SEARCH

SEARCH BY CITATION

Keywords:

  • AtNHX8;
  • Li+/H+ antiporter;
  • cation transporter;
  • lithium;
  • Arabidopsis

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The Arabidopsis monovalent cation:proton antiporter-1 (CPA1) family includes eight members, AtNHX1–8. AtNHX1 and AtNHX7/SOS1 have been well characterized as tonoplast and plasma membrane Na+/H+ antiporters, respectively. The proteins AtNHX2–6 have been phylogenetically linked to AtNHX1, while AtNHX8 appears to be related to AtNHX7/SOS1. Here we report functional characterization of AtNHX8. AtNHX8 T-DNA insertion mutants are hypersensitive to lithium ions (Li+) relative to wild-type plants, but not to the other metal ions such as sodium (Na+), potassium (K+) and caesium (Cs+). AtNHX8 overexpression in a triple-deletion yeast mutant AXT3 that exhibits defective Na+/Li+ transport specifically suppresses sensitivity to Li+, but does not affect Na+ sensitivity. Likewise, AtNHX8 overexpression complemented sensitivity to Li+, but not Na+, in sos1-1 mutant seedlings, and increased Li+ tolerance of both the sos1-1 mutant and wild-type seedlings. Results of Li+ and K+ measurement of loss-of-function and gain-of-function mutants indicate that AtNHX8 may be responsible for Li+ extrusion, and may be able to maintain K+ acquisition and intracellular ion homeostasis. Subcellular localization of the AtNHX8–enhanced green fluorescent protein (EGFP) fusion protein suggested that AtNHX8 protein is targeted to the plasma membrane. Taken together, our findings suggest that AtNHX8 encodes a putative plasma membrane Li+/H+ antiporter that functions in Li detoxification and ion homeostasis in Arabidopsis.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Transport of metal and alkali cations across plant plasma and organelle membranes is essential for plant growth, development, signal transduction and nutrient utilization, and also for extrusion and compartmentalization of toxic ions under stress conditions. Most cations are transported by proton-coupled transporters, rather than primary ion pumps (Maser et al., 2001; Ward, 2001). Among the genes encoding putative H+-coupled transporters in Arabidopsis plants, there are more than 40 genes that encode homologues of sodium–hydrogen (Na+/H+) antiporters (Brett et al., 2005a; Maser et al., 2001). Given this large number of Na+/H+ antiporter homologues, and the fact that Na+ is not an essential nutrient for plants, it is likely that these homologues have other functions in plants (Maser et al., 2001). As the substrate specificity, regulation and membrane localization of these antiporters cannot be predicted with certainty from phylogenetic relationships, they need to be functionally characterized. Researchers have begun to unveil the molecular mechanisms of ion selectivity and regulation of transporter activity. Recently, the effects have been documented on the ion selectivity and transporter activity of single-residue/domain deletions or various amino acid substitutions in Na+/H+ antiporter genes (Hamada et al., 2001; Kuwabara et al., 2004; Mitsui et al., 2004; Ohgaki et al., 2005; Waditee et al., 2001; Yamaguchi et al., 2003). Phylogenetic analysis indicates that Na+/H+ antiporters fall into three families, among which the monovalent cation:proton antiporter-1 (CPA1) family includes eight members, from AtNHX1 to AtNHX8 (Maser et al., 2001).

The Arabidopsis Na+/H+ antiporter AtNHX1 was identified based on its sequence similarity to the Saccharomyces cerevisiae NHX1 (Gaxiola et al., 1999). AtNHX1 is localized in plant vacuoles, where it functions to sequester Na+ and K+ (Apse et al., 2003; Qiu et al., 2004). AtNHX1 overexpression dramatically improves salt tolerance in transgenic Arabidopsis plants (Apse et al., 1999). The Arabidopsis AtNHX7/SOS1 gene is a homologue of the S. cerevisiae NHA1 gene, and encodes a plasma membrane Na+/H+ antiporter (Qiu et al., 2003; Shi et al., 2000). SOS1 enables Na+ efflux across the plasma membrane and controls long-distance Na+ transport from root to shoot (Shi et al., 2002). Arabidopsis sos1 mutants are hypersensitive to sodium ions (Na+) and lithium ions (Li+) (Wu et al., 1996); furthermore, overexpression of this gene improves salt tolerance in transgenic Arabidopsis (Shi et al., 2003). AtNHX2–6 are phylogenetically related to AtNHX1, and AtNHX2–5 are able to complement yeast endosomal NHX, indicating that they have functional Na+/H+ antiporter activity (Aharon et al., 2003; Yokoi et al., 2002). AtNHX8 is phylogenetically related to AtNHX7/SOS1 and predicted to encode the Na+/H+ antiporter. However, no report has been published on this.

CPA1 family members have been reported to be involved in other functions in addition to Na+ transport, such as regulation of K+ and pH homeostasis. AtNHX1 mediates K+ and Na+ transport in tonoplast vesicles isolated from transgenic tomato plants overexpressing AtNHX1 (Zhang and Blumwald, 2001). atnhx1 mutant plant leaves had markedly decreased Na+/H+- and K+/H+-exchange activity at the vacuolar membrane, and altered development (Apse et al., 2003), suggesting a role of AtNHX1 in mediating K+ and pH homeostasis. In the Japanese morning glory (Ipomoea nil or Pharbitis nil), a shift from reddish-purple buds to blue, open flowers correlates with an increase in vacuolar pH. This process is dependent on the activity of an NHX1 homologue in I. nil (Fukada-Tanaka et al., 2000; Ohnishi et al., 2005; Yamaguchi et al., 2001). Likewise, CPA2 family members have been reported to be involved in K+ and pH homeostasis (Cellier et al., 2004; Song et al., 2004), as well as in pollen development (Sze et al., 2004).

In this study we report the functional characterization of AtNHX8 with multiple approaches. nhx8 mutants are hypersensitive to LiCl stress during germination and early development. In agreement with this, plants that overexpress AtNHX8 have an extremely elevated Li+-tolerance phenotype, observed regardless of whether they have an sos1-1 or a wild-type genotype background. Furthermore, AtNHX8 expression can rescue the Li+-sensitive phenotype of the yeast triple mutant AXT3. All these phenotypes are very specific for Li+ tolerance, and were not associated with any changes in sensitivity to other ions tested. Tissue-specific expression patterns of the AtNHX8 gene and subcellular localization of the protein at the plasma membrane, together with the relationship observed between LiCl tolerance conferred by AtNHX8 and Li+ and K+ content, indicate that AtNHX8 may play a role in Li+ detoxification and maintenance of intracellular ion homeostasis.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

The nhx8 mutants are hypersensitive to LiCl

The AtNHX8 protein belongs to the Na+/H+ antiporter family, and is phylogenetically related to AtNHX7/SOS1. The predicted AtNHX8 full-length protein exhibits approximately 72% identity with the SOS1 sequence over a stretch of 760 amino acids. Hydrophobicity plot analysis showed that the N-terminal region of AtNHX8 is highly hydrophobic (Figure S1 in Supplementary Material).

Analysis of its genomic sequence has revealed that the AtNHX8 gene consists of 18 exons and 17 introns (Figure 1a). Two Arabidopsis thaliana mutant lines, named nhx8-1 and nhx8-2, harbouring T-DNA insertions within the 14th intron and the 14th exon, respectively, of the AtNHX8 gene (Salk-070497 and Salk-086336; Figure 1a) were identified in the SIGnAL database (http://signal.salk.edu/cgi-bin/tdnaexpress). nhx8-1 and nhx8-2 knockout mutant plants lack AtNHX8 transcript expression, as demonstrated by RT–PCR analysis using AtNHX8-specific primers (Figure 1b). Homozygous mutants showed a clear Li+-hypersensitivity phenotype that was apparent under low-Li+ concentrations during seed germination and during the early growth period of the seedlings (Figure 1c). While germination of wild-type seeds was not significantly disrupted by exposure to LiCl at concentrations <15 mm, germination of mutant seeds was significantly inhibited in the presence of 10 mm LiCl. When exposed to 10 mm LiCl, about 80% of the wild-type seeds germinated, whereas only 50% of the mutant seeds germinated during the 7-day period following transfer to 23°C (Figure 1c). Yet no significant difference was found in germination and subsequent seedling growth between the nhx8-1 or nhx8-2 and wild type under 50 or 100 mm NaCl treatment (Figure 1c). Root elongation of the nhx8 mutants was also more sensitive to exogenous LiCl than that of the wild-type plants (Figure 1d). nhx8 mutant hypersensitivity appears to be limited to Li+, as no hypersensitivity to other alkali metals, such as Na+, K+ and Cs+, was observed in our experiments (Figure 1d).

image

Figure 1.  Isolation of the AtNHX8 T-DNA insertional mutant and its phenotype. (a) AtNHX8 gene structure (coding region) and T-DNA location. The open reading frame and introns are indicated by filled boxes and lines, respectively. The positions of the T-DNA insertion are designated by triangles. (b) RT–PCR analysis of AtNHX8 transcripts in wild-type and mutant nhx8-1 and nhx8-2 plants. (c) Germination and subsequent growth of wild-type and nhx8-1 and nhx8-2 plants under MS agar medium or MS medium containing 0, 10 or 15 mm LiCl, or 50 or 100 mm NaCl. For germination tests, seeds on agar plates were incubated at 4°C for 3 days and allowed to germinate and grow for a week in a growth chamber at 23°C before being photographed. (d) Root growth of wild-type and nhx8-1 plants exposed to LiCl, NaCl, KCl and CsCl. Four-day-old seedlings were transferred to vertical agar plates (with roots growing upwards) containing MS medium alone or supplemented with various alkali cations, and allowed to grow for 7 days. •, Wild type; bsl00043, nhx8-1. Mean values ± SD (n = 15). (e) Li+ and K+ contents in plants grown for 6 days in control media or in media with 15 mm LiCl. These data are averages from three independent replicates. Error bars, SD. Open bars, wild type; black, nhx8-1; stippled, nhx8-2. DW, dry weight.

Download figure to PowerPoint

We further assessed the Li+ and K+ contents of wild-type, nhx8-1 and nhx8-2 seedlings under moderate LiCl treatment. Li+ contents of mutants treated with 15 mm LiCl were both slightly higher than those of the wild type, and their K+ contents were lower than the wild type (Figure 1e).The results suggest that the Li+ sensitivity of nhx8-1 and nhx8-2 is correlated with their cellular Li+ and K+ contents.

Heterologous expression of AtNHX8 in yeast mutant

Yeast has been utilized as a model organism to study transport and homeostasis of alkali metal cations (Hasegawa et al., 2000; Posas et al., 2000; Serrano and Rodriguez-Navarro, 2001). Cells of S. cerevisiae with mutations in ENA1-4 and NHA1, as well as in NHX1, are unable to grow in media containing high concentrations of toxic cations, primarily Na+ and Li+ (Quintero et al., 2002).

AXT3 cells expressing AtNHX8 exhibited no significant growth difference with respect to control cells when tested in normal medium. However, in medium with 3 mm LiCl concentration, which is normally highly toxic to AXT3 cells, AtNHX8 expression greatly improved yeast growth. No growth improvement was observed in the presence of other cations in concentrations toxic to AXT3 cells (70 mm NaCl, 1 m KCl, 30 μg ml−1 HygB; Figure 2 and data not shown). The nhx1 mutation in S. cerevisiae also causes strong specific hygromycin sensitivity in cells, which have been widely used in functional complementation studies to identify vacuolar membrane Na+/H+ antiporters. AtNHX8 expression did not suppress AXT3 hygromycin sensitivity caused by the nhx1 mutation, suggesting that AtNHX8 is involved in extruding Li+ from the cytosol at the plasma membrane, but not in sequestering Li+ into vacuoles. Our findings also suggest that AtNHX8 is important for Li+ tolerance, but not for tolerance to Na+ or other toxic ions. In summary, AtNHX8 expression selectively suppresses the Li+-sensitive phenotype of the yeast AXT3 mutant, suggesting that it directs Li+ extrusion out of the cell, and functions as an Li+/H+ transporter.

image

Figure 2. AtNHX8 overexpression specifically suppresses the Li+-sensitive phenotype of AXT3 yeast mutant. Saccharomyces cerevisiae strain AXT3 cells were transformed with an empty pYES2 vector (control), or a vector expressing AtNHX8 (+AtNHX8), and were grown overnight in liquid arginine phosphate medium. In each plate there are four 4-μl serial decimal dilutions spotted in a row for each of the two vector types. Cells in the first plate were administered 1 mm KCl only; the second 3 mm LiCl; and the third 70 mm NaCl. Plates were photographed after incubation at 28°C for 2–4 days. The same results were obtained in three separate experiments.

Download figure to PowerPoint

Gene expression analysis of AtNHX8 and subcellular localization of AtNHX8

As AtNHX8 is a putative Na+/H+ antiporter, we hypothesized that it might be inducible by cation stress. AtNHX8-specific primers were chosen to perform the RT–PCR experiment. RT–PCR results indicated that AtNHX8 transcript abundance was not significantly changed by the stress treatments of NaCl, KCl or LiCl (Figure 3a). Thus the AtNHX8 gene appears not to be transcriptionally activated under the cation stress conditions tested. This result is consistent with microarray data (Maathuis et al., 2003). RT–PCR analyses also revealed that AtNHX8 is expressed in all organs examined, including roots, shoots, leaves and flowers (Figure 3b). The transcript level appeared to be a little higher in leaves than in other organs.

image

Figure 3.  Expression patterns of the AtNHX8 gene. (a) RT–PCR analysis of AtNHX8 transcripts in Arabidopsis plants raised under normal conditions (10 days). No changes were found for treatments with 200 mm NaCl, 100 mm KCl and 40 mm LiCl for 5 h compared with the control (distilled water). (b) RT–PCR analysis of AtNHX8 transcripts in different organs of a wild-type plant (4 weeks). (c) Histochemical GUS analysis of AtNHX8 promoter–GUS transgenic plants. Left to right: 2-week-old plant; leaves; root hair.

Download figure to PowerPoint

Promoter–β-glucuronidase (GUS) analysis was performed to examine the tissue-specificity of AtNHX8 expression. GUS staining was detected in the leaves, hypocotyls, primary roots and root hairs of the transgenic seedlings. GUS staining was most pronounced in the vascular tissue cells of the leaves; robust GUS staining was also observed in the root hairs (Figure 3c).

Confocal imaging analysis of overexpressed AtNHX8–enhanced green fluorescent protein (EGFP) fusion protein in transgenic Arabidopsis young roots indicated that the protein is targeted to the plasma membrane (Figure 4b). As there are no large vacuoles in young roots, it is reasonable to believe the fusion protein was targeted to the plasma membrane rather than the tonoplast (Shi et al., 2002). The result is in agreement with plasma membrane localization data from proteomic analyses (http://psort.nibb.ac.jp/form.html). A vector-only control not including AtNHX8 was also used as shown in Figure 4(a), in which the EGFP fluorescence spread throughout the cell.

image

Figure 4.  Confocal fluorescence imaging analysis of AtNHX8–EGFP fusion protein reveals a plasma membrane localization of AtNHX8 in Arabidopsis roots. (a) Localization of EGFP fluorescence in young root cells. (b) Localization of AtNHX8–EGFP fluorescence in young root cells. Left to right: EGFP fluorescence image; transmission image; overlay of fluorescence and transmission image. Scale bar, 20 μm.

Download figure to PowerPoint

Overexpression of AtNHX8 in sos1-1 mutants complemented Li- but not Na-sensitive phenotype, and improved Li tolerance

To assess further the ion specificity of AtNHX8, we generated AtNHX8 overexpression lines in the sos1-1 mutant background. The sos1-1 background was chosen because the root growth of sos1-1 plants is obviously hypersensitive to both NaCl and LiCl (Wu et al., 1996; Zhu et al., 1998). Furthermore, the high degree of sequence similarity between the two genes suggests that the function of AtNHX8 may be similar to that of SOS1. We observed no discernible phenotypic difference under NaCl stress between sos1-1 mutant plants and the sos1-1 plants overexpressing AtNHX8 (sos1-1 35S-AtNHX8; Figure 5a). However, when the transgenic plants were subjected to the stress of Li+ (10–50 mm LiCl), the sos1-1 35S-AtNHX8 transgenic seedlings displayed greatly enhanced Li+ tolerance (Figure 5a). These results indicate that overexpression of AtNHX8 selectively complemented the Li+-sensitive phenotype of the sos1-1 mutant, and greatly improved the Li+ tolerance of sos1-1 mutant seedlings. Similar results were obtained during germination. The germination rates of lines overexpressing AtNHX8 approached nearly 100% in the presence of 20 mm LiCl, a concentration that strongly inhibited germination in wild-type plants (data not shown).

image

Figure 5.  Overexpression of the AtNHX8 gene complemented the Li+-sensitive (but not the Na+-sensitive) phenotype of the sos1-1 mutant and greatly improved the Li+ tolerance of sos1-1. (a) 5-day-old seedlings were transferred from normal MS media to MS media supplemented with LiCl (10–50 mm) or NaCl (50–100 mm), and seedlings were allowed to grow (with roots upside down) for 7 days before being photographed. (b) Root growth of wild-type, sos1-1 and sos1-1 35S-AtNHX8 seedlings exposed to LiCl and NaCl. Open bars, wild type; black, sos1-1; stippled, sos1-1 35S-AtNHX8. Mean values ± SD (n = 15).

Download figure to PowerPoint

The seedlings’ root growth was also measured under Li+ or Na+ treatments. The root growth of the sos1-1 35S-AtNHX8 exhibited no obvious difference with increased LiCl treatment, but they showed similar growth with sos1-1 when subjected to NaCl stress (Figure 5b). Although it seemed that the sos1-1 35S-AtNHX8 seedlings were slightly more tolerant to 50 mm NaCl, they were eventually killed by 100 mm NaCl as sos1-1 (Figure 5a,b).

Overexpression of AtNHX8 in wild-type Arabidopsis dramatically improved Li (but not Na) tolerance; transgenic plants maintained lower Li+ and higher K+ content under Li+ stress

Transgenic seedlings with a wild-type background expressing AtNHX8 (35S-AtNHX8) displayed no obvious differences from wild-type seedlings under KCl, CsCl or CaCl2 stress (data not shown), as well as 50 or 100 mm NaCl treatment (Figure 6a). However, the 35S-AtNHX8 seedlings displayed dramatically improved Li+ tolerance (Figure 6a). The three lines of 35S-AtNHX8 transgenic seedlings could survive with LiCl stress as high as 40–50 mm LiCl, while wild-type seedlings could not grow when Li+ concentration was >20 mm (Figure 6a).

image

Figure 6.  Overexpression of the AtNHX8 gene in the wild type dramatically improved the Li+ (but not Na+) tolerance of transgenic plants; AtNHX8-overexpressing plants maintained lower Li+ and higher K+ content under Li+ stress conditions. (a) 5-day-old homozygous T3 seedlings were transferred from normal MS media to MS media supplemented with 0, 20 or 50 mm LiCl, or 50 or 100 mm NaCl, and seedlings were allowed to grow (with roots upside down) for 7 days. (b–d) The AtNHX8 transcript levels in wild type and three independent transgenic lines (G-1, G-3, G-7) grown under normal conditions were assessed by RT–PCR (b). Li+ (c) and K+ (d) levels in whole plants grown for 6 days in control media or in media with 10, 20 or 30 mm LiCl are shown. These data are averages from three independent replicates. Error bars, SD. Open bars, wild type; black, G-1; stippled, G-3; striped, G-7. DW, dry weight.

Download figure to PowerPoint

In addition we found that, on LiCl treatment, seedlings overexpressing AtNHX8 accumulated less Li+ than wild-type plants (Figure 6c), suggesting that AtNHX8 is important for Li+ efflux in plant cells. Examination of K+ content revealed that the transgenic lines maintained substantially higher K+ content when subjected to 30 mm LiCl treatment (Figure 6d). This suggests that AtNHX8 is involved in the regulation of K+ and other ion homeostasis under Li+ stress conditions.

We also examined the effect of Li+ on the germination of transgenic seeds and subsequent seedling growth. Both the germination of wild-type seeds and subsequent seedling growth were nearly completely inhibited by 20 mm LiCl, whereas transgenic plants were not much affected. More significantly, these transgenic plants could grow even in the presence of 50 mm LiCl (Figure 7).

image

Figure 7.  Enhanced Li+ tolerance of three 35S-AtNHX8 transgenic lines during seed germination and early seedling development. Seed germination and subsequent seedling growth in MS medium without LiCl (a); or plus 20 (b); 30 (c); 40 (d); 50 (e) mm LiCl. Seeds on agar plates were allowed to germinate and grow for a week in a growth chamber at 23°C before being photographed. Seedling establishment rates of the wild type and the three transgenic lines are indicated as the percentage of seedlings that developed green cotyledons and fully expanded true green leaves a week later (f). Data are means of three replicates (each with 50 seeds) ± SD.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

AtNHX8/SOS1b is a member of the Arabidopsis CPA1 subfamily. It is phylogenetically related to AtNHX7/SOS1, and predicted to encode a plasma membrane Na+/H+ antiporter (Apse et al., 2003; Brett et al., 2005a; Maser et al., 2001; Pardo et al., 2006; Ward, 2001). Here we present evidence indicating that AtNHX8 is not an Na+/H+ antiporter, but a putative plasma membrane Li+/H+ antiporter. The nhx8 mutants were found to be hypersensitive to LiCl treatment in both root growth and seed germination, as they were clearly affected by an Li+ concentration of 10 mm, significantly below the level that seriously affected wild-type plants (20 mm). The mutants also showed higher Li+ and lower K+ contents. The ability of AtNHX8 specifically to transport Li+ was demonstrated by its ability to confer Li+ tolerance when overexpressed in yeast AXT3 and Arabidopsis sos1-1 mutants defective in Na+/Li+ extrusion; furthermore, overexpression of AtNHX8 improved Li+ tolerance in wild-type Arabidopsis seedlings. The decreased cellular Li+ content observed in transgenic plants indicates that the tolerance attained was caused by Li+ extrusion rather than compartmentalization. We also found increased K+ content in plants overexpressing AtNHX8, in accordance with previous work suggesting that K+ nutrition is of great importance for plant stress tolerance, specifically NaCl stress (Zhu et al., 1998). Our loss/gain-of-function experimental results, together with the phylogenetic analysis and subcellular localization data, strongly suggest that AtNHX8 encodes an Li+-specific plasma membrane Li+/H+ antiporter.

Examination of Arabidopsis plants transformed with a promoter–GUS construct revealed that AtNHX8 is produced throughout the plant. Such widespread expression is consistent with the possibility that it serves an important role in plant cell physiology. However, the level of the contribution of AtNHX8 to Li+/H+ exchange at the plasma membrane, as well as its role in plant cell physiology, require further study.

Under LiCl stress, nhx8 mutants showed a slightly higher Li+ content (Figure 1e), in contrast to the overexpression lines, which had a rather lower Li+ content (Figure 6c). These results suggest that AtNHX8 may function as an antiporter. It is obvious that the seedlings’ cellular Li+ content increased and K+ content decreased with the elevated LiCl treatment (Figures 1e and 6d); however, the overexpression lines had higher (Figure 6d) and nhx8 mutants lower (Figure 1e) K+ contents, indicating that K+ nutrition is also crucial for plant growth under Li+ stress, and that AtNHX8 also plays a role in plant K+ maintenance and ion homeostasis. However, the mechanisms still need further study.

Lithium specificity of AtNHX8 is unusual, as most cation/H+ antiporters transport both Na+ and Li+, and some also transport K+, Rb+ or Ca2+ (Hamada et al., 2001; Kinclováet al., 2002; Waditee et al., 2001; Yamaguchi et al., 2003). Yeast plasma membrane Na+/H+ antiporters can be classified according to substrate specificity, and probably cell function, into two distinct subfamilies: (1) a subfamily with substrate specificity for Na+ and Li+ (Schizosaccharomyces pombe sod2p, Zygosaccharomyces rouxii ZrSod2-22p), with primary function in detoxification of cells; (2) a subfamily (S. cerevisiae Nha1p, Candida albicans Cnh1p) that transports all alkali metal cations and that, in addition to elimination of toxic cations, may also play roles in regulating intracellular K+ concentration, pH and cell volume. The current classification of yeast antiporters does not reflect the levels of sequence similarity, as the closest by sequence comparison, S. cerevisiae Nha1p versus Z. rouxii ZrSod2-22p, belong by their substrate specificity to different subfamilies (Kinclováet al., 2002). Likewise, SynNhaP (Na+/H+ antiporter from Synechocystis sp. PCC 6803) exhibits both Na+(Li+)/H+ antiporter activity and high Ca2+/H+ antiporter activity (Hamada et al., 2001; Waditee et al., 2001). Recently, a member of the NHX exchanger-protein family from tomato was identified as the first plant K+/H+ antiporter (Venema et al., 2003). Heterologous expression of AtNHX1, 2 or 5 in the yeast nhx1 mutant suppressed the Na+/Li+-sensitive phenotype with different efficiency (Yokoi et al., 2002), suggesting that these NHX members may also have different ion selectivity and substrate specificity. Recent findings suggest that Na+ and Li+ are transported by separate sites on the Na+/H+ exchanger in dog red blood cells (Dunham et al., 2005), and that Na+-inhibitory sites are Li+-substrate sites (Dunham et al., 2005). These findings suggest that members of the NHX family have various transport modes and differential cation specificities.

Lithium has been widely used in laboratory research as an analogue of Na+ in salinity studies. The advantages of Li+ are that it may share transport systems and toxicity targets with Na+, and that, because of its higher inhibitory potency, it can be used at concentrations that have no osmotic effects (Nublat et al., 2001; Serrano et al., 1999). However, our results show that in some situations, for example for transgenic plants overexpressing AtNHX8, Li+ cannot be used as an analogue of Na+.

Lithium is used as treatment in human therapy of manic-depressive psychosis, but its mechanism of action is not known. In order to elucidate the mechanisms underlying the effect of Li+ therapy in bipolar disorder, many efforts have been made to identify in vivo targets of Li+ in yeast as a model organism, as well as in animals (Quiroz et al., 2004). The proposed targets of Li+ include inositol monophosphatase (Berridge et al., 1989); inositol polyphosphate 1-phosphatase (Lopez-Coronado et al., 1999; Miyamoto et al., 2000); glycogen synthase kinase-3 (O'Brien et al., 2004); fructose 1,6-bisphosphatase (Villeret et al., 1995); bisphosphate nucleotidase (Miyamoto et al., 2000; Murguia et al., 1996); and phosphoglucomutase (Csutora et al., 2005; Masuda et al., 2001). In plants, inositol monophosphatase, inositol polyphosphate 1-phosphatase and SR-like splicing proteins have been reported to be Li+ targets (Forment et al., 2002; Gillaspy et al., 1995; Xiong et al., 2004). Furthermore, there is evidence that Li+ can regulate ACC synthase gene expression in A. thaliana (Liang et al., 1996), and may trigger a hypersensitive-like response in which ethylene signalling is essential (Naranjo et al., 2003).

Despite some demonstrated effects of Li+ on plant physiology, the biological significance of why plants have specific Li+/H+ antiporters remains elusive, as Li+ is not an essential nutrient and is not present in nature in significant concentrations. It may have evolved from a non-specific Na+ (Li+)/H+ antiporter that lost transport activity for Na+. Indeed, there are reports on the effects of single-residue or domain deletions or various amino acid substitutions on different aspects of transporter activity, including ion selectivity (Hamada et al., 2001; Kuwabara et al., 2004; Mitsui et al., 2004; Ohgaki et al., 2005; Waditee et al., 2001; Yamaguchi et al., 2003).

Although the biological significance of Li+/H+ antiporters is not completely clear, possessing efficient mechanisms for toxic ion extrusion and ion homeostasis could be useful for bacteria, fungi and plants, as it would allow them to grow in a wide range of environmental conditions. On the other hand, AtNHX8 can be useful in the area of molecular biology. For example, we are now considering the feasibility of using AtNHX8 as a marker gene for screening transgenic plants. AtNHX8 can also be included in a model system for studying the cation selectivity of cation transporters.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant materials and growth treatments

Arabidopsis thaliana ecotype Columbia was used in all seedling experiments. The seeds were surface-sterilized and stratified at 4°C for 2–4 days. They were then sown on solid agar plates containing Murashige and Skoog (MS) salts, 3% (w/v) sugar, 1% agar adjusted to pH 5.7, and incubated in a growth chamber at 23°C under a 16-h light/8-h dark photoperiod at 130 μE m−2 sec−1.

Cation tolerance tests were performed using the root-bending assay (Zhu et al., 1998). Five-day-old seedlings grown on MS agar plates (1% agar, pH 5.7) containing 3% sucrose were transferred to agar plates containing the identical growth media, or plates supplemented with various concentrations of cations. Four to six plants were placed on each of three identical plates at each salt concentration. The treatment plates were incubated vertically with seedlings in the upright position. For in vitro germination tests, seeds were sown in MS agar plates supplemented with different concentrations of LiCl or NaCl.

Isolation of the nhx8 T-DNA insertional mutant

The nhx8-1 and nhx8-2 insertion alleles were isolated from the T-DNA-transformed Arabidopsis collection from the SIGnAL database (http://signal.salk.edu/cgi-bin/tdnaexpress). The nhx8-1 insertion in the gene was identified using the T-DNA left-border primer LB-1 (5′-GCCTTTTCAGAAATGGATAAATAGCCTTGCTTCCT-3′) and AtNHX8 gene-specific forward (5′-CCCTCTAACAAGCCAACCTTCTCA-3′) and reverse (5′-TCCTCACGATGGGTACGAAGC-3′) primers. The nhx8-2 insertion was identified using the T-DNA left-border primer LB-1 and AtNHX8 gene-specific forward (5′-TGGCCAATATCAGCCTATCAG-3′) and reverse (5′-ATCTTGGAGATGACGAGGAGC-3′) primers. The homozygous nhx8-1 and nhx8-2 mutants were identified by PCR analysis. Plants possessing the homozygous nhx8-1 and nhx8-2 mutant genotypes were used for further analysis.

Isolation of cDNA and genomic DNA sequence

Total RNA was isolated from 2-week-old seedlings of wild-type plants with the TRIzol® reagent (Invitrogen http://www.invitrogen.com). First-strand cDNA was synthesized from total RNA (2 μg) with M-MLV reverse transcriptase (Promega http://www.promega.com), and used as a template for subsequent PCR amplification. PCR amplification was performed using pfx DNA polymerase (Invitrogen). AtNHX8 cDNA and genomic DNA were amplified by PCR with the cDNA and genomic DNA, respectively, serving as templates. The forward primer used in the PCR amplification was 5′-ACCTCTAGAATGACGAGTATAATCGGCGCGGCG-3′, and the reverse primer was 5′-ACCGGTACCTCATGCATCTGTTTCGCCAATGGCC-3′. The forward and reverse primers were introduced into the XbaI and KpnI restriction enzyme sites, respectively, shown as underlined areas. The PCR products were cloned into pMD18-T vectors (TaKaRa) and sequenced.

Transgenic constructs and plant transformation

To construct the transgenic vectors, the EGFP coding sequence was amplified from the pEZT-NL vector using the following primers: 5′-AAATCTAGACTCGAGCGGGCCCGGGATCCGGCTG-3′ (forward), 5′-ACCGAGCTCGGTACCTTACTTGTACAGCTCGTCCATGCC-3′ (reverse). The forward and reverse primers were introduced into XbaI–XhoI and SacI–KpnI restriction enzyme sites, respectively. The fragment was ligated into the XbaI/SacI sites of the pBI121 plasmid in place of GUS to form an intermediate vector, pBI-EGFP. AtNHX8 cDNA and genomic fragments cut from the pMD18 vectors were then cloned into the XbaI and KpnI sites of pBI121–EGFP, respectively, in place of the EGFP gene. To generate an AtNHX8–EGFP fusion construct, the AtNHX8 cDNA was amplified from the pMD18-NHX8 vector containing AtNHX8 using the following primers: 5′-ACCTCTAGAATGACGAGTATAATCGGCGCGGCG-3′ (forward) and 5′-ACCCCCGGGATGCATCTGTTTCGCCAATGGCC-3′ (reverse). The PCR products were cloned into pMD18-T vectors and sequenced. The resultant cDNA fragment, in which the stop codon was eliminated, was then cut from the pMD18 vector and cloned into the XbaI and SmaI sites of the pBI121–EGFP vector. Each of these three kinds of transgenic vector were introduced into Agrobacterium GV3101 strain cells. The floral-dipping method (Clough and Bent, 1998) was used to transform sos1-1 mutant or wild-type Arabidopsis plants. Transgenic plants were screened on kanamycin plates and transformants were transferred to soil for germinating.

RT–PCR analysis

Total RNA was isolated from wild-type or transgenic plants with the TRIzol® reagent. First-strand cDNA was synthesized with the M-MLV reverse transcriptase (Promega) from total RNA (approximately 2 μg), and used as template for PCR amplification. AtNHX8 gene-specific primers were as follows: forward primer TF 5′-TCTGACTTGAAGAAGCTCCTGAG-3′ and reverse primer TR 5′-TCAAAGCCAAAAAGGATTGATTGAA-3′. β-tubulin primers were as follows: β-tub1 forward primer 5′-CGTGGATCACAGCAATACAGAGCC-3′; β-tub2 reverse primer 5′-CCTCCTGCACTTCCACTTCGTCTTC-3′ (Chinnusamy et al., 2003). We used 23 PCR cycles for β-tubulin and 27 PCR cycles for AtNHX8 amplifications. The PCR program used was as follows: 94°C for 30 sec, 55°C for 30 sec and 72°C for 45 sec. A final incubation at 72°C for 7 min was performed to complete product synthesis. At least five replications were carried out for each treatment. PCR products were separated on a 1.5% agarose gel stained with ethidium bromide and then quantified using a densitometer with IS-1000 digital imaging system software (Alpha Innotech http://www.alphainnotech.com).

Yeast strain, growth media and functional expression in yeast

The PCR-amplified products were KpnI/XbaI-digested and cloned directionally into the yeast expression vector pYES2 (Invitrogen). The construct was used to transform the yeast strain AXT3. Sequencing analysis confirmed that the inserts contained in the pYES2 plasmids were the same as the Arabidopsis AtNHX8 gene. Saccharomyces cerevisiae mutant strain AXT3 (Δena1-4::HIS3 Δnha1::LEU2 Δnhx1::TRP1) was kindly provided by J. M. Pardo, Instituto de Recursos Naturales y Agrobiología (IRNA), Sevilla, Spain. Yeast cells were grown at 30°C in yeast extract peptone dextrose medium (1% yeast extract, 2% peptone, 2% glucose) or synthetic dextrose medium (0.67% yeast nitrogen base, 2% glucose) with appropriate amino acid supplements as indicated. Yeast was transformed by a standard Li–polyethylene glycol method. For induction of gene expression under control of the GAL1 promoter, 2% glucose was replaced by 2% galactose in the growth media. Na+- and Li+-tolerance tests were performed in arginine phosphate medium, which was essentially free of alkali cations. For spot assays, aliquots (4 μl) from an exponentially growing culture (OD600 = 0.5) and serial dilutions (1:10, 1:100, 1:1000) were spotted onto APG plates with or without NaCl, KCl, LiCl or HygB. Plates were incubated at 28°C for 2–4 days.

Isolation of the AtNHX8 promoter region and GUS assays

An approximately 660-bp-long DNA fragment upstream of the ATG start codon was amplified by PCR using the following primers: S1B PF: 5′-CCCAAGCTTGAATGATGTTACACGAGACCAAGTT-3′ (added HindIII site); and S1B PR: 5′-AAGGATCCCTCCGGTAGAAAGAAGCAAACGATC-3′ (added BamHI site). Amplified DNA products were ligated via the attached BamHI/HindIII sites into the binary vector pCAMBIA1391 to obtain a transcriptional fusion of the AtNHX8 promoter and the β-glucuronidase coding sequence. Transgenic seedlings were generated as described above, and were identified on sterile agar medium containing hygromycin. For the β-glucuronidase assay, materials were stained at 37°C overnight in 1 mg ml−1 5-bromo-4-chloro-3-indolyl-β-glucuronic acid, 5 mm potassium ferricyanide, 5 mm potassium ferrocyanide, 0.03% Triton X-100 and 0.1 m sodium phosphate buffer pH 7.0.

Subcellular localization of the AtNHX8–EGFP fusion protein

Thirty independent 35S-AtNHX8–EGFP transgenic lines were generated and used for localization of the fusion protein. Confocal microscope images were taken using a Leica confocal laser microscope under a ×40 oil objective. The excitation wavelength for EGFP detection was 488 nm. The young roots of kanamycin-resistant seedlings from transgenic plants containing the 35S–EGFP and 35S–AtNHX8–EGFP fusion protein were used for direct confocal microscope analysis.

Determination of intracellular ion contents

To measure Li+ and K+ accumulation, plants were grown for 2 weeks on MS agar plates (pH 5.7), then transferred to the same medium supplemented with a range of concentrations of LiCl (0, 10, 15, 20 and 30 mm). Seedlings were collected, washed five times, then dried at 80°C until constant weights were achieved (at least for 24 h). The weighed samples were dry-ashed for 6 h in quartz dishes at 570°C in a muffle furnace, digested in 2 ml 18% HCl, and diluted in 20 ml distilled water prior to ion concentration analysis. Analyses were performed using an inductively coupled plasma atomic emission spectrometer (ICP–AES; Perkin-Elmer http://www.perkinelmer.com). The results shown are means of 30 plants.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank Drs Jian-Kang Zhu and Zhi-Zhong Gong for kindly providing sos1-1 seeds, Dr J.M. Pardo (IRNA-CSIC, Sevilla, Spain) for the yeast strain AXT3, and Drs Yan Guo, Jian-Ru Zuo and Rong-Feng Huang for their assistance and valuable suggestions. This work was supported by the National Basic Research Program of China (grant nos 2006CB100100 and 2003CB114300) and the National Science Foundation of China (grant nos 30370129 and 30421002).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Sequence data from this article have been deposited in GenBank under accession number DQ104324.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  9. Supporting Information

Figure S1. The AtNHX8 Protein Is Phylogenetically Related to AtNHX7/SOS1. (A) Alignment of AtNHX8 (At1g14660) with AtNHX7/SOS1 (At2g01980) over a stretch of 760?amino acids by the program TCoffee (http://www.ch.embnet.org/software/TCoffee.html). Identical amino acids are highlighted in black, and conservative substitutions are highlighted in gray. (B) Hydrophobicity plot of AtNHX8. The hydrophobicity values were calculated by the program TMPRED ( http://www.ch.embnet.org/software/TMPRED_form.html). (C) Phylogenetic analysis of AtNHX8 and other Na+/H+ antiporter homologs. Multiple sequence alignment and the phylogenetic tree were performed by the program TCoffee. The accession numbers and sources of each other representative Na+/H+ antiporters are as follows: NHE1 (P19634), Homo sapiens; NHA1 (NP_013239), S. cerevisiae; NHX1 (NP_010744), S. cerevisiae; SOD2 (CAA77796.1), S. pombe; (BAA31695.1), P. aeruginosa and Syn NhaP (NP_441245) Synechocystis sp. PCC 6803 . The AGI numbers of Na+/H+ antiporters homologs are as followes: AtNHX1 (At5g27150), AtNHX2 (At3g05030), AtNHX3 (At3g06370), AtNHX4 (At5g55470), AtNHX5 (At1g54370), AtNHX6 (At1g79610), AtNHX7/SOS1 (At2g01980), and AtNHX8/SOS1b (At1g14660). Supplementary information The AtNHX8 Protein Belongs to the Na+/H+ Antiporter Family, and Is Phylogenetically Related to AtNHX7/SOS1

FilenameFormatSizeDescription
TPJ_2990_sm_FIgureS1.zip2273KSupporting info item
TPJ_2990_sm_SuppInfo.doc31KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.