Hormone profiling and transcription analysis reveal a major role of ABA in tomato salt tolerance

Plant Physiology and Biochemistry 77 (2014) 23e34

Contents lists available at ScienceDirect

Plant Physiology and Biochemistry journal homepage: www.elsevier.com/locate/plaphy

Research article

Hormone profiling and transcription analysis reveal a major role of ABA in tomato salt tolerance Rongchao Yang a, Ting Yang b,1, Haijun Zhang a, Yan Qi a, Yanxia Xing a, Na Zhang a, Ren Li a, Sarah Weeda c, Shuxin Ren c, Bo Ouyang b, **, Yang-Dong Guo a, * a b c

College of Agriculture and Biotechnology, China Agricultural University, Beijing 100193, PR China College of Horticulture and Forestry Science, Huazhong Agricultural University, Wuhan 430070, PR China School of Agriculture, Virginia State University, PO Box 9061, Petersburg, VA 23806, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 September 2013 Accepted 25 January 2014 Available online 2 February 2014

The response and adaptation of plants to different environmental stresses are of great interest as they provide the key to understanding the mechanisms underlying stress tolerance. In this study, the changing patterns of four endogenous hormones and various physiological and biochemical parameters of both a salt-tolerant (LA2711) and a salt-sensitive (ZS-5) tomato cultivar were examined under salt stress and non-stress conditions. Additionally, the transcription of key genes in the abscisic acid (ABA) biosynthesis and metabolism were analyzed at different time points. The results indicated that gene expression responsible for ABA biosynthesis and metabolism coincided with the hormone level, and SlNCED1 and SlCYP707A3 may play major roles in the process. LA2711 performed superior to ZS-5 on various parameters, including seed germination, Naþ compartmentation, selective absorption of Kþ, and antioxidant enzymes activity. The difference in salt tolerance between the two genotypes could be attributed to the different levels of ABA due to differences in gene expression of key genes in ABA biosynthesis and metabolism. Although gibberellin, cytokinin and auxin were involved, our results indicated that ABA signaling plays a major role in tomato salt tolerance. As compared to ZS-5, LA2711 had a higher capability to selectively absorb and redistribute Kþ and a higher tolerance to Naþ in young leaves, which may be the main physiological mechanisms of salt tolerance. Ó 2014 Elsevier Masson SAS. All rights reserved.

Keywords: Gene expression Hormones Physiological and biochemical Salinity stress Tomato

1. Introduction High salinity is one of the most severe environmental stresses which cause crop yield loss and product quality deterioration. Plants have developed different mechanisms to adapt to salinity stress, involving complex physiological and biochemical changes. In the process of stress adaptation, hormones, especially abscisic acid (ABA), play important roles.

Abbreviations: APX, ascorbate peroxidase; ABA, abscisic acid; CTK, cytokinin; CAT, catalase; ETH, ethylene; GAs, Gibberellins; IAA, auxin; NCED, 9-cis-epoxycarotenoid dioxygenase; POD, peroxidase; PVPP, polyvinylpolypyrrolidone; SOD, superoxide dismutase; ZR, trans-zeatin-riboside. * Corresponding author. Tel.: þ86 10 6273 4845; fax: þ86 10 6273 3404. ** Corresponding author. Tel.: þ86 27 8728 1679; fax: þ86 27 8728 0016. E-mail addresses: [email protected] (B. Ouyang), [email protected] (Y.-D. Guo). 1 Present address: Plant Research International, Wageningen UR, P.O. Box 619, 6700 AP Wageningen, The Netherlands. 0981-9428/$ e see front matter Ó 2014 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.plaphy.2014.01.015

Abscisic acid is a phytohormone that regulates plant growth and development and abiotic stress response. In tomato, SlNCED1 and SlNCED2 encoding 9-cis-epoxycarotenoid dioxygenase are the primary genes responsible for ABA biosynthesis, whereas SlCYP707A1, SlCYP707A2, SlCYP707A3 and SlCYP707A4 encoding ABA 80 -hydroxylase are main genes for ABA metabolism (Sun et al., 2011; Nakaune et al., 2012). Under drought stress, a higher expression of SlNCED1 is maintained in the roots and leaves of tomato. This increases the concentration of ABA in leaves, leading to reduction of stomatal conductance and improvement of water use efficiency (Thompson et al., 2007; Tung et al., 2008). SlNCED2 is expressed mainly in vegetative tissues and young fruits, and its expression in fruits is not in step with the ABA level (Sun et al., 2012). There are still no reports of expression patterns of these tomato genes under salt stress. ABA regulates several aspects of plant response to abiotic stress. Under stress, plants produce ABA which induces the expression of genes encoding ion transporters, thus enhancing the ion selective absorption and contributing to the transfer of Naþ from the cytoplasm to vacuole or discharge it out of plant (Yu et al., 2007; Zhao

24

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

et al., 2009; Yarra et al., 2012). ABA is also involved in the modified process of the ATPase activities of the plasma membrane and tonoplast and improving proton pump activity, which will provide more power for the Naþ/Hþ antiport, thus simultaneously enhancing the selective absorption of Kþ (Janicka-Russak and Klobus, 2007; Olias et al., 2009). Under osmotic stress, ABA induces a rapid elevation of Ca2þ in apical cytoplasm, which improves the Kþ selective absorption and maintains the normal Kþ/Naþ ratio (Borsani et al., 2001; Dodd et al., 2010; Huertas et al., 2012). In addition, ABA activates the expression of genes encoding antioxidant enzymes, increasing the activity of antioxidant enzymes including superoxide dismutase (SOD), catalase (CAT) and ascorbate peroxidase (APX) (Jiang and Zhang, 2002). The increased capacity of reactive oxygen species (ROS) scavenging alleviates the damaging effect of ROS on fat/oil and protein (Hu et al., 2006). ABA also plays critical roles in reducing stomatal conductance and inducing the expression of genes responsible for various osmolytes synthesis such as proline and betaine (Jung et al., 2008; Antoni et al., 2011). Gibberellin (GA) acts as an antagonist to ABA. GA enhances seed germination under salt stress through different mechanisms such as inducing synthesis of some enzymes and stimulating Hþ-ATPase activity of the tonoplast (Maggio et al., 2010). At low salt stress, GA3 reduces the stomatal resistance of leaf, accelerates transpiration and increases water use efficiency, thus improving the salt tolerance of plants. However, GA3 does not reduce the inhibition of high salt stress on plant growth (Maggio et al., 2010; Achard et al., 2006). Other phytohormones promoting the growth of plants, such as cytokinin (CTK) and auxin (i.e. indole-3-acetic acid, IAA), can also increase the salt tolerance of plants (Wang et al., 2009; Park et al., 2011). Homeostasis of the cell is disrupted at different levels under salt stress. Salinity causes ionic stress (mainly due to Naþ, Cl and SO2 4 ), osmotic stress, and secondary stresses including nutritional imbalance and oxidative stress for glycophytes (Zhu, 2002). High concentrations of Naþ disturb the osmotic balance, causing “physiological drought”, which prevents plant water uptake. Halophytic plants are tolerant to sodium toxicity, therefore osmotic stress might be the main cause of growth inhibition. To cope with the harmful effects of salt stress, plants have evolved a series of biochemical and molecular mechanisms, mainly including selective buildup or exclusion of salt ions, control of ion uptake by roots and transport into leaves, ion compartmentalization, synthesis of compatible osmolytes, and induction of antioxidative enzymes (Rodriguez-Rosales et al., 2008). Understanding the mechanisms of plant salt tolerance will provide effective means to breed or genetically engineer salttolerant crops. Here, we report the profiling of various endogenous phytohormones (ABA, GA3, ZR and IAA) in a salt-tolerant tomato cultivar (LA2711) and a sensitive one (ZS-5), and genes involved in the biosynthesis and metabolism of ABA at the transcriptional level. We also measured the growth and various physiological and biochemical parameters of the two genotypes tested. The results would further provide insights in the response and adaptation of tomato to salt stress.

et al., 2005). Both accessions belong to cultivated tomato with large fruits, and their genetic backgrounds are relatively close. 2.2. Measurement of growth and physiological parameters Salt-stress treatment of tomato was reported previously (Ouyang et al., 2007). Briefly, tomato seeds were surface-sterilized and sown on agar-solidified MS medium (Murashige and Skoog, 1962) in triplicate, supplemented without or with different concentrations of salt (50e150 mM NaCl). Seeds were germinated at 24  C in the dark for 2 days and then were moved to an incubator at 24  C and under a lightedark regime of 16 h light/8 h darkness. The germination was recorded at 12-h intervals for 27 consecutive days using the criteria of appearance of 0.2 cm-radicle protrusion. Additionally, seed germination rate and potential, seedling fresh weight, radical length, lateral root number, cotyledon length and hypocotyl length were recorded and analyzed. Photosynthesis parameters, ions and antioxidant enzymes were measured on LA2711 and ZS-5 plants. Briefly, five-week-old plants of both genotypes grown hydroponically with one-fifth Johnson’s solution supplemented with 10 mM Fe-EDDHA were transferred either to a new nutrient solution with 150 mM NaCl for salt stress or to a nutrient solution without salt as the control (Ouyang et al., 2007; Wang et al., 2001). Net photosynthetic rate per unit area, stomatal conductance and evaporation rate of the third leaves from the top were measured after 7 d treatment with a TPS-1 photosynthesis system (PP System, UK). Fifteen days after salt treatment, the third leaves from the top, the youngest fully expanded leaves and root tips were sampled for the determination of ions, and at the same time, the root tissue were used for determination of antioxidant enzymes. Naþ, Kþ, Ca2þ and Mg2þ contents were determined by flame photometry as described in Asch et al. (2000). POD, SOD, CAT activities and proline content were determined according to previous work (Bates et al., 1973; Cavalcanti et al., 2004). 2.3. Observation of stomatal status for tomato leaves Five-week-old tomato plants grown hydroponically were transferred either to a new nutrient solution with 150 mM NaCl for salt stress treatment or without salt as the control. The third leaf from the top from different individuals was taken at the three time points: 0 h (9:30 am), 2 h (11:30 am), 24 h (9:30 am). Small pieces of leaf tissue, which were the abaxial of leaf and were sampled and fixed with 2.5% solution of glutaraldehyde in 0.1 M phosphate buffer (pH 7.2), post-fixed for about 30 min in 1% osmium tetroxide. After washing with distilled water several times, the specimens were dehydrated with ethanol series (30%, 50%, 70% and 100%) and then transferred to propylene oxide for 30 min, followed by washing with isoamyl acetate for 40 min. All specimens were critical point-dried using liquid CO2 and then coated with a thin layer of gold. The stomatal morphology was observed with a Hitachi S-450 scanning electron microscope (Hitachi, Tokyo, Japan). Stomata were counted on ten fields per sample and statistically analyzed. 2.4. Determination of hormones content in leaves and roots

2. Materials and methods 2.1. Plant materials Tomato (Solanum lycopersicum) seeds of LA2711 and ZS-5 were kindly provided by the Tomato Genetic Resource Center (TGRC, Davis USA) and the Chinese Academy of Agricultural Science (CAAS, Beijing, China), respectively. LA2711 is found to be salt-tolerant (Mahmoud et al., 1986), while ZS-5 is sensitive to salt (Wang

The culture conditions and treatments of tomato plants were the same as described above. The samplings were carried out at different time points: 0 h (9:30 am), 1 h (10:30 am), 2 h (11:30 am), 6 h (15:30 pm), 12 h (21:30 pm), 24 h (9:30 am). Root tissues and the third leaf from the top were taken at each time point, quickly preserved in liquid nitrogen and kept in an ultra-low temperature freezer (80  C). Approximately 0.5 g of tissue was weighed and homogenized in small volumes of pre-cooled 80% methanol with

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

trace amount of silver diethyldithiocarbamate that was undetectable using an electronic balance accurate to one hundred thousandth of a gram. The samples were extracted with 4e5 ml of precooled 80% methanol for 20e24 h at 0e4  C. After centrifugation at 2  104 g for 20 min at 4  C, the supernatant was collected and concentrated to the aqueous phase by placing in a 40  C water bath with Rotary Evaporator (RE-100, Bibby Sterlin LTD, Stone Staffordshire, England). The organic phase was treated with 0.2e0.3 g of polyvinylpolypyrrolidone (PVPP). After centrifugation again at 2  104 g for 20 min at 4  C, the supernatant was collected and divided into two equal parts: one part was used to extract ABA, GA and IAA, the other part for CTK. The portion for ABA, GA and IAA isolation was adjusted to pH 2.5e 3.0, extracted with an equal volume of ethyl acetate three times, and finally evaporated to dryness with Rotary Evaporator at 40  C as above. CTK was extracted from the other portion with an equal volume of n-butanol-saturated phosphate buffer (pH8.0) for three times, and evaporated to dryness at 60  C using the Rotary Evaporator. The dried samples containing ABA, GA3 and IAA were redissolved in chromatography grade methanol with 0.1 M glacial acetic acid as the mobile phase. The dried CTK sample was redissolved in ultrapure water (PH7.0) as the mobile phase. The flow rate for all analyses was adjusted to 1 ml/min. Samples of 100 ml was injected for HPLC analysis, and the detection wavelength was 280 and 260 or 267 nm for ABA/GA3/IAA, ZR, respectively. HPLC analysis was performed with an HP1100 (Agilent, Palo Alto, CA, USA) coupled with a Diode Array Detector. An Agilent ZORBAX SB-C18 clumn (5 mm 4.6*250 mm)was used in analysis and Mobile phases were 100% methanol (A) and 0.1 M acetic acid (B). The gradient elution was performed as follows: 0e 40 min, A 3% and B 97%; after 40 min, A 67.6% and B 32.2%. Flow rate was 1 ml/min and column temperature was set at 30  C. Chromatograms were used for quantification via Agilent chromatography workstation. Triplicate samples were run for each treatment. 2.5. Real-time quantitative PCR analysis Detailed methods and sampling time points for plant culture and salt treatments were the same as those for stomatal status investigation. The third leaf from the top and the whole root system of the plants were harvested, immediately frozen in liquid nitrogen and stored in a freezer (80  C). RNA isolation, reverse transcription and real-time PCR were conducted as described previously (Lu et al., 2012). The gene encoding tomato elongation factor 1a served as internal control, and the primer sequences were from a previous report (Dekkers et al., 2012). The primer sequences of genes for the biosynthesis and metabolism of ABA were also obtained from previous reports (Table S1) (Sun et al., 2011; Nakaune et al., 2012). Expression data were reported as fold change (2DDCt) relative to non-stressed controls.

25

genotype ZS-5 regarding germination rate and germination potential under salt stress (Ouyang et al., 2007). Higher germination rate of LA2711 was consistently observed under different concentration of salt stress (Fig. 1A). Here, we provided further evidence regarding growth-related parameters of the seedlings under salt stress. It showed distinct differences between the two genotypes in the growth of roots. Compared to ZS-5, LA2711 had no difference in radical length under control or 50 mM salt condition. However, under 100 mM and 150 mM salt stress, the radical length of LA2711 was significantly longer than that of ZS-5. On the whole, the radical length showed an increasing pattern between 0 and 150 mM salt stress for LA2711 but a decreasing pattern for ZS-5 (Fig. 1C). The lateral root number was similar between the two genotypes under normal condition, and the difference became larger (but without significant difference) under salt stress conditions (Fig. 1F). With increasing salt concentrations, seedlings of the two genotypes tested showed a similar decreasing trend in fresh weight, lateral root number, hypocotyl and cotyledon length. The fresh weight of the seedlings was higher for LA2711 under control or slight salt stress condition (Fig. 1B), but this difference disappeared under higher salt stress. The hypocotyl length of LA2711 was longer than that of ZS-5 under control or high salt stress (150 mM), but there was no difference under 50 and 100 mM of salt stress (Fig. 1D). The cotyledon length of LA2711 was significantly longer than that of ZS-5 under both control and different concentrations of salt stress (Fig. 1E). 3.2. Difference on leaf stomatal status between LA2711 and ZS-5 To investigate the stomatal status of the two tomato cultivars under salt stress, leaf surface of LA2711 and ZS-5 were examined by scanning electron microscope (Fig. 2). The results showed that the leaf stomatal density of LA2711 was obviously higher than that of ZS-5, both under control and stress conditions. Upon salt stress, the stomatal density initially increased, then decreased in both genotypes. The stomatal density was higher under both salt concentrations as compared to the non-stress control. After 2 h of salt stress, only 12.1% of the stoma on leaves of LA2711 was still open, while this percentage was 27.8% on the leaves of the non-stress control. For ZS-5, 49.2% of the stoma was open under salt stress, whereas 72.2% for control. Furthermore, under salt stress, a large portion of stoma of ZS-5 was severely deformed, while they were largely unaffected in LA2711 (Figs. S1 and S2). After 24 h of salt stress, the percentage of open stoma for LA2711 and ZS-5 were 49% and 61.2%, respectively, however, the corresponding percentages were 32.7% and 51.7% for their non-stressed controls. 3.3. Difference on ion contents between LA2711 and ZS-5

2.6. Statistical analysis and plotting Data were analyzed with ANOVA and Duncan’s multiple range tests, using the SAS software package (version 8.0, SAS Institute, Cary, NC, USA). Graphing was performed in Excel 2007 or SigmaPlot (Version 7.0, Systat Software, San Jose, CA), and figures were composed in Adobe Illustrator CS5 (Adobe, San Jose, CA, USA). 3. Results 3.1. Comparison of seedling growth-related parameters between LA2711 and ZS-5 In our previous report, we have shown that the salt-tolerant tomato genotype LA2711 performs better than the salt-sensitive

After salt stress, both LA2711 and ZS-5 accumulated high levels of Naþ, but the increment varied in different tissues (Fig. 3A and B). In roots, both genotypes accumulated similar levels of Naþ; the Naþ concentrations in 150 mM NaCl-treated root was roughly about four-fold greater than that in non-stressed controls. Mature leaves of salt-stressed LA2711 showed a 2.5-fold increase in Naþ, while salt-treated ZS-5 accumulated five times more Naþ than the nonstressed control, indicating mature leaves of ZS-5 accumulate more sodium. However, the situation was reversed in young leaf tissue; the level of Naþ in LA2711 was 2.7-fold greater than its nonstressed control, and the corresponding fold change was 1.7 for ZS5. Interestingly, the final level of Naþ was higher in LA2711 than in ZS-5. Overall, Naþ levels were highest in the roots and lowest in the young leaf under both control and stress conditions.

26

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

Fig. 1. Characterization of seed germination and seedling growth parameters of the tomato cultivars LA2711 and ZS-5 under salt stress or non-stressed conditions. (A) Germination rate. (B) Fresh weight. (C) Radical length. (D) Hypocotyl length. (E) Cotyledon length. (F) Lateral root number. Seeds were germinated in Petri dish with BM media (MS medium with 30 g l1 sucrose and 7 g l1 agar) supplemented with 0, 50, 100, 150 mM NaCl solution under optimal conditions in culture room. Data were collected 27 days after seed sowing. The data shown are the mean  SE (n ¼ 4). Single (*P < 0.05) and double (**P < 0.01) asterisks denote statistically significant differences between LA2711 and ZS-5 under different conditions.

Maintaining a high Kþ/Naþ ratio is considered to be important to salt tolerance; therefore we investigated Kþ content in different organs of the two tomato genotypes. The major difference between the two genotypes was the change in Kþ concentration in mature leaves. In LA2711, the concentration of Kþ increased significantly (P < 0.05) in the mature leaves after 150 mM salt stress, while in ZS5, Kþ level decreased significantly (P < 0.05) (Fig. 3C and D). This contributed to the large difference of Kþ/Naþ ratio between the two genotypes. LA2711 roots had a higher level of Kþ (10.986 mg/ 100 mg DW) under control condition, which decreased significantly (P < 0.01) after salt stress (Fig. 3C). However, the Kþ concentration in ZS-5 roots was only about half of that in LA2711 (5.393 mg/ 100 mg DW), and it increased slightly after 150 mM salt stress (Fig. 3D). The concentration of Kþ in young leaves was slightly higher in LA2711 than ZS-5, under both control and salt stress conditions. LA2711 and ZS-5 had major differences in Kþ/Naþ ratio in the various tissues (Fig. 3I and J). After 150 mM salt-stress treatment, Kþ/Naþ ratio decreased in all tissues of the two genotypes. The decrease was greater in the root and young leaf tissue of

LA2711, in which the Kþ/Naþ ratio were 12% and 36% that of the non-stressed control. However, in ZS-5, the Kþ/Naþ ratio was 28% and 60% that of the non-stressed control roots and young leaves, respectively. This trend was reversed in adult leaves, where the Kþ/ Naþ ratio decreased to 46% in LA2711 as compared to its nonstressed control, while it deceased to 16% in the mature leaves of ZS-5. Calcium signals are key components in plant stress signaling networks. After salt stress, the level of Ca2þ concentration decreased significantly both in LA2711 and ZS-5, regardless of tissue type. However, the decline was greater in ZS-5. In the root, mature and young leaf tissues of LA2711, the Ca2þ concentrations were 40%, 71% and 55% of those in the non-stressed controls, while the corresponding percentages in ZS-5 were 26%, 61% and 50%, respectively. The Ca2þ concentration level in the mature leaf of LA2711 was lower than that of ZS-5, whereas LA2711 had higher levels of Ca2þ concentration in the root and young leaf (Fig. 3E and F). Cellular magnesium (Mg2þ) is important for the functions of many enzymes. The major differences in Mg2þ concentration levels

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

27

salt stress (Figs. S3D and S3E). Proline contents increased significantly in the leaves and roots of LA2711 and ZS-5 under salt stress. However, the accumulation was higher in ZS-5 than that in LA2711, regardless of tissues tested. Under salt stress, proline contents were increased by 53.6 and 39.0 folds in the leaves of ZS-5 and LA2711, respectively. In root, the corresponding fold changes were 42.7 and 8.6 in LA2711 and ZS-5, respectively. 3.5. Dynamic changes of endogenous hormones in LA2711 and ZS-5

Fig. 2. The stomatal density and opening state on the abaxial surface of tomato leaf under 150 mM salt stress (Na) or normal condition (CK). The data shown are the mean  SE (n ¼ 10). Single (*P < 0.05) and double (**P < 0.01) asterisks denote statistically significant differences between LA2711 and ZS-5.

between LA2711 and ZS-5 were found in root tissues (Fig. 3G and H). Upon salt stress, the concentration of Mg2þ increased significantly (P < 0.01) in the root of LA2711, while it decreased significantly (P < 0.01) in the root of ZS-5. The level of Mg2þ was similar in both the mature and young leaf tissues between LA2711 and ZS-5. 3.4. Comparison of antioxidant enzyme activity and proline content between LA2711 and ZS-5 Antioxidant enzymes are involved in plant tolerance to various stresses including salinity. POD, SOD and CAT are important enzymes to eliminate ROS in plants. In this study, the POD activity increased significantly (P < 0.01) in both LA2711 and ZS-5 after salt stress (Fig. S3A), noticeably, the increase in activity was much higher in LA2711. The POD activity in LA2711 increased 2.2-fold upon salt stress, while it only increased 1.3-fold in ZS-5. The initial level of SOD activity was much higher in ZS-5 than that in LA2711, and after salt stress, it decreased to a level similar to that of LA2711. Upon 150 mM salt stress, the SOD activity level in LA2711 was similar to untreated control (Fig. S3B). The CAT activity decreased in both genotypes under salt stress, although the decline was more pronounced in ZS-5 than in LA2711 (Fig. S3C). These results indicated that LA2711 may have a greater capacity to remove ROS. Proline may function as an important osmoprotectant. The proline accumulation was investigated in LA2711 and ZS-5 under

3.5.1. Effects of salt stress on endogenous ABA ABA is considered as a plant stress hormone. Five-week-old plants were challenged with 150 mM salt, and the concentration of various hormones was measured in the root and leaf tissue for both genotypes tested at the following time points: 0 h (9:30 am), 1 h (10:30 am), 2 h (11:30 am), 6 h (15:30 pm), 12 h (21:30 pm) and 24 h (9:30 am). In unstressed root tissue, the ABA concentration decreased during the day and increased again at night, remaining less than 200 ng per gram of fresh tissue. The pattern of ABA fluctuation in root tissue of the two genotypes was similar. However, the ABA concentration was lower in LA2711 than that in ZS-5 at all times. Under salt stress, ABA levels increased dramatically within 1 h, then declined to a minimum level and re-boosted to a higher level. Upon 1 h of salt stress, the concentrations of ABA were increased by 13.7and 5.6-fold in the root tissue of LA2711 and ZS-5, respectively. ZS-5 had a higher level of ABA in the root than LA2711 after 6 h of salt stress. At 24 h, the ABA concentration in the root of ZS-5 under salt stress was 1.25-fold greater than that in LA2711, and the ABA concentration in the root of ZS-5 under the normal condition was 1.98-fold greater than that of LA2711 (Fig. 4A). In leaf tissue, the ABA levels were similar to that in root tissue under control condition, and conversely, LA2711 had a higher level of ABA all the time, except for the time point of 6 h. Upon salt stress, the ABA concentration decreased first and then dramatically increased to a maximum within 2 h. After that the ABA level decreased to a minimum and increased slightly to reach a relatively steady state. As compared to the non-stressed control, the increases of ABA in the second hour after stress were 5.7 and 6.9 folds for LA2711 and ZS-5, respectively (Fig. 4B). 3.5.2. Effects of salt stress on endogenous GA3 Gibberellin (GA) is known to antagonize the effect of ABA in various biological processes. In root tissue, when no stress was applied, ZS-5 contained a significantly higher level of GA3 than LA2711 at all the time points tested except for 6 h time point. Under salt stress, GA3 content decreased within a couple of hours in both genotypes, and increased significantly 2 h later. In the first couple of hours of salt stress, GA3 content in LA2711 was lower than that in ZS-5; however, its content in LA2711 was higher in ZS-5 after 2 h of treatment, indicating a quicker recovery of growth for LA2711. At 6 h, GA3 concentration in the root of LA2711 was about 5-fold greater than that in ZS-5 under salt-stressed conditions (Fig. 5A). In leaf tissue, the pattern was similar to that of root tissue although the level of GA3 was lower. At 2 h, the level of GA3 in LA2711 was similar to that in ZS-5. Except for the time point of 6 h, LA2711 had lower GA3 content than ZS-5 at the four other time points. Under salt stress, the level of GA3 in LA2711 increased slightly within 1 h, followed by a slight decrease at 2 h. GA3 content reached its peak at 6 h under salt stress and decreased at later time points. For ZS-5, the GA3 content decreased slightly within the first 2 h of salt stress, increased thereafter and was maintained at a higher level than that of LA2711 at the rest time points tested (Fig. 5B).

28

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

Fig. 3. Effect of salt stress on concentration of ions and Kþ/Naþ ratio in the roots, mature and young leaves of LA2711 and ZS-5. The data shown are the mean  SE (n ¼ 3). Single (*P < 0.05) and double (**P < 0.01) asterisks denote statistically significant differences between treatment (150 mM NaCl) and no-treatment (0 mM NaCl). (A, B) Naþ. (C, D) Kþ. (E, F) Ca2þ. (G, H) Mg2þ. (I, J) Kþ/Naþ ratio.

3.5.3. Effects of salt stress on endogenous ZR and IAA In root tissue, the pattern of ZR concentration in LA2711 under non-stressed condition was similar to that in ZS-5, which was decreased to a lower level at 2 h, increased to the peak at 6 h, and subsequently decreased and back to a level similar to that at 24 h. The ZR contents were slightly higher at the time points of 0 h and 24 h. Under salt stress, the ZR content was maintained at a similar level as that of the non-stressed control within the first 2 h, however, it increased significantly after 2 h and reached its peak at the

time point of 6 h, then decreased at 12 h and restored to a high level at 24 h. LA2711 under salt stress had a slightly higher level of ZR than ZS-5 at most time points tested (Fig. 5C). In leaf tissue, the ZR concentration and changing pattern in LA2711 was similar to that in ZS-5 under non-stressed condition, both genotypes had a peak in ZR content at 6 h. Under salt stress, there were obvious differences between the two genotypes in ZR concentration. In the mid-afternoon, the ZR content in salt-stressed ZS-5 increased by 1.8-fold as compared to its non-stressed control,

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

29

2 h, however, it was followed by a small peak at 6 h and another increase later (Fig. 5F). 3.6. Expression of genes involved in the biosynthesis and metabolism of ABA 3.6.1. The expression of SlNCED1 and SlNCED2 The key genes responsible for ABA synthesis in tomato, SlNCED1 and SlNCED2, were studied at the transcriptional level using real time RT-PCR. On the whole, the expression level of SlNCED2 was much lower than that of SlNCED1 (Fig. 6). In non-stressed root tissue, both SlNCED1 and SlNCED2 were down regulated within the first 2 h, while they restored the expression within 24 h. When salt stress was applied, SlNCED1 and SlNCED2 were dramatically upregulated at the first hour post stress, down-regulated at 2 h, and up-regulated again at 24 h. The expression of both genes in LA2711 was much higher than those in ZS-5 under salt stress at the first hour, while its level was lower than that in ZS-5 at 24 h (Fig. 6C and D). In leaf tissue, the expression of SlNCED1 and SlNCED2 showed a similar pattern to that in roots (Fig. 6A and B). When salt stress was applied, SlNCED1 and SlNCED2 transcripts increased several folds in both genotypes at the time point of 2 h, followed by a decrease. The expression of both genes, especially SlNCED1, in LA2711 was higher than that in ZS-5 at 2 h post salt stress.

Fig. 4. ABA content in the tomato cultivar LA2711 and ZS-5 treated with 150 mM NaCl solution (Na) and in normal condition (CK). (A) Content of ABA in root. (B) Content of ABA in leaf. Bars are mean  SD of four replications.

while there was not much change in LA2711. During 12e24 h, the ZR content deceased in salt-stressed ZS-5, however, it increased in LA2711 under salt stress (Fig. 5D). IAA concentration was also measured in the root and leaf tissues of LA2711 and ZS-5. In non-stressed LA2711 root, the IAA level reached its minimum at 1 h, increased to the peak 1 h later and decreased slightly thereafter. In non-stressed ZS-5 root, the IAA concentration was much higher than that of LA2711 at 1 h, and it also decreased to the bottom level 1 h later, followed by an increase between 1 h and 2 h and a further increase at 12 h. In salt-stressed LA2711 root, the IAA level increased during the first 2 h, followed by a slow decrease and a restoration later. While in salt-stressed ZS-5 root, IAA concentration decreased to the lowest level in the first 2 h, reached its second peak at 6 h and declined later (Fig. 5E). In leaf tissue, the IAA level in LA2711 reached the peak at 2 h, decreased between 2 h and 6 h and increased slightly later. In ZS-5, IAA level increased to its peak within 1 h, decreased continuously to the lowest level between 1 h and 12 h, and restored to the level of starting point. Under salt stress, IAA fluctuation in LA2711 was similar to that in the non-stressed control, the major difference after stress was that the IAA level was higher at most time points tested, especially at 2 h and 24 h. In ZS-5 leaf, the pattern of IAA change was very close to that in the non-stressed control in the first

3.6.2. Genes in ABA metabolism SlCYP707A1, SlCYP707A2, SlCYP707A3 and SlCYP707A4 were involved in ABA metabolism. Quantitative RT-PCR results showed that SlCYP707A1 and SlCYP707A3 were expressed much higher in the root and leaf tissue than SlCYP707A2 and SlCYP707A4 (Fig. 7). Under normal condition, the expression level of SlCYP707A1 was relatively steady in the root tissue of both genotypes. Upon salt stress, SlCYP707A1 was down-regulated in the root tissue of both genotypes within the first hour then subsequently restored (Fig. 7B). Under non-stressed condition, the expression patterns of SlCYP707A1 in leaf of both genotypes were similar, the transcripts of which were up-regulated at the time points 1 h and 2 h. During this time, the transcriptional level of SlCYP707A1 was significantly higher in ZS-5 than in LA2711. Under salt stress, SlCYP707A1 was up-regulated significantly in both genotypes in the first hour, then down-regulated at 2 h and restored to a higher level at 24 h. At 1 h after stress, the transcriptional level of SlCYP707A1 was much higher in LA2711 than in ZS-5 (Fig. 7A). The expression patterns of SlCYP707A3 were overall similar to that of SlCYP707A1 in both conditions and genotypes. In root tissue, the expression level of SlCYP707A3 was higher in LA2711 than in ZS5 (Fig. 7F). In leaf tissue, the expression was higher in LA2711 than in ZS-5 during the first hour under salt-stress treatment, however, as compared to ZS-5, the expression level was lower in LA2711 at 2 h (Fig. 7E). SlCYP707A2 and SlCYP707A4 were expressed at a very low level in tomato, and no significant difference was detected between the two genotypes (Fig. 7C, D, G and H) under both control and salt stress conditions. 4. Discussion 4.1. Indicators for screening of salt-stress tolerance in tomato In salt stress studies, dozens of candidate indicators regarding growth, physiology and biochemistry are used for evaluation of salt tolerance. Most of them were applied in our previous study (Ouyang et al., 2007), and among them, germination rate and Kþ/ Naþ ratio were critical indicators. Germination rate is the most widely-used indicator for evaluation of stress tolerance. This

30

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

Fig. 5. Contents of GA3, ZR, IAA in the tomato cultivar LA2711 and ZS-5 treated with 150 mM NaCl solution (Na) and in normal condition (CK). (A) Content of GA3 in root. (B) Content of GA3 in leaf. (C) Content of ZR in root. (D) Content of ZR in leaf. (E) Content of IAA in root. (F) Content of IAA in leaf. Bars are mean  SD of four replications.

indicator has been used in quantitative trait loci (QTLs) mapping of tomato salt tolerance (Foolad and Lin, 1997, 1999). At seedling stage, ion content, especially Naþ and Kþ (and thus Kþ/Naþ ratio) has been employed to evaluate salt tolerance in tomato (Ouyang et al., 2007; Saranga et al., 1993; Rubio et al., 2004; Juan et al., 2005). Our present results showed that Naþ content in mature leaves of the sensitive genotype (ZS-5) was significantly higher than that of the salttolerant genotype (LA2711) (Fig. S4), suggesting that the photosynthetic capacity was higher in LA2711 than in ZS-5. Additionally, our results demonstrated that Naþ content in young leaves of the salt-tolerant genotype was higher than that of the sensitive genotype. This might be a typical characteristic of halophyte plants, which tolerate higher Naþ levels, as Naþ could serve as the cheapest solutes for osmotic adjustment (Rush and Epstein, 1981; Cuartero et al., 2006). This also consistent with previous research results that a major QTL (over 40%) detected on chromosome 7 which is responsible for the concentrations of Naþ and Kþ in stems and leaves (Villalta et al., 2008). There can be other indicators for salt-tolerance evaluation in tomato. From the analysis between LA2711 and ZS-5, we found that radical length might serve as a good candidate indicator for evaluation of tomato salt tolerance. However, our results indicated that fresh weight and cotyledon length are not salt-stressed indicators.

The difference between the two genotypes regarding these two parameters appeared even without salt treatment and remained in a series of salt stress (Fig. 1B and E). Faster germination of LA2711 seeds could be the main reason underlying these phenomena (unpublished). Also, salt-tolerant genotypes may need to absorb more water to offset the negative effects of Naþ, thus increasing the fresh weight. Although LA2711 seedlings tended to produce more lateral roots under stress than those of ZS-5, the difference was not statistically significant. Therefore, using lateral root number for screening of salt-tolerance needs further evaluation. Photosynthesis parameters are also commonly used in stress tolerance evaluation, as abiotic stresses can significantly inhibit the photosynthesis of plants. Indeed, our results indicated that, compared to ZS-5, LA2711 showed no advantage in photosynthesis under control condition, but its photosynthesis efficiency was higher than that of ZS-5 under salt stress (Table S2). This was well in agreement with the results of stomatal observation, which indicated that the extent of the leaf damage in LA2711 was lower than that in ZS-5. We also tested biochemical indicators such as proline content and three ROS scavenging enzymes (POD, SOD and CAT) between LA2711 and ZS-5 under salt stress. Our results showed that ZS-5 accumulated a higher level of proline than LA2711. The role of proline in salt stress physiology remains controversial. Tal et al.

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

31

Fig. 6. Expression level of SlNCED genes in tomato leaves and roots under 150 mM salt treatment (Na) and normal condition (CK). (A, C) Relative mRNA transcript level of SlNCED1 in the leaves and roots of LA2711 and ZS-5. (B, D) Relative mRNA level of SlNCED2 in the leaves and roots of LA2711 and ZS-5. The gene encoding tomato elongation factor 1a is used as the internal control. The data shown are the mean  SE (n ¼ 4). Single (*P < 0.05) and double (**P < 0.01) asterisks denote statistically significant differences between LA2711 and ZS-5.

(1979) reported that the salt-sensitive tomato accumulated more proline under salt stress than that of the salt-tolerant tomato. Aziz et al. (1998) reported that the salt tolerance of tomato was negatively correlated with the content of proline. However, Melchor et al. (2005) found that the salt-tolerant tomato accumulated more proline than that of the salt-sensitive one. Our study showed that both LA2711 and ZS-5 accumulated proline under salt stress, but the increase for LA2711 was less than that of ZS-5, which agreed with the results of Tal et al. (1979) and Aziz et al. (1998). Under this circumstance, we think that proline is more likely to be an indicator for the degree of damage under salt stress. The changes in POD, SOD and CAT activities were consistent with the previous results (Mittova et al., 2002). Our results showed that the SOD activity in the root of salt-tolerance genotype (LA2711) was not inhibited under 150 mM salt stress, but it was significantly decreased in the salt-sensitive genotype (ZS-5) under salt stress (Fig. S3). Also, the extent of POD activity increase upon salt stress in LA2711 was larger than that in ZS-5 (Mittova et al., 2002; Shalata et al., 2001). Although salt-tolerant germplasms are superior in many aspects, the indicators for each aspect used for screening of salt-tolerance should be determined on a case-by-case basis, as different germplasms may use different resistance mechanisms. 4.2. Roles of hormones in tomato salt tolerance Our results provided the first-hand proof of dynamic change of ABA in tomato seedlings within a 24-h range, and thus could explain the discrepancies observed in previous reports. A few studies reveal that endogenous ABA increases significantly in salttreated tomato (Chen and Plant, 1999), while other reports contrast this finding (Achuo et al., 2006). The discrepancy could be well-explained by our results. Within 24 h, there existed dynamic changes in ABA content in tomato both under control and salt stress

conditions. ABA could transiently increase by 13 folds in roots under 150 mM salt stress (Fig. 4A). Although ABA was observed changing with time during drought stress, it is detected at one-day resolution (Cohen et al., 1999). Our results would be very informative for experimental design regarding ABA and plant stress. Our results suggested that ABA plays important roles in abiotic stress in tomato. Under salt stress, ABA levels increased dramatically in tomato root within 1 h, and then in leaf 1 h later (Fig. 4), suggesting that ABA acts as a signal in tomato salt response. Additionally, the peaks of ABA level were detected at 1 h in roots but 2 h in leaf (Fig. 4), this time lag strongly supported that the ABA signal is produced first in root and then transduced or transferred to the aboveground tissue (Wilkinson and Davies, 2002). And it also coordinated with the expression of genes regulating the synthesis (SlNCEDs) and metabolism (SlCYP707As) of ABA (Figs. 6 and 7). Noticeably, LA2711 maintained a lower level of ABA under control condition, but the level of ABA increased sharply under salt stress. This indicated that LA2711 exhibited a stronger response to salt stress. ABA is involved in various physiological and biochemical processes under salt stress. ABA contributes to a higher ratio of Kþ/Naþ, by enhancing expression of genes encoding ion transporters and proton pumps which help in the Naþ compartmentation and selective absorption of Kþ (Janicka-Russak and Klobus, 2007). In addition, ABA contributes to stomatal closure by enhancing Ca2þ signal along with other signal molecules such as H2O2 and NO. This mechanism, which is very important to reduce water evaporation under salt stress, could be also applicable to tomato. Under stress, ABA enhances the enzyme activity of SOD, POD, CAT and APX, thus increasing the potential to scavenge oxygen free radical. When the two tomato genotypes were exposed to salt stress, the POD activity in root tissue increased significantly, whereas it was significantly higher in LA2711 than that in ZS-5

32

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

Fig. 7. Expression level of SlCYP707A genes in tomato leaves and roots under 100 mM salt treatment (Na) and normal condition (CK). (A, B) Relative mRNA transcript level of SlCYP707A1 in the leaves and roots of LA2711 and ZS-5. (C, D) Relative mRNA level of SlCYP707A2 in the leaves and roots of LA2711 and ZS-5. (E, F) Relative mRNA level of SlCYP707A3 in the leaves and roots of LA2711 and ZS-5. (G, H) Relative mRNA level of SlCYP707A4 in the leaves and roots of LA2711 and ZS-5. The gene encoding tomato elongation factor 1a is used as the internal control. The data shown are the mean  SE (n ¼ 4). Single (*P < 0.05) and double (**P < 0.01) asterisks denote statistically significant differences between LA2711 and ZS-5.

(Fig. S3A). In addition, the CAT activity decreased in both genotypes tested (Fig. S3C), however, its activity in LA2711 was less influenced than in ZS-5. This difference may be caused by the different level of ABA between LA2711 and ZS-5. The SOD activity in LA2711 was very stable, while it was significantly inhibited in ZS-5, and this might be also related to ABA. In our study, the different hormones were measured simultaneously. Our results showed that the changing pattern of GA3 was quite opposite to the one for ABA, consistent with reports on the antagonist effect between ABA and GA3 (Fig. 5A and B). The difference of GA3 concentration between LA2711 and ZS-5 may also reflect their different ability to tolerate salt stress. In the early stage of salt stress, GA3 content in the root of LA2711 was lower than that in ZS-5, however, this difference was reversed after 6 h and the GA3 content in LA2711 was five folds greater than that in ZS-5. This suggested that LA2711 recovered faster than ZS-5 under salt stress.

It was worthwhile to note that our results were somewhat contradictory with previous results reporting that the GAs level decreases under salt stress and consequently inhibits plant growth to facilitate the fight against the external stress (Achard et al., 2006). Our results might reflect the short-term response of tomato to salt stress, and GA3 level may decrease after long-term of salt treatment. We also measured the levels of IAA and ZR at the corresponding time points, but further evaluation will help to address the significance of pattern difference between the two genotypes. 4.3. Coincidence of ABA and transcriptional level of key genes in ABA biosynthesis and metabolism Real-time quantitative PCR results indicated that the expression of SlNCED1 and SlNCED2 in both roots and leaves of the two tomato cultivars tested were induced significantly under salt stress. The

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

33

expression of SlNCED1 was significantly stronger than that of SlNCED2, indicating that SlNCED1 might play a major role in the process of ABA synthesis. During the early stages of salt stress, the expression levels of SlNCED1 and SlNCED2 in the leaves of LA2711 were significantly higher than those of ZS-5. After 24 h, the expression levels of SlNCED1 and SlNCED2 in LA2711 were slightly lower than those of ZS-5. This difference in gene expression of ABA biosynthesis shall contribute to the difference in ABA content between the two genotypes. Furthermore, the differential expression of SlCYP707A1 and SlCYP707A3 in LA2711 and ZS-5 would also contribute to the difference of the overall ABA levels between the two genotypes. Therefore, the difference in ABA levels between the salt-tolerant and salt-sensitive tomato genotypes in this study could be largely explained by the difference of the expression of key genes in ABA biosynthesis and metabolism.

(Nonprofit industry research projects “Seed production and seed quality control for vegetable crops”).

4.4. Mechanisms of salt tolerance in LA2711

References

LA2711 is more tolerant to salt stress than ZS-5. This could be explained by the superior performance in various parameters such as vigorous germination, higher Kþ/Naþ ratio, its level of basic SOD activity not being inhibited under salt stress, and the level of basic POD activity being higher than that of ZS-5 under salt stress. Here we would like to highlight the following features: 1) Compared to ZS-5, LA2711 showed a higher capability to absorb and redistribute Kþ. This was evident from the data showing that the basal level of Kþ in the root tissue of LA2711 was much higher than that in ZS-5 under control condition. And under salt stress, the level of Kþ content in the root of LA2711 significantly decreased, however, the Kþ level in mature leaf of LA2711 increased significantly and its content was much higher than that in ZS-5. This indicated that more Kþ was transported from root to leaf. This hypothesis can be tested by growing the two genotypes in Kþ deficient hydroponic solution. 2) Leaves of LA2711, especially the young leaves, could tolerate higher level of Naþ. LA2711 may have a higher capability to relocate or compartmentalize Naþ to vacuoles. Validation could be conducted by assaying the Naþ distribution on a cellular resolution. These physiological processes are likely regulated by the stress hormone, ABA. Furthermore, at the molecular level, the ABA dynamic change and distribution were fine tuned in a temporal and spatial way by key genes in ABA biosynthesis and metabolism. The fine tuning of ABA should also contribute to stomatal movement. With the accomplishment of the tomato genome sequencing and the ongoing de novo sequencing of a large number of tomato germplasms, it may be possible to check the sequence differences and their biological significance in the promoter regions of these key genes between stress-tolerant and -sensitive genotypes. Of course, differences in trans-acting factors might also be responsible for the differential expression of these genes. In addition, genes encoding Kþ transporter, Naþ/Hþ antiporter or exchanger would be a promising target for genetic engineering for improving salt tolerance (Cuartero et al., 2006; Huertas et al., 2012). Actually, significant improvement of salt tolerance has been demonstrated by ectopic expression of an Arabidopsis vacuole-type Naþ/Hþ antiporter into cultivated tomato, while transgenic tomato plants with reduced expression of SlSOS1, a homolog of Arabidopsis AtSOS1 which encodes a plasma membrane-localized Naþ/Hþ antiporter, exhibits reduced growth rate as compared to wild-type plants in saline conditions (Olias et al., 2009).

Achard, P., Cheng, H., Grauwe, L.D., Decat, J., Schoutteten, H., Moritz, T., Straeten, D.V.D., Peng, J., Harberd, N.P., 2006. Integration of plant responses to environmentally activated phytohormonal signals. Science 311, 91e94. Achuo, E.A., Prinsen, E., Höfte, M., 2006. Influence of drought, salt stress and abscisic acid on the resistance of tomato to Botrytis cinerea and Oidium neolycopersici. Plant Pathol. 55, 178e186. Antoni, R., Rodriguez, L., Gonzalez-Guzman, M., Pizzio, G.A., Rodriguez, P.L., 2011. News on ABA transport, protein degradation, and ABFs/WRKYs in ABA signaling. Curr. Opin. Plant Biol. 14, 1e7. Asch, F., Dingkuhn, M., Miezan, K., Dörffling, K., 2000. Leaf Kþ/Naþ ratio predicts salinity induced yield loss in irrigated rice. Euphytical 113, 109e118. Aziz, A., Martin-Tanguy, J., Larher, F., 1998. Stressed-induced changes in polyamine and tyramine levels can regulate proline accumulation in tomato leaf discs treated with sodium chloride. Physiol. Plant 104, 195e202. Bates, L.S., Waldren, R.P., Teare, I.D., 1973. Rapid determination of free proline for water stress studies. Plant Soil 39, 205e207. Borsani, O., Cuartero, J., Fernandez, J.A., Valpuesta, V., Botella, M.A., 2001. Identification of two loci in tomato reveals distinct mechanisms for salt tolerance. Plant Cell 13, 873e887. Cavalcanti, F.R., Oliveira, J.T.A., Martins-Miranda, A.S., Viégas, R., Silveira, J.A.G., 2004. Superoxide dismutase, catalase and peroxidase activities do not confer protection against oxidative damage in salt-stressed cowpea leaves. New Phytol. 163, 563e571. Chen, C.C.S., Plant, A.L., 1999. Salt-induced protein synthesis in tomato roots: the role of ABA. J. Exp. Bot. 50, 677e687. Cohen, A., Moses, M.S., Plant, Á.L., Bray, E.A., 1999. Multiple mechanisms control the expression of abscisic acid (ABA)-requiring genes in tomato plants exposed to soil water deficit. Plant Cell Environ. 22, 989e998. Cuartero, J., Bolarin, M.C., Asins, M.J., Moreno, V., 2006. Increasing salt tolerance in the tomato. J. Exp. Bot. 57, 1045e1058. Dekkers, B.J., Willems, L., Bassel, G.W., Van Bolderen-Veldkamp, R.P., Ligterink, W., Hilhorst, H.W., Bentsink, L., 2012. Identification of reference genes for RTqPCR expression analysis in Arabidopsis and tomato seeds. Plant Cell Physiol. 53, 28e37. Dodd, A.N., Kudla, J., Sanders, D., 2010. The language of calcium signaling. Annu. Rev. Plant Biol. 61, 593e620. Foolad, M.R., Lin, G.Y., 1997. Genetic potential for salt tolerance during germination in Lycopersicon species. Hortic. Sci. 32, 296e300. Foolad, M.R., Lin, G.Y., 1999. Relationships between cold- and salt-tolerance during seed germination in tomato: germplasm evaluation. Plant Breed. 118, 45e48. Hu, X.L., Jiang, M.Y., Zhang, J.H., Zhang, A.Y., Lin, F., Tan, M.P., 2006. Calciumcalmodulin is required for abscisic acid-induced antioxidant defense and functions both upstream and downstream of H2O2 production in leaves of maize (Zea mays) plants. New Phytol. 173, 27e38. Huertas, R., Olias, R., Eljakaoui, Z., Galvez, F.J., Li, J., De Morales, P.A., Belver, A., Rodriguez-Rosales, M.P., 2012. Overexpression of SlSOS2 (SlCIPK24) confers salt tolerance to transgenic tomato. Plant Cell Environ. 35, 1467e1482. Janicka-Russak, M., Klobus, G., 2007. Modification of plasma membrane and vacuolar Hþ-ATPases in response to NaCl and ABA. J. Plant Physiol. 164, 295e302. Jiang, M., Zhang, J., 2002. Role of abscisic acid in water stress-induced antioxidant defense in leaves of maize seedlings. Free Radic. Res. 36, 1001e1015. Juan, M., Rosa, M., Rivero, M.R., Romero, L., Ruiz, J.M., 2005. Evaluation of some nutritional and biochemical indicators in selecting salt-resistant tomato cultivars. Environ. Exp. Bot. 54, 193e201. Jung, C., Seo, J.S., Han, S.W., Koo, Y.J., Kim, C.H., Song, S.I., Nahm, B.H., Choi, Y.D., Cheong, J.J., 2008. Overexpression of AtMYB44 enhances stomatal closure to confer abiotic stress tolerance in transgenic Arabidopsis. Plant Physiol. 146, 623e635. Lu, Y., Ouyang, B., Zhang, J., Wang, T., Lu, C., Han, Q., Zhao, S., Ye, Z., Li, H., 2012. Genomic organization, phylogenetic comparison and expression profiles of annexin gene family in tomato (Solanum lycopersicum). Gene 499, 14e24.

Acknowledgments This work was partly supported by the Fundamental Research Funds (2013PY079, 2012AA100104) to Ouyang and the grant to Guo

Contributions R. Yang, B. Ouyang and Y.-D. Guo designed research; R. Yang, T. Yang, H. Zhang, Y. Qi, Y. Xing, N. Zhang, R. Li performed research; R. Yang, S. Ren, B. Ouyang and Y.-D. Guo analyzed data; R. Yang, S. Weeda, S. Ren, B. Ouyang and Y.-D. Guo wrote the paper. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.plaphy.2014.01.015.

34

R. Yang et al. / Plant Physiology and Biochemistry 77 (2014) 23e34

Maggio, A., Barbieri, G., Raimondi, G., Pascale, S.D., 2010. Contrasting effects of GA3 treatments on tomato plants exposed to increasing salinity. J. Plant Growth Regul. 29, 295e302. Mahmoud, M.H., Jones, R.A., El Beltagy, A.S., 1986. A comparative responses to high salinity between salt sensitive and salt tolerant genotypes of the tomato. Acta Hort. 190, 533e543. Melchor, J., Rosa, M.R., Luis, R., Juan, M.R., 2005. Evaluation of some nutritional and biochemical indicators in selecting salt-resistant tomato cultivars. Environ. Exp. Bot. 54, 193e201. Mittova, V., Guy, M., Tal, M., Volokita, M., 2002. Response of the cultivated tomato and its wild salt-tolerant relative Lycopersicon pennellii to salt-dependent oxidative stress: increased activities of antioxidant enzymes in root plastids. Free Radic. Res. 36, 195e202. Murashige, T., Skoog, F., 1962. A revised medium for rapid growth and bioassays with tobacco cultures. Physiol. Plant 15, 473e497. Nakaune, M., Hanada, A., Yin, Y.G., Matsukura, C., Yamaguchi, S., Ezura, H., 2012. Molecular and physiological dissection of enhanced seed germination using short-term low-concentration salt seed priming in tomato. Plant Physiol. Biochem. 52, 28e37. Olias, R., Eljakaoui, Z., Li, J., De Morales, P.A., Marin-Manzano, M.C., Pardo, J.M., Belver, A., 2009. The plasma membrane Naþ/Hþ antiporter SOS1 is essential for salt tolerance in tomato and affects the partitioning of Naþ between plant organs. Plant Cell Environ. 32, 904e916. Ouyang, B., Yang, T., Li, H.X., Zhang, L., Zhang, Y.Y., Zhang, J.H., Fei, Z.J., Ye, Z.B., 2007. Identification of early salt stress response genes in tomato root by suppression subtractive hybridization and microarray analysis. J. Exp. Bot. 58, 507e520. Park, J., Kim, Y.S., Kim, S.G., Jung, J.H., Woo, J.C., Park, C.M., 2011. Integration of auxin and salt signals by the NAC transcription factor NTM2 during seed germination in Arabidopsis. Plant Physiol. 156, 537e549. Rodriguez-Rosales, M.P., Jiang, X., Galvez, F.J., Aranda, M.N., Cubero, B., Venema, K., 2008. Overexpression of the tomato Kþ/Hþ antiporter LeNHX2 confers salt tolerance by improving potassium compartmentalization. New Phytol. 179, 366e377. Rubio, L., Rosado, A., Linares-Rueda, A., Borsani, O., Garcia-Sanchez, M.J., Valpuesta, V., Fernandez, J.A., Botella, M.A., 2004. Regulation of Kþ transport in tomato roots by the TSS1 locus. Implications in salt tolerance. Plant Physiol. 134, 452e459. Rush, D.W., Epstein, E., 1981. Comparative studies on the sodium, potassium, and chloride relations of a wild halophytic and a domestic salt-sensitive tomato species. Plant Physiol. 68, 1308e1313. Saranga, Y., Zamir, D., Marani, A., Rudich, J., 1993. Breeding tomatoes for salt tolerance-variations in ion concentrations associated with response to salinity. J. Am. Soc. Hortic. Sci. 118, 405e408. Shalata, A., Mittova, V., Volokita, M., Guy, M., Tal, M., 2001. Response of the cultivated tomato and its wild salt-tolerant relative Lycopersicon pennellii to saltdependent oxidative stress: the root antioxidative system. Physiol. Plant 112, 487e494.

Sun, L., Wang, Y.P., Chen, P., Ren, J., Ji, K., Li, Q., Li, P., Dai, S.J., Leng, P., 2011. Transcriptional regulation of SlPYL, SlPP2C, and SlSnRK2 gene families encoding ABA signal core components during tomato fruit development and drought stress. J. Exp. Bot. 62, 5659e5669. Sun, L., Yuan, B., Zhang, M., Wang, L., Cui1, M., Wang, Q., Leng, P., 2012. Fruit-specific RNAi-mediated suppression of SlNCED1 increases both lycopene and b-carotene contents in tomato fruit. J. Exp. Biol. 63, 3097e3108. Tal, M., Katz, A., Heiken, H., Dehan, K., 1979. Salt tolerance in the wild relatives of the cultivated tomato: proline accumulation in Lycopersicon esculentum Mill., Peruvianum Mill. and Solanum pennellii Cor. treated with NaCl and polyethylene glycol. New Phytol. 82, 349e360. Thompson, A.J., Andrews, J., Mulholland, B.J., McKee, J.M.T., Hilton, H.W., Horridge, J.S., Farquhar, G.D., Smeeton, R.C., Smillie, I.R.A., Black, C.R., et al., 2007. Overproduction of abscisic acid in tomato increases transpiration efficiency and root hydraulic conductivity and influences leaf expansion. Plant Physiol. 143, 1905e1917. Tung, S.A., Smeeton, R., White, C.A., Black, C.R., Taylor, I.B., Hilton, H.W., Thompson, A.J., 2008. Over-expression of LeNCED1 in tomato (Solanum lycopersicum L.) with the rbcS3C promoter allows recovery of lines that accumulate very high levels of abscisic acid and exhibit severe phenotypes. Plant Cell Environ. 31, 968e981. Villalta, I., Reina-Sanchez, A., Bolarin, M.C., Cuartero, J., Belver, A., Venema, K., Carbonell, E.A., Asins, M.J., 2008. Genetic analysis of Naþ and Kþ concentrations in leaf and stem as physiological components of salt tolerance in tomato. Theor. Appl. Genet. 116, 869e880. Wang, Y.H., Garvin, D.F., Kochian, L.V., 2001. Nitrate-induced genes in tomato roots. Array analysis reveals novel genes that may play a role in nitrogen nutrition. Plant Physiol. 127, 345e359. Wang, Y., Wisniewski, M., Meilan, R., Cui, M., Webb, R., Fuchigami, L., 2005. Overexpression of cytosolic ascorbate peroxidase in tomato confers tolerance to chilling and salt stress. J. Am. Soc. Hortic. Sci. 130, 167e173. Wang, Y., Li, K., Li, X., 2009. Auxin redistribution modulates plastic development of root system architecture under salt stress in Arabidopsis thaliana. J. Plant Physiol. 166, 1637e1645. Wilkinson, S., Davies, W.J., 2002. ABA-based chemical signalling: the coordination of responses to stress in plants. Plant Cell Environ. 25, 195e210. Yarra, R., He, S.J., Abbagani, S., Ma, B., Bulle, M., Zhang, W.K., 2012. Overexpression of a wheat Naþ/Hþ antiporter gene (TaNHX2) enhances tolerance to salt stress in transgenic tomato plants (Solanum lycopersicum L.). Plant Cell Tiss. Organ Cult. 111, 49e57. Yu, J.N., Huang, J., Wang, Z.N., Zhang, J.S., Chen, S.Y., 2007. An Naþ/Hþ antiporter gene from wheat plays an important role in stress tolerance. J. Biosci. 32, 1153e1161. Zhao, Q., Zhao, Y.J., Zhao, B.C., Ge, R.C., Li, M., Shen, Y.Z., Huang, Z.J., 2009. Cloning and functional analysis of wheat V-Hþ-ATPase subunit genes. Plant Mol. Biol. 69, 33e46. Zhu, J.K., 2002. Salt and drought stress signal transduction in plants. Annu. Rev. Plant Biol. 53, 247e273.