The seeds of the Qinghai-Tibetan annual wild barley line XZ166 (H. spontanum) and cultivated barley CM72 (H. vulgare L. ssp. Vulgare), Arabidopsis thaliana ecotype Columbia-0 (Col-0), and rice Zhonghua11 (ZH11, Oryza sativa) were obtained from the Provincial Key Laboratory of Gene Resources, Zhejiang University, Hangzhou, China. The seeds of XZ166 and CM72 were surface sterilized in 5% NaClO for 15 min, rinsed with tap water, and germinated on moistened fine sand in plastic pots (15 cm × 20 cm × 15 cm) for 2 days at 25°C in the dark. The plastic pots with germinated seeds were then transferred to a greenhouse under a 14 h light (420 μmol m⁻² s⁻¹) period at 25°C and 10 h dark period at 23°C daily. We adopted 70% relative humidity and watered the plants with nutrient solution (pH 6.0, mg L⁻¹) that contained (NH₄)₂SO₄ (48.2), MgSO₄ (154.88), K₂SO₄ (15.9), KNO₃ (18.5), KH₂PO₄ (24.8), Ca(NO₃)₂ (86.17), Fe-citrate (7), MnCl₂·4H₂O (0.9), ZnSO₄·7H₂O (0.11), CuSO₄·5H₂O (0.04), HBO₃ (2.9), and H₂MoO₄ (0.01). The solution was renewed at 6-day intervals, and its value was adjusted to pH 6.0 ± 0.1 using 1 M HCl daily. CaCl₂ was added with NaCl to maintain a molar ratio of Na⁺:Ca²⁺ = 10:1.

The Arabidopsis ecotype Col-0 and the T₂ plants (HsCBL8_Promoter-GUS) were grown for 3 weeks in the greenhouse (16 h light, 110 μmol m⁻² s⁻¹), subsequently transferred to half strength MS liquid media, and allowed to grow for 1 week on the medium. After the plants became adapted to the MS media, different treatments, including 100 μM ABA, 100 mM NaCl, and 20% PEG6000 with 5% sucrose were applied at 4°C for 12 h. Moreover, plants in different developmental stages were used for β-glucuronidase GUS histochemical staining.

The seed dormancy of ZH11 and transgenic plants (35S-HsCBL8) was terminated with treatment of 0.5% H₂O₂ for 12 h. The seeds were then sterilized in 5% NaClO for 20 min, rinsed with tap water, and germinated on moistened filter papers in an incubator in a 14 h light period (110 μmol m⁻² s⁻¹) and 10 h dark period at 25°C. To calculate the germination rate, 20 rice seeds per biological repeat (four repeats independently) were treated with 125 mM NaCl in 1/4 nutrient solution. Moreover, the solution was renewed daily. Data on germination were collected from the “specific day” until 70 to 80% was reached per treatment for each genotype. Then the germination rate was calculated.

Seedlings of the XZ166 with three leaves were treated with 200 mM NaCl for 48 h, and their roots were selected for RNA extraction. Total RNA was extracted by a kit method (OMEGA, Norcross, GA RNA kit). PrimeScript^TM First Strand cDNA Synthesis Kit (Takara, Dalian China) was used to synthesize the cDNA. The genomic DNA of the XZ166 leaves was extracted using the CTAB method. The degenerate primers (Table S1) were designed based on cDNA sequences available on the National Center for Biotechnology Information (NCBI) database with accession IDs AL713904, FJ901265.1, EU085040.1, AB378095.1, XM_002524988.1, DQ907707.1, DQ201200.1, FJ901264.1, and FJ901259.1. The 25 μL PCR reaction mix used for amplification included L-Taq polymerase (Takara, Dalian China) (1.5 units), genomic DNA (25 ng), per primer 10 μM, dNTP (20 mM). PCR amplification was performed as follows: 95°C for 3 min, 10 cycles of 94°C for 30 s, and 60°C for 30 s at −0.5°C per cycle; 72°C for 60 s; and 25 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 60 s, and 72°C for 5 min. The PCR products were separated by 1.0% agarose gel electrophoresis. We obtained four wild barley CBL sequence candidates; one of which was identified as cultivated barley EST FLbaf27i23 under the Barley Database (http://www.shigen.nig.ac.jp/barley/). The coding sequences (CDSs) of its cDNA and genomic DNA were amplified using primers designed from FLbaf27i23, which contained additional Xba I and BamH I restriction sites (underlined, Table S1), respectively. The HsCBL8 promoter was cloned using inverse PCR (IPCR, http://dps.plants.ox.ac.uk/langdalelab/protocols) combined with nested PCR. Genomic DNA (2.5 μg) was cut by Xho I and Cla I independently; the DNAs ligated by T4 DNA ligase (Takara, Daliang China) were employed for amplification. The primers used are shown in Table S1. The prepared 25 μL PCR reaction mix included LTaq polymerase (Takara, Dalian China) (1.5 units), genomic DNA (50 ng), per primer (10 μM), and dNTPs (20 mM). PCR amplification was performed as follows: 95°C for 3 min, 10 cycles of 94°C for 30 s, 60°C for 30 s, −0.5°C per cycle, and 72°C for 3 min; 25 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 3 min; and 72°C for 5 min. PCR products were separated by 1.0% agarose gel electrophoresis. Then, the HsCBL8 promoter region was amplified using primers (Table S1) that contained additional Hind III and Xba I restriction sites (underlined).

Nucleotide and protein sequences (Table S2) were analyzed by BioEdit 7.2.5 software (Abbott, Carlsbad, Canada) and NCBI (http://blast.ncbi.nlm.nih.gov/). Amino acid sequence alignment was analyzed by Clustal X, and phylogenetic trees were established by the neighborhood-joining bootstrap method; using the default parameters in the software MEGA 6.06 (http://www.megasoftware.net/index.php). The cis elements of the HsCBL8 promoter sequence (the upstream 1000 bp from the initiation codon) were predicted by PlantCARE online (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/).

The transgenic Arabidopsis plants harboring HsCBL8_Promoter-GUS were generated according to the methods we previously described (Wang et al., 2014). In brief, a 2160 bp HsCBL8 promoter and GUS was inserted into pCAMBIA1300 binary vector. After verification by restrictive digestion and DNA sequencing, the construct was applied to transform wild type plants by floral dipping using Agrobacteria tumefaciens strain GV3103. Transgenic plants (T₀) survived on solid Murashige and Skoog (MS) medium containing 30 mg ml⁻¹ hygromycin were further verified by PCR genotyping. T1 and T2 seedlings were selected by PCR genotyping with the primers listed in Table S1.

pCAMBIA1300 was modified by inserting the CaMV 35S promoter-GUS (35S-GUS) region cut from pBI121 using Hind III and EcoR I and named as pCAMBIA1300-35S-GUS. The genomic DNA of the HsCBL8 coding region was inserted into the Xba I and BamH I sites of the pCAMBIA1300-35S-GUS plasmid to form expression vector pCAMBIA1300-35S-HsCBL8 (Figure 1A). This construct was transferred into Agrobacterium tumefaciens EHA105 and used for rice ZH11 transformation. Callus induction and subculture were performed on the medium supplemented with 2 mg/L 2,4-D, 3% sucrose, and 0.55% agar, named NB medium by Ozawa (2012). After a number of subcultures with an interval of 4 weeks, calli (between 1 and 3 mm in diameter) were used for transformation by co-culture with EHA105 harboring pCAMBIA1300-35S-HsCBL8. Then, the co-cultured calli were transferred onto pre-selection medium (NB supplemented with 30 mg/L hygromycin B), and incubated in the dark for 7 days. Subsequently, the differentiated calli were placed on the selection medium (NB with 50 mg/L hygromycin B) for another 14–17 days. The resistant calli were then transferred to pre-regeneration medium (NB supplemented with 2 mg/L 6-BA, 1 mg/L NAA, 5 mg/L ABA and 50 mg/L hygromycin B) and cultured in the dark for 7 days, followed by plant regeneration on medium NB supplemented with 3 mg/L 6-BA, 0.5 mg/L NAA and 50 mg/L hygromycin B, for 14–21 days under a 16 h photoperiod until shoots appeared. Regenerated plants were subsequently transferred into greenhouse and self-pollinated to produce progenies. The transgenic rice was first selected by hygromycin and PCR analysis using primers listed in Table S1, and then confirmed by Southern blot. The 35S region of the pCAMBIA1300-35S-GUS plasmid was replaced by the HsCBL8 promoter (2170 bp before the initiation codon; HsCBL8_promoter) to form the vector pCAMBIA1300-ProHsCBL8-GUS using Hind III and Xba I (Figure 1B). This construct was transferred into A. tumefaciens GV3101 and used for Arabidopsis (Col-0) transformation by the floral dip method with slight modification. The Arabidopsis plants were grown in a greenhouse for about 4 weeks with a cycle of 16 h light and 8 h dark periods before transformation. GV3101 cells, cultured in Luria–Bertani (LB) medium for 18–20 h at 28°C (optical density at 600 nm [OD₆₀₀]: 1.6–2.0), were collected by centrifugation and re-suspended in infiltration medium (half-strength MS medium, 5% sucrose, 0.044 μM benzylamino purine, and 0.2% Silwet L-77) to an OD₆₀₀ of 0.8–1.0. Plant inflorescences were dipped in the infiltration solution for 2 min, and transferred to a dark chamber in a greenhouse for 2 days, and then cultured under normal conditions. Transgenic plants (T₀) survived on solid Murashige and Skoog (MS) medium containing 30 mg ml⁻¹ hygromycin, were further verified by PCR genotyping. T1 and T2 seedlings were selected by PCR genotyping with the primers listed in Table S1.

Both genomic and cDNA were identified. The specific primers (Table S1) from HsCBL8 genomic DNA were used to separately amplify 679 bp fragments. The specific primers (Table S1) from the HsCBL8 cDNA were used to amplify 596 bp fragments, and OsGAPDH was used as reference gene (425 bp). The prepared 20 μL PCR reaction mix contained rTaq polymerase (Takara, Dalian China) (1.0 units), genomic DNA (20 ng), per primer (10 μM), and dNTP (10 mM). The following PCR program was used: 95°C for 4 min; 25 cycles of 94°C for 30 s, 58°C for 30 s, and 72°C for 60 s; and 72°C for 5 min. The PCR products were separated by 1.0% agarose gel electrophoresis.

Approximately 10 μg DNA per sample was separately and completely digested by Hind III and EcoR I, separately, and the restriction fragments were separated with 0.8% (w/v) agarose gel. Then, DNA was transferred onto a Hybond N⁺ nylon membrane (Boehringer, Mannheim, Germany) for 12 h, and the membrane was hybridized using a specific probe, including HsCBL8 and an NOS sequence (Figure 1A). The probe (679 bp) was then amplified in accordance with primers (Table S1) using plasmid pCAMBIA1300-35S-HsCBL8. The probe was labeled with the enzyme horseradish peroxidase (Roche DIG High Prime DNA Labeling and Detection Starter kit II) for 10–12 h at 65°C. After hybridization, the membrane was washed twice with 2 × standard saline citrate (SSC) plus 0.1% sodium dodecyl sulfate (SDS) at 65°C for 10 min and washed twice with 0.5 × SSC plus 0.1% SDS for 10 min at room temperature. The membrane was covered with the detection reagent mixture for 30 min in accordance with the supplier's instructions. Finally, the membrane was wrapped in Saran wrap with 1 ml CSPD ready-to-use for 5 min and subsequently detected by a Fuji Photo Film Releases LAS-3000 Imaging System (Tokyo, Japan).

Histochemical staining of GUS activity was performed as follows. Plant tissues were incubated at 37°C overnight in GUS staining buffer (2 mM 5-bromo-4-chloro-3-indolyl-β-glucuronic acid in 50 mM sodium Pi buffer, pH 7.2) containing 0.1% Triton X-100, 2 mM K₄Fe(CN)₆, 2 mM K₃Fe(CN)₆, and 10 mM EDTA. The stained seedlings were sequentially transferred to 50, 70, and 100% (v/v) ethanol to remove chlorophyll. The stained materials were examined with a dissecting microscope (Leica MZ95) with a digital camera. GUS activity was also determined by fluorometric assay (Gallagher, 1992).

The young rice seedlings (30 days) were treated with 125 mM NaCl in half-strength nutrient solution. Fresh weights were calculated at 0, 8, 16, and 24 h; plant phenotypes were photographed at 2, 4, and 6 days, and all leaves at 0, 24, and 72 h were used for analysis of proline, malondialdehyde (MDA), and Na⁺ and K⁺ contents, and rate of electrolyte leakage (REL). Plants without NaCl treatment were used as control. Meanwhile, proline content was determined as follows: 0.2 g fresh leaves were chopped and immersed in 5 ml of 3% sulphosalicylic acid at 100°C to extrac free proline. Proline content was then measured by UV Spectrophotometer (UV1000, LabTech) at 520 nm. MDA content was determined by the thiobarbituric acid (TBA) reaction. Leaves (0.2 g) were homogenized with 10 ml of 10% trichloroacetic acid (TCA) and centrifuged at 4000 rpm for 10 min. Part of the supernatant (2 ml) was mixed with 2 ml 0.6% TBA and incubated in 100°C water for 15 min, then cooled immediately on ice and centrifuged at 4000 rpm for 10 min. OD₄₅₀, OD₅₃₂, and OD₆₀₀ were determined by UV spectrophotometry (UV1000, LabTech). MDA concentration (MDAC) were calculated through the formula; MDAC (μmol/L) = [6.45 (OD₅₃₂ − OD₆₀₀) − 0.56 OD₄₅₀] × V × m, where V is the volume of the total extraction liquid and m is the sample weight. The leaves and roots were collected and dehydrated in a Muffle furnace at 500°C for 6 h. These ash samples were then extracted in 50% HNO₃ overnight. The Na⁺ and K⁺ contents were measured by atomic absorption spectrophotometry (Z-5000 Polarized Zeeman, Hitachi, Japan). Membrane permeability can be reflected by the REL. Fresh leaves were collected and washed three times with deionized water to remove surface-adhered electrolytes. Each sample was divided equally and placed into two sealed vials containing 10 ml of deionized water. One vial was incubated at 25°C on a rotary shaker for 3 h, and then the electrical conductivity (EC1) was measured with a conductivity meter (162A, Thermo Orion, USA). Another vial was autoclaved at 100°C for 20 min, and the electrical conductivity (EC2) was determined by a conductivity meter. The REL can then be defined as REL = (EC1/EC2).

In these analyses, four biological replicates were performed independently, and four plants were used in per replicate.

XZ166 and CM72 plants (with three leaves, about 10 days) were treated with 200 mM NaCl, and the roots and shoots of three plants were separately collected at 0, 1, 6, 12, 24, and 48 h. Similarly, ZH11 and transgenic (35S-HsCBL8) seedlings (30 days) were treated with 125 mM NaCl, and shoot of three plants were separately collected at 0, 3, and 6 h. Three biological replicates were carried out independently. The primers adopted are listed in Table S1. HvGapdh and actin mRNAs were separately employed as internal controls for barley and rice. Real-time PCR was performed using the CFX96 system (Bio-Rad, America) with iTaq Universal SYBR® Green Supermix (Bio-Rad). The PCR program was implemented as follows: 95°C for 3 min and 40 cycles of 95°C for 15 s, 58°C for 15 s, and 72°C for 15 s.

Statistical analysis for real-time PCR was conducted using the software CFX Manager (version 2.1), and the relative expression level of per gene was calculated by the 2^−ΔΔCT method (Livak and Schmittgen, 2001). Moreover, student's t-test was applied for each parameter studied. Differences were considered significant at P <0.05 and highly significant at P <0.01.

The first 494 bp cDNA sequence amplified with degenerate PCR primers (Table S1) was found similar to the full-length cDNA sequence of the gene FLbaf27i23 in barley (H. vulgare subsp. vulgare) (http://www.shigen.nig.ac.jp/barley/). BLASTX (http://blast.ncbi.nlm.nih.gov/Blast.cgi) analysis of the coding sequence (CDS) predicted 915 nucleotides, which code HvCBL8 with 305 amino acids (Figure 2A). A fragment of 873 bp at the CDS was cloned from wild barley XZ166 using primers based on FLbaf27i23, which encodes the HsCBL8 protein of 291 amino acids (Figure 2A, NCBI, AEM44693.1). We performed IPCR of the 2388 bp fragment (NCBI ID, HQ696007.1) upstream the initial codon. Moreover, 31 cis elements of the upstream 1000 bp from initiation codon were predicted by PlantCARE. Ten of these elements were light response elements, five were MeJA response elements, four were ABA response elements, four were endosperm expression elements, three were drought response elements, two were GA-responsive and circadian control elements, and one was a salicylic acid (SA) response element (Table 1). The predicted cis elements of the HsCBL8 promoter sequence indicated that HsCBL8 could be induced by different stresses (Table 1).

To study the phylogenetic relationship of HsCBL8, the amino acid sequence of HsCBL8 was compared with 71 CBL genes from plant species including monocot (rice, sorghum, maize and barley), dicot (Arabidopsis, grape and populous), and protozoan (Naegleria fowleri). The corresponding IDs are presented in Table S2. Accordingly, members of the family CBL family could be clustered into two groups (Figure 3). Group I only comprised four members. Three belonged to simple plants and N. fowleri, and one represented Arabidopsis (AtCBL5). By contrast, Group II was divided into two sub-clusters: II-I and II–II. Moreover, II-I was grouped into II-I-I and II-I-II, whereas II-I-I was sub-clustered into II-I-I-I and II-I-I-II. HsCBL8 was found closely related to rice OsCBL9 (63.70%) and thus placed in the group II-I-I-II, which included 12 CBLs (Figure 3). Results revealed that most of CBLs belonging to monocots were slightly deviant phylogenetically from dicot species, whereas the CBLs of lower plants were found in close relation to higher plant species in each group (Figure 3). Moreover, HsCBL8 was significantly different to HvCBL8 in terms of a 14-amino-acids deletion at its nitrite end, and five other changed sites, including three changed, one deletion, and one more amino acid (Figure 2A). Normally, the CBL protein contains four EF–hand domains for Ca²⁺ ion binding (Figure 2B), but the last domain was lost in HsCBL8 and HvCBL8 (Figure 2A). Obviously, a conserved myristoylation motif (MGCXXS/T) in the N-terminus of AtCBL1 was not discovered in the group II-I-I-II (Figure 2), including HsCBL8 (Figure 2A).

To evaluate whether HsCBL8 in XZ166 and HvCBL8 in CM72 could respond to salt, we conducted real time reverse-transcription PCR analysis to test their expression patterns in the control and under salt stress. Results revealed that the expression of both the genes in their respective plant species, were significantly induced in their shoots and roots, respectively, in response to 200 mM Nacl stress (Figure 4). Overall, HsCBL8 was expressed to significantly higher levels in both shoots and roots than that in HvCBL8. The expression profile of both the genes in the shoot generally followed a sigmoid curve in general. The expression level was enhanced in the first hour of NaCl application, reaching its peak after 6 h and then gradually declining with prolonged salt stress from 12 to 48 h, respectively (Figure 4A). In the roots, the expression of the CBL8s in both XZ166 and CM72 was greatly suppressed (Figure 4B) during first hour of NaCl treatment. However, both the genes were highly expressed at 3 and 6 h of stress treatment. Afterward, a gradual decline was observed in the expression pattern after up to 48 h of stress treatment. However, significant differences in transcriptional levels were observed between HvCBL8 and HsCBL8 at 3, 24, and 48 h in the roots (Figure 4B), revealing the variation in the degree of sensitivity of these genes to salt stress.

To study the role of HsCBL8 in rice, an Agrobacterium strain EHA105 harboring the plasmid pCAMBIA1300-35S-HsCBL8, was transformed into rice ZH11 using embryo calli induced from mature seeds. A Southern blot analysis using two restriction enzymes, Hind III and EcoR I, revealed the presence of two copies (T1-1 and T1-3) of the T1 transgenic line of rice (Figure 5). Of these transgenic plants, 20 T2 seedlings of T1-1 and 20 T2 seedlings T1-3 were verified by PCR and resulted in the detection of 9 homozygous transgene (35S-HsCBL8) plants with 679 bp band length. Furthermore, T3 seeds from T2 homozygous lines namely T2-1-3 (L1) and T2-3-9 (L2) (Figure S1), were selected for further analysis to evaluate the functional attributes of the gene for its tissue-specific expression.

To evaluate the possible role of HsCBL8 in salt tolerance, the seedling growth and morphology of the transgenic rice lines (L1 and L2) were compared with that of the control (ZH11) plants (Figure 6). The comparison was conducted under normal and 125 mM NaCl stress conditions. We found insignificant differences in seed germination (Figure 6A) and seedling fresh weight (Figure 6C) between the transgenic and ZH11 plants under the normal growth conditions as evident from the seedling growth (Figure 6B). By contrast, fairly obvious and significant variations in seed germination rate as well as seedling fresh weight were observed between the transgenic lines and non-transgenic ZH11 under 125 mM NaCl stress (Figures 6A,C). The overall germination rate of both transgenic and ZH11 seeds were reduced under salt stress relative to that of the untreated seeds. However, the lines L1 and L2 thrived best at around 75% germination which was 20% higher than that of ZH11 (Figure 6A). This finding was evident from the stunted growth of the radicles and plumules of the ZH11 seedlings under salt stress (Figure 6B). Moreover, the rate of water loss from the transgenic lines was significantly lower than that of ZH11 when the application of salt stress was prolonged from 0 to 24 h. Obvious differences between the fresh weights of seedlings of transgenic lines and ZH11 were observed at 16 (p <0.05) and 24 h (p <0.01) of salt stress (Figure 6C). Furthermore, we tested salt tolerance in the plants at the well-grown vegetative stage. We then planted the seedlings of L1, L2 and ZH11 for 30 days in normal conditions hydroponically. After 30 days of growth, we applied 125 mM NaCl stress, initially for 2 days, 4 days, and finally 6 days (Figure 6D). The affects of NaCl treatment were significantly prominent on the leaves of the ZH11 plants, especially on the older leaves. The degree of severity of the effect was increased with the exposure time of the ZH11 plants to the treatment from 2 to 6 days compared with that in the transgenic lines. This effect was manifested as withering of older leaves, increased water loss, and curling of young leaves in the ZH11 plants compared with those in the L1 and L2 plants. The transgenic plants showed better performance as exhibited by the strong vitality of the top new leaves, even after 6 days of NaCl treatment (Figure 6D). These results indicate the obvious role of HsCBL8 in salt tolerance.

Generally, the accumulation of compatible osmolytes such as proline, are associated to the stress tolerance of the corresponding plant (Ahmed et al., 2013; Mekawy et al., 2015). Meanwhile, quantification of MDA concentrations and REL is considered as an efficient indicator of the structural integrity of the plant membranes in response to abiotic factors, such as salt stress. To evaluate the functional attributes of overly expressed HsCBL8 for rice tolerance against salt stress, we analyzed the proline and MDA concentrations, and rate of REL. Accordingly, the HsCBL8 overexpression in the transgenic rice resulted in the accumulation of 35 and 56% additional proline content in the L1 and L2 lines, respectively, after 24 h of salt stress compared with that in ZH11 (Figure 7A). Moreover, plant exposure to plants to salt stress for 72 h resulted in an overall increase in proline content in both the transgenic and ZH11 plants compared with those in 24 h. Even so, 33% and 25% accumulation of additional proline content was observed in L1 and L2 respectively, relative to that in ZH11. This result indicates the possible role of HsCBL8 in salt tolerance. On the contrary, HsCBL8 overexpression significantly reduced the MDA content and REL in the transgenic lines compared with the non-transgenic ZH11 plants under salt stress. This reduction in MDA content was in the order of 36% (in both transgenic lines) at 24 h, but 31 and 29% in L1 and L2, respectively, at 72 h of salt stress compared with the contents in ZH11 (Figure 7B). Similarly, 33 and 46% reduction in REL was observed in L1 and L2 respectively, at 24 h of salt treatment, whereas, reduction of 23% and 24% reduction in REL was noted in L1 and L2, respectively, at 72 h of salt stress compared with that in the ZH11 plants (Figure 7C).

Generally, the ability of a crop variety to inhibit the uptake of potentially toxic ions such as Na⁺ and preferential uptake of K⁺ to keep the balance between ions is regarded as a desirable trait for salt tolerance. To investigate whether HsCBL8 could induce the salt tolerance trait in transgenic rice, we analyzed the status of Na⁺ and K⁺ ions and their comparative ratio in roots and shoots under control, as well as salt stress conditions. Plants of both transgenic as well as ZH11 pre-grown for 30 days under normal condition were exposed to 125 mM NaCl stress for 6 days. After 6 days, analysis of the ions revealed insignificant differences in Na⁺ and K⁺ concentration among unstressed transgenic lines and ZH11. By contrast, under salt stress, Na⁺ accumulation was generally noted at higher content in the shoots than in the roots in either of the transgenic lines and ZH11 plants. However, the Na⁺ contents in ZH11 plants were significantly higher in the shoots (42%) as well as the roots (18.5%) than in both of the transgenic lines (Figure 7D). On the other hand, salt stress induced a slight change in root and shoot K⁺ contents between the transgenic lines and ZH11 plants applied with or without NaCl treatment (Figure 7E). Consequently, the Na⁺/K⁺ ratio in both shoot and root was lower in transgenic lines than that in ZH11 plants (Figure 7F), indicating that the overexpression of HsCBL8 in rice could regulate and induce salt tolerance.

To investigate whether HsCBL8 could regulate the salt-responsive genes in rice, we performed the gene expression profiling of the 35S-HsCBL8 transgenic lines (L1 and L2) in comparison with the non-transgenic plants under normal conditions and salt stress. The genes selected were either specifically salt responsive (OsSOS2/OsCIPK24, OsNHX1, and OsSOS1) and/or responsive to multiple stresses (abiotic or biotic) (OsCIPK15, OsRD29A, and OsDREB2A). Under normal growth conditions (unstressed, 0 h), we observed extremely weak expression, as well as insignificant variations in the transcriptional levels of all the selected genes in both ZH11 and transgenic plants. This result indicates that HsCBL8 exerted an almost negligible interaction with these genes (Figures 8A–F) in the unstressed environment. At the onset of stress application (125 mM NaCl) initially for 3 h, the overall relative expression level of the genes was enhanced relative to that in the unstressed plants. However, the difference between the transgenic lines and ZH11 was insignificant. Moreover, duration of salt stress exposure was prolonged to 6 h, all the genes were found substantially up-regulated compared with the unstressed plants (0 h) or those subjected to salt stress for 3 h. However, a a comparison between the transgenic lines and ZH11 at 6 h of salt stress revealed that the overexpression of HsCBL8 in both the transgenic rice lines up-regulated some of the genes, such as OsSOS2/OsCIPK24, OsCIPK15, and OsRD29A, to a higher level than that in ZH11 (Figures 8A,B,E). Even so, the overall variation was insignificant, similar to the one observed in plants exposed to 3 h of salt stress or the untreated plants. These results indicate that the overexpression of HsCBL8 in rice could not affect the regulation of salt-responsive genes significantly, showing the negligible interaction of HsCBL8 with all the selected genes expressed in response to salt stress.

To have more detailed perspective of expression pattern of HsCBL8 in response to various abiotic stresses, the promoter fragment of HsCBL8 was fused with the GUS reporter gene and transferred to Arabidopsis to develop HsCBL8_Pro-GUS transgenic lines (Figure 1B). The expression of GUS activity was examined in 17 independent transgenic lines. Except for three lines (with zero GUS activity), the other 14 lines exhibited similar tissue-specific expression pattern. However, the intensity of the colors varied among lines, probably because of the position effect. The GUS expression pattern of a representative line (TGUS-7) is presented herein (Figures 9A–J). Under normal conditions, GUS signals were extremely weak in the cotyledon apex at the second day after germination (Figure 9A), whereas, the hypocotyl stele was stained light blue at the third day (Figure 9B), showing the activity of GUS promoted in hypocotyl and cotyledons (Figure 9C) and further strengthened in the vascular bundles of elder rosette leaves (Figures 9D,J). However, no activity was observed in the newly formed root tissues, including lateral root tips (Figures 9D,J), instead the blue stain was observed within the root stele tissue (Figure 9J) and the junction tissue of tap and lateral roots (Figure 9J). Furthermore, GUS activity at the reproductive phase was detected in sepals, filaments, and tissues between the style and stigma, and tissues between the silique and silique petiole but no GUS signal was observed in the petals and seeds (Figures 9E–H).

Some cis elements in the HsCBL8 promoter sequence predicted by PlantCARE, such as ABRE and MBS (Table 1), normally respond to ABA signal and drought stress, respectively. To evaluate the response of the HsCBL8 gene to ABA, drought, NaCl and cold stress, GUS activity was measured by fluorometric assay (Figure 9K). Four-week-old Arabidopsis seedlings were supplemented with 100 μM ABA, 20% PEG6000, and 100 mM NaCl to apply ABA, drought and salt stress, respectively. For cold treatment, the seedlings were transferred to a growth chamber under dark conditions at 4°C. GUS activity was assayed after 12 h treatment. Accordingly, compared with untreated plants, GUS activity was observed to significantly (p <0.01) higher levels in the plants treated with ABA (294%), NaCl (230%), and PEG6000 (p <0.05; 205%), respectively, whereas, an increase of only 43% was obtained with cold treatment, indicating the possible response of HsCBL8 to these stresses.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.