An Arabidopsis mutant that requires increased calcium for potassium nutrition and salt tolerance

Abstract

Potassium (K⁺) nutrition and salt tolerance are key factors controlling plant productivity. However, the mechanisms by which plants regulate K⁺ nutrition and salt tolerance are poorly understood. We report here the identification of an Arabidopsis thaliana mutant, sos3 (salt-overly-sensitive 3), which is hypersensitive to Na⁺ and Li⁺ stresses. The mutation is recessive and is in a nuclear gene that maps to chromosome V. The sos3 mutation also renders the plant unable to grow on low K⁺. Surprisingly, increased extracellular Ca²⁺ suppresses the growth defect of sos3 plants on low K⁺ or 50 mM NaCl. In contrast, high concentrations of external Ca²⁺ do not rescue the growth of the salt-hypersensitive sos1 mutant on low K⁺ or 50 mM NaCl. Under NaCl stress, sos3 seedlings accumulated more Na⁺ and less K⁺ than the wild type. Increased external Ca²⁺ improved K⁺/Na⁺ selectivity of both sos3 and wild-type plants. However, this Ca²⁺ effect in sos3 is more than twice as much as that in the wild type. In addition to defining the first plant mutant with an altered calcium response, these results demonstrate that the SOS3 locus is essential for K⁺ nutrition, K⁺/Na⁺ selectivity, and salt tolerance in higher plants.

Soil salinity presents an increasing threat to plant agriculture as more and more of the world’s arable land becomes salinized (1). High salinity exerts its detrimental effects on plants because of ion toxicity as well as osmotic stress (2). In most saline soils, Na⁺ is the major toxic cation. One harmful effect of Na⁺ is that it disrupts K⁺ nutrition (3, 4). K⁺ is one of the three major components in fertilizers applied to soils and a key factor controlling crop productivity (5). Physiologic studies have established that K⁺ uptake by plant roots is mediated by at least two mechanisms, i.e., the mechanism 1 (high-affinity) and 2 (low-affinity) transport systems (3, 6, 7). Because soil solutions often contain <1 mM K⁺, the high-affinity system is thought to play the predominant role in plant potassium nutrition (5). High-affinity K⁺ uptake is a tightly regulated process. When plant roots experience K⁺ deficiency, the high-affinity system is induced and becomes essential for plant growth (8–10). Both high- (11, 12) and low-affinity (13–15) K⁺ transporters have been cloned from plants. However, it remains to be shown which of these cloned transporters actually function in root K⁺ uptake from the soil solutions. Furthermore, the mechanisms by which roots detect K⁺ deficiency and the signaling pathway that up-regulates the high-affinity system are still unknown.

The importance of the mechanism 1 transport system in plant salt tolerance is underlined by recent genetic evidence suggesting that it is essential for plant survival and growth under salinity stress (10). The Arabidopsis sos1 mutation which causes reduced mechanism 1 K⁺ uptake also leads to increased sensitivity of plant growth to inhibition by NaCl (10). Kinetic analysis suggested that mechanism 1 uptake has a higher K⁺/Na⁺ selectivity than mechanism 2 (3). One factor known to be involved in the regulation of K⁺/Na⁺ selectivity of K⁺ transport during NaCl stress is Ca²⁺ (16). Under NaCl stress, Ca²⁺ increases the selectivity of root K⁺ transport systems (16). High external Ca²⁺ has been shown to improve plant salt tolerance (17). However, the cellular components that mediate these Ca²⁺ effects have not been identified genetically or biochemically.

We report here the identification of a genetic locus, SOS3, in Arabidopsis thaliana that is necessary for salt tolerance, K⁺ nutrition, and Ca²⁺ response. A mutation in this locus (sos3) results in plant hypersensitivity toward NaCl inhibition and in an increased requirement for K⁺. Surprisingly, high external Ca²⁺ can suppress the mutant phenotype. sos3 represents the first higher plant mutant with an altered response to Ca²⁺. The mutation reveals an important link between plant responses to Na⁺ stress, K⁺ deficiency, and Ca²⁺ regulation. Our results suggest that Ca²⁺ plays a fundamental role in regulating root potassium nutrition and plant salt tolerance.

MATERIALS AND METHODS

Isolation of the sos3 Mutant.

A. thaliana (ecotype Columbia) carrying the homozygous recessive gl1 mutation was the parental strain of ethyl methanesulfonate- or fast neutron-mutagenized seeds. M2 populations from ethyl methanesulfonate- or fast neutron-mutagenized seeds or T4 pools from Agrobacterium-transformed lines (18), were screened using the root-bending assay of Wu et al. (10). Seeds were surface-sterilized and germinated on medium containing Murashige and Skoog (MS) salts (19), 3% (wt/vol) sucrose, and 1.2% (wt/vol) agar, pH 5.7. When appropriate, seedlings were transplanted to pot medium and grown to maturity. Growth conditions were as described (10).

Genetic Analysis.

The sos3 mutant was backcrossed with the wild-type Columbia gl1 background, and the F1 seedlings were allowed to self-pollinate. F1 and F2 seedlings were scored for salt sensitivity using the root-bending assay (10). For mapping of the SOS3 locus, homozygous sos3 plants in the Columbia gl1 background were crossed to plants of the Landsberg erecta background. From the segregating F2 generation, 426 homozygous sos3 mutants were selected for mapping with molecular markers that are polymorphic between Columbia and Landsberg erecta.

Growth Measurement.

Wild-type and sos3 mutant seedlings (4 days old) grown on vertical MS agar plates were transferred to various agar media for stress treatment and growth measurements (10). To determine the K⁺ requirement of mutant plants, seedlings were transferred to a modified MS medium A supplemented with various levels of KCl. Modified MS medium A contains potassium-free 1/20 strength MS major salts and 1× MS minor salts. For the determination of the Ca²⁺ requirement, seedlings were transferred to a modified MS medium B supplemented with various levels of CaCl₂. Modified MS medium B contains potassium- and calcium-free 1/20 strength MS major salts, 1× MS minor salts, and 20 μM K⁺ unless stated otherwise.

Determination of Ion Contents.

Four-day-old seedlings grown on MS plates were transferred to 250-ml flasks containing a 50-ml solution of half-strength MS salts and 2% sucrose. The flasks were shaken at 120 rpm under constant fluorescent light for 5 days. The seedlings were transferred to 50-ml treatment solutions after being washed briefly twice with deionized water and once with appropriate treatment solutions. The treatment solutions were prepared by adjusting the K⁺, Na⁺, and Ca²⁺ concentrations of a calcium–potassium-free modified MS medium to the stated levels.

The calcium–potassium-free modified MS medium consists of the following: 1650 mg/liter NH₄NO₃, 370 mg/liter MgSO₄⋅7H₂O, 165 mg/liter (NH₄)₂HPO₄, 27.8 mg/liter FeSO₄⋅7H₂O, 37.3 mg/liter disodium EDTA, 0.7495 mg/liter NaI, 6.3 mg/liter H₃BO₃, 16.9 mg/liter MnSO₄⋅H₂O, 8.6 mg/liter ZnSO₄⋅7H₂O, 0.25 mg/liter Na₂MO₄⋅2H₂O, 0.016 mg/liter CuSO₄⋅5H₂O, and 0.0267 mg/liter CoSO₄⋅6H₂O.

The seedlings were grown in the treatment solutions for 4 days, briefly rinsed three times with deionized water, and dried in a 80°C oven. The samples were digested in 3 ml of concentrated nitric acid overnight followed by boiling for ∼1 h until the solution became completely clear. The solution was brought to 25 ml by adding 0.25% lanthanum in 2.5% nitric acid. The K⁺, Na⁺, and Ca²⁺ contents in the solution were determined by atomic absorption spectrophotometry.

RESULTS

Isolation of the sos3 Mutant.

The sos3 mutant was identified from a large-scale screening of sos mutants. Altogether, ∼260,000 seedlings from ethyl methanesulfonate- or fast neutron-mutagenized M2 seeds, or T4 seeds from T-DNA-mutagenized plants were screened on 50 mM or 75 mM NaCl by using the root-bending assay of Wu et al. (10). Of 41 sos mutants isolated (J.K.Z., unpublished data), one line defines a new locus, designated as SOS3. The sos3 mutant was identified from fast neutron-mutagenized M2 seeds. Fig. 1 shows the phenotype of the sos3 mutant under the conditions of NaCl stress or low K⁺. On regular nutrient medium (MS salts), sos3 seedlings were indistinguishable from the wild type. However, under high NaCl or low K⁺ stresses, the growth of sos3 plants was inhibited to a greater extent than that of the wild type (Fig. 1).

Figure 1.

Representative plants of wild-type (WT), sos1, and sos3 mutants that were exposed to high-salt (100 mM NaCl) or low-K⁺ (20 μM K⁺) stresses. Seedlings were grown on vertical MS agar plates for 4 days (root length ≅ 1.5 cm) and then transferred to vertical agar plates containing control (A, MS nutrients), 100 mM NaCl (B, MS + 100 mM NaCl), or modified MS (see Materials and Methods) with 20 μM K⁺ and 0.15 mM (C) or 3 mM Ca²⁺ (D) for 7 days. The growth of both sos1 and sos3 was inhibited by either NaCl (B) or low K⁺ (C). However, the growth of sos3 on low K⁺ was restored by increased Ca²⁺ (D). (Bar = 1 cm.)

Genetic Analysis.

The sos3 mutant was backcrossed with the wild-type background. The resulting F1 plants all exhibited the wild-type phenotype in response to NaCl stress. F2 progeny from the cross segregated 925:310 (WT:sos3) (Table 1). This indicated that sos3 is a monogenic recessive mutation in a nuclear gene. The sos3 mutant was also crossed to sos1-1. The resulting F1 plants exhibited the wild-type phenotype under NaCl stress, indicating that the sos3 mutant defines a locus that is different from SOS1.

Table 1.

Genetic analysis of the sos3 mutant

Parental genotype	Seedlings tested, n	Resistant*	Sensitive*
SOS3/SOS3 × sos3/sos3	131	131	0
SOS3/sos3 × SOS3/sos3	1235	925	310
sos3/sos3 × sos1-1/sos1-1	34	34	0

*Resistance or sensitivity was determined in the root-bending assay, using 100 mM NaCl. 

Chromosome Mapping of sos3.

To map the sos3 mutation, the mutant in the Columbia background was crossed with wild-type Landsberg erecta plants. From the segregating F2 population, 426 salt hypersensitive plants were selected, and genomic DNA was extracted from each of the plants. Initial screening with simple sequence length polymorphism markers (20) from each of the five chromosomes of Arabidopsis indicated that the sos3 mutation is linked to the marker nga76 on chromosome V. Further analysis with additional markers on chromosome V showed that sos3 is very tightly linked to nga139. Only 4 recombination events were detected among the 426 plants (equivalent of 852 chromosomes) examined. Therefore, the SOS3 locus is ∼0.5 cM away from nga139. Based on a recently published physical map of chromosome V (21), this genetic distance likely corresponds to <100 kb. As shown in Fig. 2, sos3 is ∼1.6 cM (14 recombination events detected among the 426 plants) from marker CDPK9 which represents a locus encoding a calcium-dependent protein kinase (22).

Figure 2.

Genetic map showing the position of the SOS3 locus relative to selected physical markers on chromosome V.

sos3 Is Specifically Hypersensitive to Na⁺ and Li⁺.

Root elongation is a convenient and accurate indicator of Arabidopsis seedling growth (10). Measurement of root elongation showed quantitatively that the sos3 mutant is hypersensitive to NaCl stress (Fig. 3). The concentration of NaCl that decreased the root elongation rate by 50% relative to medium without salt (I₅₀) was estimated. The I₅₀ for sos3 and the wild-type seedlings are ∼37 and 100 mM, respectively (Fig. 3). To determine whether the sos3 mutant is hypersensitive to osmotic stress or to specific ions, the seedlings were also treated on media supplemented with KCl, LiCl, CsCl, or mannitol. The results show that sos3 is also hypersensitive to LiCl, but not to KCl, CsCl, or mannitol (Fig. 3). Therefore, the sos3 mutation does not result in hypersensitivity to general osmotic stress. Rather, the hypersensitivity is restricted to Na⁺ and Li⁺.

Figure 3.

The sos3 mutant is hypersensitive to Na⁺ and Li⁺ but not to K⁺, Cs⁺, or general osmotic stress induced by mannitol. Wild-type (•) and sos3 (○) seedlings were grown for 4 days on vertical agar plates containing MS nutrients and then transferred to vertical agar plates containing MS supplemented with NaCl, KCl, LiCl, CsCl, or mannitol. Root growth was measured 6 days after the transfer. Growth on MS nutrients was considered as 100%. Values are the averages of 15 seedlings.

Potassium Requirement of sos3.

Because the salt-hypersensitive phenotype of sos1 mutants has been found to cosegregate with the inability to grow on low K⁺ culture medium (10), we tested whether sos3 is similarly affected. Although the sos3 mutant seedlings grew normally on medium containing 20 mM K⁺, they failed to grow on a medium containing 20 μM K⁺ (Fig. 1). In contrast, wild type grew well on both 20 μM and 20 mM K⁺ media (Fig. 1). Forty-five sos3 seedlings were selected from a segregating F2 population based on their NaCl sensitivity. These seedlings all failed to grow on 20 μM K⁺, indicating that the inability of sos3 seedlings to grow on low K⁺ cosegregates with their salt-hypersensitive phenotype. The results suggest an essential role of the SOS3 gene in potassium nutrition under limiting K⁺ availability.

To determine whether sos3 seedlings require more or less K⁺ than sos1-1, their root growth was measured over a wide range of K⁺ levels. As shown in Fig. 4A, while sos1-1 required 20 mM K⁺ for significant root growth, ∼1 mM was sufficient for sos3. For wild-type seedlings, only 0.1 mM K⁺ was necessary. Both sos3 and the wild type had maximal root growth at 10 mM K⁺. K⁺ levels above 10 mM inhibited sos3 and wild-type root growth, whereas sos1 root growth was improved by these high levels of K⁺.

Figure 4.

High Ca²⁺ restores the growth of sos3 on low-K⁺ growth medium. (A) Optimal growth of wild type (•), sos1 (○), and sos3 (▪) seedlings requires different external K⁺ concentrations in the media. (B) Ca²⁺ dose-response curve of wild type (•), sos1 (○), and sos3 (▪) seedlings. The medium contained 20 μM K⁺. Media were modified from MS (see Materials and Methods). Growth measurement was similar to that in Fig. 3 except that the measurement was taken 3 days after the seedling transfer to minimize the effect of residual K⁺ and Ca²⁺ carried by the plants from MS medium. [Bars = SD (n = 15).]

Calcium Requirement of sos3.

Surprisingly, we found that the growth defect of sos3 seedlings on low potassium growth medium could be corrected by raising the external Ca²⁺ concentration to >2 mM (Figs. 1D and 4B). Calcium also stimulated root growth of wild-type plants. However, the wild type required only 125 μM Ca²⁺ for maximal growth, whereas 1.25 mM Ca²⁺ was not enough for sos3 mutant plants (Fig. 4B). In comparison, up to 10 mM Ca²⁺ did not significantly increase the growth of sos1 plants on low potassium growth medium (Fig. 4B).

We determined whether increased Ca²⁺ could similarly correct the growth defect of sos3 on medium containing high levels of NaCl. Indeed, Ca²⁺ substantially increased the salt tolerance of sos3 plants (Fig. 5). On medium containing 50 mM NaCl, root growth of sos3 was completely inhibited when no Ca²⁺ was supplemented. However, 1.5 mM or higher Ca²⁺ resulted in significant growth of sos3 plants (Fig. 5). Although elevated Ca²⁺ also increased the salt tolerance of the wild type, the effect was not as great as for sos3 mutant plants (Fig. 5). The wild type showed significant growth even without any added Ca²⁺. Although <1.5 mM Ca²⁺ reduced the survival rate of sos1 plants, high Ca²⁺ was not able to increase the growth of sos1 on 50 mM NaCl (Fig. 5). Increased Ca²⁺ did not substantially increase the growth of sos3, sos1, or the wild-type plants on 100 mM NaCl (data not shown). Perhaps at 100 mM NaCl, the ameliorating effect of Ca²⁺ was simply swamped out.

Figure 5.

Effect of Ca²⁺ on salt tolerance of wild-type (•), sos1 (▪, □), and sos3 (○) seedlings. ▪, sos1 seedlings that were killed by the treatments. Four-day-old seedlings grown on MS plates were transferred to grow on nutrient media supplemented with 50 mM NaCl and various levels of Ca²⁺. The nutrient media were the same as for MS except that Ca²⁺ was removed. Growth measurement was similar to that in Fig. 3. [Bars = SD (n = 15).]

Ca²⁺ Improves K⁺/Na⁺ Selectivity of sos3 Seedlings.

It has been suggested that high external Ca²⁺ mitigates NaCl effect by improving the K⁺:Na⁺ ratio of plant cells (16). To determine whether high Ca²⁺ rescues sos3 growth under NaCl stress through such a mechanism, K⁺ and Na⁺ contents were measured in 50 mM NaCl-stressed sos3 seedlings under high or low external Ca²⁺. Similar to sos1-1, sos3 seedlings exhibited K⁺ deficiency under NaCl stress (Table 2). However, sos3 seedlings accumulated more Na⁺ than the wild type. In comparison, NaCl-stressed sos1-1 accumulated less Na⁺ than the wild type (23). There was virtually no difference between the K⁺ or Na⁺ content in sos3 and wild-type seedlings that were grown on MS medium alone without NaCl (data not shown).

Table 2.

Effects of external Ca²⁺ levels on K⁺/Na⁺ selectivity of sos3 and wild-type seedlings treated with 50 mM NaCl

Ca2+ level	K+ content, mg/g dry wt % increase		Na+ content, mg/g dry wt % decrease		K+/Na+ ratio % increase
	Low*	High†	Low*	High†	Low*	High†
WT	53.4 ± 2.0	60.9 ± 0.4 (+14%)	29.8 ± 2.2	27.0 ± 1.3 (−9%)	1.8	2.26 (+26%)
sos3	36.5 ± 0.7	47.6 ± 2.0 (+30%)	51.8 ± 1.8	43.4 ± 3.0 (−16%)	0.7	1.10 (+57%)

Data represent mean ± SE (n = 3). 

*Low Ca²⁺ medium contains 0.3 mM CaCl₂. 

^†High Ca²⁺ medium contains 2 mM CaCl₂. 

High external Ca²⁺ increased cellular K⁺ content in both sos3 and wild- type plants (Table 2). This Ca²⁺ effect was more than twice as large in sos3 (30%) than in the wild type (14%). High Ca²⁺ also reduced Na⁺ content in both sos3 and the wild type. Again, this effect was more pronounced in sos3 plants. In summary, high Ca²⁺ increased the K⁺:Na⁺ ratio by 57% and 26% in sos3 and the wild type, respectively. The data suggest that the Ca²⁺ effect observed on sos3 may be mediated at least partially through the improvement of potassium nutrition and K⁺/Na⁺ selectivity.

sos1 Is Epistatic to sos3.

The sos3 and sos1 mutants exhibit similar phenotypes in that they are hypersensitive to Na⁺ and Li⁺ and not capable of growing on low K⁺ media. Thus, the two SOS genes may function in the same pathway that regulates K⁺ nutrition and salt tolerance. To determine the epistatic relationship between sos3 and sos1, we constructed sos1sos3 double mutant. Thirty sos mutants were selected from the selfed F2 progeny of a cross between sos3sos3 and sos1-1sos1-1. To identify a double mutant, these sos mutants were each test crossed to sos3sos3 and sos1-1sos1-1. One sos1sos3 double mutant line was identified because the F1 progenies from both of the test crosses showed a NaCl hypersensitive phenotype.

The response of sos1sos3 to NaCl is similar to that of sos1 (Fig. 6A). Without NaCl, the growth rate of sos1sos3 seedlings was similar to that of sos1. With 10 or 25 mM NaCl, growth of sos1sos3 seedlings was reduced to similar levels as sos1. The response of sos1sos3 seedlings to Ca²⁺ is also the same as that of sos1. As shown in Fig. 6B, the growth of sos3 on low K⁺ culture medium was restored by high Ca²⁺, whereas high Ca²⁺ had little effect on the growth of sos1 or sos1sos3. Therefore, increased Ca²⁺ suppressed sos3 but not sos1 or sos1sos3 mutations. Taken together, these results suggest that sos1 is epistatic to sos3.

Figure 6.

sos1sos3 double mutant resembles sos1 both in salt sensitivity and in response to Ca²⁺. (A) Growth of wild-type (□), sos1 (▪), sos3 (░⃞), and sos1sos3 (▥) seedlings on MS medium, MS medium supplemented with extra 10 or 25 mM NaCl. (B) Growth of wild-type (□), sos1 (▪), sos3 (▥), and sos1sos3 (▥) seedlings on MS (20 mM K⁺, 3 mM Ca²⁺), low K⁺–low Ca²⁺ (20 μM K⁺, 0.15 mM Ca²⁺), or low K⁺–high Ca²⁺ (20 μM K⁺, 3 mM Ca²⁺) media. Growth measurement was conducted on the seventh day after the transfer. For Ca²⁺-responsive treatment (B), the measurement was taken 3 days after seedling transfer to minimize the effect of residual K⁺ and Ca²⁺ carried by the plants from MS medium. [Bars = SD (n = 15).]

DISCUSSION

The sos3 mutation defines a genetic locus that is essential for salt tolerance as well as growth on low-K⁺ culture media. It is surprising that increased external Ca²⁺ could completely suppress the growth defect on low-K⁺ culture medium and partially suppress the salt-hypersensitive phenotype of the sos3 mutant. The sos3 mutation thus uncovers a fundamental role of calcium in regulating potassium nutrition and salt tolerance in plant roots. A beneficial effect of external Ca²⁺ on plant salt tolerance was reported decades ago (16, 17). Despite of intensive efforts, the underlying mechanisms of this effect remain unclear. The identification of the sos3 mutation provides an excellent opportunity to resolve the molecular nature of this important calcium response.

The calcium effect is specific to sos3 and was not observed for sos1 (Figs. 4B and 5). Because sos1 and sos3 exhibit similar phenotypes, the two genes probably function in a common pathway that regulates potassium nutrition and salt tolerance. Genetic evidence suggests that sos1 is epistatic to sos3. Thus, SOS1 may encode a component upstream of SOS3.

The failure of sos3 seedlings to grow on low-K⁺ culture medium suggests that the mutant may be defective in high-affinity K⁺ uptake. We measured K⁺ uptake (10) in sos3 seedlings or excised seedling roots over a wide range of external K⁺ concentrations with low or high Ca²⁺. No substantial difference in K⁺ uptake between sos3 and the wild type was found (data not shown). This could be a result of technical limitations of the uptake assay if the SOS3 gene controls a K⁺ uptake system with an extremely high affinity (e.g., submicromolar range) or if the altered uptake is restricted to particular cells or tissues responsible for the growth defect. Alternatively, the sos3 mutation may affect K⁺ utilization and does not affect K⁺ uptake. We also found that there is no substantial difference in K⁺ efflux between sos3 and wild-type seedlings (data not shown).

The sos3 mutation clearly causes a decrease in K⁺ content and K⁺/Na⁺ selectivity in NaCl-stressed seedlings. High external Ca²⁺ greatly improves tissue K⁺ content and K⁺/Na⁺ selectivity of sos3 plants. This may explain the observed partial Ca²⁺ suppression of NaCl hypersensitivity of sos3. The SOS3 gene could encode a calcium-sensing protein such as calmodulin, calcium-dependent–calmodulin-independent protein kinase or calcium/calmodulin-dependent protein phosphatase (i.e., calcineurin) (24, 25). Calmodulin is known to mediate calcium regulation of potassium channels in Paramecium (26). Of particular relevance is that, in higher plants, it has been shown that the activation of a vacuolar K⁺ channel requires 10 times more Ca²⁺ without calmodulin than in the presence of calmodulin (27). Calcineurin is an essential component for salt tolerance in yeast by controlling sodium efflux and potassium uptake (28). Data suggest that in NaCl-stressed yeast cells, calcineurin functions to switch off low-affinity potassium uptake and turn on the high-affinity potassium uptake system thereby increase K⁺/Na⁺ selectivity (28). Pharmacologic evidence indicates that a calcineurin homolog is also present in plant cells and mediates the calcium inhibition of inward-rectifying K⁺ channels in guard cells (29). However, the molecular nature of calcineurin remains elusive in plants.

Alternatively, the observed Ca²⁺ effect could be mediated through an extracellular target. Aside from abundant Ca²⁺ ligands in the pectic component of the cell wall and lipids of the plasma membrane, no Ca²⁺-binding protein is known in the wall or the outer surface of the plasma membrane. A general beneficial effect of external Ca²⁺ on plant salt tolerance is often attributed to the notion that Ca²⁺ is necessary for maintaining plasma membrane integrity and improving K⁺/Na⁺ selectivity of potassium uptake systems (16). The general beneficial effect of Ca²⁺ was also observed here for wild type and sos1 plants (Fig. 5). However, the effect on sos3 is much more dramatic and specific. If the Ca²⁺ effect reported here is extracellular, then there could be a specific cell wall or plasma membrane calcium sensor controlling potassium acquisition. Extracellular Ca²⁺-sensing receptors that respond to millimolar concentrations of Ca²⁺ have been found in mouse keratinocyte as well as bovine parathyroid (30, 31). Whatever Ca²⁺ sensor SOS3 may encode, the sos3 mutation appears to increase the threshold of Ca²⁺ activation of the sensor or its downstream target.

Yet another possibility is that sos3 is defective in general cellular Ca²⁺ homeostasis as a result of impaired uptake or insufficient intracellular pools of stored Ca²⁺. This appears unlikely because defective cellular Ca²⁺ homeostasis would be expected to cause pleiotropic phenotypes, since calcium is thought to be required for numerous signaling reactions critical for normal plant growth and development (24, 32). sos3 plants are normal except when challenged with low potassium or salt stress (Fig. 1). We have also measured Ca²⁺ uptake in sos3 and wild-type seedlings using ⁴⁵Ca²⁺. No difference was detected between sos3 and the wild type (data not shown). There was also no significant difference in Ca²⁺ content between NaCl-stressed sos3 and wild-type seedlings (data not shown). Cloning of the SOS3 gene is expected to provide insights into the structure and function of the gene product. We have taken a big step toward the positional cloning of SOS3 by placing the mutation within a distance of probably <100 kb from the physical marker nga139 (Fig. 2).