INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The sensitivity of crop plants toward salinity is one of the major factors causing agricultural losses in arid regions (Greenway and Munns, 1980; Maas 1986). Studies analyzing the mechanisms of salinity have shown specific toxic effects of Na⁺ ions (Kingsbury and Epstein, 1986). Under saline conditions, the large electrochemical Na⁺ gradient results in passive Na⁺ uptake into root cells (Smith and Walker, 1989; Allen et al., 1995; Tyerman and Skerrett, 1999). Long-term Na⁺ influx by Na⁺-permeable channels/transporters can elevate the cytoplasmic Na⁺ concentration to toxic levels and trigger a variety of detrimental cellular effects (for review, see Volkmar et al., 1999). Kinetic analyses of low-affinity Na⁺ uptake show multiple phases with distinguishable low-affinity components, leading to the suggestion that multiple low-affinity Na⁺ uptake pathways exist in roots (Epstein and Rains, 1965).

Several ion transport and channel currents have been described in wheat (Triticum aestivum L.) root cells, including an outward-rectifying K⁺ channel current (I_K⁺_out) (Schachtman et al., 1991), inward-rectifying K⁺ channel currents (I_K⁺_in) (Findlay et al., 1994; Gassmann and Schroeder, 1994), a spiky inward K⁺ channel (Findlay et al., 1994), an aluminum-activated anion channel current (Ryan et al., 1997), and a Na⁺-dependent instantaneous current (Tyerman et al., 1997). Only the spiky K⁺ current and the Na⁺-dependent current were shown to carry Na⁺ uptake currents (Tyerman et al., 1997). In contrast to animal cells, no specific Na⁺ channel has been isolated in plants so far.

Inward Na⁺ channel in root cells have also been reported from other species (for review, see Amtmann and Sanders, 1998; Tyerman and Skerrett, 1999). Several non-selective cation channels from rye root plasma membrane have been characterized in lipid bilayers, and one of these might be responsible for instantaneous currents measured in protoplasts of epidermal rye root cells (White and Lemtiri-Chlieh, 1995; White 1997). A Na⁺-permeable channel was characterized in protoplasts of maize root cortical cells (Roberts and Tester, 1997). In barley suspension-cultured cells, instantaneously activating inward-rectifying currents that were permeable to Na⁺ and K⁺ were identified (Amtmann et al., 1997). From guard cells of two Aster species, non-rectifying cation channels that possess similar characteristics as the Na⁺-dependent inward currents from roots have been reported (Véry et al., 1998).

Several earlier studies reported that K⁺ starvation induces increased low-affinity Na⁺ uptake (Pitman, 1967; Pitman et al., 1968; Ding and Zhu, 1997) and low-affinity K⁺ uptake (Pitman et al., 1968; Benlloch et al., 1989; Maathuis and Sanders, 1995). In addition, K⁺ starvation has been shown to reduce the selectivity of root membranes for K⁺ over Na⁺ (Pitman et al., 1968; Kochian et al., 1985). Enhancement of mRNA levels by K⁺ starvation was also found for the high-affinity K⁺ transporters HKT1 in barley, wheat, and rice (Golldack et al., 1997; Wang et al., 1998), HvHAK1 in barley (Santa-Maria et al., 1997), and AtKUP3 in Arabidopsis (Kim et al., 1998). Interestingly, these K⁺ transporters have been shown to mediate low-affinity Na⁺ uptake under saline conditions (Rubio et al., 1995; Gassmann et al., 1996; Santa-Maria et al., 1997; Fu and Luan, 1998).

The characterization of K⁺ channels proposed to contribute to low-affinity K⁺ uptake is far more advanced than for the pathways of low-affinity Na⁺ uptake. Plant inward-rectifying K⁺ channels have been cloned from Arabidopsis and potato (Sentenac et al., 1992; for review, see Schroeder et al., 1994; Fox and Guerinot, 1998; Czempinski et al., 1999), but so far no inward-rectifying K⁺ channel genes have been characterized from wheat.

In the present study we addressed the question of whether Na⁺ and K⁺ uptake currents in wheat root cells are affected by external K⁺ supply during growth. We found that instantaneous Na⁺ currents and time-dependent I_K⁺_in are enhanced by K⁺ starvation. In addition, we observed a correlation between induced I_K⁺_in and the induction of an isolated wheat root K⁺ channel gene.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Isolation of the TaAKT1 cDNA

A wheat (Triticum aestivum L.) root-specific cDNA library constructed from wheat plants grown under K⁺ starvation conditions (Schachtman and Schroeder, 1994) was used as template DNA for PCR using the following primers: 5'-TG GAA TTC ATG YTI MGI YTI TGG MG-3' as the forward primer and 5'-ATR ACC WSI TAI TGI TG CCT AGG AA-3' as the reverse primer where M represents A or C; R: A or G; S: C or G; W: A or T; Y: C or T; and I: inosine triphosphate. The resulting 270-bp PCR fragment, which was cloned and sequenced, shared homology to plant K⁺ channels. With this PCR product as a probe, the wheat root cDNA library was screened to identify 16 putative clones. The cDNA with the largest insert size, named TaAKT1, was used for further studies. The TaAKT1 cDNA was sequenced on both strands with an automatic sequencer (ABI, Sunnyvale, CA). Sequence analysis was performed with Lasergene 99 software from DNASTAR (Madison, WI). Homologous sequences were identified by searching within the GenBank database using BLAST (Altschul et al., 1990). The amino acid sequence was analyzed with the SMART program (Schultz et al., 1998). Putative transmembrane spans were predicted by the TopPred2 program (von Heijne, 1992).

RNA Isolation and Reverse Transcriptase (RT)-PCR Analysis of TaAKT1 Expression

Total RNA was isolated from wheat roots grown hydroponically in either 1 mM CaCl₂ and 0 mM KCl or 1 mM CaCl₂ and 5 mM KCl, as described below. Total RNA was quantitated spectrophotometrically. Dilutions of the RNA were electrophoresed on an RNA formaldehyde gel, and the intensity of the rRNA bands was compared to confirm that equal quantities of RNA were taken for first-strand cDNA syntheses. First-strand cDNA was synthesized from 5 µg of total RNA with a first-strand cDNA synthesis kit (Amersham-Pharmacia Biotech, Uppsala), diluted 1:10, and used as the template for the PCR reaction. For competitive PCR analyses, a PCR fragment was amplified from wheat genomic DNA using the same primer pair as for the RT-PCR. Due to the presence of introns, this fragment was 900 bp longer than the fragment amplified from the cDNA. Ten to 600 fg of competitive DNA was added to the PCR reactions, and 30 cycles of 94°C, 30 s; 60°C, 30 s; and 72°C, 1 min were performed. PCR products were analyzed by agarose gel electrophoresis. To confirm the results, total RNA was isolated from another set of plants grown under the same conditions and the RT-PCR experiment was repeated.

Growth of Plants and Protoplast Preparation

Wheat (cv Atlas 66) seeds were surface-sterilized with 0.5% (w/v) NaOCl, germinated on wetted filter paper for 4 d in the dark, and then transferred to hydroponic solutions containing either 1 mM CaCl₂ and 0 mM KCl or 1 mM CaCl₂ and 5 mM KCl. After 5 to 7 d, protoplasts were isolated as described by Schachtman et al. (1991). About 1 g of root tissue was chopped, washed, and then enzymatically digested in 0.8% (w/v) cellulase (Onozuka RS, Yakult Honsha, Tokyo) and 0.08% (w/v) pectolyase (Sigma Chemical Co., St. Louis) for 3 to 4 h. Root protoplasts were collected by a Suc density gradient. Large protoplasts with visible cytoplasmic streaming were selected for patch clamping. Previous studies have shown that protoplasts greater than 30 µm in diameter originate predominantly from the root cortex (Tyerman et al., 1997).

Electrophysiology

Patch-clamp pipettes were pulled from glass capillaries (Kimax 51, Kimble, Toledo, OH) on a multistage puller (P-87, Sutter Instrument, Novato, CA) and fire-polished to a tip resistance of 5 to 10 M. Whole-cell currents across wheat root protoplasts were measured using the patch-clamp technique (Hamill et al., 1981) with an amplifier (Axopatch-1D, Axon Instruments, Foster City, CA). pClamp software (Axon Instruments) was used for the generation of voltage pulses, data recording, data storage, and data analysis. Patching of protoplasts was performed in a flow chamber of about 0.2-mL volume. To minimize offset potentials, agar salt bridges were used for the reference electrode. Ionic activities were calculated and corrected using GEOCHEM software (D.R. Parker, University of California, Riverside). Liquid junction potentials were calculated and corrected using JPCalc software (Barry, 1994). Data are presented as means ± SE, and the statistical significance of differences between currents was determined by the Student's t test.

Experimental Solutions

The standard intracellular solution used in patch-clamp experiments was composed of: 30 mM KCl, 70 mM K-Glu, 2 mM MgCl₂, 6.7 mM EGTA, 3.35 mM CaCl₂, 5 mM MgATP, and 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)/Tris, pH 7.1. The standard bath solutions contained: 100 mM KCl, 10 mM CaCl₂, 2 mM MgCl₂, 5 mM 2-(N-morpholino)-ethanesulfonic acid (MES)/Tris, pH 5.6 (standard K⁺ solution), or 50 mM Na₂SO₄, 2 mM MgCl₂, 0.05 mM CaCl₂, and 5 mM MES/Tris, pH 5.6 (standard Na⁺ solution).

In addition, low-Cl intracellular and extracellular solutions were used in some experiments to exclude possible Cl efflux currents. These solutions included: (a) low-Cl intracellular solution: 100 mM K-Glu, 2 mM MgCl₂, 2 mM EGTA, 5 mM MgATP, and 5 mM HEPES/BTP, pH 7.1; (b) K-Glu extracellular solution: 100 mM K-Glu, 10 mM CaCl₂, 2 mM MgCl₂, and 10 mM MES/Tris, pH 5.8; and (c) Na-Glu extracellular solution: 100 mM Na-Glu, 2 mM MgCl₂, 0.05 mM CaCl₂, 10 mM MES/Tris, pH 5.8. For selectivity examinations, the following solutions were used: 100 mM LiCl, 100 mM RbCl, or 100 mM CsCl, plus 2 mM MgCl₂, 0.05 mM CaCl₂, and 10 mM MES/Tris, pH 5.8. Sorbitol was used to adjust the osmolality of the bath solutions to 625 mOsmol/kg and the pipette solutions to 650 mOsmol/kg.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Enhancement of I_K⁺_in by K⁺ Starvation

In previous studies on wheat root cortex protoplasts from plants grown with regular K⁺ supply (Schachtman et al., 1991; Findlay et al., 1994; Tyerman et al., 1997), I_K⁺_out were more frequently observed (79%) than I_K⁺_in (23%). We examined K⁺ currents in wheat roots grown under K⁺ starvation conditions (hydroponics with 1 mM CaCl₂ and 0 mM KCl) and found I_K⁺_in in 91% of the protoplasts (Fig. 1A) and I_K⁺_out in only 45% (n = 101 total protoplasts). Patch-clamp experiments in which the extracellular K⁺ concentration was shifted from 100 to 30 mM followed by current reversal ("tail") analyses showed that I_K⁺_in were carried by K⁺ (data not shown) and had similar properties to the I_K⁺_in previously described by Findlay et al. (1994). The increased frequency of I_K⁺_in under K⁺-starved conditions suggested an effect of the growth conditions on the K⁺ channel activity. We tested different K⁺-supplemented growth solutions with regard to their effect on I_K⁺_in. In general, we found a reduction of I_K⁺_in occurrences and magnitudes for wheat root protoplasts grown with K⁺. I_K⁺_out were observed in 79% of the protoplasts and I_K⁺_in in only 54% of the measured root protoplasts grown with 1 to 5 mM K⁺ (n = 56 total protoplasts). The average magnitude of I_K⁺_in from K⁺-starved root protoplasts was higher (19.96 ± 1.60 pA/pF at 148 mV, n = 18) than the average magnitude of I_K⁺_in from K⁺-pretreated root protoplasts (9.85 ± 1.39 pA/pF at 148 mV, n = 11; only protoplasts with I_K⁺_in were included for this analysis). The effect of adding 5 mM K⁺ to the growth medium, enzyme solution, and protoplast storage solution is shown in Figure 1, A through C (Fig. 1C, P < 0.001 at 148 mV). Interestingly, for 6-d-old wheat seedlings grown in 1 mM CaCl₂ and 5 mM KCl, the root/shoot length ratio was about one-half the value of seedlings grown in 1 mM CaCl₂ and 0 mM KCl.

View larger version (15K): [in this window] [in a new window]   Figure 1.   Whole-cell K+in currents in the plasma membrane of wheat root cortex protoplasts are enhanced by K+ starvation during growth. A, Whole-cell currents from a wheat root cortical cell grown 6 d in 1 mM CaCl2 and 0 mM KCl (capacitance = 30 pF). B, Whole-cell currents from a wheat root cortical cell grown for 6 d in 1 mM CaCl2 and 5 mM KCl (capacitance = 34 pF). C, Average current-voltage relationships of whole-cell currents from wheat root cortical cells grown for 6 d in 1 mM CaCl2 and 0 mM KCl (, n = 11) or in 1 mM CaCl2 and 5 mM KCl (, n = 8). Means ± SE of time-dependent currents obtained by voltage pulses from +60 to 160 mV in 20 mV steps are shown in C. The holding potential was 30 mV. A standard K+ bath solution and a standard pipette solution were used.

Further control experiments were performed with a low-Cl pipette solution (see "Material and Methods") to exclude the possibility that inward-rectifying anion currents (Ryan et al., 1997) contributed to I_K⁺_in. We could not observe a significant difference in the kinetics (e.g. Fig. 6A) or current density of I_K⁺_in in response to the changed pipette solution (standard pipette solution: 19.4 ± 1.6 pA/pF at 148 mV, n = 11; low-Cl pipette solution: 19.6 ± 2.8 pA/pF at 148 mV, n = 8).

Cloning of the TaAKT1 Gene and K⁺-Dependent mRNA Levels

The inward-rectifying K⁺ channel AKT1 plays a role in K⁺ uptake from roots, as evidenced by its preferential localization in roots (Lagarde et al., 1996) and reduced root growth in Arabidopsis mutants disrupted in the AKT1 gene under low-K⁺ and high-NH₄⁺ conditions (Hirsch et al., 1998). However, there was no induction of AKT1 RNA levels by low-K⁺ conditions in Brassica napus roots (Lagarde et al., 1996). To examine whether a K⁺ uptake channel gene is induced by K⁺ starvation in wheat, we first cloned a wheat homolog of the Arabidopsis AKT1 channel gene. Degenerate PCR primers were based on the sequence MLRLWR in the S4 domain for the forward primer and on the sequence YWSITT in the pore domain of K⁺_in channels for the reverse primer. These primers were used to amplify a fragment from a K⁺-starved wheat root cDNA library (Schachtman and Schroeder, 1994) that showed homology to the corresponding region of AKT1. The PCR product was used as a probe to screen the wheat root cDNA library. A full-length cDNA of 3,300 bp (accession no. AF207745), encoding a deduced polypeptide of 897 amino acids, having a predicted molecular mass of 100.9 kD, and showing 76% similarity to the Arabidopsis AKT1 sequence (Sentenac et al., 1992), 84% similarity to a K⁺ channel cDNA from maize (accession no. Y07632), and 77% to SKT1 from potato (Zimmermann et al., 1998) at the amino acid level, was isolated. The deduced polypeptide exhibited all of the structural features that are shared by the plant inward-rectifying K⁺ channels: six transmembrane domains (S1-S6) (Uozumi et al., 1998), a K⁺-selective pore-forming domain (P) between S5 and S6, a putative cyclic nucleotide-binding domain, and five ankyrin repeat sequences that are present only in the AKT subfamily of plant inward-rectifying K⁺ channels (Sentenac et al., 1992). We therefore designated this gene TaAKT1 (T. aestivum AKT1-like) (Fig. 2).

View larger version (50K): [in this window] [in a new window]   Figure 2.   Predicted amino acid sequence of TaAKT1. Transmembrane segments S1 to S6, pore region P, and the putative cyclic nucleotide binding site cNMP are underlined and labeled below the respective sequence; ankyrin-like repeats A1 to A5 are boxed. The protein sequence on which degenerate PCRs were based is boxed.

Since standard northern-blot experiments did not give us a clear signal with the TaAKT1 probe, we performed RT-PCR experiments with TaAKT1-specific primers to determine whether the TaAKT1 gene was induced by K⁺ starvation in wheat roots. Fragments of the expected size were detected in both K⁺-starved (Fig. 3B) and 5 mM K⁺-grown roots (Fig. 3A). Figure 3C shows the amount of RNA used for the first-strand cDNA synthesis. To analyze the level of induction of the TaAKT1 transcript, we used the wheat cDNA together with different amounts of competitor DNA in PCR reactions. The comparison of the band intensity indicates a 3- to 10-fold induction in the TaAKT1 message in replicate experiments in wheat roots grown at K⁺-starved conditions (Fig. 3).

View larger version (66K): [in this window] [in a new window]   Figure 3.   Enhancement of TaAKT1 mRNA levels in response to K+ deprivation. Total RNA extracted from wheat roots grown in 1 mM CaCl2 and 5 mM KCl (+K+ in A) or in 1 mM CaCl2 and 0 mM KCl (K+ in B) was used for first-strand cDNA synthesis and subsequent PCR reactions. C shows the quantity of total RNA from K+-starved (K+) and K+-grown (+K+) wheat roots used for first-strand cDNA synthesis. Total RNA electrophoresed on a formaldehyde gel was stained by ethidium bromide. For the competitor DNA, a PCR-amplified fragment isolated from the genomic DNA was used. This fragment was approximately 900 bp longer than the cDNA-amplified fragment because of introns. The indicated amounts of competitor DNA were added to the PCR reactions. Aliquots were analyzed by agarose gel electrophoresis.

In our studies, K⁺ starvation induced an increased expression of TaAKT1, and patch-clamp studies also showed an increased K⁺_in current in K⁺-starved wheat roots. AKT1 has been shown to function as a K⁺_in channel in yeast (Bertl et al., 1997) and in insect cells (Gaymard et al., 1996). Therefore, we propose that the TaAKT1 channel might contribute to induced I_K⁺_in in wheat root protoplasts.

Enhancement of I_Na⁺ by K⁺ Starvation

After exchanging the K⁺ bath solution with a Na⁺-containing bath solution (100 mM Na⁺), we found at low extracellular Ca²⁺ concentrations (0.05 mM) instantaneous inward-directed currents in root cortex protoplasts from K⁺-starved wheat seedlings (Fig. 4; n = 46 protoplasts). When roots and root protoplasts were pretreated with K⁺, only reduced instantaneous currents were observed (n = 21 protoplasts). Adding 5 mM KCl to the growth medium, enzyme solution, and protoplast storage solution caused about a 3-fold decrease of the instantaneous currents (Fig. 4C; P < 0.01 at 147 mV). On the other hand, increasing the bath Na⁺ concentration from 100 to 180 mM Na⁺ increased the current and shifted the whole-cell reversal potential to more positive values (Fig. 4, C and D, P < 0.05 at 147 mV). Control measurements with low-Cl pipette solutions were performed to exclude effects of inward-rectifying anion currents (n = 9). The time dependence of activation of the Na⁺-dependent currents (e.g. Fig. 6B) and the current density were not affected by Cl ions (standard pipette solution: 4.6 ± 0.6 pA/pF at 147 mV, n = 11; low-Cl pipette solution: 4.3 ± 1.2 pA/pF at 147 mV, n = 9). We examined the selectivity of the instantaneous currents with respect to monovalent cations. For Li⁺ and Cs⁺ ions (100 mM), we found only small background currents at hyperpolarized membrane potentials (Fig. 5, A and C). When protoplasts were extracellularly exposed to 100 mM Rb⁺, we observed small time-dependent currents that might have been mainly mediated by I_K⁺_in (data not shown). The positive shifts in reversal potentials upon increasing extracellular Na⁺ concentration, the Cl independence, and the reduction of clearly resolved currents in the presence of Li⁺ or Cs⁺ show that the observed instantaneous currents were mainly carried by Na⁺ ions which were named I_Na⁺.

View larger version (27K): [in this window] [in a new window]   Figure 4.   K+ deprivation induces instantaneous Na+ currents measured across the plasma membrane of protoplasts isolated from the cortex of wheat roots. A, Whole-cell currents from a K+-starved wheat root cortex cell measured in a bath solution with 50 mM Na2SO4, 2 mM MgCl2, and 0.05 CaCl2 (capacitance = 36pF). Wheat seedlings were grown 6 d in 1 mM CaCl2. B, Whole-cell currents from a 5 mM K+-grown wheat root cortex cell measured in a bath solution with 50 mM Na2SO4, 2 mM MgCl2, and 0.05 CaCl2 (capacitance = 52pF). Wheat seedlings were grown 6 d in 1 mM CaCl2 and 5 mM KCl. C, Average current-voltage relationships of whole-cell currents from wheat root cortex cells measured as in A and B. Wheat seedlings were grown for 6 d in 1 mM CaCl2 and 0 mM KCl (, n = 11) or in 1 mM CaCl2 and 5 mM KCl (, n = 11). D, Average current-voltage relationships of whole-cell currents from wheat root cortex cells measured in bath solution with 90 mM Na2SO4, 2 mM MgCl2, and 0.05 CaCl2. Wheat seedlings were grown for 6 d in 1 mM CaCl2 and 0 mM KCl (, n = 9) or in 1 mM CaCl2 and 5 mM KCl (, n = 6). Holding potential was 30 mV. Voltage pulses from +40 to 140 mV in 20 mV steps were applied.

View larger version (12K): [in this window] [in a new window]   Figure 5.   Alkali cation selectivity of instantaneous whole-cell inward currents from wheat root protoplasts under low external Ca2+ conditions (0.05 mM CaCl2). A, Whole-cell currents from wheat root protoplasts measured with 100 mM LiCl and 0.05 mM CaCl2 in the bath solution and using standard pipette solution (C = 65pF). B, Whole-cell currents from wheat root protoplasts measured with 100 mM Na-Glu and 0.05 mM CaCl2 in the bath solution and standard pipette solution (same protoplast as in 5A). C, Average steady-state current-voltage relationships of whole-cell currents from wheat root cortex cells measured in bath solutions with 100 mM Na+ (, n = 11), Cs+ (×, n = 6), and Li+ (, n = 5). Holding potential was 30 mV. Voltage pulses from 0 to 140 mV in 20 mV steps were applied. Wheat seedlings were grown for 6 d in 1 mM CaCl2.

We pursued experiments to determine whether I_Na⁺ was related to I_K⁺_in. The time dependencies of activation and deactivation of I_K⁺_in and I_Na⁺ were found to be clearly different. I_K⁺_in showed a time-dependent activation characteristic for K⁺_in channels, whereas I_Na⁺ activated with a major rapid component (Fig. 6, B and C). In addition, the activation curve of I_K⁺_in could be well described by a Boltzmann curve, as reported for other I_K⁺_in in roots (e.g. White, 1997). In contrast, the conductance-voltage relationship for I_Na⁺ was nearly linear (Fig. 6D). A similar conductance-voltage relationship for Na⁺-dependent currents has been reported in maize roots (Roberts and Tester, 1997). In addition, we found, in agreement with other studies (Tyerman et al., 1997), that recordings of I_Na⁺ were possible in protoplasts without I_K⁺_in (n = 8). Furthermore, I_K⁺_in, but not I_Na⁺, could be blocked by addition of 10 mM CsCl to the bath solution. Thus, I_Na⁺ shared several biophysical properties that could be clearly distinguished from I_K⁺_in, showing that different proteins account for the two currents.

View larger version (25K): [in this window] [in a new window]   Figure 6.   Comparison of whole-cell K+- and Na+-dependent inward currents from wheat root cortex cells. A, Whole-cell currents from a wheat root protoplast measured with 100 mM K+ bath solution low-Cl pipette solution (C = 65 pF). Voltage pulses from +20 to 160 mV in 20 mV steps were applied. B, Whole-cell currents from a wheat root protoplasts measured with Na-Glu bath solution and low-Cl pipette solution (C = 31 pF). Voltage pulses from +40 to 140 mV in 20 mV steps were applied. C, Average fast time constants of activation for IK+in (, n = 6) and INa+ (, n = 5). D, Average steady-state chord conductance as a function of membrane potential. The chord conductance density was calculated as g = I/(EM  EX), where I is the steady-state inward current density, EM the clamped membrane potential, and EX the reversal potential of the inward currents. A Boltzmann equation of the form g = gmax/(1+exp[{EM  E1/2}/{RT/zgF}]) was fitted to the data, where gmax is the maximal chord-conductance density, E1/2 is the voltage at which the chord conductance density is half-maximal, R, T, and F have their usual thermodynamic meanings, and zg is the valency of the gating charge. Fitting was done by a simplex method. For IK+in we obtained gmax = 171.1 pS/pF, E1/2 = 132.5 mV, RT/zgF = 14.4 mV (zg = 1.75). Holding potential was 30 mV. Wheat seedlings were grown for 6 d in 1 mM CaCl2.

Ca²⁺ is known to have ameliorative effects on plants under high-salinity conditions (LaHaye and Epstein, 1969) and to reduce low-affinity Na⁺ uptake (Rengel, 1992). Therefore, we tested the effect of increased external Ca²⁺ on I_Na⁺. Protoplasts recorded in the presence of 4 mM Ca²⁺ (Fig. 7A) showed reduced I_Na⁺ in comparison with I_Na⁺ recorded in the presence of 0.05 mM Ca²⁺ (Fig. 7, B and C, P < 0.02 at 147 mV). Reversibility of Ca²⁺ reduction in I_Na⁺ was tested next. When the Na⁺ bath solution with 4 mM Ca²⁺ (Fig. 7A) was exchanged by a Na⁺ bath solution with 0.05 mM Ca²⁺ (same protoplast), the magnitude of I_Na⁺ increased (Fig. 7B, n = 2).

View larger version (13K): [in this window] [in a new window]   Figure 7.   Reduction of INa+ from wheat root protoplasts by increased external Ca2+ concentration. A, Whole-cell currents from a wheat root cortical cell measured with 100 mM Na-Glu and 4 mM CaCl2 bath solution and standard pipette solution (C = 50 pF). B, Whole-cell currents from the same protoplasts as in Figure 7A measured with 100 mM Na-Glu and 0.05 mM CaCl2 bath solution and standard pipette solution. C, Average current-voltage relationships of whole-cell currents from wheat roots cortex cells measured with 100 mM Na-Glu and 4 mM CaCl2 (, n = 5) or 0.05 mM CaCl2 (, n = 6) bath solution and standard pipette solution. Holding potential was 30 mV. Voltage pulses from +40 to 140 mV in 20 mV steps were applied. Wheat seedlings were grown for 6 d in 1 mM CaCl2.

Protein kinases and phosphatases are involved in the response to salt stress in yeast and plant cells (Lee et al., 1999; Serrano et al., 1999). We tested the effect of externally applied deltamethrin (2 nM), an inhibitor of type 2B protein phosphatases, and okadaic acid (100 nM), an inhibitor of type 2A protein phosphatases. In both cases we observed only small changes in the current density of I_K⁺_in (not shown) or I_Na⁺ (standard Na⁺ bath solution: 4.6 ± 0.6 pA/pF at 147 mV, n = 11; standard Na⁺ bath solution with 2 nM deltamethrin: 5.9 ± 1.1 pA/pF at 147 mV, n = 7; standard Na⁺ bath solution with 100 nM okadaic acid: 5.3 ± 0.8 pA/pF at 147 mV, n = 5).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Induction of K⁺ Currents and of TaAKT1 mRNA

Inward-rectifying K⁺ channels provide a pathway for K⁺ uptake in root cells (see the introduction). In an earlier study on wheat roots grown with regular K⁺ supply, I_K⁺_in could be found in only 23% (n = 184) of the protoplasts (Findlay et al., 1994). Several explanations have been discussed, and it has been suggested that the occurrence of I_K⁺_in is limited to specific locations in the root cortex. On the other hand, the outward-rectifying K⁺ channel could be observed in 66% to 79% of the wheat root protoplasts (Schachtman et al., 1991; Findlay et al., 1994). Here we show that I_K⁺_in are induced by K⁺ starvation. We found I_K⁺_in in 91% and I_K⁺_out in 45% of the K⁺-starved wheat root protoplasts (n = 101 protoplasts). However, when protoplasts were pretreated with 5 mM K⁺ we observed I_K⁺_in in 54% and I_K⁺_out in 79% of the measured root protoplasts (n = 56 protoplasts). In addition, the K⁺-pretreated root protoplasts were found to have a decreased magnitude of I_K⁺_in. Interestingly, in a study on Arabidopsis root protoplasts low external K⁺ supply also caused an enhanced activity of I_K⁺_in (Maathuis and Sanders, 1995).

We examined the possibility that a K⁺ uptake channel gene is induced by K⁺ starvation in wheat roots. We isolated TaAKT1, a complementary DNA from wheat that shows high sequence homology to the Arabidopsis inward-rectifying K⁺ channel gene AKT1 (Sentenac et al., 1992). The deduced polypeptide sequence of TaAKT1 shows the typical features of the plant K⁺_in channels and animal shaker K⁺ channels. Each of the four subunits of a functional K⁺ channel consists of six transmembrane domains and a short hairpin segment called the P loop that is located between S5 and S6 and determines the selectivity of the pore. The AKT family of plant K⁺ uptake channels contain five ankyrin repeats in their C-terminal halves, which might represent potential domains for interactions of membrane proteins with the cytoskeleton, as has been shown in animal cells (Sentenac et al., 1992; for review, see Czempinsky et al., 1999).

Earlier models suggested that K⁺ absorption is mediated by cotransporters at micromolar K⁺ concentrations and by channels at higher concentrations. The level of the AKT1 gene in B. napus roots was unaffected by external K⁺ concentrations, suggesting that AKT1 is a constitutive component of K⁺ uptake (Lagarde et al., 1996). On the other hand, models have been proposed in which K⁺_in channels could contribute to both low- and high-affinity K⁺ uptake (Hedrich and Schroeder, 1989; Gassmann et al., 1993; Schroeder et al., 1994). AKT1 expression in yeast (Sentenac et al., 1992) and native expression in Arabidopsis roots (Hirsch et al., 1998; Spalding et al., 1999) have demonstrated that AKT1 can mediate high-affinity K⁺ uptake at micromolar K⁺. An Arabidopsis mutant disrupted in the AKT1 channel gene showed reduced growth at 100 µM external K⁺ when ammonium was added to the growth medium (Hirsch et al., 1998). These results, together with the induction of the TaAKT1 mRNA and K⁺_in in wheat roots by K⁺ starvation, suggest that K⁺_in channels might be involved not only in constitutive, but also in inducible, K⁺ uptake in wheat roots.

I_Na⁺ in Wheat Root Cells

We studied a Na⁺-dependent inward current found in a bath solution with high Na⁺ and low Ca²⁺ concentrations. I_Na⁺ have been reported in previous studies. In wheat roots, Na⁺ currents that show a fast-activating and a slow-activating component have been described (Tyerman et al., 1997). It is likely that the instantaneous currents reported here are related to these currents, although a comparison is complicated by the use of different varieties of wheat, different growth conditions, and the fact that in the previous study all Na⁺ bath solutions contained 10 mM KCl. Under the conditions reported here, we found that the magnitude of I_Na⁺ from K⁺-starved wheat root cortex protoplasts was about 3-fold enhanced compared with protoplasts from K⁺-supplemented roots. This increase in I_Na⁺ could be mediated by transcriptional regulation and/or by post-translational modification.

Previous studies reported that Na⁺-dependent inward currents from different root plasma membranes are insensitive to the K⁺ channel blockers Cs⁺ (up to 10 mM), TEA⁺ (up to 20 mM), and TTX⁺ (up to 50 µM) (Roberts and Tester, 1997; Tyerman and Skerrett, 1999). Tyerman et al. (1997) determined permeability ratios of Na⁺-dependent currents in wheat root protoplasts showing a relatively low whole-cell P_K+/P_Na+ (approximately 1.7:1). In addition, Cs⁺ ions were found to be more permeable than Na⁺ ions. As pointed out previously (Tyerman et al., 1997), whole-cell reversal potentials of relatively small I_Na⁺ provide only approximate permeability ratio values, because proton ATPases hyperpolarize root cells and other background conductances are included. Under the K⁺ starvation conditions used here, we found a K⁺/Na⁺ permeability ratio of about 2.6 and a Cs⁺/Na⁺ permeability ratio of about 0.12 based on shifts in the whole-cell reversal potential. It is possible that the previously described Na⁺-dependent currents (Tyerman et al., 1997) and K⁺-starvation-induced I_Na⁺ are mediated by different transport proteins, or that under these different conditions several channels/transporters make different relative contributions to an observed whole-cell I_Na⁺.

Ca²⁺ is known to have ameliorative effects on plants under high-salinity conditions (LaHaye and Epstein, 1969; Greenway and Munns, 1980; Rengel, 1992). In agreement with previous Na⁺ flux measurements (Allen et al., 1995; Davenport et al., 1997) and patch-clamp studies (White and Lemtiri-Chlieh, 1995; Roberts and Tester, 1997; Tyerman et al., 1997), we found a suppression of I_Na⁺ by high external Ca²⁺ concentrations. The correlation between the protective effect of external Ca²⁺ under high-salinity conditions for the whole plant and the suppression of Na⁺ uptake by Ca²⁺ indicates a physiological role for these Na⁺-dependent currents in salt stress. Note that the Na⁺ uptake rate into the roots of a salt-sensitive wheat species and a salt-tolerant wheat species did not differ significantly (Davenport et al., 1997), suggesting that Na⁺ tolerance is mediated by other mechanisms.

Protein phosphorylation and dephosphorylation constitute a general mechanism by which plant cells regulate cellular mechanisms in response to changes in the extracellular environment (Stone and Walker, 1995; Luan, 1998). Several protein kinases and phosphatases are involved in the response to salt stress in yeast and plant cells (Lee et al., 1999; Serrano et al., 1999). In Arabidopsis, the SOS3 gene was found to exhibit sequence similarity to the Ca²⁺-binding domain of a 2B-type protein phosphatase. Recessive sos3 mutants show a Ca²⁺-dependent reduction in high-affinity K⁺ uptake and hypersensitivity to NaCl stress (Liu and Zhu, 1998). We found small effects of externally applied phosphatase inhibitors deltamethrin (2 nM) and okadaic acid (100 nM) on I_Na⁺ under the imposed conditions, and further studies will be pursued.

Candidate Molecular Mechanisms and Physiological Roles of Na⁺ Currents

Thus far, the Na⁺-dependent currents in plant root cells have not been directly linked to specific genes or proteins, but several candidate genes have been characterized. It is possible that channels/transporters, which under non-saline conditions mediate the transport of other ions, become Na⁺ permeable under saline conditions (Khakh and Lester, 1999). This possibility has also been examined for K⁺_in channels. In several studies it was found that I_K⁺_in are highly selective for K⁺ over Na⁺, and therefore are not likely to represent Na⁺ uptake pathways (Gassmann and Schroeder, 1994; Tyerman et al., 1997; Amtmann and Sanders, 1998).

However, some of the recently isolated cation uptake transporters could be involved in Na⁺ uptake (for review, see Schachtman and Liu, 1999). The K⁺ transporter HKT1 is rapidly induced at the transcriptional level in wheat and barley roots by K⁺ starvation (Wang et al., 1998) and mediates low-affinity Na⁺ uptake in oocytes and in yeast (Rubio et al., 1995). In addition, the voltage dependence, time dependence, and cation selectivity of I_Na⁺ show correlations to HKT1-mediated low-affinity Na⁺ currents. However, HKT1-mediated low-affinity Na⁺ currents were reported to be insensitive to external Ca²⁺ in Xenopus oocytes (Tyerman and Skerrett, 1999). The absence of HKT1 modulation by Ca²⁺ in oocytes does not exclude the possibility of Ca²⁺ regulation of HKT1 in plants via post-translational mechanisms, as recent research suggests that Ca²⁺ regulation is mediated by phosphorylation/dephosphorylation (Liu and Zhu, 1998).

Other channels/transporters may also contribute to I_Na⁺. Recently, a non-specific cation channel NSC1 that is regulated by Ca²⁺ was found in yeast (Bihler et al., 1998), but the underlying gene remains unknown. LCT1 (Schachtman et al., 1997), AtKUP1 (Fu and Luan, 1998; Kim et al., 1998), and HvHAK1 (Santa-Maria et al., 1997) have also been shown to mediate low-affinity Na⁺ uptake in yeast. The low-affinity Na⁺ transport capabilities of these transporters seemed to be lower than HKT1 under the imposed conditions in yeast. Electrophysiological recordings of I_Na⁺ via LCT1, AtKUP1, and HvHAK1 have not yet been reported, and therefore a comparison to I_Na⁺ in root cells is not yet possible.

Single-channel currents that can account for the Na⁺-dependent whole-cell currents have been reported (Roberts and Tester, 1997; Tyerman et al., 1997; White, 1997). In animal systems it was discovered in recent years that transporters also can exhibit channel behavior, and that the classical distinction between transporters and channels does not always hold at the molecular level (Fairman et al., 1995; Cammack and Schwartz, 1996; Lin et al., 1996; Sonders and Amara, 1996; Su et al., 1996). As discussed previously, it seems likely that plant transporters such as AtKUPs and HKT1 could show channel conductance states (Gassmann et al., 1996; Fu and Luan, 1998; Chrispeels et al., 1999, Durell et al., 1999). Further research is needed to directly identify the different membrane channel/transporter genes contributing to inward-directed Na⁺ currents in root cells.

The physiological role of Na⁺ uptake has been the subject of discussion, because Na⁺ is not essential for most plants and excessive low-affinity Na⁺ uptake can result in toxic cytoplasmic Na⁺ concentrations. Our results, together with other studies (Pitman et al., 1968; Zhu et al., 1998), suggest that the availability of K⁺ in the growth medium is a crucial parameter in inducing Na⁺ uptake. The recent findings that high-affinity K⁺ uptake transporters are induced by K⁺ starvation, and that these transporters also mediate low-affinity Na⁺ uptake provide a possible model for the inducibility of I_Na⁺ reported here. Another non-exclusive explanation would be that low-affinity Na⁺ uptake pathways are activated in K⁺-starved plants, because the plants can to a certain degree compensate K⁺ deficiency by Na⁺ uptake (Mengel and Kirkby, 1982; Flowers and Läuchli, 1983; Rodriguez-Navarro, 2000). Further research should allow the establishment of relative contributions of individual genes responsible for Na⁺ uptake and I_Na⁺ induced by K⁺ starvation. Furthermore, the finding that both I_K⁺_in and TaAKT1 mRNA are induced by K⁺ starvation indicates that K⁺_in channels may contribute to both constitutive and inducible K⁺ uptake in wheat roots.