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

This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details) Journal of Experimental Botany, Vol. 66, No. 3 pp. 863–878, 2015 doi:10.1093/jxb/eru448 Advance Access publication 11 November, 2014 ReseaRch PaPeR Ectopic overexpression of the cell wall invertase gene CIN1 leads to dehydration avoidance intomato AlfonsoAlbacete1,2, ElenaCantero-Navarro1, Dominik K.Großkinsky2,3, Cintia L.Arias4, María EncarnaciónBalibrea1, RoqueBru5, LenaFragner6, Michel E.Ghanem1, María de la CruzGonzález7, Jose A.Hernández8, CristinaMartínez- Andújar1, Ericvan der Graaff2,3, WolframWeckwerth6, GüntherZellnig2, FranciscoPérez-Alfocea1,* and ThomasRoitsch2,3,9 1 Department of Plant Nutrition, CEBAS-CSIC, Campus de Espinardo, 30100 Murcia, Spain 2 Institute of Plant Sciences, Department of Plant Physiology, University of Graz, 8010 Graz, Austria 3 Department of Plant and Environmental Sciences, Copenhagen Plant Science Centre, University of Copenhagen, Højbakkegård Allé 13, DK-2630 Taastrup, Denmark 4 Centro de Estudios Fotosintéticos y Bioquímicos, Universidad Nacional de Rosario, Suipacha 531, 2000 Rosario, Argentina 5 Departamento de Agroquímica y Bioquímica, Facultad de Ciencias, Universidad de Alicante, 03080 Alicante, Spain 6 Department of Molecular Systems Biology, Faculty of Life Sciences, University of Vienna, 1090 Vienna, Austria 7 Instituto de Bioquímica Vegetal y Fotosíntesis, Universidad de Sevilla, CSIC, 41092 Sevilla, Spain 8 Department of Fruit Breeding, CEBAS-CSIC, Campus de Espinardo, 30100 Murcia, Spain 9 Global Change Research Centre, Czech Globe AS CR, v.v.i., Drásov 470, Cz-664 24 Drásov, Czech Republic * To whom correspondence should be addressed. E-mail: [email protected] Received 3 July 2014; Revised 15 September 2014; Accepted 7 October 2014 Abstract Drought stress conditions modify source–sink relations, thereby influencing plant growth, adaptive responses, and consequently crop yield. Invertases are key metabolic enzymes regulating sink activity through the hydrolytic cleavage of sucrose into hexose monomers, thus playing a crucial role in plant growth and development. However, the physiological role of invertases during adaptation to abiotic stress conditions is not yet fully understood. Here it is shown that plant adaptation to drought stress can be markedly improved in tomato (Solanum lycopersicum L.) by overexpression of the cell wall invertase (cwInv) gene CIN1 from Chenopodium rubrum. CIN1 overexpression limited stomatal conductance under normal watering regimes, leading to reduced water consumption during the drought period, while photosynthetic activity was maintained. This caused a strong increase in water use efficiency (up to 50%), markedly improving water stress adaptation through an efficient physiological strategy of dehydration avoidance. Drought stress strongly reduced cwInv activity and induced its proteinaceous inhibitor in the leaves of the wild-type plants. However, the CIN1-overexpressing plants registered 3- to 6-fold higher cwInv activity in all analysed conditions. Surprisingly, the enhanced invertase activity did not result in increased hexose concentrations due to the activation of the metabolic carbohydrate fluxes, as reflected by the maintenance of the activity of key enzymes of primary metabolism and increased levels of sugar-phosphate intermediates under water deprivation. The induced sink metabolism in the leaves explained the maintenance of photosynthetic activity, delayed senescence, and increased source activity under drought stress. Moreover, CIN1 plants also presented a better control of production of reactive oxygen species and sustained membrane protection. Those metabolic changes conferred by CIN1 overexpression were accompanied by increases in the concentrations of the senescence-delaying hormone trans-zeatin and decreases in the senescence-inducing ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC) in the leaves. Thus, cwInv critically functions at the This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. © The Author 2014. Published by Oxford University Press on behalf of the Society for Experimental Biology. 864 | Albacete etal. integration point of metabolic, hormonal, and stress signals, providing a novel strategy to overcome drought-induced limitations to crop yield, without negatively affecting plant fitness under optimal growth conditions. Key words: Cell wall invertase, cytokinins, drought stress, ethylene, source–sink relationships, tomato. 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., 2000 2003; Roitsch 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-toler- ant 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., 2010 2011 2014b; Pérez-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 2003; Roitsch, 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 Cell wall invertase leads to dehydration avoidance in tomato | 865 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. T2 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) T2 plants was determined in their respective T3 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 weight– dry 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 2003). 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. CO2 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 CO2 surrounding the leaf (Ca). 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-tem- poral 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, 866 | Albacete etal. tomato leaves (2 g) were cut into pieces (~2 cm2) 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 (Albacete etal., 2008). 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. Results 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 T3 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 (gs), chlorophyll fluorescence (Fv/Fm), 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, gs 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, gs 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 Cell wall invertase leads to dehydration avoidance in tomato | 867 values of (Fig.3B) Fv/Fm 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 Fig.1. Expression and regulation of the CIN1 gene (A) and cell wall invertase activity (B) in mature leaves of WT and CIN1 plants under normal watering regimes and after 9 d of drought stress. Comparative images of WT and CIN1 lines at the vegetative (7 d of drought stress, left panel) and flowering (2 d of drought stress, right panel) stages (C), and between WT and CIN1 seedlings growing in a greenhouse under non-controlled conditions after 15 d of withholding water (left panel) and re-watering for 1 d (right panel) (D). Dry weight (E) and leaf area (F) of WT and CIN1 plants under normal watering regimes and after 9 d of drought stress. Data are presented as means± SE, *P<0.05, one-way ANOVA, n=3. 868 | Albacete etal. 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 Fig.2. Soil water potential (A) and leaf relative water content (B) of WT and CIN1 plants during the drought period. Accumulated transpiration (C), water-use efficiency (D), and stomatal conductance (E) in WT and CIN1 plants after 9 d of drought stress. Data are presented as means ±SE., *P<0.05, one-way ANOVA, n=3. Cell wall invertase leads to dehydration avoidance in tomato | 869 Fig.3. Chlorophyll fluorescence imaging indicating the maximum quantum yield (Fv/Fm) of photosystem II in leaves of WT and CIN1 plants under normal watering regimes and after 9 d of drought stress (A), and evolution of leaf Fv/Fm during the drought period (B). Photosynthetic rate in leaves of WT and CIN1 plants under normal watering regimes and after 9 d of drought stress (C). Transmission electron microscopy images of tomato leaves subjected to 9 d of drought stress showing starch formation (D). Data are presented as means ±SE, *P<0.05, one-way ANOVA, n=3. 870 | Albacete etal. 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-glucose- pyrophosphorylase 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. Fig.4. Vacuolar invertase (A), cytoplasmic invertase (B), and sucrose synthase (C) activities, hexose (glucose+fructose) concentrations (D), and invertase inhibitor activity (E) in mature leaves of WT and CIN1 plants under normal watering regimes and after 9 d of drought stress. Data are presented as means ±SE, *P<0.05, one-way ANOVA, n=3. Cell wall invertase leads to dehydration avoidance in tomato | 871 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-car- boxylic 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 Fig.5. Aldolase (A), phosphofructokinase (B), phosphoglucoisomerase (C), phosphoglucomutase (D), UDP-glucose pyrophosphorylase (E), glucose-6- phosphate dehydrogenase (F), hexokinase (G), fructokinase (H), and ADP-glucose pyrophosphorylase (I) activities in mature leaves of WT and CIN1 plants under normal watering regimes and after 9days of drought stress. Data are presented as means ±SE, *P<0.05, one-way ANOVA, n=3. 878 | Albacete etal. extreme drought tolerance in a flowering plant. Proceedings of the National Academy of Sciences, USA 104, 19631–19636. Roitsch T. 1999. Source–sink regulation by sugar and stress. Current Opinion in Plant Biology 2, 198–206. Roitsch T, Balibrea ME, Hofmann M, Proels R, Sinha AK. 2003. Extracellular invertase: key metabolic enzyme and PR protein. Journal of Experimental Botany 54, 513–524. Roitsch T, Bittner M, Godt DE. 1995. Induction of apoplastic invertase of Chenopodium rubrum by d-glucose and a glucose analog and tissue-specific expression suggest a role in sink–source regulation. Plant Physiology 108, 285–294. Roitsch T, Ehness R. 2000. Regulation of source/sink relations by cytokinins. Plant Growth Regulation 32, 359–367. Roitsch T, González MC. 2004. Function and regulation of plant invertases: sweet sensations. Trends in Plant Science 9, 606–613. Ruan YL, Jin Y, Yang YJ, Li GJ, Boyer JS. 2010. Sugar input, metabolism, and signaling mediated by invertase: roles in development, yield potential, and response to drought and heat. Molecular Plant 3, 942–955. Scherling C, Ulrich K, Ewald D, Weckwerth W. 2009. A metabolic signature of the beneficial interaction of the endophyte Paenibacillus sp. isolate and in vitro-grown poplar plants revealed by metabolomics. Molecular Plant-Microbe Interactions 22, 1032–1037. Schreiber U. 2004. Pulse-amplitude-modulation (PAM) fluorometry and saturation pulse method: an overview. In: Papageorgiou GC, Govindjee R, eds. Chlorophyll fluorescence: a signature of photosynthesis . Berlin: Springer, 2279–2319. Seki M, Umezawa T, Urano K, Shinozaki K. 2007. Regulatory metabolic networks in drought stress responses. Current Opinion in Plant Biology 10, 296–302. Shinozaki K, Yamaguchi-Shinozaki K. 2007. Gene networks involved in drought stress response and tolerance. Journal of Experimental Botany 58, 221–227. Skirycz A, Vandenbroucke K, Clauw P, etal. 2011. Survival and growth of Arabidopsis plants given limited water are not equal. Nature Biotechnology 29, 212–214. Stitt M. 1991. Rising CO2 levels and their potential significance for carbon flow in photosynthetic cells. Plant, Cell and Environment 14, 741–762. Sulpice R, Pyl ET, Ishihara H, etal. 2009. Starch as a major integrator in the regulation of plant growth. Proceedings of the National Academy of Sciences, USA 106, 10348–10353. Sun L, Yang D-l, Kong Y, Chen Y, Li X-Z, Zeng L-J, Li Q, Wang E-T, He Z-H. 2014. Sugar homeostasis mediated by cell wall invertase GRAIN INCOMPLETE FILLING 1 (GIF1) plays a role in preexisting and induced defence in rice. Molecular Plant Pathology 15, 161–173. Tang GQ, Lüscher M, Sturm A. 1999. Antisense repression of vacuolar and cell wall invertase in transgenic carrot alters early plant development and sucrose partitioning. The Plant Cell 11, 177–189. Van Houtte H, Vandesteene L, López-Galvis L, etal. 2013. Overexpression of the trehalase gene AtTRE1 leads to increased drought stress tolerance in Arabidopsis and is involved in abscisic acid-induced stomatal closure. Plant Physiology 161, 1158–1171. Webster H, Keeble G, Dell B, etal. 2012. Genome-level identification of cell wall invertase genes in wheat for the study of drought tolerance. Functional Plant Biology 39, 569–579. Weschke W, Panitz R, Gubatz S, Wang Q, Radchuk R, Weber H, Wobus U. 2003. The role of invertases and hexose transporters in controlling sugar ratios in maternal and filial tissues of barley caryopses during early development. The Plant Journal 33, 395–411. Zechmann B, Müller M, Zellnig G. 2007. Membrane associated qualitative differences in cell ultrastructure of chemically and high pressure cryofixed plant cells. Journal of Structural Biology 158, 370–377.