Wheat CBL-interacting protein kinase 25 negatively regulates salt tolerance in transgenic wheat

CBL-interacting protein kinases are involved in plant responses to abiotic stresses, including salt stress. However, the negative regulating mechanism of this gene family in response to salinity is less reported. In this study, we evaluated the role of TaCIPK25 in regulating salt response in wheat. Under conditions of high salinity, TaCIPK25 expression was markedly down-regulated in roots. Overexpression of TaCIPK25 resulted in hypersensitivity to Na+ and superfluous accumulation of Na+ in transgenic wheat lines. TaCIPK25 expression did not decline in transgenic wheat and remained at an even higher level than that in wild-type wheat controls under high-salinity treatment. Furthermore, transmembrane Na+/H+ exchange was impaired in the root cells of transgenic wheat. These results suggested that TaCIPK25 negatively regulated salt response in wheat. Additionally, yeast-one-hybrid, β-glucuronidase activity and DNA-protein-interaction-enzyme-linked-immunosorbent assays showed that the transcription factor TaWRKY9 bound W-box in the TaCIPK25 promoter region. Quantitative real-time polymerase chain reaction assays showed concomitantly inverted expression patterns of TaCIPK25 and TaWRKY9 in wheat roots under salt treatment, ABA application and inhibition of endogenous ABA condition. Overall, based on our results, in a salt stress condition, the negative salt response in wheat involved TaCIPK25 with the expression regulated by TaWRKY9.

acid (ABA)-dependent and -independent pathways, several transcription factors and promoter elements have also been identified in salt-stress signaling of plants 28 . However, the detailed regulation of CIPKs in these signaling pathways is less characterized. In a previous study, we isolated a wheat CIPK, i.e., TaCIPK25 (homology to AtCIPK12, 18 and 19) 29 . In this study, based on our results, the overexpression of TaCIPK25 impaired salt tolerance, which was mediated by a WRKY transcription factor in an ABA-dependent pathway under saline conditions.

Results
Expression patterns of TaCIPK25 in response to salt and ABA application. To analyze the expression patterns of TaCIPK25, the relative expression levels of TaCIPK25 in different wheat tissues were detected using fluorescent quantitative analysis (qRT-PCR). TaCIPK25 was expressed in all assayed tissues, including coleoptile, root, stem, leaf, pistil and stamens (Fig. 1A). To determine whether TaCIPK25 plays role in responses to abiotic stresses in wheat, the expression patterns of TaCIPK25 in seedling roots and leaves were detected under ABA (100 μ M) and NaCl (200 mM) treatments (Fig. 1B). The expression level of TaCIPK25 in leaves initially decreased slightly, then was up-regulated at 6 h and increased to peak levels of expression under ABA and NaCl treatments. For gene expression in roots, expression of TaCIPK25 decreased rapidly in 1 h and then decreased consistently to the lowest level of expression at 24 h in both NaCl and ABA treatments.
ABA is a well-known signal molecule in abiotic stress responses and regulates many stress-responsive genes 30 . To explore whether the expression of TaCIPK25 in response to NaCl was related to the ABA-dependent pathway, we used fluridone to inhibit endogenous ABA biosynthesis. Under high salt treatment, the down-regulation of TaCIPK25 expression was impaired in fluridone pretreated seedling roots compared with untreated seedling roots (Fig. 1C). Based on these results, TaCIPK25 down-regulated expressions induced by NaCl were involved in ABA signaling pathway.
TaWRKY9 binding to the promoter region of TaCIPK25. Cis-elements play critical roles in the regulation of gene expression. Therefore, we isolated the 1,000-bp upstream sequence from the start codon of TaCIPK25 and searched for putative elements using PLACE. We identified many putative cis-elements, including five putative WRKY transcription factor-binding sites (W1-W5; W-box core sequence, TGAC; Fig. 2A). WRKY transcription factors are both positive and negative regulators of ABA signaling and rapidly activate gene expression in plant defense processes 31,32 . In a previous study, we demonstrated that TaWRKYs were involved in improving stress tolerance in transgenic tobacco 33,34 ; therefore, we selected three representative TaWRKY transcription factors (TaWRKY1, TaWRKY9 and TaWRKY44) to verify the interactions with promoters of TaCIPK25.
In a yeast one-hybrid (Y1H) assay, yeast cells transformed with three combinations (TaWRKY9-W4, TaWRKY9-W5, and TaWRKY44-W5) of TaWRKYs, and PHIS2-W-boxes grew well on the selective media (50-100 mM 3-AT; Fig. 2A and Supplementary Fig. 1), which was an indication that TaWRKY9 and TaWRKY44 physically interacted with the promoter of TaCIPK25. In our previous study using the Y1H method (positive control), TaWRKY44 interacted with the typical TTGACC element 34 , but this type of element was not found in the promoter region of TaCIPK25. No interaction between TaWRKY1 and W-boxes was detected in this analysis. Additionally, we used DPI-ELISA to further confirm these interactions; however, we did not analyze TaWRKY44 in this assay because we could not purify the recombinant protein effectively. As shown in Fig. 2B, TaWRKY1 gave a significant light absorption signal with the W1-W5 boxes, whereas TaWRKY9 had a strong signal only with the W5 element. Therefore, specific binding between TaWRKY9 and the W5 element was verified by both Y1H and DPI-ELISA methods. Notably, DPI-ELISA assay results were not completely consistent with those of the Y1H method, particularly for TaWRKY1. These differences might be related to the difference in protein purification method (purification in a prokaryotic expression system) with DPI-ELISA. We selected Gus gene as a reporter driven by the TaCIPK25 promoter and transiently expressed the promoter-Gus construct in transgenic tobacco lines overexpressing TaWRKY1, TaWRKY9, and TaWRKY44. As shown in Fig. 2C, the relative Gus activities in the TaWRKY1 and TaWRKY44 transgenic lines had no significant differences compared with control lines. However, the relative Gus activity in TaWRKY9 transgenic lines decreased compared with control lines, which again confirmed that TaWRKY9 acted on the promoter region of TaCIPK25 by binding the W5 element and down-regulating the expression.
Expression patterns of TaWRKYs in response to NaCl and ABA treatments. To confirm the correlations of expression profiles between TaCIPK25 and TaWRKYs (TaWRKY1, 9 and 44), the expression of TaWRKYs was also determined by qRT-PCR in wheat. Under ABA and NaCl treatments, TaWRKY1 and TaWRKY44 were up-regulated in leaves and down-regulated in roots ( Supplementary Fig. 2). TaWRKY9 was rapidly up-regulated and reached peak levels in 1 h in leaves and roots of wheat plants following ABA or NaCl treatment (Fig. 3A). Notably, the expression patterns of TaWRKY9 were in contrast to those of TaCIPK25 (Fig. 3B). Whereas the transcript levels of TaWRKY9 rapidly accumulated in 1 h after treatment with NaCl and ABA, expression of TaCIPK25 was down-regulated, particularly in roots. However, in saline conditions, expression patterns of TaWRKY9 were inverted, when endogenous ABA was inhibited by fluridone (Fig. 3C). Based on these results, TaWRKY9 was a candidate transcription factor for the negative regulation of the expression of TaCIPK25 by the ABA-dependent pathway under salt stress.
Overexpression of TaCIPK25 decreased salt tolerance and affected Na + accumulation in Arabidopsis. Considering the low success rate of obtaining positive transgenic wheat for transformed genes, we first confirmed the salt tolerance performance in a model plant species. Transgenic Arabidopsis plants overexpressing TaCIPK25 driven by CaMV 35S promoter were generated. Among these transgenic lines, OE7, OE17 and OE23 were randomly selected for further analysis. Overexpression of TaCIPK25 in Arabidopsis was verified by RT-PCR; the three lines had different levels of TaCIPK25 expression ( Supplementary Fig. 3). For the salt tolerance assay, four-leaf stage, MS-germinated seedlings were transferred to MS medium or MS medium supplied with 150 mM NaCl. After 20 d, the growth phenotype of the transgenic lines was weaker than that of WT and vector control (VC) plants (Fig. 4B). For further analysis, 10-day-old seedlings of WT, VC and transgenic lines grown in soil were irrigated with 200 mM NaCl at 5-day intervals. As shown in Fig. 4A, the difference in growth status of transgenic lines and control plants began to appear 10 d after NaCl treatment. With an increase in time, leaf death was greater in transgenic lines than that in control plants. These results suggested that the tolerance of transgenic lines was impaired by overexpression of TaCIPK25 in Arabidopsis. Anti-oxidative systems are typically used to determine the salt tolerance of plants subjected to salt stress. In this study, the H 2 O 2 content of transgenic lines was higher than those of controls ( Supplementary Fig. 4). Expression of TaCIPK25 was induced when wheat plants were treated with H 2 O 2 . To explore whether TaCIPK25 participated in oxidative stress, the transgenic and control plants were treated with methyl viologen (causing oxidative damage); no obvious difference in the growth of plants between transgenic and control lines was observed within 20 d of treatment ( Supplementary Fig. 5). These data suggested that the peroxides caused by salt stress were not responsible for the excessively withered leaves of transgenic Arabidopsis under a saline condition; thus, TaCIPK25 did not improve the tolerance of Arabidopsis to oxidative damage.
The phenotype with impaired salt tolerance of transgenic lines might be explained by the accumulation of Na + /K + . To determine whether this assumption was valid, the Na + /K + content in leaves of the transgenic lines and control plants was measured following salt treatment. Based on the results, the accumulation of Na + was higher in transgenic plants than that in controls, whereas the content of K + was not significantly different ( Fig. 4C). Therefore, the overexpression of TaCIPK25 resulted in a greater accumulation of Na + in transgenic Arabidopsis. Generation of transgenic wheat with overexpression of TaCIPK25. The ectopic expression of TaCIPK25 in Arabidopsis provided a preliminary view of TaCIPK25 function, whereas the actual role of TaCIPK25 depends on performance in wheat. Therefore, the CDS of TaCIPK25 was constructed into an overexpression vector (OV-TaCIPK25) driven by the maize ubiquitin promoter. To detect transcripts of TaCIPK25 derived from OV-TaCIPK25, specific primers were added to bilateral sites of the insertion gene (Fig. 5A). The OV-TaCIPK25 DNA was bombarded into wheat (Chinese spring cultivar) immature embryos to generate TaCIPK25-overexpressing transgenic wheat plants. Five positive transgenic T 0 plants were identified by PCR with transgene-specific primers, and PCR products were sequenced to confirm the validity. When the transgenic lines were proliferated to T 2 progenies, two stable transgenic lines, OE40-2 and OE40-3, were obtained with transcripts of TaCIPK25 (derived from OV-TaCIPK25) detected in both lines by PCR assay. We found expression of TaCIPK25 in root and leaf tissues (Fig. 5B). Based on qRT-PCR results, the TaCIPK25 transgenic lines had much higher levels of TaCIPK25 expression than that of the wild type under normal conditions (Fig. 5C). Notably, the transcript levels of TaCIPK25 in transgenic lines did not decrease following treatment with 150 mM NaCl, which suggested that these two transgenic wheat lines could be used in further analyses.

Overexpression of TaCIPK25 impaired salt tolerance in wheat.
To evaluate the effect of TaCIPK25 on salinity tolerance in wheat, wild type and transgenic lines were germinated and grown in MS liquid medium, and then 10-day-old seedlings were transferred to MS liquid medium containing 150 mM NaCl. Transgenic wheat plants withered on MS liquid medium containing 150 mM NaCl after 24 h treatment and were almost dead after 96 h, whereas WT plants grew normally (Fig. 5D). Based on these results, the overexpression of TaCIPK25 significantly impaired salt tolerance in wheat. Twenty-four hours after salt treatment, the Na + /K + content in roots and leaves of wild type and transgenic seedlings was measured. As shown in Fig. 5E, the accumulation of Na + in the roots and leaves of the transgenic lines increased compared with that in those of wild-type controls (the K + content was not significantly different). These results were similar to those observed in transgenic Arabidopsis (Fig. 4C); thus, the excess accumulation of Na + was an indication of impaired salt tolerance in transgenic wheat. TaCIPK25 specific regulation of Na + accumulation in wheat. The sodium transport mechanism in plants primarily includes Na + specific channels and nonselective cation channel pathways 1,9,35 . Therefore, whether the higher Na + content in transgenic lines was related to Na + specific channels or to nonselective cation channels must be determined. In transgenic Arabidopsis and wheat overexpressing TaCIPK25, no obvious changes were observed in the K + content between transgenic and control lines under salt treatment (Figs 4C and 5E). For further analysis, the transgenic and wild type wheat were treated with 150 mM KCl for 24 h, and no obvious difference in K + content was observed (Fig. 6D). When seedlings were treated with 75 mM NaCl and 75 mM KCl, transgenic lines accumulated more Na + than that in the wild type, but no significant changes occurred in K + content. Under normal conditions (MS-treated), the Na + /K + content in transgenic and wild type seedlings was not different.
In a previous study, the wheat gene TaLCT1, a low affinity cation transporter, generated hypersensitivity to Na + in yeast and mediated the uptake of Na + , Cs + and Li + 36 . In this study, 24 h after treatment with 150 mM CsCl or 150 mM LiCl, the transgenic lines overexpressing TaCIPK25 did not accumulate more Cs + and Li + than wild-type controls (Fig. 6B,C). However, the addition of 75 mM NaCl with 75 mM CsCl restrained the TaCIPK25-induced Na + accumulation, and Na + did not affect the accumulation of Cs + (Fig. 6B). Additionally, transgenic lines that overexpressed TaCIPK25 accumulated more Li + only when NaCl was added (Fig. 6C), indicating a likely Na + -induced Li + accumulation pathway. In general, based on these results, Na + homeostasis regulated by TaCIPK25 in transgenic wheat lines did not involve nonselective cation channel pathways.
TaCIPK25 involvement in regulation of Na + /H + exchanges. In a previous study using the yeast two-hybrid method, we found that TaCIPK25 interacted with TaCBL1 29 . To further confirm this interaction and determine the localization, TaCBL1 and TaCIPK25 were respectively constructed to the C-and N-terminals of yellow fluorescent protein (YFP) for a bimolecular fluorescence complementation (BiFC) assay and to GFP fusion constructs for intracellular locations. As shown in Supplementary Fig. 6, TaCBL1 was primarily localized at the cell membrane and TaCIPK25 was distributed ubiquitously within the cell. Notably, yellow fluorescent signal was observed at the cell membrane in the BiFC analysis, which indicated the interaction between TaCBL1 and TaCIPK25 in plant cells and a potential functional site.
To explore whether co-localization of TaCBL1 and TaCIPK25 indicated a potential functional site for ion flux regulation at the cell membrane, we measured the transmembrane ion fluxes noninvasively using the scanning ion-selective electrode technique (SIET). For the measurement of Na + efflux, wheat plants were pretreated with 50 mM NaCl for 3 h and then transferred to measure buffer for Na + flux measurements. As shown in Fig. 7A, the net immediate Na + efflux was lower in transgenic lines than that in control seedlings with NaCl treatment; therefore, overexpression of TaCIPK25 in wheat likely decreased Na + efflux from cells. For net H + efflux, we found no significant differences between WT and transgenic wheat lines, when wheat plants were put into measuring buffer. However, the H + effluxes of the wild type were significantly more impaired than those of transgenic wheat lines after the addition of 50 mM NaCl (Fig. 7B), which indicated lower H + influxes in transgenic wheat lines. Thus, based on these results, TaCIPK25 was likely involved in regulating Na + /H + exchange.

Discussion
High soil salinity is one of the factors that most restricts plant growth and development, and plants have developed many mechanisms to cope with salt stress. An intriguing question is how this salinity response mechanism is initiated and then transmitted to downstream functional proteins when plants are exposed to a saline environment. ABA is an important signal molecule that plants use to respond to abiotic stress, which also regulates many stress-responsive genes 30 . WRKY transcription factors are both positive and negative regulators of ABA signaling and rapidly activate gene expression in plant defensive processes 31,32 . In this study, we found interactions among ABA, WRKY and CIPK in the regulation of ABA pathways. Expression levels of TaCIPK25 were regulated by ABA and high salinity, and when endogenous ABA was inhibited by fluridone, expression recovered in roots (Fig. 1C), which indicated expression was regulated by an ABA-dependent pathway. Notably, the expression patterns of TaWRKY9 were in contrast to those of TaCIPK25 under salinity and ABA application with levels of expression also affected by eliminating endogenous ABA (Fig. 3B). We showed that TaWRKY9 bound directly to the promoter of TaCIPK25 and negatively regulated the expression levels of TaCIPK25 (Fig. 2). Therefore, TaWRKY9 likely negatively regulated the expression of TaCIPK25 through an ABA-dependent pathway under salt stress.
To explore the effects on wheat plants caused by the regulation of TaCIPK25 expression by TaWRKY9 under salt treatment, transgenic Arabidopsis and wheat plants were generated that overexpressed TaCIPK25. In early studies, overexpression of CIPKs increased salt tolerance in many plants 25,[37][38][39] . In this study, overexpression of TaCIPK25 in Arabidopsis and wheat had negative effects on salt tolerance with overexpression in transgenic lines leading to the accumulation of more Na + than that in wild-type controls, which indicated a different mechanism for involvement of CIPK in the response to salt stress. At the cellular level, Na + accumulation in plants primarily occurs through two distinct processes, Na + influx and efflux. For Na + uptake, many nonselective cation channels (NSCCs) are identified as Na + permeable channels [40][41][42] , which also have other cation transportation capabilities. For example, TaLCT1 is a low-affinity cation transporter that mediates the uptake of Na + , Li + and Cs + , and the uptake of Na + through LCT was inhibited by K + and Cs + in yeast 36 . Our results showed that transgenic wheat lines overexpressing TaCIPK25 specifically accumulated Na + , whereas the transport of Li + and Cs + was unaffected. Moreover, neither Li + nor K + inhibited Na + uptake in these plants (Fig. 6). For Cs + , the addition of CsCl eliminated the difference in Na + content between control and transgenic lines (Fig. 6B), implying a Bars represent the mean ± SD (n = 3, * P < 0.05, * * P < 0.01 by Student's t-test).
Scientific RepoRts | 6:28884 | DOI: 10.1038/srep28884 potential function related to TaLCT1 or other ion homeostasis mechanism. However, no significant differences in Cs + content between control and transgenic lines were observed under the 150 mM CsCl treatment. Therefore, based on current data, TaCIPK25 was apparently not involved in the regulation of TaLCT1 or other NSCCs. In other studies, the durum wheat SOS1 and Arabidopsis NHX8 (SOS) proteins efficiently mediated not only Na + efflux but also Li + efflux 43,44 . Notably, transgenic wheat overexpressing TaCIPK25 accumulated Li + through a Na + -dependent pathway (Fig. 6C), and low levels of Na + effluxes and H + influxes were observed in the TaCIPK25 overexpressed wheat lines (Fig. 7), which suggested a plausible relationship between TaCIPK25 and SOS pathways with different roles in regulating the Na + /H + antiporter. However, based on the current data, we cannot conclude that TaCIPK25 participated in negative regulation of the SOS pathway. Nevertheless, specific Na + transportation was a reasonable explanation for the TaCIPK25-induced salt sensitive phenotype, which also provided a clue for the exploration of potential target proteins of TaCIPK25 in future work.
In conclusion, this study provides insights into the mechanisms of gene expression regulation and salt response of TaCIPK25 in wheat (Fig. 8). Under nonsaline conditions, TaCIPK25 is persistently expressed in wheat roots and might be involved in uptake of nutritional Na + . When plants encounter high-salinity stress, ABA, as an important signal molecule, regulates expression of many stress-responsive genes, including TaWRKY9. The ABA-dependent pathway to induce the transcription factor gene TaWRKY9 which represses TaCIPK25 expression in wheat roots rapidly induced and further affected Na + transportation. The sensitive phenotypic characteristics that were observed in transgenic wheat were closely related to the expression levels of TaCIPK25, which did not decline and even maintained a higher level of expression than that of wild-type wheat controls under the high salinity treatment. Therefore, the down-regulation of TaCIPK25 mediated by TaWRKY9 is a protection mechanism to cope with high salinity environments in wheat, which also provides useful information for the genetic manipulation of wheat to improve salt tolerance. Studies are also underway to examine specific Na + channels that directly interact with TaCIPK25 protein with the aim to better understand the involvement of TaCIPK25 in the salt response pathway.

Methods
Wheat treatments and qRT-PCR. The wheat cultivar Chinese spring was used in this study. For expression of TaCIPK25 in wheat tissues, seedling stage tissues (coleoptiles, roots, stems and leaves) were obtained from sterile seedlings, and flowering stage tissues (roots, stems, leaves, flag leaves, pollen and pistils), embryos (approximately 15 d after anthesis) and endosperm (approximately 15 d after anthesis) were cut from wheat plants grown in soil. For stress response analyses, 7-day-old wheat seedlings were treated with ABA (100 μ M), H 2 O 2 (10 μ M) and NaCl (200 mM). For treatments with inhibitors, seedlings were first pretreated with fluridone (10 μ M) for 2 h Assay to determine binding of WRKYs to cis-elements of TaCIPK25 promoter region. To analyze putative cis-elements in the promoters of TaCIPK25 gene, a 1,000 bp upstream region of the gene was extracted from wheat genomic sequences and was searched for a cis-elements using PLACE (http://www.dna. affrc.go.jp/PLACE/signalup.html). The putative W-boxes (W1-W5, each containing three tandem repeat copies) were subcloned to PHis2.0 vectors, and WRKYs were constructed into pGADT7 vectors. The detailed experimental procedure followed that in a previous publication by Wang et al. 34 . DPI-ELISA assays followed the description in a previous publication 45 . Recombinant proteins were obtained by a prokaryotic expression system (expressed in E. coli BL21). The biotinylated complementary oligonucleotides were ordered from a company (BGI.tech, China). The relative GUS activity assays were performed, primarily, as described elsewhere 46,47 . All tested tobacco plants (TaWRKY1, TaWRKY9 and TaWRKY44) were transformed by constructs (proTaCIPK25:GUS, 35S:LUC) using the Agrobacterium-mediated transient expression procedure. The 35S:LUC construct was used as a transformation efficiency control. The level of expression (relative to the housekeeping gene) was similar in all transgenic tobacco lines used in this study.

Plant transformations.
To generate transgenic Arabidopsis overexpressing TaCIPK25, the CDS of TaCIPK25 was constructed to pBI121 vector driven by cauliflower mosaic virus 35S (CaMV 35S) promoter, and pBI121-TaCIPK25 was transferred into Agrobacterium tumefaciens strain LBA4404. The transformation of Arabidopsis was performed with the floral dip method 48,49 . T 0 positive seeds were selected on MS medium containing kanamycin (50 mg L −1 ), and T 2 generation seeds were selected for further analysis.
The wheat transformation protocol was conducted based on a particle bombardment method 50 . Genomic DNA extracted from leaf tissue by a PlantGen DNA Kit (Beijing ComWin Biotech, Beijing, China) was used for T 0 positive transgenic lines with specific primers with PCR (Supplementary Table 1). PCR products were confirmed by sequencing.

Assay of stress tolerance phenotype in transgenic lines and control plants. Seeds of WT, VC and
transgenic lines were germinated on MS medium after surface sterilization. Four-leaf stage seedlings (10 d) were transferred to MS medium containing 150 mM NaCl for salt stress analysis. For salt assays in soil, 10-day-old seedlings were irrigated with 200 mM NaCl at 5-day intervals. For anti-oxidation experiments in soil, 10-day-old seedlings were uniformly sprayed with 10 μ M methyl viologen at 5-day intervals.
After germination of wheat seeds, wheat seedlings were grown in water for 10 d. For salt tolerance assays, 10-day-old wheat seedlings were transferred to MS liquid medium with 150 mM NaCl with the medium refreshed daily.  Scientific RepoRts | 6:28884 | DOI: 10.1038/srep28884 0.5 mM NaCl, 0.2 mM Na 2 SO 4 , 0.1 mM MgCl 2 , 0.1 mM CaCl 2 , 0.3 mM MES, pH 6.0). The procedure used to measure ion fluxes was described in previous reports 51,52 . Briefly, electrodes were made from glass micropipettes and filled with selective liquid ion-exchange cocktails (Na: 100 mM NaCl; H: 40 mM KH 2 PO 4 ; and 15 mM NaCl, pH 7.0) for Net Na + /H + measurements. Only electrodes with slopes > 50 mV/decade were used in this study. Wheat seedlings were placed in measuring buffer without or with 50 mM NaCl for H + flux measurements. For net Na + , wheat seedlings pretreated with 50 mM NaCl for 3 h were transferred to measuring buffer (without NaCl). The fluxes of target ions were determined by ion concentration gradients between two positions close to the plant material in a preset excursion (30 μ m). Further details on data processing and calibration of SIET are found on the following web site: http://marlin.bio.umass.edu/biology/kunkel/nvp_umass.html.

Measurements of Na
For Arabidopsis, the leaves of seedlings treated with 150 mM NaCl were harvested at 10 d for ion and H 2 O 2 measurements. For ion assays in wheat plants, 10-day-old seedlings were transferred to MS liquid medium with or without ions (Na + , K + , Li + and Cs + ) and grown for 24 h. To measure the ion content, all samples were dried at 75 °C for 3 d, weighed and extracted with 1 N HNO 3 + 15% H 2 O 2 at 121 °Cfor 30 min. Ion concentrations were determined with an atomic absorption spectrophotometer (Analytik, Jena, Germany). The H 2 O 2 content was assessed using a commercially available kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

Subcellular localizations and biomolecular fluorescence complementation (BiFC). The cDNA
of TaCBL1 and TaCIPK25 was subcloned into modified pBI121 vectors (fused with GFP) and BiFC vectors (YNE-TaCBL1 and YCE-TaCIPK25) using specific primers (Supplementary Table S1) for transient expression in onion epidermal cells. These vectors and PM-RK (a plasma membrane-localized marker based on the full-length coding region of AtPIP2A) were transiently expressed in onion epidermal cells using a Biolistic Particle Delivery System (PDS-1000/He; Bio-Rad, CA, USA) 53 . Fluorescence was observed by inverted fluorescence microscopy (IX71; Olympus, Tokyo, Japan) after incubation at 25 °Cfor 16-24 h on MS medium.