Functional characterization of a type 2 metallothionein gene, SsMT2, from alkaline-tolerant Suaeda salsa

A type 2 metallothionein gene, SsMT2, was cloned from Suaeda salsa, a salt- and alkali-tolerant plant, which is dominant species on the saline/alkali soil of northeast China. The SsMT2 gene was expressed in all organs except the flower and its expression was induced by various stresses such as CdCl2, NaCl, NaHCO3, and H2O2 treatments. SsMT2-transgenic yeast (Saccharomyces cerevisiae) and plants (Arabidopsis thaliana) showed significantly increased resistance to metal, salt and oxidant stresses. These transgenics accumulated more Cd2+, but less Na+ than their wild type counterparts. SsMT2 transgenic Arabidopsis maintained lower level of H2O2 than wild type plants did in response to the stress treatments. These results demonstrated that the SsMT2 gene plays an important role in reactive oxygen species scavenging and confers enhanced metal and oxidant tolerance to plants.

The ectopic expression of OsMT1e-P enhanced tolerance of salt stresses in transgenic tobacco, and the resultant plants survived and set viable seeds under saline conditions 26 . A SbMT2 gene was used to transform tobacco, and transgenic lines had better phenotypic performance under salt (NaCl) stress conditions compared to wildtype plants 27 . Overexpression of OsIFL in transgenic tobacco plants conferred salinity stress tolerance. Screening of a rice cDNA library revealed OsIFL strongly interacted with metallothionein protein 28 .
Cadmium (or Cd 2+ ), among the most toxic non-essential elements with high mobility in plants, directly or indirectly inhibits primary physiological processes 29 . The photosynthetic apparatus appears to be particularly sensitive to Cd 2+ toxicity, even at very low concentrations 30 . MTs were first isolated as Cd-binding protein from horse kidney in 1957 31 . This family of proteins detoxifies metal ions through direct binding Cd 2+ 9 .
The production of reactive oxygen species (ROS) occurs at all times during plant growth and development 32 , and increases when plants are exposed to biotic and abiotic stresses 33 . The cysteines in MTs directly involved in the removal of ROS and thus, protect against cellular injury, and indirectly reduce the production of cellular ROS 34 . MTs may act as an antioxidant by mitigating ROS-induced cellular injury independent of a function in metal sequestration 35 .
Each kind of MT may have a unique function and plays an important role against abiotic stress. Because S. salsa grows in saline or alkaline soil habitat and persists 3,4 , the biological function of MTs in anti-alkali plants has not been elucidated. Therefore, we cloned an open reading frame of a type 2 MT, designated as SsMT2, from S. salsa and investigated its function under the stress induced by Cd 2+ , Na + and H 2 O 2 in transgenic yeast (Saccharomyces cerevisiae) and Arabidopsis thaliana. The results enhance our insights into the SsMT2 gene function when halophyte plants are grown under environmental stresses.

Results
Cloning of an open reading frame of SsMT2 in S. salsa. The open reading frame (ORF) of SsMT2 was obtained from the cDNA in the S. salsa. The full-length fragment contains of 234 bp and encodes a 77-amino acid polypeptide GenBank accession number MF447531). The amino acid sequence of this transcript had the highest similarity (91%) with that of the SbMT protein (GenBank accession number: JF780913) from Salicornia brachiata, followed by AcMT from Amaranthus cruentus (AF268027) (79%), SnMT from Silene niceensis (ADP92404) (75%), and SmMT from Salvia miltiorrhiza (ABR92329) (60%) (Fig. S1).
SsMT2 gene expression in S. salsa. Northern blot detected strong signals in roots, leaves, stems and seed, but no signal in flowers, indicating that the SsMT2 gene expressed in all organs except flowers (Fig. 1A). The expression of the SsMT2 gene was significantly induced under CdCl 2 and H 2 O 2 stresses in S. salsa. NaCl stress caused a moderate increase of the transcript and NaHCO 3 stress caused slight increase of the transcript (Fig. 1B). The results indicated that different stresses affect the SsMT2 expression differentially in S. salsa.
SsMT2-transgenic yeast responses to Cd 2+ , Na + and H 2 O 2 stresses. Northern blot showed that one distinct band was detected in the transgenic yeast and no signal in the control and indicated that the SsMT2 gene was expressed in the transgenic yeast (Fig. S2A). The quantification of SsMT2 protein in yeast was analyzed using Western blot (Fig. S2B). Stronger signals were detected in SsMT2 transformed yeast, compared to weak signal in WT yeast (non-SsMT2 transformed). This result indicated that some other MT proteins present in the yeast, and SsMT2 transformed yeast has more MT protein than WT yeast.
The cell growth of transgenic and non-transgenic yeasts was compared at five serial dilutions for each treatment (corresponding to the five columns in each panel in Fig. 2). Without stress (control), the growth of both transgenic and non-transgenic yeasts showed no significant difference (upper left panel in Fig. 2). However, growth was affected when the stresses were applied. The transgenic yeasts grew better than the non-transgenic yeasts in the presence of 140 µM CdCl 2 , 600 mM NaCl, 22 mM NaHCO 3 or 2.8 mM H 2 O 2 (Fig. 2). When the concentration was increased to 160 µM CdCl 2 , 1 M NaCl, 26 mM NaHCO 3 , or 3.2 mM H 2 O 2 , the transgenic yeasts grew, but non-transgenic yeasts did not grow (Fig. 2).

SsMT2-transgenic
Arabidopsis responses to Cd 2+ , Na + and H 2 O 2 stresses. The copy numbers of the SsMT2 gene in the transgenic lines were indicated by one or more distinct bands in the transgenic Arabidopsis. There were four plants (#1, #3, #5 and #6 in Fig. S2C) that had one copy, one plant (#2) that had three copies (Fig. S2C), and one plant (#4) that had nine copies (Fig. S2C). No positive signal was detected in WT Arabidopsis plants (Fig. S2C). The expression of SsMT2 gene in transgenic Arabidopsis was detected by Northern blot. Of these transgenic Arabidopsis plants, three (#1, #5 and #6 in Fig. S2D) were positive, and indicated the SsMT2 gene was highly expressed in these transgenic plants.
The effects of CdCl 2 , NaCl, NaHCO 3 and H 2 O 2 on seed germination were examined in the above three selected transgenic Arabidopsis and wild type plants (Fig. 3A). Seeds of wild type and transgenic plants were germinated on medium, each containing 100 µM CdCl 2, 100 mM NaCl, 2 mM NaHCO 3 or 1 mM H 2 O 2 , with 3 days later for wild type than transgenic lines. In the presence of 150 mM NaCl or 4 mM NaHCO 3 , only 40% or 20% wild type lines seed germination respectively, while 100% transgenic lines seed germination. In the presence of 180 µM CdCl 2 or 5 mM H 2 O 2 , no wild type seeds were germinated. Although transgenic plant seeds were also heavily affected, 48% or 72% seeds were germinated respectively. The transgenic lines extended germination until the cotyledon turned white under 5 mM H 2 O 2 . On the control (no stress) media, seed germination showed no significant difference between wild type and three selected transgenic lines (Fig. 3A).
The effects of CdCl 2 , NaCl, NaHCO 3 and H 2 O 2 on seedling growth were examined at the early stage of growth of transgenic plants #1 and #5 (#5 and #6 had very similar phenotype, so only #5 plant was selected for analysis) (Fig. 3B). No significant phenotypic difference was observed between the transgenic lines and WT plants on the control medium. However, the growth of transgenic and WT lines was inhibited when the medium contained 100 µM CdCl 2 , 100 mM NaCl, 2 mM NaHCO 3 , or 1 mM H 2 O 2 . However, the transgenic plants grew better than their WT counterparts. The growth of young leaves of the SsMT2-transgenic lines were less affected under the 180 µM CdCl 2 , 150 mM NaCl, or 4 mM NaHCO 3 stress compared to the wild type plants. There were no significant differences in the dry weights of the SsMT2 transgenic lines and WT plants without stresses and 5 mM H 2 O 2 . However, green leaves in transgenic plant and white leaves in WT plants were observed when grown on the medium with 5 mM H 2 O 2 . Dry weight (Table 1) of the SsMT2 transgenic lines was higher than WT plants under other stress conditions. Additionally, there was no significant difference with root length among plants between transgenic and wild type plants under stress (data not shown). These results showed that the SsMT2 gene expression in Arabidopsis transgenic plants increased metal, salt or oxidant tolerance during early stage of seedling growth.
The effects of CdCl 2 , NaCl, NaHCO 3 and H 2 O 2 on #1 and #5 transgenic plants were examined during adult stage of plant growth (Fig. 3C). No phenotypic differences were observed between the transgenic and WT plants under normal conditions. After exposing both sets of plants to 100 mM CdCl 2 , 400 mM NaCl, 500 mM NaCl, 400 mM NaHCO 3 or 500 mM NaHCO 3  Metal ion uptake in SsMT2-transgenic yeast. SsMT2-transgenic yeast accumulated higher amounts of Cd 2+ (Table 3) and lower amounts Na + (Table 4) than non-transgenic yeast (control) when exposed to 140 µM CdCl 2 , 160 µM CdCl 2 , 600 mM NaCl, 1 M NaCl, 22 mM NaHCO 3 , or 26 mM NaHCO 3 stresses. No significant differences in the amount of Cd 2+ and Na + accumulation were observed between transgenic and non-transgenic yeast on the YPG (1% yeast extract + 2% peptone + 2% galactose) medium without any stresses.   (Table 5) and Na + (Table 6) in the shoots and roots did not differ significantly between transgenic and WT seedlings. The concentration of Cd 2+ in SsMT2-transgenic and WT plants increased dramatically, with a relatively higher level in roots and shoots of SsMT2-transgenic plants when seedlings were grown on medium containing either 100 µM CdCl 2 or 180 µM CdCl 2 . When exposed either to 100 or 150 mM NaCl, 2 or 4 mM NaHCO 3 , the Na + concentrations in the transgenic and WT plants dramatically increased, but with a relatively lower level in both roots and shoots of SsMT2-transgenic lines.   Effects of treatments on the production of H 2 O 2 in plant leaves. Hydrogen peroxide in leaves was detected in situ using 3, 3′-Diaminobenzidine (DAB) histochemical staining method (Fig. 4A) Table 6. Na + accumulation (mg/g dry weight) in shoots and roots of wild-type and transgenic Arabidopsis lines in the presence of NaCl or NaHCO 3 . Results are presented as means ± SE (n = 3). Low case letters a and b indicate significant differences among mean values within each plant at p ≤ 0.05. CK, control; WT, wild type; #1 is SsMT2-transgenic plants.

Discussion
The expression of the SsMT2 gene was increased significantly after S. salsa plants were grown under various stresses and indicated that the SsMT2 gene may be involved in adaptation to these stresses. Similar expression pattern of MTs was induced when CdCl 2 was applied 19,20,36 , and when salt stresses presented in plants 23,24,37 . The SsMT2-transgenic yeast showed higher tolerance to CdCl 2 , NaCl, and NaHCO 3 stress than the non-transgenic yeast in present study. In plants, different MTs often showed different expression patterns in different plant organs. For example, type 2 MTs were preferentially expressed in the leaves 11,38 , type 1 MTs were found mainly in roots 39,40 . SsMT2 was expressed in most organs of Arabidopsis, including leaves and seeds, and its expression level increased when the S. salsa plants were exposed to the stresses conditions. Increased expression implies the SsMT2 gene transcript may affect plant seed germination and development, which were inhibited under the stressful environments 41,42 . MTs had significant impacts on plant growth when the plant suffered various abiotic stresses 40,43 . In this study, transgenic Arabidopsis plants had significantly higher seed germination rates and more vigorous seedling growth than non-transgenic plants under high concentrations of metals, salts or hydrogen peroxide. These results indicated that the SsMT2 gene was involved in the transgenic Arabidopsis accommodation of metal, salt and/or oxidant stresses. SsMT2 transgenic yeast and Arabidopsis plants increased tolerance to CdCl 2 stress. However, Cd 2+ accumulation in cells were elevated and indicated that the SsMT2 expression and Cd 2+ accumulation have positive linear correlation. The SsMT2 gene has the same function with the CeMT2b gene, which greatly increased Cd 2+ tolerance and Cd 2+ accumulation in E.coli and tobacco 44 . Arabidopsis MT1 knock-down lines were hypersensitive to Cd 2+ and accumulated lower amounts of Cd 2+ when compared with WT plants 45 . Compared with the wild type, transgenic plants of Ziziphus jujuba overexpressing the ZjMT gene and accumulate more Cd 2+ in the roots 43 . However, there are some exceptions, for example, BcMT2 46 and TcMT2 47 transgenic lines did not increase tolerance to Cd 2+ nor did they increase Cd 2+ accumulation. In this study, the Cd 2+ accumulation was higher in the transgenic yeast and Arabidopsis, and more tolerance to Cd 2+ than WT plants. The SsMT2 protein chelates the Cd 2+ in the cytoplasm, and thus blocks Cd 2+ from freely interacting with cytoplasmic components or entering into organelles. Via this mode of action, decreased Cd 2+ does limited damage transgenic yeast cells and plants, whereas Cd 2+ damages WT yeast and plants. The full function of MTs to influence Cd 2+ tolerance and Cd 2+ accumulation in cells requires further investigation to elucidate its function.
Sodium ion accumulation in SsMT2-overexpressed yeast and plants was significantly lower than that in WTs under high NaCl or NaHCO 3 environments. There are three mechanisms to prevent excess Na + accumulation in the plant. First, Na + in plant cells may be reduced once Na + influx transporter genes are activated. Second, Na + can be transported and stored in vacuoles. Third, Na + in the cytoplasm can be exported to external medium or the apoplast via plasma membrane Na + /H + antiporters 48 . The plant MTs do not contain signal peptides and do not have Na + transportation function. The reason for resulting in lower Na + concentration in SsMT2-transgenic lines and enhancing the tolerance of transgenic organism to salt stress may be that the SsMT2 gene interacted with transporter genes. Overexpression of SsMT2 in transgenic lines induced the transport Na + out of plant. Lower Na + concentration in the SsMT2-transgenic lines probably decreased damage to the plant and increased the tolerance of transgenic yeasts and plants to Na stress.
The exposure of plants to heavy metals and salts can induce ROS to be produced and thus change the balance between ROS production and scavenging 49 52 , and Gossypium hirsutum GhMt3a 13 the transgenic seedlings of these species had less H 2 O 2 than that in control plants under various stresses. The SsMT2 gene is involved in the mediation of H 2 O 2 scavenging during the abiotic stress and resulted in much lower level of H 2 O 2 accumulated in the transgenic plants. Therefore, the SsMT2 gene plays an important role in reactive oxygen species scavenging under the stresses imposed in this study. The present study also provided evidence that SsMT2 may decrease the impact by induced H 2 O 2, and protected plants from damage.
In conclusion, SsMT2 was expressed from seed germination and increased tolerance to stress in transgenic plants. H 2 O 2 content in transgenic lines was lower than the control. These results suggest that the role of SsMT2 to influence plant or yeast tolerance to heavy metal and salt stresses may directly bind ion and trigger other genes' function, or indirectly improve ROS-scavenging ability.

Cloning of full-length open reading frame (ORF) region of SsMT2.
We have identified some candidate salt-responsive genes in S. salsa using the full-length cDNA over-expressing gene (FOX)-hunting system 53 . The SsMT2 gene was one of those genes identified. Seeds of S. salsa plants were collected from an alkaline soil area in Northeast China and germinated on MS medium 54 at 28 °C under 2000 Lux irradiation with a 16 h light/8 h dark photoperiod in an illuminated incubator. Total RNA was isolated from 4-week old seedlings using RNeasy Plant Mini Kit (Qiagen, Hilden, Germany). cDNA was synthesized from l µg of the total RNA with Prime-Script Reverse Transcriptase (Takara, Tokyo, Japan) using an oligo (dT) primer. MT cDNA sequence from FOX-hunting system was obtained and open reading frame (ORF) was found by blasting in the NCBI database. A transcript fragment was amplified by PCR from the cDNA with the forward primer (5′-ATGTCTTGCTGTGGTGGTAACTGTGG-3′) and reverse primer (5′-TCATTTGCAGGTGCATGGGTTG-3′), which were designed from the MT ORF sequence. The PCR product was purified from agarose gel using the DNA Gel Extraction Kit (Generay, Shanghai, China) and cloned into plasmid pMD18-T (Takara, Tokyo, Japan) and sequenced. A new gene was designated as SsMT2 and its ORF nucleotide sequence and protein sequence was deposited to GenBank database (MF447531). Construction of expression and transformation vectors. Construction of yeast expression and transformation vectors. The coding region of the SsMT2 gene was amplified from pMD18T-SsMT2 plasmid DNA with BamHI sense primer 5′-GGATCCATGTCTTGCTGTGGTGGTAA-3′ (restriction site underlined for all restriction enzymes below) and XhoI antisense primer 5′-CTCGAGTCATTTGCAGGTGCATGGGT-3′. The PCR amplified fragments were digested with two restriction enzymes BamHI and XhoI and then ligated into the BamHI/XhoI sites of the vector pYES2 (Takara, Tokyo, Japan) to get pYES2-SsMT2 construct. The plasmid DNA of pYES2-SsMT2 was transformed into competent yeast strain INVSc1 (S. cerevisiae) (Takara, Tokyo, Japan) using the electric impulse method following the manufacturer's instructions (Invitrogen)and the transformants were selected based on their growth on uracil deficient synthetic complete (SC-Ura) solid medium (6.7 g/L Yeast Nitrogen Base, 0.77 g/L -Ura Do supplement, PH = 5.8).

Construction of plant expression and transformation vectors.
The coding region of SsMT2 gene was amplified from pMD18T-SsMT2 plasmid DNA with the previously described BamHI sense primer and SacI antisense primer 5′-GAGCTCTCATTTGCAGGTGCATGGGT-3′. The PCR fragments were digested with BamHI and SacI and then ligated into the BamHI/SacI site of pBI121 binary vector (Takara, Tokyo, Japan), the plasmid DNAs of pBI121-SsMT2 was transformed into the Agrobacterium tumefaciens strain EHA105 (Takara, Tokyo, Japan) and then Arabidopsis (ecotype: Columbia) was transformed using the floral dip method 55 .
The SsMT2 gene expression level in S. salsa seedling under different stresses was detected by Northern blot. The seeds of S. Salsa were sown onto the MS medium, then the 4-week-old S. Salsa seedlings were treated with various stresses (100 µM CdCl 2 , 180 µM CdCl 2 , 100 mM NaCl, 150 mM NaCl, 2 mM NaHCO 3 or 4 mM NaHCO 3 , 1 mM H 2 O 2 and 5 mM H 2 O 2 ) for 48 h.Total RNA was isolated from leaves. Northern blot was conducted as above procedure.
Stress tolerance of the transgenic yeast. The expression of SsMT2 gene in transgenic yeast was analyzed using Northern blot. Total RNA from yeast was extracted using the RNeasy Yeast Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instruction. Northern blot was conducted as above procedure.
Western blot was used to investigate the SsMT2 protein amount in yeast. Protein extraction from yeast followed Zhang's protocol 56 . Yeast cells (1.5 mL in YPG) were harvested prior to stationary phase (OD 600 = 1.0) by centrifugation. Cells were first pre-treated with 2 M LiAc and then treated with 0.4 M NaOH for 5 min on ice. Finally, cells were centrifuged and yeast whole proteins were extracted with SDS-PAGE sample buffer. Western blot was conducted according to Ohkuni's protocol 57 . Equal volume of samples was lysed in SDS sample buffer. These samples were separated by 12% SDS-PAGE and subsequently transferred the proteins from gel to a polyvinylidene difluoride (PVDF) membrane using a transfer apparatus at 30 V for 90 min. After blocked in PBST (Phosphate Buffered Saline with Tween20) containing 5% skimmed milk for 1 h at room temperature, membrane was incubated with MT antibody (1: 3,000) overnight at 4 °C and wash membrane 3 times for 10 min each time with 1x PBST, then incubated with alkaline phosphatase-conjugated goat anti-rabbit immunoglobulins (1: 5,000; Sigma) at 37 °C for 1 h. Wash membrane 3 times for 10 min each time with 1x PBST. The signals were detected with CDP-Star detection reagent using Biotech Image Master VDS-CL Multifunction Bio-imaging Station.
Stress tolerance of the transgenic Arabidopsis. The southern hybridization of genomic DNA of transgenic Arabidopsis was conducted to investigate the copy number of the SsMT2 gene in the transgenic lines. Genomic DNA from 2-week-old Arabidopsis (wild type, transgenic lines#1, #2, #3, #4, #5 and #6) leaves was isolated using the CTAB method and then digested with HindIII at 37 °C for 60 min. The digested fragments were separated on 1% (w/v) agarose gel and then transferred to the hybrid. Expression of SsMT2 gene in transgenic Arabidopsis was analyzed using Northern blot. The SsMT2 gene expression level in transgenic Arabidopsis lines (#1, #5 and #6) was detected by Northern blot.
The seeds of wild type and the third generation(homozygous) transgenic Arabidopsis plants (#1, #5, #6) were surfaced-sterilized with 70% ethanol for 1 min, followed by 1% NaClO solution for 3 min, and then rinsed three times in sterile water. The seeds were sown onto agar plates that contained MS basal medium, 1% (w/v) sucrose, and 0.8% (w/v) agar, supplemented with either filter-sterilized 100 µM CdCl 2 , 180 µM CdCl 2 , 100 mM NaCl, 150 mM NaCl, 2 mM NaHCO 3 , 4 mM NaHCO 3  medium were used as control, three times repeat. Photos were taken on the 14 th day after the stress treatments. Germination rate was calculated when the transgenic and wild type Arabidopsis no longer sprouted.
To find out whether the SsMT2 gene impacts the early seedling development under the different stresses, the seeds of wild type and transgenic Arabidopsis (#1, #5) were germinated on MS medium. The 14-day-old seedlings were transplanted onto MS medium (as a control) and MS medium supplemented with different concentrations of metals (100 µM CdCl 2 , 180 µM CdCl 2 ), salts (100 mM NaCl, 150 mM NaCl, 2 mM NaHCO 3 , 4 mM NaHCO 3 ), or oxidant (1 mM H 2 O 2 , 5 mM H 2 O 2 ), respectively. The plates were positioned vertically on shelves in order to compare root growth visually. Root length and dry weight were measured after stresses applied 7 th and 14 th day, three times repeat. Photos were taken between the 7 th and 14 th day after the stress treatments.
In addition, we examined the stress tolerance at the plant adult stage. Briefly, wild type and transgenic seeds (#1, #5) were grown on MS medium. One-week-old plants were transferred to pots filled with 3:1 mixture of nutrition soil: peat in a chamber (22 °C, 100 M photons·m −2 ·s −1 , 60% relative humidity, 16/8 h day-night cycles). The soil-grown plants were watered with 50 mM CdCl 2 , 100 mM CdCl 2 , 400 mM NaCl, 500 mM NaCl, 400 mM NaHCO 3 , 500 mM NaHCO 3, 1.5 M H 2 O 2 , or 2 M H 2 O 2 solution respectively every 4 days for a total of 12 days, three times repeat. The plants survived rate was calculated on the 12 th day after treatment and we took photos at the same time.
Ion uptake in transgenic yeast. To examine whether the SsMT2 gene involves in the accumulation of metals in yeast cells, the Cd 2+ or Na + content was measured with the method previously described 58 . In brief, yeast cells cultured in the YPG liquid medium containing 140 µM CdCl 2 , 160 µM CdCl 2 , 600 mM NaCl, 1 M NaCl, 22 mM NaHCO 3 or 26 mM NaHCO 3 and maintained at 30 °C with shaking at 160 rpm for 12 h. After treatment, 200 mg (dry weight) of cells were collected and analyzed using atomic absorption spectrophotometer (AA800, Perkin Elmer, America). Blank sample was used 10 times to calculate the standard deviation, then the measured standard deviation value was put into the regression equation to figure out that Atomic Absorption Spectrometry (AAS) detection limitation for Cd 2+ was 0.0003 μg/g and Na + was 0.0005 μg/g. The samples were divided into two groups, two samples per group. In each group, one sample was added the standards, another one as a control. Every time the two samples were measured in parallel. The recovery rate was calculated according to the additive amount and the detectable quantity of the ions. The recovery rate for Cd 2+ in the standard reference material (GSB 04-1721-2004 Beijing, China) was 95% and Na + in the standard reference material (GSB 04-1738-2004 Beijing, China) was 98%, indicating that this method is accurate.

Ion uptake in Arabidopsis plants.
Fourteen-day-old WT and transgenic Arabidopsis plants (#1 transgenic plant) were treated without (control) or with each of following solution: 100 µM CdCl 2 , 180 µM CdCl 2 , 100 mM NaCl, 150 mM NaCl, 2 mM NaHCO 3 or 4 mM NaHCO 3 respectively for 48 h. Roots and shoots were harvested and washed in deionized water. Desorption of shoot and root was performed with 1 mM MES-Tris (pH 6.0) containing 0.5 mM CaCl 2 . The samples were dried at 80 °C for 2 days for dry weight measurement. The dried plant materials were digested in a 5 mL mixture of HNO 3 and HClO (87:13, v/v) overnight at room temperature, diluted with 5 mL of 2.5% HNO 3 , and then measured for ion contents by an atomic absorption spectrophotometer.

Reaction to H 2 O 2 stress in transgenic Arabidopsis plants. Fourteen-day-old WT and transgenic
Arabidopsis plants (#1) were treated without (control) or with each of 100 µM CdCl 2 , 180 µM CdCl 2 , 100 mM NaCl, 150 mM NaCl, 2 mM NaHCO 3 , 4 mM NaHCO 3 , 1 mM H 2 O 2 or 5 mM H 2 O 2 respectively for 48 h. H 2 O 2 accumulation in plant leaves was visualized by histochemical staining with 3, 3′-Diaminobenzidine (DAB). DAB is oxidized by H 2 O 2 in presence of peroxidases and produces reddish brown precipitate 59 . The treated leaves were immersed in 1 mg·mL −1 DAB solution, vacuum-infiltrated for 10 min, and then incubated at room temperature for 12 h in the absence of light until the appearance of blown spots. The stain solution was poured off and the chlorophyll was removed by incubating the samples in absolute ethanol overnight. Staining of the rosette leaf was photographed with a microscopy (Olympus). The H 2 O 2 content was also measured using Plant H 2 O 2 ELISA Kit (America Rapid Bio).