Diversity in Expression Patterns and Functional ... - Plant Physiology

Diversity in Expression Patterns and Functional Properties in the Rice HKT Transporter Family1[W] Mehdi Jabnoune, Sandra Espeout, Delphine Mieulet, Cećile Fizames, Jean-Luc Verdeil, Genevie`ve Cone´je´ro, Alonso Rodrı´guez-Navarro, Herve´ Sentenac, Emmanuel Guiderdoni, Chedly Abdelly, and Anne-Alieńor Ve´ry* Biochimie et Physiologie Molećulaire des Plantes, UMR 5004 CNRS/INRA/SupAgro-M/UM2, Campus SupAgro-M/INRA, 34060 Montpellier cedex 1, France (M.J., S.E., D.M., C.F., G.C., H.S., A.-A.V.); Laboratoire d’Adaptation des Plantes aux Stress Abiotiques, Centre de Biotechnologie de Borj-Ce´dria, Hammam-Lif 2050, Tunisia (M.J., C.A.); UMR De´veloppement et Ame´lioration des Plantes, Centre de Coope´ration Internationale en Recherche Agronomique pour le De´veloppement/INRA/SupAgro-M/UM2, 34398 Montpellier cedex 5, France (S.E., D.M., J.-L.V., E.G.); and Departamento de Biotecnologıá, Universidad Politećnica de Madrid, 28040 Madrid, Spain (A.R.-N.) Plant growth under low K+ availability or salt stress requires tight control of K+ and Na+ uptake, long-distance transport, and accumulation. The family of membrane transporters named HKT (for High-Affinity K+ Transporters), permeable either to K+ and Na+ or to Na+ only, is thought to play major roles in these functions. Whereas Arabidopsis (Arabidopsis thaliana) possesses a single HKT transporter, involved in Na+ transport in vascular tissues, a larger number of HKT transporters are present in rice (Oryza sativa) as well as in other monocots. Here, we report on the expression patterns and functional properties of three rice HKT transporters, OsHKT1;1, OsHKT1;3, and OsHKT2;1. In situ hybridization experiments revealed overlapping but distinctive and complex expression patterns, wider than expected for such a transporter type, including vascular tissues and root periphery but also new locations, such as osmocontractile leaf bulliform cells (involved in leaf folding). Functional analyses in Xenopus laevis oocytes revealed striking diversity. OsHKT1;1 and OsHKT1;3, shown to be permeable to Na+ only, are strongly different in terms of affinity for this cation and direction of transport (inward only or reversible). OsHKT2;1 displays diverse permeation modes, Na+-K+ symport, Na+ uniport, or inhibited states, depending on external Na+ and K+ concentrations within the physiological concentration range. The whole set of data indicates that HKT transporters fulfill distinctive roles at the whole plant level in rice, each system playing diverse roles in different cell types. Such a large diversity within the HKT transporter family might be central to the regulation of K+ and Na+ accumulation in monocots.

Although it is not clear what levels of Na+ are toxic in the plant cell cytosol and actually unacceptable in vivo, the hypothesis that this cation must be excluded from the cytoplasm is widely accepted. The most abundant inorganic cation in the cytosol is K+, in plant as in animal cells. This cation has probably been selected during evolution because it is less chaotropic 1 This work was supported by the Agropolis Fondation under the Rice Functional Genomics platform (Montpellier, France), by an Agence Universitaire de la Francophonie graduate fellowship and a national doctoral fellowship from the Tunisian Ministry of Higher Education, Scientific Research, and Technology (LR02CB02 to M.J.), by the European Research Area Network Plant Genomics Programme (grant no. ERA–PG FP/06.018B to E.G. and H.S.), and by the Biotechnology and Biological Sciences Research CouncilInstitut National de la Recherche Agronomique (grant to H.S. and A.-A.V.). * Corresponding author; e-mail [email protected]. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Anne-Alieńor Ve´ry ([email protected]). [W] The online version of this article contains Web-only data. www.plantphysiol.org/cgi/doi/10.1104/pp.109.138008

than Na+ (i.e. more compatible with protein structure even at high concentrations; Clarkson and Hanson, 1980). Its selection might also be due to the fact that in primitive cells, which originated in environmental conditions (seawater) where Na+ was more abundant than K+, a straightforward process to energize the cell membrane was to accumulate the less abundant cation and to exclude the most abundant one. In the cell, K+ plays a role in basic functions, such as regulation of cell membrane polarization, electrical neutralization of anionic groups, and osmoregulation. Concerning the latter function, K+ uptake or release is the usual way through which plant cells control their water potential and turgor. Although toxic at high concentrations, Na+ can be used as osmoticum and substituted for K+, mainly in the vacuole, when the plant is facing low K+ conditions and Na+ is available in the soil solution. This use of Na+, however, requires a tight regulation of K+ and Na+ transport and compartmentalization that becomes crucial in conditions of high Na+ concentrations in the soil solution. Control of Na+ and K+ uptake, long-distance transport in the xylem and phloem vasculatures, accumulation in aerial parts, and compartmentalization at the cellular

Plant PhysiologyÒ, August 2009, Vol. 150, pp. 1955–1971, www.plantphysiol.org Ó 2009 American Society of Plant Biologists

1955

Jabnoune et al.

and tissue levels have actually been shown to be essential in plant adaptation to salt stress (Greenway and Munns, 1980; Flowers, 1985; Hasegawa et al., 2000; Mu¨hling and Laüchli, 2002). Thus, accumulation of Na+ as osmoticum during K+ shortage or plant adaptation to salt stress requires integration at the whole plant level of Na+ and K+ membrane transport system activities (Apse et al., 1999; Shi et al., 2002; Qi and Spalding, 2004; Ren et al., 2005; Maathuis, 2006; Pardo et al., 2006; Horie et al., 2007). This report concerns transport systems named HKT upon first identification (for High-Affinity K+ Transporters) that are active at the plasma membrane and permeable to either K+ and Na+ or to Na+ only (Schachtman and Schroeder, 1994; Rodrı´guez-Navarro and Rubio, 2006). Several members of the HKT family have already been shown, by genetic approaches, to play important roles in plant salt tolerance (Berthomieu et al., 2003; Ren et al., 2005; Huang et al., 2006; Byrt et al., 2007) or growth in conditions of K+ shortage (Horie et al., 2007). In Arabidopsis (Arabidopsis thaliana), the HKT family comprises a single member, AtHKT1;1, which is permeable to Na+ only (Uozumi et al., 2000) and contributes to Na+ removal from the ascending xylem sap and recirculation from the leaves to the roots via the phloem vasculature (Berthomieu et al., 2003; Sunarpi et al., 2005). Interestingly, the HKT family comprises a much larger number of members in rice (Oryza sativa), with seven to nine genes depending on the cultivar (Garciadebla´s et al., 2003). In line with previous reports using rice as a model species to decipher the roles that HKT transporters can play in the plant, we have analyzed the expression patterns of three rice HKT genes, OsHKT2;1, OsHKT1;1, and OsHKT1;3, and investigated the functional properties of these transporters after heterologous expression, revealing new patterns of expression for HKT transporters and striking functional diversity. RESULTS Expression Patterns of OsHKT2;1, OsHKT1;1, and OsHKT1;3

The expression patterns of OsHKT2;1, OsHKT1;1, and OsHKT1;3 were analyzed using the in situ hybridization technique in 30-d-old rice plants grown in standard conditions, K+ shortage, or salt stress conditions. Cross sections of roots (Fig. 1) and leaves (Fig. 2) were hybridized with antisense or sense RNA probes specific of OsHKT2;1, OsHKT1;1, or OsHKT1;3 transcripts. Positive controls using an 18S ribonucleic RNA antisense probe indicated that strong hybridization signals could be observed in all root and leaf tissues in our experiments (Figs. 1G and 2G). Negative controls with HKT or 18S sense probes did not lead to any significant labeling (Figs. 1, B, D, F, and H, and 2, B, D, F, and H). In roots, labeling of OsHKT2;1 expression with the antisense probe was strongest in peripheral layers (epidermis, exodermis, and cortex differentiated into 1956

an aerenchyma). It was also detected in the stele, mainly in phloem (sieve elements and companion cells; Fig. 1A). The expression pattern of OsHKT1;1 in roots was reminiscent of that of OsHKT2;1 (Fig. 1C); however, the labeling required longer incubation times in color development buffer, suggesting that the level of OsHKT1;1 expression in roots was lower than that of OsHKT2;1. Expression of OsHKT1;3 was also detected both in peripheral layers (mainly in the cortex) and stele (vascular tissues), but compared with OsHKT2;1 and OsHKT1;1, the labeling was strongest in the phloem (Fig. 1E). Whatever the HKT gene studied, changes in growth conditions (see “Materials and Methods”) did not seem to affect its expression pattern (data not shown). In mature leaves, transcripts of the three HKT genes were identified in bulliform cells, large highly vacuolated cells of the adaxial epidermis involved in leaf rolling (Fig. 2, A, C, and E). Staining of OsHKT1;3 transcripts was particularly strong in these cells. Expression of the three HKT genes was also detected in vascular tissues, phloem, and xylem (parenchyma cells and differentiating vessels; Fig. 2, A, C, and E). To a lower extent, OsHKT2;1 and OsHKT1;3 expression was found in mesophyll (Fig. 2, A and E). As in the roots, the expression patterns of the three HKT genes in the leaves were not modified by the different growth conditions (data not shown). Staining, on the other hand, was dependent on the developmental stage: young immature leaves still folded in the sheath displayed a broader expression pattern of the three HKT genes. In particular, high labeling of mesophyll was observed in sheath for the three HKT genes (data not shown). Rectification Properties of HKT Transporters Expressed in Xenopus laevis Oocytes

Functional diversity among the rice HKT transporter family was investigated by characterizing OsHKT2;1, OsHKT1;1, and OsHKT1;3 in parallel experiments after heterologous expression in Xenopus oocytes. OsHKT2;1 had already been partly characterized in yeast (Saccharomyces cerevisiae) and Xenopus oocytes, but conclusions were controversial, the transporter being described either as a highly selective Na+ transporter (Horie et al., 2001; Garciadebla´s et al., 2003) or a weakly selective alkali cation transporter (Golldack et al., 2002). OsHKT1;1 had been reported to function as a Na+-selective transporter in yeast (Garciadebla´s et al., 2003). For each of the three transporters, OsHKT2;1, OsHKT1;1, and OsHKT1;3, injection of 50 ng of transporter copy RNA (cRNA) per oocyte resulted in a high level of expression 1 to 2 d after the oocyte injection. Figure 3 shows typical examples of current traces and mean currents recorded in control oocytes injected with water and in oocytes expressing OsHKT2;1, OsHKT1;1, or OsHKT1;3. Currents were recorded in Na+-containing medium, since the three rice HKT transporters were already reported or could be prePlant Physiol. Vol. 150, 2009

Diversity among Rice HKT Transporters Figure 1. Localization of OsHKT2;1, OsHKT1;1, and OsHKT1;3 expression in rice root by in situ hybridization. Twelveday-old plants were subjected to reduced K+ (100 mM) and mild salt stress (50 mM NaCl) conditions for 18 d before tissue collection. Hybridizations of root cross sections were revealed with VectorBlue Kit III. A and B, Localization of OsHKT2;1 transcripts. C and D, Localization of OsHKT1;1 transcripts. E and F, Localization of OsHKT1;3 transcripts. G and H, Control: hybridization with 18S ribonucleic RNA probe. A, C, E, and G, Antisense probes. B, D, F, and H, Sense probes. a, Aerenchyma; c, cortex cells; en, endodermis; ep, epidermis; ex, exodermis; pe, pericycle; ph, phloem; sc, sclerenchyma; xy, xylem. Bars = 100 mm.

dicted (based on phylogenic relationships; see “Discussion”) to be permeable to Na+. For each HKT transporter, the level of inward currents recorded in oocytes after 1 to 2 d of expression was at least 25 times larger than that recorded in control oocytes. Thus, each of the three transporters was by far the main transport system active at the oocyte membrane. Currents mediated by OsHKT2;1, OsHKT1;1, or OsHKT1;3 in Na+-containing solutions activated instantaneously and remained stable or slightly decreased over a 1-s voltage clamp stimulation (Fig. 3, B–D). Surprisingly, the three rice HKT transporters exhibited marked differences in rectification properties. Two of them, OsHKT2;1 and OsHKT1;3, were able to mediate both inward and outward currents, as already observed in members of the plant HKT family characterized so far (Rubio et al., 1995; Fairbairn et al., 2000; Uozumi et al., 2000; Horie et al., 2001; Ren et al., 2005), but the former almost did not rectify while the latter displayed weak inward rectification, its outward currents saturating when the driving force increased Plant Physiol. Vol. 150, 2009

(Fig. 3, B, D, F, and H). The third transporter, OsHKT1;1, strongly differed from the two others, since it was strongly inwardly rectifying, mediating no outward current (Fig. 3, C and G). Measurements of oocyte membrane resting potentials in bath solutions containing a few millimolar external Na+ yielded values consistent with the rectification properties of the three transporters. For example, the membrane potential in the presence of 3 mM external Na+ was much more negative in oocytes expressing OsHKT2;1 or OsHKT1;3 (272 6 1 mV and 290 6 4 mV) than in oocytes expressing OsHKT1;1 (219 6 2 mV) or control water-injected oocytes (217 6 2 mV). A strong inward rectification such as that observed in OsHKT1;1 had not, to our knowledge, been previously reported in the plant HKT family. Affinity for Na+

Na+ uptake kinetics with respect to external Na+ concentration was studied in the three rice HKT trans1957

Jabnoune et al. Figure 2. Localization of OsHKT2;1, OsHKT1;1, and OsHKT1;3 expression in rice leaf by in situ hybridization. Twelve-day-old plants were subjected to reduced K+ (100 mM) and mild salt stress (50 mM NaCl) conditions for 18 d before tissue collection. Hybridizations of mature leaf blade cross sections were revealed with VectorBlue Kit III. A and B, Localization of OsHKT2;1 transcripts. C and D, Localization of OsHKT1;1 transcripts. E and F, Localization of OsHKT1;3 transcripts. G and H, Control: hybridization with 18S ribonucleic RNA probe. A, C, E, and G, Antisense probes. B, D, F, and H, Sense probes. abe, Abaxial epidermis; ade, adaxial epidermis; bc, bulliform cells; bs, bundle sheath cells; m, mesophyll; ph, phloem; xy, xylem. Bars = 50 mm.

porters (Fig. 4). External Na+ concentration was varied in the range 4 mM to 100 mM, external K+ being kept as traces only (approximately 2 mM). Increasing external Na+ concentration increased the macroscopic inward conductance of the three transporters in a saturable manner (Fig. 4). In the two nonrectifying or weakly rectifying transporters OsHKT2;1 and OsHKT1;3, increasing external Na+ also resulted in positive shifts of the zero current potential (Fig. 4, A and E). Fitting the inward conductance versus external Na+ with hyperbolic functions gave rise to apparent half-saturation constants (K1/2) that were strikingly different in the three transporters: 1 order of magnitude separated the K1/2 of OsHKT2;1 (0.15 6 0.07 mM Na+ [n = 5]) from that of OsHKT1;3 (3.5 6 2 mM Na+ [n = 5]), which was itself separated by more than 1 order of magnitude from that of OsHKT1;1 (76 6 8 mM Na+ [n = 6]). Hence, the rice HKT transporter family comprises members displaying high, low, or very low Na+ uptake affinity. It should also be noted that the inward conductance versus external Na+ relationships could be fitted well with a single hyperbolic function in OsHKT1;1 and OsHKT1;3 but not in OsHKT2;1 (Fig. 4B). Fitting conductance data in OsHKT2;1 with the sum of two hyperboles resulted in high-affinity and low-affinity half-saturation constants (9.5 6 0.2 mM Na+ and 2.2 6 0.2 mM Na+, respectively), suggesting that OsHKT2;1 may possess a dual mode of Na+ uptake (see below). 1958

Cation Selectivity

The cation selectivity of the three rice HKT transporters was examined by comparing the currents mediated in the presence of different monovalent cations (Fig. 5). Figure 5, A and B, shows mean OsHKT2;1 currents recorded successively in the presence of Na+, K+, Rb+, Cs+, and Li+. The external cation concentration was either 10 mM (Fig. 5A) or 1 mM (Fig. 5B). In the presence of 10 mM external Na+, OsHKT2;1 could mediate large inward and outward currents reversing close to 250 mV, as shown previously (Fig. 4A). When Li+ replaced Na+ at the same concentration, both inward and outward OsHKT2;1 currents could still be observed, but the current-voltage relationship was shifted 50 mV negatively and the inward conductance was reduced (Fig. 5A). This indicates that OsHKT2;1 is selective for Na+ against Li+, probably by favoring Na+ at a conducting site. When oocytes were bathed with 10 mM K+, Rb+, or Cs+, both inward and outward currents through OsHKT2;1 almost disappeared (Fig. 5A) and the reversal potential of residual currents remained close to that observed in the presence of Na+. OsHKT2;1, therefore, is also highly selective for Na+ against K+, Rb+, and Cs+ in these conditions, but the mechanism of selectivity seems different from that occurring between Na+ and Li+. The strong reduction of both inward and outward currents and the absence Plant Physiol. Vol. 150, 2009

Diversity among Rice HKT Transporters Figure 3. Rice HKT transporters differ in their rectification properties. A to D, Examples of currents recorded in Xenopus oocytes. E to H, Mean currentvoltage (I-V) curves. A and E, Control (water-injected) oocytes. B and F, Oocytes expressing OsHKT2;1. C and G, Oocytes expressing OsHKT1;1. D and H, Oocytes expressing OsHKT1;3. Currents were measured in a bath solution containing either 3 mM Na-Glu (3 Na; B, D, F, and H) or 30 mM Na-Glu (30 Na; A, C, E, and G). Applied voltages varied in the range 2180 to 215 mV with an increment of +15 mV (most positive and negative voltages applied are indicated at right of the corresponding current traces). The dashed lines in A to D mark the zero current level. Data presented in I-V relationships (E–H) correspond to total oocyte currents (means 6 SE; n $ 4).

of any negative shift of the reversal potential when external K+, Rb+, or Cs+ at 10 mmol L21 replaced Na+ suggest that the selectivity for Na+ against the latter cations is not due to high discrimination but that external K+, Rb+, or Cs+, present at 10 mmol L21, renders OsHKT2;1 very poorly conductive. Similar current patterns were observed when external cations were present at 100 mmol L21 (data not shown): (1) the reversal potential of OsHKT2;1 currents in bath containing Na+ was then close to zero (5 6 2 mV [n = 16]) and it was shifted by 281 6 1 mV when Li+ was substituted for Na+; (2) almost no current was observed in the presence of K+, Rb+, or Cs+. Interestingly, a different current pattern was observed when the external cation concentration was reduced to 1 mmol L21 (Fig. 5B). Indeed, when K+, Rb+, and Cs+ were present at this concentration, contrary to what was observed when they were present at 10 or 100 mM, noticeable inward (Fig. 5B) and outward (data not shown) currents could be recorded. These currents reversed at a potential more positive than in the presence of 1 mM Na+, excluding the possibility that they could be due to Na+ contamination. Thus, at high external cation concentration, OsHKT2;1 would be almost only permeable to Na+, but at low external cation concentration it would be permeable also to K+, Rb+, and Cs+. Concerning Li+, the zero current potential was more negative than that recorded in the presence of Na+, as already observed in 10 or 100 mM solution, confirming that Li+ was less permeant than Na+. Furthermore, the zero current potential poorly depended on the concentration of Li+ in the external medium (1, 10, or Plant Physiol. Vol. 150, 2009

100 mM), suggesting that OsHKT2;1 is not significantly permeable to Li+. The inward currents measured in LiCl solutions are thus likely mostly carried by Na+ and K+ present as contaminants (see Fig. 5B legend and Fig. 7A below). In order to further analyze the permeability to K+ in OsHKT2;1, currents were recorded in solutions containing both Na+ and K+ at different concentrations, 0.3, 3, and 10 mM (Fig. 5, C and D). When external K+ was present at 0.3 mM, increasing external Na+ from 0.3 to 3 mM and then to 10 mM led to slight increases in OsHKT2;1 inward and outward conductance (1.8 6 0.02 times between 0.3 and 3 mM and 1.4 6 0.02 times between 3 and 10 mM, the conductances being determined close to the current reversal potentials) and positive shifts of the current reversal potential (shift of 61 6 3 mV between 0.3 and 10 mM; Fig. 5C), in line with what had been observed when K+ was present at nominal concentration (Fig. 4A). In the symmetrical experiment, when external Na+ was present at 0.3 mM, increasing external K+ from 0.3 to 3 mM and then to 10 mM led to quite similar shifts of current reversal potential (54 6 2 mV between 0.3 and 10 mM) but to strong decreases in OsHKT2;1 inward and outward conductances (4.8 6 1 times between 0.3 and 3 mM and 8.3 6 1 times between 0.3 and 10 mM; Fig. 5D). This indicates that when Na+ and K+ are both present in the bath, at least in the concentrations used in this experiment, OsHKT2;1 carries both cations, but external K+ exerts an inhibitory effect on OsHKT2;1 macroscopic inward and outward conductances. 1959

Jabnoune et al. Figure 4. Rice HKT transporters differ in their affinity for Na+. A, C, and E, Current-voltage (I-V) curves from oocytes expressing OsHKT2;1 (A), OsHKT1;1 (C), or OsHKT1;3 (E) in the presence of varying external Na+ concentrations. The background solution contained 4 mM Na+ and 2 mM K+ as contaminants. Additional Na+ was present as Glu salt (final concentrations as indicated). Currents flowing through HKT transporters were extracted from total oocyte currents by subtracting mean currents obtained in water-injected oocytes from the same batch in the same ionic conditions (see “Materials and Methods”). B, D, and F, Variation of rice HKT inward conductance with external Na+ concentration. Macroscopic conductances were defined as slopes of I-V relationships, either close to reversal potentials in the nonrectifying OsHKT2;1 and OsHKT1;3 transporters (determination with three I-V points) or between 2150 and 2120 mV in the inwardly rectifying OsHKT1;1 transporter. Macroscopic inward conductances in B, D, and F were extracted from I-V data shown in A, C, and E, respectively. Inward conductance values obtained for each oocyte were plotted versus external Na+ concentration, then normalized by the maximal conductance determined with a hyperbolic fit (sum of two Michaelis-Menten hyperboles in B and a single hyperbole in D and F) to suppress variability caused by differences in oocyte levels of expression. Mean normalized inward conductances were fitted again with the same kind of hyperbolic equations (solid lines) to determine the apparent K1/2 (half-saturation concentration regardless of the fitting equation used). Fit parameters of the two hyperboles in B were as follows: Km1 = 9.5 mM, Km2 = 2.2 mM, Gmax1 = 50%, Gmax2 = 50%. In D, it should be noted that OsHKT1;1 conductance in 100 mM Na+ (the highest concentration used) was still far from saturation. Therefore, the apparent K1/2 obtained for this transporter is probably not very accurate. The inset in B is a magnification of conductance data and fit at low Na+ concentrations. Data are means 6 SE (n = 5 in A, B, E, and F and n = 6 in C and D) and are representative of at least two experiments performed on different oocyte batches.

The selectivity to movovalent cations in OsHKT1;1 and OsHKT1;3 was analyzed in a similar way (Fig. 5, E–H). When oocytes were bathed successively in 100 mmol L21 Na+, K+, Rb+, Cs+, or Li+, significant OsHKT1;1 inward currents were recorded in the presence of Na+ only (currents in Na+ were about 20 times greater than those in the presence of the other cations; Fig. 5E). Absence of permeability to K+ in OsHKT1;1 in solutions containing only traces of Na+ was further checked at lower K+ concentration. OsHKT1;1 currents were negligible and almost constant when external K+ concentration was varied in the range 3 to 50 mM, while they strongly increased when external Na+ concentration was raised within the same range (Fig. 5F). This confirmed that OsHKT1;1, at least in the absence of Na+, is not permeable to K+. The question of OsHKT1;1’s possible permeability to K+ in the pres1960

ence of Na+ was difficult to address, because of the strong inward rectification and hence the absence of detectable variation with external concentration of the reversal potential in this transporter. However, we did not observe any current increase when 5 to 70 mM K+ was added to 30 mM external Na+ (rather, the current decreased; Fig. 6, G and I). Thus, no indication that OsHKT1;1 might work as a Na+-K+ cotransporter was obtained. In OsHKT1;3, as in OsHKT1;1, no indication of a significant permeability to another cation than Na+ was obtained (Fig. 5, G and H). When Na+, K+, Rb+, Cs+, and Li+ were present separately at 10 mmol L21 in the bath, OsHKT1;3 currents reversed either close to 250 mV (247 6 1 mV) in the case of Na+, as in OsHKT2;1, or about 270 mV more negatively (2116 6 2 mV) in the case of K+, Rb+, Cs+, or Li+ (Fig. 5G). The Plant Physiol. Vol. 150, 2009

Diversity among Rice HKT Transporters Figure 5. Rice HKT transporters display different cation selectivities. A to D, Oocyte currents mediated by OsHKT2;1. E and F, Oocyte currents mediated by OsHKT1;1. G and H, Oocyte currents mediated by OsHKT1;3. A, B, E, and G, Bath solutions with standard background successively supplemented with Na+, K+, Rb+, Cs+, or Li+ (chloride salts) at concentrations of 10 mM (A and G), 1 mM (B), or 100 mM (E). C, D, F, and H, Bath solutions with varying concentrations, eventually mixed, of Na+ and K+ (Glu salts). Na+ and K+ concentrations are indicated (in mM). Bath solutions containing 1 mM (B), 10 mM (A and G), and 100 mM (E) cations contained approximately 10, 10, and 25 mM K+ and 10, 15, and 20 mM Na+, respectively. Data are means 6 SE (n = 4 in A, C, and D, n = 7 in B and H, and n = 6 in E and F) and are representative of at least two experiments performed on different oocyte batches.

negative shift in reversal potential when K+, Rb+, Cs+, or Li+ replaced Na+ was accompanied by a reduction in OsHKT1;3 inward and outward conductance (5 6 1-fold reduction; Fig. 5G). In solutions containing both Na+ and K+, no indication of K+ permeability was obtained either (Fig. 5H): increasing K+ concentration from 0.3 to 10 mM in the presence of 0.3 mM Na+ was without any significant effect on OsHKT1;3 reversal potential and on the inward and outward conductances, whereas increasing Na+ concentration in the symmetrical experiment (K+ concentration fixed to 0.3 mM) increased OsHKT1;3 conductance and shifted OsHKT1;3’s reversal potential positively by 70 6 3 mV. K+ uptake in the wheat (Triticum aestivum) TaHKT2;1 transporter was first reported to be energized by H+ (Schachtman and Schroeder, 1994), and several HKTPlant Physiol. Vol. 150, 2009

related transporters in fungi (Trk family), which have their activity regulated by external pH, are thought to work as H+-K+ symporters, at least in certain conditions (Lichtenberg-Frate et al., 1996; Bihler et al., 1999; Rodrı´guez-Navarro, 2000). Therefore, we investigated whether external pH regulates K+ and/or Na+ transport in the three rice HKT transporters (Supplemental Fig. S1). Experiments were performed with solutions containing either K+ “only” (about 1 mM contaminant Na+) or Na+ “only” (approximately 2 mM contaminant K+) for OsHKT2;1 (Supplemental Fig. S1, A and B), since this transporter was shown to be permeable to both cations, or Na+ only for OsHKT1;1 and OsHKT1;3 (Supplemental Fig. S1, C and D). In every case, to favor observation of H+-Na+ or H+-K+ cotransport behavior, if any, external Na+ concentrations were chosen far 1961

Jabnoune et al. Figure 6. Inhibition of Na+ transport by monovalent cations in rice HKT transporters. Effect of external monovalent cations on Na+ transport through OsHKT2;1 (A–F), OsHKT1;1 (G–I), and OsHKT1;3 (J). Na+ external concentration (chloride salt) was 30 mM. Added concentrations of K+ (A, G, and J), Cs+ (B, H, and J), Rb+ (C and J), and Li+ (D and J; chloride salts) are indicated (in mM). Chloride concentration was kept constant in all bath solutions by adding, when necessary, HCl titrated with N-methyl-D-glucamine (NMDG). A to D, G, H, and J, Current-voltage (I-V) curves. E, F, and I, Kinetics of inhibition by external cations of inward (E and I) and outward (F) rice HKT conductances. Inward and outward conductances were determined in the nonrectifying OsHKT2;1 transporter using I-V data at potentials between 2150 and 2105 mV and between 0 and +15 mV, respectively. Inward conductances were determined for the inwardly rectifying OsHKT1;1 transporter between 2150 and 2120 mV. External cation concentrations leading to half conductance inhibition (Ki, as indicated) were obtained by mass action law fits (solid lines). Data are means 6 SE (n = 6 in A–F and n = 5 in G–J).

from values saturating the conductance (Na+ concentration lower than K1/2 for each transporter), and external K+ concentration for OsHKT2;1 was fixed very low (12 mM). Decreasing external pH from 7.5 to 6.5 and 5 did not affect the conductance of the three transporters and did not induce any significant shift in the current-voltage relationships. Thus, OsHKT2;1, OsHKT1;1, and OsHKT1;3 would not cotransport H+ or be regulated by external pH. 1962

Inhibition of Na+ Transport by Monovalent Cations in Rice HKT Transporters

Inhibition by external K+ of high-affinity Na+ uptake through OsHKT2;1 had been evidenced by flux analysis after expression of the transporter in yeast (Garciadebla´s et al., 2003). In Figure 5D, we show that when external Na+ was 0.3 mM, increasing external K+ concentration to the millimolar range strongly dePlant Physiol. Vol. 150, 2009

Diversity among Rice HKT Transporters

creased both inward and outward OsHKT2;1 macroscopic conductances. In these conditions, the increases in external K+ concentration also resulted in positive shifts of OsHKT2;1 reversal potential, indicative of K+ permeability. Sensitivity to external K+ in OsHKT2;1 was further examined using a 10 times higher Na+ basis than in Figure 5D: 30 mM (Fig. 6A). External K+ was varied in the range approximately 10 mM (nominal K+ concentration) to 70 mM, the ionic strength of the solutions being kept constant by addition of the nonpermeant large organic cation N-methyl-D-glucamine, to which OsHKT2;1 was insensitive (data not shown). In 30 mM Na+, as in 0.3 mM Na+, external K+ at a concentration of a few millimolar efficiently inhibited both the inward and the outward OsHKT2;1 macroscopic conductances (Fig. 6A). Half-inhibition concentration, derived from mass action law fit, was about the same for the inward and the outward conductances: close to 1 mM (Fig. 6, E and F). Contrary to what was observed in 0.3 mM Na+, however, additions of K+ to the 30 mM Na+ external solution did not induce positive shifts of OsHKT2;1 reversal potential (Fig. 6A), indicating that in the presence of high external Na+ concentration, OsHKT2;1 is not significantly permeable to K+. Thus, in these conditions, external K+ inhibits OsHKT2;1 without being transported. OsHKT2;1 sensitivity to external Cs+, Rb+, and Li+ was similarly examined in the presence of the same 30 mM Na+ background (Fig. 6, B–D). Cs+ (Fig. 6B) and Rb+ (Fig. 6C), like K+, inhibited OsHKT2;1 inward and outward conductances with about the same efficiency in the inward and outward directions but without modifying the current reversal potential, suggesting a mechanism of inhibition similar to that occurring with external K+. Among these three inhibitors, the most efficient was K+ (Ki approximately 1 mM in presence of 30 mM Na+), then Cs+ (Ki approximately 2 mM) and Rb+ (Ki approximately 3.5 mM; Fig. 6, E and F). In contrast to K+, Cs+, and Rb+, Li+ at a concentration up to 70 mM in the 30 mM Na+ medium did not have any effect on OsHKT2;1 inward and outward conductances (Fig. 6D). Li+ did not show any sign of permeation through OsHKT2;1 either. Interestingly, these results are reminiscent of those obtained with K+, Cs+, Rb+, or Li+ “alone” (in the presence of a nominal concentration of Na+; Fig. 5, A and B), where Li+ was already noticed to behave differently from K+, Cs+, and Rb+ in permeation of OsHKT2;1. Thus, both blockage and permeation experiments suggest the presence of binding site(s) with higher affinity for K+, Cs+, and Rb+ than for Li+ in the OsHKT2;1 translocation pathway. Sensitivity to external K+, Cs+, Rb+, and Li+ in the presence of 30 mM Na+ was also analyzed in OsHKT1;1 (Fig. 6, G–I) and OsHKT1;3 (Fig. 6J). OsHKT1;1 currents were inhibited by K+ (Fig. 6G) and Cs+ (Fig. 6H), although with a strongly lower sensitivity (Ki approximately 85 mM; Fig. 6I) than OsHKT2;1 currents. OsHKT1;3 currents were not sensitive to any of the four alkali cations (Fig. 6J). Plant Physiol. Vol. 150, 2009

Interactions between Na+ and K+ in OsHKT2;1 at Micromolar and Millimolar External Concentrations

Figures 5 and 6 show that permeation properties in OsHKT2;1 were highly dependent on external concentrations of Na+ and K+. OsHKT2;1 was found to be permeable to both Na+ and K+ in the presence of 0.3 mM Na+ and a few millimolar external K+ (Fig. 5D). It was permeable to Na+ only when external Na+ concentration was 30 mM (Fig. 6A). Whatever the Na+ concentration, it displayed a very weak macroscopic conductance when external K+ concentration was more than 10 mM (Figs. 5, A and D, and 6A). OsHKT2;1 permeation properties were further analyzed in the presence of very low external Na+ concentrations (4 mM Na+), external K+ being varied from the micromolar to the millimolar range (Fig. 7, A and B). Increasing K+ concentration induced positive shifts of OsHKT2;1 zero current potential. The magnitude of the shift was about 52 mV upon a 100-fold increase in K+ concentration from 10 mM to 1 mM (Fig. 7A). The whole set of results indicated that in the presence of low external Na+ concentrations, even as low as a few micromolar, OsHKT2;1 is permeable to K+ within a K+ concentration range from at least a few micromolar to a few millimolar. In this experiment, an inhibitory effect of external K+ on OsHKT2;1 macroscopic conductance was also clearly apparent (Fig. 7, A and B), as already mentioned with 0.3 mM external Na+ background (Fig. 5D). However, a detailed investigation of the effect of varying K+ concentration in the 4 mM Na+ background revealed that addition of K+ at very low concentration first increased OsHKT2;1 conductance before it had, at higher concentration, an inhibitory effect, with the maximal conductance being observed at approximately 10 mM external K+ (Fig. 7B). It is worth noting that the initial increase in conductance while increasing external K+ (up to 10 mM) provides further support to the conclusion drawn from reversal potential analysis that OsHKT2;1 at low external Na+ is permeable to K+. It should also be noted that the modulation (stimulation and inhibition) of OsHKT2;1 macroscopic conductance by external K+ was observed in both the inward and the outward directions. Compiling experiments performed in the presence of various external Na+ backgrounds highlights complex relationships between OsHKT2;1 conductance and external K+ or Na+ concentrations (Fig. 7C). Inhibition of OsHKT2;1 conductance by external K+ was determined neither by the sole external K+ concentration nor by the Na+/K+ concentration ratio (Fig. 7C). Diverse Conduction Modes in OsHKT2;1

Putative coupling between Na+ and K+ transport in OsHKT2;1 was examined by analyzing more precisely the effect of external K+ and Na+ activities on the OsHKT2;1 zero current potential (Fig. 7D). In the presence of 4 mM external Na+ and K+ activity in the 2 to 900 mM range, variation of OsHKT2;1 zero current 1963

Jabnoune et al.

Figure 7. Interaction between Na+ and K+ in OsHKT2;1. A and B, Effect of increasing external K+ on OsHKT2;1 transport properties in the presence of low external Na+. The external concentration of Na+ was 4 mM (Na+ present as contaminant). External K+ concentrations (2 mM to 1 mM, 2 mM being the nominal concentration) are indicated in A. Additional K+ was present as its Glu salt. A, Current-voltage (I-V) curves. B, Conductances extracted from I-V data shown in A. Inward and outward conductances were determined close to reversal potentials (three-point and two-point determinations, respectively). C, Absence of unequivocal relationship between the sensitivity of OsHKT2;1 inward conductance to K+ and the K+-to-Na+ concentration ratio in the bath solution. GNa and GNaK refer to OsHKT2;1 conductance in the absence and presence, respectively, of added K+ (contaminant K+ concentration of 2–10 mM). The concentration of Na+ in the bath solution was 0.004 (contamination in background), 0.05, 0.3, 1, or 30 mM. D, Sensitivity of OsHKT2;1 and OsHKT1;3 zero current potential (Erev) to external Na+ or K+ activity. In each set of solutions, either the concentration of K+ was fixed and that of Na+ was variable or that of Na+ was fixed and that of K+ was variable (activities indicated in mM). Solid lines correspond to logarithmic fits of Erev versus Na+ activity or K+ activity. Black symbols represent OsHKT2;1. Slope of the logarithmic fits in the presence of 4 mM Na+ was 26 mV per decade of K+ activity. In the presence of 2 mM K+ and varying Na+ concentrations, regressions were performed on Erev values at activities either lower than 1 mM or higher than 1 mM, leading to slope values of 22 or 48 mV per decade of Na+ activity, respectively. White symbols represent OsHKT1;3. Slope of the fits was 46 mV per decade of Na+ activity, independent of the K+ activity (0.002 or 0.25 mM). Data are means 6 SE (n = 7 in A and B and n = 5–7 in C and D).

potential was linear on a logarithmic scale, with a slope of 26 mV per decade of K+ activity (Fig. 7D, black diamonds). Linear variations of OsHKT2;1 zero current potential, with similar slope values, were also observed in symmetrical experiments (i.e. the external activity of K+ being kept constant at 2 mM and that of Na+ varied in the submillimolar range, 4–900 mM; Fig. 7D, black triangles). The slope values were then of about 22 mV per decade of Na+ activity. In summary, variations with external K+ or external Na+ of OsHKT2;1 zero current potential at submillimolar K+ and Na+ concentrations were very similar, linear on a logarithmic scale with slope values (22–26 mV) close to that expected for a 1964

Na+-K+ symport with a stoichiometry of 1:1 (slope of 29 mV per decade of Na+ or K+ activity). That such slope values were slightly lower than the theoretical ones for a symport with a stoichiometry of 1:1 may indicate a Na+:K+ stoichiometry slightly divergent from 1:1 and/or a slight permeability to another ion (Kuroda et al., 2004). Overall, these results strongly suggest that the transport of Na+ and K+ by OsHKT2;1 at submillimolar K+ and Na+ concentrations is coupled. The fact that the coupled transport behavior is still observed with [Na+]-to-[K+] or [K+]-to-[Na+] ratios of about 500 (Fig. 7D) suggests the presence of high-affinity and selective binding sites for Na+ and K+, able to discriminate efficiently between Plant Physiol. Vol. 150, 2009


one ion and the other in the OsHKT2;1 ion translocation pathway. When the external concentration of Na+ was increased in the millimolar range (Na+ activity from 1 to 25 mM, K+ kept constant at 2 mM), the slope of OsHKT2;1 zero current potential variation changed from approximately 22 mV to approximately 50 mV per decade of Na+ activity (Fig. 7D, black triangles, dotted line), suggesting that OsHKT2;1 then behaves like a Na+ uniport (slope expected for a selective uniport: 58 mV per decade). This is in agreement with the conclusion previously drawn from Figure 6A that at 30 mM external Na+, OsHKT2;1 is not permeable to K+. It is worth noting that OsHKT1;3, which was found to be permeable to Na+ and not to K+ (Fig. 5, G and H), displayed zero current potential variations with a single slope from the submillimolar to the millimolar Na+ concentration range, close to 50 mV per decade of Na+ activity (Fig. 7D, white squares and triangles). External K+ in OsHKT2;1 symport or uniport modes inhibits the macroscopic conductance. We propose that this inhibition, which concerns both the inward and outward directions (Figs. 5D, 6A, and 7A), is due to the presence in the OsHKT2;1 population of transporters in nonconductive (or very weakly conductive) states. Since external K+ has almost the same effect on inward and outward currents, K+ binding leading to nonconductive states should involve relatively external binding sites. Similarly, the presence of an external

binding site is expected to explain Na+ control of the OsHKT2;1 conduction mode (symport or uniport) from the external side. Figure 8 summarizes the different conduction modes proposed to be encountered by OsHKT2;1, with examples of external concentrations of K+ and Na+ at which each mode is preponderant. Schematically, the symport mode is expected to operate mainly at submillimolar external Na+ and K+, the Na+ uniport mode at external Na+ in the millimolar range and external K+ in the submillimolar range, and nonconductive states at external K+ in the millimolar to 10 mM range. Thus, the different conduction modes displayed by OsHKT2;1 while expressed in Xenopus oocytes all operate at physiological concentrations of Na+ and K+. Operation of these different conduction modes in planta, therefore, is likely. DISCUSSION OsHKT2;1 Permeability to Monovalent Cations

The question of OsHKT2;1 ionic selectivity was debated (Horie et al., 2001; Golldack et al., 2002; Garciadebla´s et al., 2003). Our data show that OsHKT2;1 is permeable to Na+, K+, Cs+, and Rb+ but not (or weakly) to Li+ (Fig. 5B) at external alkali cation concentrations lower than 1 mM, while it is significantly permeable to Na+ only at higher alkali cation

Figure 8. Scheme of OsHKT2;1 conduction modes in Xenopus oocytes depending on external K+ and Na+ concentrations. Three types of conduction mode proposed to be encountered by OsHKT2;1 (Na+-K+ symport, Na+ uniport, and nonconductive states) are positioned in the scheme according to external K+ and Na+ concentrations. Sets of external Na+ and K+ concentrations (in mM) at which each mode is expected to be preponderant are indicated with reference to supporting figures below each mode. Note that the two independent arrows present in the diagrammatic representation of the symport mode do not mean that Na+ and K+ could not move through the same pore.

Plant Physiol. Vol. 150, 2009

1965

Jabnoune et al.

concentrations (Figs. 5A, 6, A–D, and 7D). However, in the presence of high Na+ concentrations, it displays a very low conductance when K+, Cs+, or Rb+ is also present at a high concentration (.1 mM) in the external solution (Figs. 5A, 6, A–C, and 7B). Thus, OsHKT2;1 displays complex permeation properties and ionic selectivity, differing deeply when the external concentrations of cations (Na+, K+, Rb+, and Cs+) are varied from the micromolar to the 10 to 100 mM range. Initial and apparently contradictory analyses of its ionic selectivity (at least concerning Na+ and K+ in the Xenopus oocyte system) are thus likely to describe distinct transport behaviors restricted to partial concentration ranges. Na+-K+ Interactions and Transport Mechanism in OsHKT2;1

As outlined in Figure 8 and further discussed below, our electrophysiological analyses in Xenopus oocytes indicate (see last paragraph of “Results”) that OsHKT2;1 displays three conductive modes depending on external K+ and/or Na+: a Na+-K+ symport, which is the dominating mode at submillimolar external Na+ and K+; a Na+ uniport, preponderant when the external concentration of Na+ is within or above the millimolar range and external K+ is in the submillimolar range; and nonconductive states at external K+ in the millimolar to 10 mM range. This behavior is reminiscent of (but probably more complex than) that of the wheat TaHKT2;1 transporter, which has been shown to work as a Na+-K+ symport when external Na+ and K+ concentrations are balanced and as a Na+-selective uniport when external Na+ is in excess (Rubio et al., 1995; Gassmann et al., 1996; see below). Kinetic models were extensively developed in the 1980s to account for transport properties in various ion transport systems, including symporters (Hansen et al., 1981; Sanders et al., 1984; Blatt et al., 1987). In these models, symporter behavior is classically described by six-state schemes (when stoichiometry is 1:1) comprising steps of successive binding of the two transported ions, voltage-dependent ion translocation across the membrane, release of the transported ions, and voltage-independent recycling in the membrane. The attractiveness of these models relies on their simplicity, since the six-state model can be operationally reduced to a two- to four-state model when the concentration of only one transported ion is varied at one side of the membrane and on the absence of presupposition concerning the rate-limiting steps (McCulloch et al., 1990; Meharg and Blatt, 1995; Maathuis et al., 1997; Haro and Rodrı´guez-Navarro, 2002). Such models do not include the eventuality of competitive binding by the cotransported ion at one or each of the two binding sites. It was not possible, within this framework, to fit OsHKT2;1 current-voltage relationships determined over a large range of Na+ concentration (up to 30 mM; Fig. 4A) with the switch from the symport to the uniport mode or to fit the 1966

complex effect of external K+ (stimulation or inhibition, depending on K+ concentration) on OsHKT2;1 currents (data not shown). To account for the dual mode of transport in TaHKT2;1 (Na+-K+ symport or Na+ uniport), Rubio et al. (1995) and Gassmann et al. (1996) have taken into account the eventuality of competitive binding by the cotransported ion at the binding sites. They proposed that two high-affinity binding sites, one for K+ and the other one for Na+, named K+ and Na+ coupling sites, respectively, are present in the transporter and need to be occupied for uptake to occur. Competitive binding of K+ and Na+ at the K+ “coupling” site determines the permeation mode of the transporter, namely symport or Na+ uniport. Our data regarding the effects of varying external Na+ (Fig. 7D) are compatible with this model. Since the symport behavior could be observed in the presence of only a few micromolar K+ and concentrations of Na+ up to 500 times higher (e.g. 2 mM K+ and 1 mM Na+; Fig. 7D), the K+ coupling site in OsHKT2;1 must display very high affinity and selectivity for K+ against Na+ (higher than in TaHKT2;1, where symport behavior was only observed in quite balanced Na+ and K+ concentrations). Furthermore, experiments with other alkali cations suggest that the K+ coupling site in OsHKT2;1 would poorly discriminate between K+, Cs+, and Rb+ but would be selective against Li+ (Fig. 5B). Inhibition of symport activity by “high” external K+ (Figs. 5D and 7A) could also fit the coupling site model, assuming that the binding of K+ at the Na+ coupling site results in a nonconductive (or weakly conductive) state in OsHKT2;1. Similarly, concerning the Na+ coupling site in the framework of the discussed model, the fact that symport activity was still observed in the presence of a few micromolar Na+ and concentrations of K+ up to 200 times higher (e.g. 4 mM Na+ and 1 mM K+; Fig. 7D) indicates that the Na+ coupling site must display high affinity and selectivity for Na+ against K+. The assumption of only two (Na+ and K+) highly selective coupling sites controlling the transporter mode, however, does not help explain the inhibition by external K+ of OsHKT2;1 in the Na+ uniport mode (Fig. 6A). Indeed, addition of K+ in the millimolar range in the presence of 30 mM Na+, assuming the presence of a highly selective K+ coupling site, should restore symport activity, a phenomenon that can be discarded owing to the absence of a shift of OsHKT2;1 reversal potential upon addition of K+ in these conditions. The presence of a supplementary site, therefore, should be proposed. Binding of K+ to this regulation site would prevent conduction. Although this coupling and regulation site model remains speculative, our analysis strongly suggests the presence of at least three binding sites able to interact with both Na+ and K+. The presence of a set of sites displaying distinctive affinities for Na+ and K+ is not incompatible with the hypothesis that permeation occurs through a narrow pore where several ions move simultaneously in a single file, hopping from one binding site to the next Plant Physiol. Vol. 150, 2009


one. Within the framework of this hypothesis, the three binding sites might be located within the conduction pathway itself. It should be noted, however, that our data clearly indicate that the OsHKT2;1 conduction mode is controlled from the external side of the transporter (at a location not accessible from the internal side), since the three conduction modes (symport, uniport, and nonconductive states) are determined by the external concentrations of K+ and Na+, even when polarization conditions result in net effluxes through OsHKT2;1 (Figs. 6A, and 7, A and D). Functional Diversity in the HKT Family

The HKT family was described as comprising two major types of transporters when expressed in yeast or Xenopus oocytes (Rodrı´guez-Navarro and Rubio, 2006): those permeable to Na+ only, like AtHKT1;1, the sole member of the HKT family present in Arabidopsis; and those permeable to both Na+ and K+ and eventually endowed with Na+-K+ symport activity, as evidenced for two of them, TaHKT2;1 (Rubio et al., 1995; Gassmann et al., 1996) and HvHKT2;1 (Haro et al., 2005). The presence in rice of nine HKT genes (Garciadebla´s et al., 2003) raised the question of the functional diversity within this transporter family. Four rice members were already partly characterized: OsHKT1;1 and OsHKT1;5 (reported to be permeable to Na+ only; Garciadebla´s et al., 2003; Ren et al., 2005); OsHKT2;1, the permeability of which was debated (see above); and OsHKT2;2, described as permeable to both K+ and Na+ (Horie et al., 2001). This report provides detailed electrophysiological analyses for two of these members, OsHKT2;1 and OsHKT1;1, and for a new member, OsHKT1;3, that was not previously characterized. OsHKT2;1 is shown to be permeable to both Na+ and K+ and to be able to mediate symport activity in some conditions (Figs. 5, A–D, 7D, and 8; see above). OsHKT1;1 and OsHKT1;3 are shown to be permeable to Na+ only (Fig. 5, E–H). Thus, the new data are consistent with the view that HKT transporters can be sorted into two types when their ionic permeability to Na+ and K+ is considered, all transporters being permeable to Na+ and some also to K+. In HKT transporters of the latter type, the question of interactions between the two transported species has not been systematically addressed, but when such analyses have been carried out (i.e. for TaHKT2;1, HvHKT2;1, and OsHKT2;1), they have always led to the conclusion that the transporters were able to behave as Na+-K+ symporters, at least in some environmental conditions (Rodrı´guez-Navarro and Rubio, 2006; Fig. 7D). This may indicate that every HKT transporter permeable to K+ is endowed with the capacity to function as a Na+-K+ symport. The rice K+-permeable OsHKT2;2 transporter has not been characterized in detail, but the available information suggests that its apparent affinity for Na+ is lower than that of OsHKT2;1 and its sensitivity to (blockage by) external K+ is weaker (Horie et al., 2001; Plant Physiol. Vol. 150, 2009

Figs. 4, A and B, 5D, and 6A). Similarly, as discussed above, OsHKT2;1 displays a higher affinity for Na+ and a stronger sensitivity to K+ than the wheat TaHKT2;1. Our data on OsHKT1;1 and OsHKT1;3 also indicate that rice HKT transporters permeable to Na+ only strongly differ in terms of affinity for Na+, rectification, and sensitivity to external K+ (Ren et al., 2005; Figs. 3, C and D, 4, C–F, and 6, G–J). The physiological implication of differences in rectification among HKT transporters is so far unknown. Evidence that HKT transporters are not all able to mediate inward and outward Na+ fluxes raises the question of the possible occurrence and role of cation effluxes through nonrectifying members in vivo. Phylogenetic Relationships and Functional Properties in HKT Transporters

Phylogenetic analyses of publicly available HKT sequences revealed, 2 years ago, that the family splits into two major branches, which were named subfamilies 1 and 2 (Platten et al., 2006). This led to the proposal of a new nomenclature to sort out members according to their subfamily membership, the names HKT1;x or HKT2;y referring to transporters belonging to subfamily 1 or 2, respectively (Platten et al., 2006). Based on available data, it was suggested that subfamily 1 would correspond to HKT transporters permeable to Na+ only and subfamily 2 to transporters permeable to both Na+ and K+. The analysis also suggested differences between dicots and monocots concerning the number and functional properties of HKT transporters. Dicotyledonous species appeared to constitute a low number of HKT genes, 1 (as in Arabidopsis or poplar [Populus trichocarpa]) or 2 (as in Eucalyptus camaldulensis). Furthermore, HKT transporters permeable to both Na+ and K+ (subfamily 2) were present only in monocotyledonous species. A much larger number of HKT sequences are available now (see the phylogenetic tree in Supplemental Fig. S2). The data indicate that some dicotyledonous species can display a larger number of HKT members, for example, at least four in grapevine (Vitis vinifera). However, in agreement with previous hypotheses, all newly identified HKT transporters from dicotyledonous species, including the four grapevine members, belong to subfamily 1, based on sequence analysis. The phylogenetic tree shown in Supplemental Figure S2 and our electrophysiological analyses also suggest that distinct subgroups might exist in monocot subfamily 1. Indeed, OsHKT1;1 and OsHKT1;3, which are lowaffinity slightly or strongly rectifying transporters, appear to belong to the same subbranch, distinct from that of OsHKT1;5, which seems to be a highaffinity nonrectifying transporter (Ren et al., 2005). Expression Patterns and Roles in the Plant

Information on tissue localization was available for three members of each HKT subfamily so far: 1967

Jabnoune et al.

AtHKT1;1 from Arabidopsis (Berthomieu et al., 2003; Sunarpi et al., 2005), OsHKT1;5 from rice (Ren et al., 2005), and McHKT1;1 from ice plant (Mesembryanthemum crystallinum; Su et al., 2003) for subfamily 1; TaHKT2;1 from wheat (Schachtman and Schroeder, 1994), OsHKT2;1 (Golldack et al., 2002; Kader et al., 2006; Horie et al., 2007), and OsHKT2;2 (Kader et al., 2006) from rice for subfamily 2. The whole set of data emphasized that HKT transporters are strongly expressed in tissues (root epidermis and cortex, xylem and phloem, vascular bundle regions) playing crucial roles in ion uptake or long-distance transport and redistribution in the plant. The subfamily 1 transporters have been constantly reported to be expressed in the plant vasculature and rarely in other tissues. Expression patterns of the subfamily 2 transporters have always been found to include root periphery cells and often tissues in or close to the plant vasculature. The tissue specificity of OsHKT2;1 expression presented in this study on the japonica rice ‘Nipponbare’ is consistent with the pattern previously reported in the same cultivar by Horie et al. (2007) and overlaps with, but also differs slightly from, the expression patterns reported in indica rice varieties (Golldack et al., 2002; Kader et al., 2006). Our study also provides the first information on two other HKT transporters, OsHKT1;1 and OsHKT1;3. Taken as a whole, the new data reveal broad expression patterns, not restricted to root periphery cells and vascular regions. For example, hybridization results shown in Figure 2 indicate that the three HKTs under investigation are all expressed in bulliform cells, enlarged epidermal cells present in longitudinal rows in leaves of grasses and involved in leaf rolling and unrolling. It is tempting to speculate that these transporters are involved in ion fluxes triggering turgor changes in bulliform cells, allowing them to control leaf folding in response to environmental conditions. Thus, the roles of HKT transporters would not be restricted to ion uptake from the soil solution and long-distance transport and redistribution in the plant. Our data also indicate that root periphery cells can express HKT transporters from both subfamilies 1 and 2. For example, data shown in Figure 1 indicate that root cortical cells express HKT2;1, HKT1;1, and at a lower level HKT1;3. Also, the three HKTs are expressed in root phloem. In other words, the information available in monocots at present, where the two subfamilies of HKT transporters coexist, points to complex expression patterns resulting in intriguing overlaps of transporters endowed with clearly distinctive functional properties in terms of permeation and selectivity (uniport and/or symport activity) and affinity. It is tempting to speculate that this complexity and diversity allow HKT transporters to fulfill highly integrated roles at the whole plant level. Reverse genetics approaches in Arabidopsis and analysis of quantitative trait loci for salt tolerance in rice have highlighted the roles in planta of two HKT transporters from subfamily 1, AtHKT1;1 and 1968

OsHKT1;5. In Arabidopsis, AtHKT1;1 is expressed in stelar tissues of both roots and shoots and plays a role in preventing Na+ overaccumulation in shoots by unloading Na+ from the ascending xylem sap and loading it into the descending phloem (Berthomieu et al., 2003; Sunarpi et al., 2005; Davenport et al., 2007). Similarly, in rice, OsHKT1;5 expression was detected mainly in shoot and root vascular tissues, shown to play a role in Na+ transport by the xylem vasculature and control Na+ accumulation in leaves (Ren et al., 2005). At the whole plant level, each of these two systems contributes to adaptation to salt stress. The roles of HKT transporters from subfamily 2 remain more mysterious. A role for OsHKT2;1 in highaffinity Na+ uptake by rice roots has recently been demonstrated using a reverse genetics approach (Horie et al., 2007). When grown under low Na+ and nominal K+ (“K+-free”) conditions, knockout mutant rice lines disrupted in OsHKT2;1 displayed strongly reduced high-affinity Na+ uptake by roots compared with wild-type plants. They also displayed reduced size and biomass, suggesting a beneficial role of Na+ nutrition on rice growth at low Na+ concentration in conditions of severe K+ starvation. In wheat, a role for TaHKT2;1 in root Na+ uptake from external solutions containing high Na+ concentrations (100–200 mM) was deduced from comparative analyses of wild-type and antisense wheat lines (Laurie et al., 2002). Thus, depending on their affinity for this cation, HKT transporters can contribute to Na+ uptake in a large concentration range. On the other hand, no evidence is available so far that these systems contribute to K+ transport in the plant. Furthermore, no indication of Na+-K+ symport activity has been obtained in higher plant tissues yet (Maathuis et al., 1996; Walker et al., 1996; Haro et al., 2005). The absence of evidence that subfamily 2 HKT transporters contribute to K+ fluxes in planta, while these systems are clearly permeable to K+, highly expressed in plant cells, and functional in the cell membrane (as shown by their strong contribution to Na+ transport) might be due to redundancy in K+ transport systems from different families. For instance, in root periphery tissues, when plants are grown in classical environmental conditions and/or when experiments are performed in classical bath solutions, K+ transport activity of HKT transporters might be masked by that of K+ channels from the Shaker family (Sentenac et al., 1992; Hirsch et al., 1998; Santa-Marıá et al., 2000; Pilot et al., 2003; Ve´ry and Sentenac, 2003) and/or H+-K+ symporters from the KUP/HAK family (Santa-Marıá et al., 2000; Ban˜uelos et al., 2002; Gierth and Ma¨ser, 2007). The question of the existence of Na+-K+ symport activity in plants, owing to its importance, clearly deserves to be readdressed. Two HKT transporters, OsHKT2;1 and TaHKT2;1, are now clearly demonstrated to be able to mediate Na+-K+ symport activity. Furthermore, they are both induced upon K+ starvation and can complement yeast mutants defective in K+ uptake (Schachtman Plant Physiol. Vol. 150, 2009


and Schroeder, 1994; Wang et al., 1998; Horie et al., 2001; Golldack et al., 2002). The eventuality that HKT transporters can actually function as Na+-K+ uptake systems in planta is thus worth being further examined, taking advantage of the important progress that has been made recently in molecular identification of K+ transport systems and using rice as a model plant. MATERIALS AND METHODS Plant Material Seeds of rice (Oryza sativa japonica ‘Nipponbare’) were provided by the French Agricultural Research Centre for International Development. They were treated with 70% ethanol for 1 min, rinsed with sterile distilled water, treated with 15% potassium hypochlorite solution for 20 min, and rinsed three times again with sterile distilled water. They were then dipped in water for 12 h. For germination, five seeds were placed on one layer of Whatman germination paper in a plastic petri dish into which 10 mL of sterile distilled water had been introduced. The dishes were then stored in a growth chamber (14 h of light per day, 500 mE m22 s21, 28°C/25°C day/night). Seedlings were transferred into aerated hydroponic tanks 12 d after germination. The growth solutions used were based on that described by Yoshida et al. (1976). The medium used for control treatment contained 0.7 mM KNO3, 0.8 mM KH2PO4, 1.2 mM Ca(NO3)2, 0.5 mM (NH4)2SO4, 1.6 mM MgSO4, 60 mM NaFeEDTA, 45 mM H3BO3, 20 mM MnSO4, 1.6 mM CuSO4, 1.4 mM ZnSO4, and 0.3 mM (NH4)6Mo7O24. In experiments with reduced K+ (“2K+”), NaH2PO4 was substituted for KH2PO4 and 0.35 mM Ca(NO3)2 was substituted for 0.7 mM KNO3. For salt stress treatment, 50 mM NaCl was added to Yoshida or 2K+ medium. The pH of the growth solutions was adjusted between 5.3 and 5.5 every 2 d. Concentrations of contaminant K+ and Na+ were about 100 mM. Plants were grown in a phytotron with a photoperiod of 12 h of light/12 h of dark at 500 mE m22 s21 at day/night temperatures of 28°C/24°C under 65% hygrometry. Experiments were performed on 1-month-old plants.

In Situ Hybridization RNA Probe Synthesis Primers were designed for PCR amplification of specific regions of OsHKT2;1, OsHKT1;1, or OsHKT1;3 cDNAs, respectively, as follows: OsHKT2;1-Up and OsHKT2;1-Down (5#-CATACTCGTTGGCTCGTTGC-3# and 5#-GGTTGCTTGGCTTCAGGAAC-3#), OsHKT1;1-Up and OsHKT1;1Down (5#-GGAAGTCAATTTCTTCTGATCCA-3# and 5#-TCATTTCAGGATGAACTCCTTG-3#), and OsHKT1;3-Up and OsHKT1;3-Down (5#-CTCCAAACTTTCCGCACATT-3# and 5#-TTGACTTGTCTCGTGGCTTG-3#). Primers were also designed for PCR amplification of an 18S ribosome cDNA region (18S ribosome probe used as a positive control): Rib-Up (5#-CCGACCCTGATCTTCTGTGAAGGG-3) and Rib-Down (5#-CCAAGTCAGACGAACGATTTGCACG-3#). Primers containing the above sequences but extended at their 5# ends with the T7 RNA polymerase promoter sequence (5#-GCGAAATTAATACGACTCACTATAGGGAGA-3#) were also designed and were named OsHKT2;1-T7-Up and OsHKT2;1-T7-Down, OsHKT1;1-T7-Up and OsHKT1;1T7-Down, OsHKT1;3-T7-Up and OsHKT1;3-T7-Down, and RibT7-Up and RibT7-Down. Finally, one primer corresponding to the T7 end was also designed, E-T7 (5#-GCGAAATTAATACGACTCAC-3#). Sense and antisense probes were synthesized in two steps: a first PCR was performed with one primer Up and one primer T7-Down or with one primer Down and one primer T7-Up. The PCR was carried out with 2 ng of DNA, 13 Taq polymerase reaction buffer, 1.5 mM MgCl2, 10 mM dATP, dCTP, dGTP, and dTTP, 0.4 mM forward and reverse oligonucleotide primers, and 5 units of Taq polymerase in a final volume of 50 mL. Samples were maintained for 3 min at 95°C, subjected to 35 cycles (95°C for 30 s, 55°C for 30 s, and 72°C for 1 min), and then maintained for 7 min at 72°C. A second PCR was performed using 2003 dilutions of the first PCR products, E-T7 primer, and primer Up or Down to obtain the DNA template for the desired sense or antisense probe. The PCR was carried out with 1 mL of DNA dilution using the same protocol as in the first PCR. Amplification products were purified by electrophoresis (on a 1% agarose gel). After purification in ethanol, the fragments were used to generate sense or antisense digoxigeninlabeled transcripts by in vitro transcription (T7 MAXIScript Kit; Ambion). Final


purification of the probes was done using ethanol precipitation. Probes were labeled using the dIG Oligonucleotide Tailing Kit (Roche).

Tissue Preparation Root and leaf samples were fixed with 4% paraformaldehyde in phosphate buffer (0.2 M, pH 7.5) for 16 h. They were washed in phosphate buffer containing 0.1 M Gly and then dehydrated through a series of incubations in ethanol-butanol solutions: ethanol 50%, twice for 7 min; ethanol 70%, 30 min and then 1 h; ethanol 95%, twice for 30 min; butanol 100%, twice for 1 h; butanol 100%, 3 d. They were then progressively embedded in paraffin (ParaplastPlus; Structure Probes) through several baths: butanol:Safesolv (Labonord; 2:1), 1 h; butanol:Safesolv (1:2), 1 h; Safesolv:Paraplast (1:1), 1 h; Paraplast, twice for 1 h and then 48 h. After inclusion, samples were cut into 8-mm sections and mounted on silanized slides (Dako). Sections were subjected to dewaxing with Safesolv and rehydration through an ethanol series: ethanol 100%, twice for 10 min; ethanol 70%, twice for 1 min; ethanol 50%, 2 min; and two 2-min baths into diethyl pyrocarbonate (DEPC) sterile water. Slides were then subjected to protein digestion with proteinase K (0.1 unit mL21) for 30 min at 37°C, the reaction being stopped with an arrest buffer brought twice for 5 min at room temperature. Slides were washed with phosphate-buffered saline (PBS) plus 0.2% Gly for 2 min and then twice for 2 min in PBS only, then dehydrated through an ethanol series: ethanol 50%, 1 min; ethanol 70%, 1 min; ethanol 100%, twice for 1 min.

Hybridization Hybridization was done overnight at 45°C in a humidity chamber with a denatured (5 min at 65°C) hybridization mixture: 50% deionized formamide, 10% 103 SSC, 10% dextran sulfate, 2% Denhardt’s reagent, 0.1 mg mL21 tRNA, approximately 200 ng of the probe of interest, and DEPC water such that the final volume for one hybridization was 100 mL. Slides were washed (23 SSC twice for 5 min, 0.23 SSC twice for 30 min at 50°C, and 13 NaCl-TrisEDTA buffer [NTE] for 5 min at 37°C), subjected to RNase treatment (13 NTE, 20 mg mL21 RNase A for 30 min at 37°C), washing (13 NTE twice for 5 min at 37°C, 0.23 SSC twice for 30 min at 55°C), blocking (13 Blocking Reagent [Roche] for 30 min at 37°C), incubation with anti-digoxigenin antibody conjugated with alkaline phosphatase (1:500 dilution [Roche]; 1 h at 37°C), washing (PBS three times for 10 min at room temperature), and incubation in the color development buffer (100 mM Tris-HCl, pH 9.5, 100 mM NaCl) twice for 10 min at room temperature. Hybridization signals were detected with VectorBlue Kit III (Vector Laboratories) in color development buffer in the dark at room temperature (incubation until signal appears). Color development was stopped by incubation in water. Sections were then mounted in Mowiol. Slides were observed with a Leica DM6000 microscope under white light. Photographs were taken with a Retiga 2000R camera (QImaging), and images were processed through Volocity 4.0.1 (Improvision). In situ hybridization experiments have been conducted on the Plate-Forme d’Histocytologie et Imagerie Cellulaire Ve´ge´tale (http://phiv.cirad.fr/) using microscopes of the Montpellier Rio Imaging platform (www.mri.cnrs.fr).

Expression in Xenopus laevis Oocytes Rice HKT cDNAs were subcloned into the pGEMDG vector (D. Becker) downstream from the T7 promoter and between the 5# and 3# untranslated regions of the Xenopus b-globin gene. Capped and polyadenylated cRNAs were synthesized in vitro from linearized vector using the mMESSAGE mMACHINE T7 kit (Ambion). Oocytes were isolated as described previously (Ve´ry et al., 1995), injected with 50 ng of OsHKT cRNA in 50 nL of DEPCtreated water or with 50 nL of DEPC-treated water (for control oocytes), and then kept at 18°C in ND96 solution (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl2, 1 mM MgCl2, 2.5 mM Na-pyruvate, and 5 mM HEPES-NaOH, pH 7.4) supplemented with 0.5 mg L21 gentamycin until electrophysiological recordings.

Electrophysiology Whole oocyte currents were recorded using the two-electrode voltageclamp technique 1 to 3 d after cRNA injection. The voltage-clamp amplifier was an Axoclamp 2A (Axon Instruments). Voltage-pulse protocols, data

1969

Jabnoune et al.

acquisition, and data analyses were performed using pClamp9 (Axon Instruments) and Sigmaplot8 (Jandel Scientific) software. Both membrane potential and current were recorded. Correction was made for voltage drop through the series resistance of the bath and the reference electrode using a voltagerecording microelectrode in the bath close to the oocyte surface (potential difference between the local electrode and the reference subtracted from the potential of intracellular electrodes in the amplifier for real-time series resistance correction). All electrodes were filled with 1 M KCl. The oocyte was continuously perfused during the voltage-clamp experiment. All bath solutions contained a background of 6 mM MgCl2, 1.8 mM CaCl2, and 10 mM MES-1,3-bis[tris(hydroxymethyl)methylamino]propane, pH 5.5. Monovalent cations were added as Glu or chloride salts (chloride concentration constant in each set of solutions). D-Mannitol was added when necessary to adjust the osmolarity (same osmolarity in each set of solutions in the range 220–280 mosmol). In experiments where external pH was changed to 7.5, HEPES replaced MES in the background solution. K+ and Na+ actual concentrations in solutions were systematically measured by flame photometry. K+ and Na+ nominal concentrations were in the 2 to 25 mM range (see figure legends for detailed values). To extract OsHKT-mediated currents from total oocyte currents, mean currents recorded in water-injected control oocytes (n = 4–7) from the same batch in the same ionic conditions were subtracted from those recorded in OsHKT-expressing oocytes. Unless otherwise stated, presented current-voltage relationships were constructed with OsHKT-extracted currents. Sequence data from this article can be found in the GenBank/EMBL data libraries under accession numbers AB061311 (OsHKT2;1 mRNA), AJ491816 (OsHKT1;1 mRNA), and AJ491818 (OsHKT1;3 mRNA).

Supplemental Data The following materials are available in the online version of this article. Supplemental Figure S1. Insensitivity of OsHKT2;1, OsHKT1;1, and OsHKT1;3 to extracellular pH. Supplemental Figure S2. Phylogenetic relationships among HKT transporters.

ACKNOWLEDGMENTS We are grateful to Mike Blatt and Jean-Baptiste Thibaud for helpful discussions and to Julia Davies for critical reading of the manuscript. Received March 3, 2009; accepted May 23, 2009; published May 29, 2009.

LITERATURE CITED Apse MP, Aharon GS, Snedden WA, Blumwald E (1999) Salt tolerance conferred by overexpression of a vacuolar Na+/H+ antiport in Arabidopsis. Science 285: 1256–1258 Ban˜uelos MA, Garciadebla´s B, Cubero B, Rodrı´guez-Navarro A (2002) Inventory and functional characterization of the HAK potassium transporters of rice. Plant Physiol 130: 784–795 Berthomieu P, Cone´je´ro G, Nublat A, Brackenbury WJ, Lambert C, Savio C, Uozumi N, Oiki S, Yamada K, Cellier F, et al (2003) Functional analysis of AtHKT1 in Arabidopsis shows that Na+ recirculation by the phloem is crucial for salt tolerance. EMBO J 22: 2004–2014 Bihler H, Gaber RF, Slayman CL, Bertl A (1999) The presumed potassium carrier Trk2p in Saccharomyces cerevisiae determines an H+-dependent, K+-independent current. FEBS Lett 447: 115–120 Blatt MR, Rodrı´guez-Navarro A, Slayman CL (1987) Potassium-proton symport in Neurospora: kinetic control by pH and membrane potential. J Membr Biol 98: 169–189 Byrt CS, Platten JD, Spielmeyer W, James RA, Lagudah ES, Dennis ES, Tester M, Munns R (2007) HKT1;5-like cation transporters linked to Na+ exclusion loci in wheat, Nax2 and Kna1. Plant Physiol 143: 1918–1928 Clarkson DT, Hanson JB (1980) The mineral nutrition of higher plants. Annu Rev Plant Physiol 31: 239–298 Davenport RJ, Munõz-Mayor A, Jha D, Essah PA, Rus A, Tester M (2007)

1970

The Na+ transporter AtHKT1 controls xylem retrieval of Na+ in Arabidopsis. Plant Cell Environ 30: 497–507 Fairbairn DJ, Liu W, Schachtman DP, Gomez-Gallego S, Day SR, Teasdale RD (2000) Characterisation of two distinct HKT1-like potassium transporters from Eucalyptus camaldulensis. Plant Mol Biol 43: 515–525 Flowers TJ (1985) Physiology of halophytes. Plant Soil 89: 41–56 Garciadebla´s B, Senn ME, Ban˜uelos MA, Rodrı´guez-Navarro A (2003) Sodium transport and HKT transporters: the rice model. Plant J 34: 788–801 Gassmann W, Rubio F, Schroeder JI (1996) Alkali cation selectivity of the wheat root high-affinity potassium transporter HKT1. Plant J 10: 852–869 Gierth M, Ma¨ser P (2007) Potassium transporters in plants: involvement in K+ acquisition, redistribution and homeostasis. FEBS Lett 581: 2348–2356 Golldack D, Su H, Quigley F, Kamasani UR, Munõz-Garay C, Balderas E, Popova OV, Bennett J, Bohnert HJ, Pantoja O (2002) Characterization of a HKT-type transporter in rice as a general alkali cation transporter. Plant J 31: 529–542 Greenway H, Munns R (1980) Mechanisms of salt tolerance in nonhalophytes. Annu Rev Plant Physiol 31: 149–190 Hansen UP, Gradmann D, Sanders D, Slayman CL (1981) Interpretation of current-voltage relationships for “active” ion transport systems. I. Steady-state reaction-kinetic analysis of class-I mechanisms. J Membr Biol 63: 165–190 Haro R, Ban˜uelos MA, Senn ME, Barrero-Gil J, Rodrı´guez-Navarro A (2005) HKT1 mediates sodium uniport in roots: pitfalls in the expression of HKT1 in yeast. Plant Physiol 139: 1495–1506 Haro R, Rodrı´guez-Navarro A (2002) Molecular analysis of the mechanism of potassium uptake through the TRK1 transporter of Saccharomyces cerevisiae. Biochim Biophys Acta 1564: 114–122 Hasegawa PM, Bressan RA, Zhu JK, Bohnert HJ (2000) Plant cellular and molecular responses to high salinity. Annu Rev Plant Physiol Plant Mol Biol 51: 463–499 Hirsch RE, Lewis BD, Spalding EP, Sussman MR (1998) A role for the AKT1 potassium channel in plant nutrition. Science 280: 918–921 Horie T, Costa A, Kim TH, Han MJ, Horie R, Leung HY, Miyao A, Hirochika H, An G, Schroeder JI (2007) Rice OsHKT2;1 transporter mediates large Na+ influx component into K+-starved roots for growth. EMBO J 26: 3003–3014 Horie T, Yoshida K, Nakayama H, Yamada K, Oiki S, Shinmyo A (2001) Two types of HKT transporters with different properties of Na+ and K+ transport in Oryza sativa. Plant J 27: 129–138 Huang S, Spielmeyer W, Lagudah ES, James RA, Platten JD, Dennies ES, Munns R (2006) A sodium transporter (HKT7) is a candidate for Nax1, a gene for salt tolerance in durum wheat. Plant Physiol 142: 1718–1727 Kader MA, Seidel T, Golldack D, Lindberg S (2006) Expressions of OsHKT1, OsHKT2, and OsVHA are differentially regulated under NaCl stress in salt-sensitive and salt-tolerant rice (Oryza sativa L.) cultivars. J Exp Bot 57: 4257–4268 Kuroda T, Bihler H, Bashi E, Slayman CL, Rivetta A (2004) Chloride channel function in the yeast TRK-potassium transporters. J Membr Biol 198: 177–192 Laurie S, Feeney KA, Maathuis FJ, Heard PJ, Brown SJ, Leigh RA (2002) A role for HKT1 in sodium uptake by wheat roots. Plant J 32: 139–149 Lichtenberg-Frate H, Reid JD, Heyer M, Ho¨fer M (1996) TKHp in Schizosaccharomyces pombe. J Membr Biol 152: 169–181 Maathuis FJM (2006) The role of monovalent cation transporters in plant responses to salinity. J Exp Bot 57: 1137–1147 Maathuis FJM, Sanders D, Gradmann D (1997) Kinetics of high-affinity K+ uptake in plants, derived from K+-induced changes in current-voltage relationships: a modelling approach to the analysis of carrier-mediated transport. Planta 203: 229–236 Maathuis FJM, Verlin D, Smith FA, Sanders D, Fernańdez JA, Walker NA (1996) The physiological relevance of Na+-coupled K+-transport. Plant Physiol 112: 1609–1616 McCulloch SR, Beilby MJ, Walker NA (1990) Transport of potassium in Chara australis. II. Kinetics of a symport with sodium. J Membr Biol 115: 129–143 Meharg AA, Blatt MR (1995) NO32 transport across the plasma membrane of Arabidopsis thaliana root hairs: kinetic control by pH and membrane voltage. J Membr Biol 145: 49–66 Mu¨hling KH, Laüchli AE (2002) Effect of salt stress on growth and cation



compartmentation in leaves of two plant species differing in salt tolerance. J Plant Physiol 159: 137–146 Pardo JM, Cubero B, Leidi EO, Quintero FJ (2006) Alkali cation exchangers: roles in cellular homeostasis and stress tolerance. J Exp Bot 57: 1181–1199 Pilot G, Pratelli R, Gaymard F, Meyer Y, Sentenac H (2003) Five-group distribution of the Shaker-like K+ channel family in higher plants. J Mol Evol 56: 418–434 Platten JD, Cotsaftis O, Berthomieu P, Bohnert H, Davenport RJ, Fairbairn DJ, Horie T, Leigh RA, Lin HX, Luan S, et al (2006) Nomenclature for HKT transporters, key determinants of plant salinity tolerance. Trends Plant Sci 11: 372–374 Qi Z, Spalding EP (2004) Protection of plasma membrane K+ transport by the Salt Overly Sensitive1 Na+-H+ antiporter during salinity stress. Plant Physiol 136: 2548–2555 Ren ZH, Gao JP, Li LG, Cai XL, Huang W, Chao DY, Zhu MZ, Wang ZY, Luan S, Lin HX (2005) A rice quantitative trait locus for salt tolerance encodes a sodium transporter. Nat Genet 37: 1141–1146 Rodrı´guez-Navarro A (2000) Potassium transport in fungi and plants. Biochim Biophys Acta 1469: 1–30 Rodrı´guez-Navarro A, Rubio F (2006) High-affinity potassium and sodium transport systems in plants. J Exp Bot 57: 1149–1160 Rubio F, Gassmann W, Schroeder JI (1995) Sodium-driven potassium uptake by the plant potassium transporter HKT1 and mutations conferring salt tolerance. Science 270: 1660–1663 Sanders D, Hansen UP, Gradmann D, Slayman CL (1984) Generalized kinetic analysis of ion-driven cotransport systems: a unified interpretation of selective ionic effects on Michaelis parameters. J Membr Biol 77: 123–152 Santa-Marıá GE, Danna CH, Czibener C (2000) High-affinity potassium transport in barley roots: ammonium-sensitive and -insensitive pathways. Plant Physiol 123: 297–306 Schachtman DP, Schroeder JI (1994) Structure and transport mechanism of a high-affinity potassium uptake transporter from higher plants. Nature 370: 655–658


Sentenac H, Bonneaud N, Minet M, Lacroute F, Salmon JM, Gaymard F, Grignon C (1992) Cloning and expression in yeast of a plant potassium ion transport system. Science 256: 663–665 Shi H, Quintero FJ, Pardo JM, Zhu JK (2002) The putative plasma membrane Na+/H+ antiporter SOS1 controls long-distance Na+ transport in plants. Plant Cell 14: 465–477 Su H, Balderas E, Vera-Estrella R, Golldack D, Quigley F, Zhao C, Pantoja O, Bohnert HJ (2003) Expression of the cation transporter McHKT1 in a halophyte. Plant Mol Biol 52: 967–980 Sunarpi, Horie T, Motoda J, Kubo M, Yang H, Yoda K, Horie R, Chan WY, Leung HY, Hattori K, et al (2005) Enhanced salt tolerance mediated by AtHKT1 transporter-induced Na+ unloading from xylem vessels to xylem parenchyma cells. Plant J 44: 928–938 Uozumi N, Kim EJ, Rubio F, Yamaguchi T, Muto S, Tsuboi A, Bakker EP, Nakamura T, Schroeder JI (2000) The Arabidopsis HKT1 gene homolog mediates inward Na+ currents in Xenopus laevis oocytes and Na+ uptake in Saccharomyces cerevisiae. Plant Physiol 122: 1249–1259 Ve´ry AA, Gaymard F, Bosseux C, Sentenac H, Thibaud JB (1995) Expression of a cloned plant K+ channel in Xenopus oocytes: analysis of macroscopic currents. Plant J 7: 321–332 Ve´ry AA, Sentenac H (2003) Molecular mechanisms and regulation of K+ transport in higher plants. Annu Rev Plant Biol 54: 575–603 Walker NA, Sanders D, Maathuis FJM (1996) High-affinity potassium uptake in plants. Science 273: 997–998 Wang TB, Gassmann W, Rubio F, Schroeder JI, Glass ADM (1998) Rapid up-regulation of HKT1, a high-affinity potassium transporter gene, in roots of barley and wheat following withdrawal of potassium. Plant Physiol 118: 651–659 Yoshida S, Forno DA, Cock JH, Gomez KA (1976) Routine procedure for growing rice plants in culture solution. In S Yoshida, DA Forno, JH Cock, KA Gomez, eds, Laboratory Manual for Physiological Studies of Rice, Ed 2. International Rice Research Institute, Los Banõs, Philippines, pp 61–66

1971