linking the local induction and/or maintenance of the sink strength to stress responses by derepression of the invertase activity present.
In summary, the results show that although the promoter employed should confer predominantly expression in developing tomato fruits (Albacete et al., 2014a), a weak vegetative CIN1 expression was already sufficient to confer tolerance towards drought stress without affecting plant fitness under optimal growth conditions. A reduced water use and an increased source activity explained by changes in metabolic fluxes, and hormonal and redox status suggest that the physiological mechanisms regulated by cell wall invertase play a key role in abiotic stress adaptation. The results presented in this study are very promising because enhanced WUE and plant adaptation to drought stress would contribute to the development of new varieties with reduced yield penalties in the most economically important horticultural areas, where water resources are scarce.