Salt Tolerance Conferred by Overexpression of a Vacuolar Na+/H+ Antiport in Arabidopsis

See allHide authors and affiliations

Science  20 Aug 1999:
Vol. 285, Issue 5431, pp. 1256-1258
DOI: 10.1126/science.285.5431.1256


Agricultural productivity is severely affected by soil salinity. One possible mechanism by which plants could survive salt stress is to compartmentalize sodium ions away from the cytosol. Overexpression of a vacuolar Na+/H+antiport from Arabidopsis thaliana in Arabidopsisplants promotes sustained growth and development in soil watered with up to 200 millimolar sodium chloride. This salinity tolerance was correlated with higher-than-normal levels of AtNHX1transcripts, protein, and vacuolar Na+/H+(sodium/proton) antiport activity. These results demonstrate the feasibility of engineering salt tolerance in plants.

Salinity stress is one of the most serious factors limiting the productivity of agricultural crops. The detrimental effects of salt on plants are a consequence of both a water deficit resulting in osmotic stress and the effects of excess sodium ions on critical biochemical processes (1). In order to tolerate high levels of salt, plants should be able to utilize ions for osmotic adjustment and internally distribute these ions to keep sodium away from the sites of metabolism (1). Plant cells are structurally well suited for the sequestration of ions because of the presence of large, membrane-bound vacuoles. It has been proposed that in salt-tolerant plants, the compartmentation of Na+ into vacuoles, through the operation of a vacuolar Na+/H+ antiport, provides an efficient mechanism to avert the deleterious effects of Na+in the cytosol and maintains osmotic balance by using Na+(and chloride) accumulated in the vacuole to drive water into the cells (2). This Na+/H+antiport transports Na+ into the vacuole by using the electrochemical gradient of protons generated by the vacuolar H+-translocating enzymes, H+–adenosine triphosphatase (ATPase) and H+–inorganic pyrophosphatase (PPiase) (3, 4). Vacuolar Na+/H+ antiport activity was shown first in tonoplast vesicles from red beet storage tissue (5) and later in various halophytic and salt-tolerant glycophytic species (6, 7). Chloride transport into the vacuole is mediated by anion channels (8). InArabidopsis, a vacuolar chloride channel, At-CLCd, similar to the yeast Gef1, has been cloned (9). The analysis of genes involved in cation detoxification in yeast led to the identification of a novel Na+/H+antiport (Nhx1). Nhx1 was localized to a prevacuolar compartment and showed a high degree of amino acid sequence similarity to Na+/H+ antiports fromCaenorhabditis elegans and humans (NHE6, mitochondrial) (10). Recently, the Arabidopsis thalianagenome-sequencing project has allowed for the identification of a plant gene (AtNHX1) homologous to the Saccharomyces cerevisiae Nhx1 gene product (11–13). Both Nhx1 and Gef1 are localized to the yeast prevacuolar compartment, suggesting a role for this compartment in salt tolerance. Overexpression of AtNHX1 suppresses some of the salt-sensitive phenotypes of the Δnhx1 yeast strain (13), suggesting that the plant and the yeast gene products might be functionally similar.

AtNHX1 transcripts are found in root, shoot, leaf, and flower tissues (11). To determine the subcellular localization of AtNHX1, we immunoblotted membrane fractions (14) isolated from wild-type plants and plants overexpressing AtNHX1 (15) with antibodies raised against the COOH-terminus of AtNHX1 (Fig. 1). A protein of an apparent molecular mass of 47 kD was detected mainly in the tonoplast- and Golgi/endoplasmic reticulum (ER)–enriched fractions, and was more abundant in the transgenic plants. No noticeable cross-reactivity with these antibodies was observed in plasma membrane– and mitochondria-enriched fractions, which suggests that the plant NHX1 does not share the same subcellular localization as that determined for NHE6 (16). It remains unclear whether the immunoreactive protein observed in the Golgi/ER-enriched fractions represents contamination of this fraction with tonoplast proteins (5) or presence of the target protein in prevacuolar compartments. Nonetheless, immunoblots of purified leaf vacuoles (17) from transformed plants showed more AtNHX1 gene product than those from wild-type plants (Fig. 1B). The apparent molecular mass (47 kD) of AtNHX1 is somewhat lower than that predicted by the amino acid sequence of the AtNHX1 open reading frame (58 kD) and may reflect anomalous migration in the SDS–polyacrylamide gel electrophoresis (SDS-PAGE) gel (18) or specific cleavage (or degradation).

Figure 1

Subcellular localization of AtNHX1. (A) Membrane fractions were isolated from shoots of 5-week-old seedlings (14). Protein (5 μg) was separated on a 10% SDS-PAGE and electroblotted. Polyclonal antibodies, raised against the COOH-terminal portion of AtNHX1 (21), were used for the immunoblot, which was developed using chemiluminescence. The lane numbers correspond as follows: Lanes 1 and 2, mitochondria; lanes 3 and 4, tonoplast-enriched fraction; lanes 5 and 6, Golgi/ER-enriched fraction; lanes 7 and 8, plasma membrane–enriched fraction. Odd-numbered lanes correspond to membrane fractions isolated from plants overexpressing AtNHX1, while even-numbered lanes correspond to fractions isolated from wild-type plants. (B) Enrichment of the fractions with tonoplast membranes was assessed with antibodies raised against the vacuolar H+-PPiase and the vacuolar V-ATPase. (C) Purified vacuoles (17) were pelleted and solubilized in SDS-PAGE sample buffer. Protein (3 μg) was resolved on 10% SDS-PAGE and immunoblotted. Lanes 1 and 3 correspond to vacuolar membranes isolated from plants overexpressing AtNHX1, while lanes 2 and 4 correspond to vacuolar membranes isolated from wild-type plants. Lanes 1 and 2 show immunoreaction with antibodies raised against AtNHX1. Lanes 3 and 4 show immunoreaction to antibodies raised against the vacuolar H+-PPiase. Relative molecular masses are indicated on the left.

In order to assess whether AtNHX1 provides a Na+/H+ exchange function, Na+-dependent H+ movements (19) were measured in vacuoles isolated from the leaves of wild-type plants and plants overexpressing AtNHX1 (Fig. 2). The Na+/H+ exchange rates were very low in vacuoles from wild-type plants (Fig. 2A, trace 1). In contrast, Na+/H+ exchange rates were much higher in vacuoles from the transgenic plants (Fig. 2A, trace 2). The Na+/H+ antiport activity was not affected by the presence of 30 mM K+ ions in the assay medium (20), indicating selectivity for Na+. Electroneutral Na+/H+ exchange is suggested by the similar rates of Na+-dependent H+ movements obtained when the membrane potential was clamped by the addition of 1 μM valinomicyn and 2 mM K+to the assay medium (20). The Na+ dependence of the H+ flux is evidence of a Na+/H+ antiport mechanism. The Na+/H+ exchange displayed Michaelis-Menten kinetics with respect to extravacuolar Na+concentrations (Fig. 2B). Similar apparent K mvalues of the exchanger for Na+ were obtained with two independent transgenic lines, K m = 7 mM (Fig. 2C) and 6.1 mM (20). These values are of the same order of magnitude as those reported for other plant species (7). The relative increase in protein abundance (Fig. 1B) is less than the increase in Na+/H+ antiport activity measured in vacuoles from transgenic plants (Fig. 2, A and B). These observations suggest that in wild-type plants under normal growth conditions, AtNHX1 function may be repressed. Overexpression of AtNHX1 may overcome this endogenous repression mechanism.

Figure 2

The Na+/H+exchange activity in leaf vacuoles from wild-type plants and plants overexpressing AtNHX1. (A) The fluorescence quenching of acridine orange was used to monitor the dissipation of inside-acid pH gradients across the vacuoles. Vacuoles (5 × 104) were added to 0.8 ml buffer containing 0.3 M mannitol, 2 mM dithiothreitol, 5 mM Tris/MES buffer (pH 8.0), 5 mM glucose, 30 mM tetramethylammoniuim chloride, 1.5 mM Tris-ATP, and 5 μM acridine orange. Proton translocation was initiated with the addition of 3 mM Mg2+ and the change of fluorescence with time was monitored as described (19). When a steady-state pH gradient (acidic inside) was formed, the ATP-dependent H+ transport activity was stopped by the addition of hexokinase (HK) (19). After a constant rate of fluorescence recovery was obtained, aliquots of 5 M NaCl (Na+) were added, and the changes in fluorescence recovery were determined (19). The addition of monensin (mon), an artificial Na+/H+ antiport, abolished the pH gradient and the fluorescence was fully recovered. The figure shows a typical recording; gaps in the lines indicate the addition of reagents. Curves labeled 1 and 2 indicate vacuoles from wild-type plants and from plants overexpressing AtNHX1, respectively. (B) Kinetics of Na+/H+ exchange in vacuoles from wild-type plants (○) and plants overexpressing AtNHX1 (•). Each point represents the mean ± SD (n = 3). (C) Double reciprocal plot of the Na+-dependent recovery of acridine orange fluorescence quenching. Each point represents the mean ± SD (n = 3). The apparentK m was calculated by the intersection of the fitted line with the abscissa. % Q, percentage of fluorescence quenching.

Salt tolerance was tested in wild-type and transgenic plants overexpressing AtNHX1 (Fig. 3). Wild-type plants displayed progressive chlorosis, reduced leaf size, and a general growth inhibition when watered with a NaCl-containing solution. These inhibitory effects increased progressively with the increasing NaCl concentration in the watering solution. The transgenic plants were unaffected by up to 200 mM NaCl (Fig. 3, lower panel) and plant development was not compromised, as the transgenic plants bolted and set seed in all salt treatments. However, transgenic plants grown at 300 mM NaCl displayed a reduction in leaf size and chlorosis (20). The three independent transgenic lines showed a similar increase of AtNHX1 transcript levels (Fig. 3, inset) and similar salt tolerance (20). The Na+ content of both wild-type and transgenic plants increased with exposure to high NaCl (Fig. 4). The higher Na+content of the transgenic plants growing in 200 mM NaCl, together with the sustained growth (Fig. 3) and the increased vacuolar antiport activity (Fig. 2), is consistent with increased vacuolar compartmentation of Na+ in the transgenic plants. We did not detect an increase in AtNHX1 transcript levels in response to NaCl (50 to 250 mM) (13) or upon the application of exogenous ABA (20). We have analyzed RNA from young seedlings and mature plants, and from the roots, shoots, and leaves of NaCl-stressed plants at different time points (6 to 36 hours) grown in petri dishes or in soil. No increase in the AtNHX1 protein product was detected either by immunoblotting or by vacuolar Na+/H+ activity assays in response to NaCl-stress (20). These results would suggest that the induction of AtNHX1 protein synthesis or vacuolar Na+/H+ antiport activity in response to NaCl-stress in Arabidopsis wild-type plants requires conditions which are currently unknown. Since Arabidopsis is a glycophytic plant with a sensitivity to salt similar to most crop plants, our findings suggest the feasibility of genetic engineering crop plants with improved salt tolerance.

Figure 3

Salt treatment of wild-type plants and plants overexpressing AtNHX1. Transgenic T3 plants from the lines overexpressing AtNHX1 were used in assessing salt tolerance. Twenty wild-type plants and 20 of each of the three lines of transformed plants overexpressing AtNHX1 were grown on a short-day cycle (8 hours light, 16 hours dark). Each set of 20 plants was divided into five groups (labeled A through E) of four plants each. We applied 25 ml of a diluted nutrient solution (1/8 MS salts) every other day over the 16-day watering treatment. The control group received no NaCl supplementation. The remaining groups were watered with nutrient solution supplemented with NaCl. The concentrations of NaCl supplementation were increased stepwise by 50 mM every 4 days for each group, to the indicated maximum: (A) control, (B) 50 mM NaCl, (C) 100 mM NaCl, (D) 150 mM NaCl, and (E) 200 mM NaCl. The salt-tolerance phenotype was observed in three independent transgenic lines tested. The transgenic line shown (2′) is representative of the three tested lines as are the plants from each treatment group. (A) (upper panel) Wild-type plants. (B) (lower panel) Plants overexpressing AtNHX1. (Inset) Northern blot of RNA isolated from leaves of wild-type (wt) and three independent lines (2′; 3′; 4′) of transgenic plants grown in the absence of NaCl. RNA was probed with AtNHX1 cDNA. Equal amounts of total RNA were present in each sample (30 μg). In all plants, the endogenous 2.1 kilobase (kb) transcript was detected. An additional transcript of ∼1.7 kb that corresponds to the predicted open reading frame of AtNHX1 can be seen only in the transgenic plants. Apparent molecular size (kb) of the transcripts is indicated to the left.

Figure 4

The Na+ content in wild-type (black bars) and transgenic plants (white bars) grown in the absence or presence of 200 mM NaCl. The above-ground parts of the plants were harvested at the end of the salt treatment. Dry weight was measured after 24 hours at 70°C, and Na+ content was determined by atomic absorption spectrophotometry. Values are the mean ± SD (n = 4).

  • * These authors contributed equally to this work.

  • To whom correspondence should be addressed. E-mail: blumwald{at}


Stay Connected to Science

Navigate This Article