Research ArticlePlant biology

Stomatal Closure by Fast Abscisic Acid Signaling Is Mediated by the Guard Cell Anion Channel SLAH3 and the Receptor RCAR1

See allHide authors and affiliations

Science Signaling  17 May 2011:
Vol. 4, Issue 173, pp. ra32
DOI: 10.1126/scisignal.2001346


S-type anion channels are direct targets of abscisic acid (ABA) signaling and contribute to chloride and nitrate release from guard cells, which in turn initiates stomatal closure. SLAC1 was the first component of the guard cell S-type anion channel identified. However, we found that guard cells of Arabidopsis SLAC1 mutants exhibited nitrate conductance. SLAH3 (SLAC1 homolog 3) was also present in guard cells, and coexpression of SLAH3 with the calcium ion (Ca2+)–dependent kinase CPK21 in Xenopus oocytes mediated nitrate-induced anion currents. Nitrate, calcium, and phosphorylation regulated SLAH3 activity. CPK21-dependent SLAH3 phosphorylation and activation were blocked by ABI1, a PP2C-type protein phosphatase that is inhibited by ABA and inhibits the ABA signaling pathway in guard cells. We reconstituted the ABA-stimulated phosphorylation of the SLAH3 amino-terminal domain by CPK21 in vitro by including the ABA receptor–phosphatase complex RCAR1-ABI1 in the reactions. We propose that ABA perception by the complex consisting of ABA receptors of the RCAR/PYR/PYL family and ABI1 releases CPK21 from inhibition by ABI1, and then CPK21 is further activated by an increase in the cytosolic Ca2+ concentration, leading to its phosphorylation of SLAH3. Thus, the identification of SLAH3 as the nitrate-, calcium-, and ABA-sensitive guard cell anion channel provides insights into the relationship among stomatal response to drought, signaling by nitrate, and nitrate metabolism.


Stomata, microscopic pores in the epidermis of plants, are each surrounded by a pair of guard cells. Turgor changes in these cells can open or close the stomata to balance the uptake of CO2 from the atmosphere and the concomitant loss of water from leaves (15). Influx of K+ ions triggers stomatal opening by increasing the osmotic pressure within the guard cells; stomatal closure is initiated by the activity of S-type anion channels, resulting in a release of anions and K+ from guard cells. The decrease in guard cell turgor and volume leads to stomatal closure, reducing transpirational loss of water from the leaf.

Chloride (Cl) and malate are the primary anions that counterbalance the influx of K+ during stomatal opening. Current models propose that anion transport during stomatal opening is primarily mediated by chloride transporters (68). However, nitrate (NO3) may also directly contribute to stomatal movement (9). For example, CHL1, which encodes a plasma membrane nitrate transporter (10, 11), is required for stomatal opening. Anion channels that contribute to fast stomatal closure in response to drought stress exhibit a high permeability for nitrate over chloride, with a ratio [P(NO3)/P(Cl)] of ~21 (12).

Abscisic acid (ABA) is a plant stress hormone that is produced in response to drought. In guard cells, ABA triggers a fast response and a slow response. The fast ABA response mediates “fast” stomatal closure within seconds or a few minutes by the activation of S-type anion channels and the inhibition of inward-rectifying K+ channels, such as KAT1 (13, 14). Over the course of several minutes to hours, the “slow” ABA signaling pathway, which involves bZIP (basic leucine zipper) transcription factors of the AREB and ABF families, leads to the induction of genes involved in long-term drought tolerance (15). This slow ABA signaling pathway has been identified (16) and involves a plasma membrane ABC [adenosine triphosphate (ATP)–binding cassette] transporter for ABA uptake into guard cells (17). In contrast, the fast stomatal ABA signal transduction pathway is still fragmentary. The initial steps in this fast ABA signaling pathway activate guard cell anion channels in a calcium-dependent and -independent manner (13, 1822). The SLAC1 gene encodes a plant plasma membrane nitrate-permeable S-type anion channel with a P(NO3)/P(Cl) ratio of 10 (23). ABA activates SLAC1 by stimulating its phosphorylation by the Ca2+-independent protein kinase OST1, as well as by Ca2+-dependent kinases of the AtCPK family (2224). Although guard cells of SLAC1 loss-of-function mutants measured in chloride-based buffers had reduced S-type anion currents, the stomata of the slac1-3 mutant still responded to day-to-night transitions (25, 26).

We report that, in addition to SLAC1, the gene encoding SLAC1 homolog 3, SLAH3, is expressed in Arabidopsis guard cells. Characterization of the channels expressed in Xenopus oocytes indicated that both channels were activated by a distinct set of protein kinases and were stimulated by ABA through the ABA receptor–phosphatase complex RCAR1-ABI1. Binding of ABA to this receptor-phosphatase complex inhibits the activity of the phosphatase (27, 28). In contrast to SLAC1, we found that SLAH3 resembled a voltage-dependent, nitrate-activated S-type anion channel with a P(NO3)/P(Cl) ratio of 20. These results suggest that together with SLAC1, SLAH3 is responsible for stomatal closure upon drought stress. Because SLAH3 required nitrate for activation and was stimulated by ABA, we propose that SLAH3 in guard cells might connect the stomatal response to drought stress with nitrate metabolism and nitrate signaling.


SLAH3 is present in guard cells and mediates S-type currents in the presence of nitrate

Guard cell protoplasts of slac1-3 loss-of-function plants exhibit reduced S-type anion currents in the presence of chloride (23), but normal anion currents appeared in the presence of 3 mM nitrate (Fig. 1A). To identify the missing NO3-permeable component of the S-type currents, we performed quantitative real-time polymerase chain reaction (PCR) for SLAC1 and the SLAC1 homologous (SLAH) anion channel genes SLAH1, SLAH2, SLAH3, and SLAH4 (25). We detected SLAC1 in guard cells (Fig. 1B), which is consistent with previous reports (25, 26). In addition to SLAC1, we, but not Negi et al. (25), found that SLAH3 was expressed in guard cells (Fig. 1B). SLAH3 expression was weaker in guard cells than in mesophyll cells. With the exception of a weak signal for SLAH2 transcripts, we did not detect transcripts for SLAH1 or SLAH4 in guard cells or mesophyll cells. To confirm our results, we also analyzed three samples of manually collected, contamination-free guard cell protoplasts and found SLAH3 together with SLAC1 in guard cells (fig. S1). We additionally verified that SLAH3 was present in guard cells and mesophyll cells of plants grown under our conditions using the SLAH3-promoter:GUS lines (25), which showed GUS staining of guard cells and mesophyll cells (fig. S1).

Fig. 1

SLAH3, a nitrate-dependent S-type anion channel in Arabidopsis guard cells. (A) Steady-state current densities (ISS/Cm) of slac1-3 guard cell protoplasts plotted against the clamped voltages. Experiments were performed with chloride-based media in the absence and presence of 3 mM external nitrate. Data points represent means ± SE of a total of seven and eight experiments (at least three independent ones) in the absence and presence of nitrate. ISS, steady-state current; Cm, membrane capacitance. (B) Quantification of the abundance of the indicated transcripts in guard cells and mesophyll cells. Transcripts were normalized to 10,000 molecules of actin2/8 with standard curves calculated for the individual PCR products before transcript quantification. The length of the individual PCR fragments was considered in the calculation (n.d., not detectable; n = 4 experiments, mean ± SD). rel., relative. (C) Abundance of the indicated transcripts in slac1-3 guard cells and wild-type (WT) guard cells (n = 5 experiments, mean ± SD; *P < 0.01). (D) Macroscopic guard cell anion currents recorded from WT and slah3-1 mutant plants in nitrate-based media. Representative current responses to 7.5-s voltage pulses from +44 to −136 mV in 30-mV decrements are shown. (E and F) Steady-state current densities of slah3-1 and WT guard cell protoplasts (E) and slac1-3 and WT guard cell protoplasts (F). In (E) and (F), whole-cell currents were measured with nitrate-based media in the pipette (162 mM) and the bath solution (36 mM). Data points represent means ± SE. The number of experiments (at least three independent) was five for slah3-1 and six for WT in (E) and six for slac1-3 and WT in (F).

Moreover, in the slac1-3 mutant, the SLAH3 transcripts appeared to be increased by a factor of 2 (Fig. 1C). In contrast to SLAH3, the abundance of the transcripts for the guard cell potassium release channel, encoded by GORK (29, 30), remained unchanged in the slac1-3 mutant relative to the wild type (Fig. 1C). Thus, in the presence of nitrate, SLAH3 might compensate for the loss of SLAC1, which is consistent with a study by Negi et al. (25), showing that expression of SLAH3 under the control of the slac1 promoter rescued the SLAC1 knockout phenotype.

Patch-clamp experiments with nitrate-based media of guard cells in slah3-1 loss-of-function plants revealed that anion current amplitudes in guard cells were reduced relative to the wild type (Fig. 1, D and E). Under the same conditions, guard cell protoplasts of the SLAC1 mutant (slac1-3) mediated currents similar to those in wild-type protoplasts (Fig. 1F), indicating that SLAH3 accounted for the guard cell S-type anion currents in the slac1-3 mutant. Besides this electrical phenotype, we did not discover any obvious differences in growth and development under standard greenhouse conditions.

In the presence of nitrate, SLAH3 mediates S-type anion currents when coexpressed with CPK21 in Xenopus oocytes

We expressed SLAH3 in Xenopus laevis oocytes to functionally characterize the putative nitrate channel in the absence of other SLAH or SLAC channels. SLAC1 is activated in a calcium-dependent manner by Ca2+-dependent protein kinases of the CPK family (for example, CPK21), as well as in a Ca2+-independent manner by the protein kinase OST1 (23). To assess whether these kinases also regulate SLAH3, we monitored the activity of the channel in the presence of the kinases OST1, CPK21, or CPK23 in oocytes with the double-electrode voltage-clamp technique (22, 23). We used truncated CPK variants (CPK21ΔEF and CPK23ΔEF) lacking a junction and calcium-binding domain, which renders these kinases Ca2+-independent and constitutively active (3133), to avoid the influence of different kinase-specific Ca2+ sensitivities on SLAH3 activation.

Upon oocyte injection with SLAH3 alone, we could not detect anion currents in nitrate-based media (Fig. 2, A and B). Coexpression of SLAH3 with OST1 also did not result in S-type anion currents (Fig. 2B). However, when SLAH3 was coexpressed with CPK21ΔEF or CPK23ΔEF, typical S-type anion currents were detected (Fig. 2, A and B). We further focused on the activation of SLAH3 by CPK21 because coexpression of CPK21ΔEF evoked larger SLAH3-mediated anion currents than did coexpression with CPK23ΔEF (Fig. 2B). In 3 mM external NO3, mimicking the high-nitrate nutrition status in the xylem sap and around the guard cells (34), the application of 20-s voltage pulses starting from a holding potential of 0 mV led to steady-state SLAH3 currents (ISS) that deactivated at negative membrane potentials (Fig. 2A). These slow deactivation characteristics (Fig. 2A) are reminiscent of S-type anion currents in intact guard cells and of currents present in oocytes expressing SLAC1 (23, 35). We examined the currents of SLAH3 in oocytes expressing both SLAH3 and CPK21ΔEF in buffers containing chloride or chloride and nitrate (Fig. 2C). Currents were low in the presence of 3 mM extracellular chloride in the absence of nitrate. Relative to the current amplitude detected in the presence of solutions containing 3 mM nitrate in combination with either 3 or 100 mM chloride, the SLAH3 current amplitude was substantially reduced at negative membrane potentials in the presence of only 100 mM chloride (Fig. 2C). The reversal potential was shifted to negative membrane potentials with 100 mM chloride relative to the currents detected with low chloride concentrations (3 mM) (Fig. 2C), indicating that SLAH3 is permeable to chloride. The voltage-activation threshold of SLAH3, however, remained unaffected by chloride but was substantially shifted to negative membrane potentials in the presence of NO3 (Fig. 2D), indicating that gating is modified by nitrate. In contrast, chloride buffers did not reduce CPK23ΔEF-activated SLAC1 currents in oocytes (fig. S2, A and B).

Fig. 2

Electrical properties of SLAH3 coexpressed with CPK21ΔEF in Xenopus oocytes. (A) Whole-oocyte current recordings in standard bath solution with or without 3 mM NO3 (indicated by +NO3 or −NO3; pH 7.5) upon 20-s voltage pulses ranging from +40 to −180 mV in 20-mV decrements followed by a 3-s voltage pulse to −120 mV. The holding potential was 0 mV. No current responses of oocytes expressing CPK21ΔEF or SLAH3 alone could be recorded. Coexpression of CPK21ΔEF with SLAH3 resulted in macroscopic anion currents slowly deactivating at negative membrane potentials in the presence of nitrate only. Recordings from representative cells are shown. (B) Instantaneous currents (IT) recorded at −100 mV of oocytes expressing SLAH3 in the presence of CPK21ΔEF, CPK23ΔEF, OST1, or a kinase-deficient mutant (D204A) of CPK21ΔEF in nitrate-based buffers (n = 5 experiments, mean ± SD). (C) NO3 and voltage dependence of steady-state currents (ISS) of SLAH3- and CPK21ΔEF-coexpressing oocytes. Currents were normalized to +40 mV in 3 mM NO3 + 100 mM Cl and plotted against the membrane potential. With 3 mM NO3 in the bath medium, SLAH3-mediated currents could be observed, which decreased at negative membrane potentials. Addition of 100 mM Cl to the bath medium in the presence of 3 mM NO3 shifted the reversal potential (Vrev) to negative voltages relative to the situation with 3 mM NO3 alone or 3 mM Cl (n = 5 experiments, mean ± SD). (D) Relative voltage-dependent open probabilities (rel. PO) of SLAH3- and CPK21-mediated currents under conditions used in (C). Rel. PO was calculated from a −120-mV voltage pulse following the test pulses in the voltage range +60 to −200 mV in 20-mV decrements. Data points were fitted by a Boltzmann equation (continuous line; n = 5 experiments, mean ± SE). (E) Slow activation kinetics of SLAH3-mediated anion currents. SLAH3 was activated by test pulses in the range +40 to −200 mV in 20-mV decrements from a holding potential of −180 mV in the absence or presence of the indicated concentrations of NO3 at depolarized membrane potentials. Fitting the current response at +40 mV and an external NO3 concentration of 10 mM with a single-exponential equation yielded the indicated relaxation of current activation time constant (τact), and fitting the current response at −180 mV and an external NO3 concentration of 10 mM yielded the indicated deactivation time constant (τdeact) (n = 5 experiments, mean ± SD). (F) Reversal potentials Vrev of SLAH3- and CPK21ΔEF-expressing oocytes are shown as a function of the logarithmic external nitrate concentration. As expected for an anion-selective channel, the reversal potential shifted with a slope of 52 mV per decade to more negative values with increasing anion concentrations (n ≥ 4 experiments, mean ± SD).

Stepwise alteration of the membrane potential in 20-mV increments from a holding voltage of −180 mV (mimicking the resting potential of open stomata guard cells) to depolarized membrane potentials [mimicking ABA action (13)] revealed SLAH3 activation with a slow half-activation time of τact ~10 s at +40 mV (Fig. 2E). In contrast, channel deactivation kinetics (τdeact) appeared 10 times faster at −180 mV. Reduction of the nitrate concentration in the bath shifted the reversal potential (Vrev) to positive membrane voltages. A factor of 10 change in the external nitrate concentration resulted in a 52-mV shift of the reversal potential (Fig. 2F).

SLAH3 is activated by extracellular nitrate and shows strong selectivity for nitrate

We examined the permeability of SLAH3 for physiological anions relative to nitrate by replacing external NO3 with Cl, SO42−, HCO3, or the dicarboxylate malate (Fig. 3A). The obtained relative anion permeability sequence of NO3 1 >> Cl 0.05 ± 0.01 > malate 0.01 ± 0.01, HCO3 0.01 ± 0.00, SO42− 0.01 ± 0.00 characterized SLAH3 as a preferentially NO3-permeable channel that is 20 times more permeable to NO3 than to Cl (or the other anions tested) (12).

Fig. 3

Selectivity and nitrate dependence of SLAH3. (A) Relative permeability of SLAH3 coexpressed with CPK21ΔEF in Xenopus oocytes (permeability for NO3 was set to 1). Standard bath solution contained 50 mM of the respective anion (pH 5.6) (n = 5 experiments). (B) NO3 dependence of steady-state currents (ISS) of SLAH3- and CPK21ΔEF-coexpressing oocytes was normalized to the values at +40 mV in 10 mM NO3 and plotted against the membrane potential. Currents were measured at the indicated voltages in the presence of the indicated concentrations of external NO3 concentrations and 3 mM Cl (n = 4 experiments, mean ± SD). (C) The relative open probability (rel. PO) of SLAH3 in various NO3 concentrations was plotted against the membrane potential. Data points were fitted with a single Boltzmann equation (solid lines, n = 4 experiments, mean ± SE). (D) The half-maximal PO (V1/2) calculated from the data in (C) was plotted against the NO3 concentration. A Michaelis-Menten equation was used to calculate K0.5 = 8.3 mM NO3 (n = 4 experiments, mean ± SE).

Measurements of currents in SLAH3- and CPK21ΔEF-coexpressing oocytes exposed to increasing external nitrate concentrations ([NO3]) revealed a shift of the peak efflux current and the relative open probability toward negative membrane potentials (Fig. 3, B and C). These observations indicate that under the physiological [NO3] of 2 to 5 mM (34), this signaling metabolite is not only conducted by the channel, but can also increase the activity of SLAH3 [similar to the activation of animal ClC channels by Cl (36)] and thereby increase the plasma membrane anion conductance. The nitrate sensitivity (K0.5 = 8.3 mM) of the SLAH3 gate (Fig. 3D) was also close to the range of physiological concentrations present in the extracellular fluid of plants. In contrast, the voltage gate of SLAC1 was much less sensitive to external nitrate than to chloride, resulting in a shift of the SLAC1 relative open probability to more negative membrane potentials only at nitrate concentrations above 10 mM (K0.5 = 117 mM; fig. S2, C to F). These findings indicate that nitrate shifted the voltage-dependent open probability (PO) of SLAH3 toward the physiological resting potentials (−180 mV) of guard cells with open stomata (Figs. 2D and 3, C and D) (13).

Bimolecular fluorescence complementation assays reveal a direct interaction between SLAH3 and CPK21 and between CPK21 and ABI1/2

Transpiration is reduced by KNO3 and ABA in a synergistic manner (37). Furthermore, nitric oxide (NO) generated by nitrate reductase in guard cells appeared to affect ABA-dependent stomatal closure (38). Experiments with ABA-insensitive abi1-1 and abi2-1 mutants revealed that the type 2C protein phosphatases (PP2Cs) ABI1 and ABI2 are downstream of NO in the ABA signal transduction cascade (38). To study the interaction of SLAH3 with ABA signaling components, we used the bimolecular fluorescence complementation (BiFC) technique. Upon expression of SLAH3::YFPC (yellow fluorescent protein C terminus) alone or with YFPN (yellow fluorescent protein N terminus) (fig. S3, A and B), no specific YFP fluorescence was emitted from oocytes. Similar to the interaction detected between SLAC1 and OST1, CPK21, or CPK23 with this technique (22, 23), when SLAH3::YFPC and YFPN fused to CPK21 were coexpressed, YFP fluorescence indicated the direct interaction between SLAH3 and CPK21 (Fig. 4A). BiFC experiments in mesophyll protoplasts confirmed those from the oocyte system, indicating direct interaction of CPK21 with SLAH3 at the plasma membrane (Fig. 4B).

Fig. 4

BiFC-based interaction of SLAH3 and its regulatory components in oocytes and mesophyll protoplasts. (A) SLAH3::YFPC coexpressed with full-length CPK21::YFPN in oocytes. (B) SLAH3::YFPC coexpressed with CPK21::YFPN in mesophyll protoplasts. (C) ABI11::YFPC coexpressed with CPK21::YFPN in oocytes. (D) CPK21::YFPN coexpressed with ABI1::YFPC in mesophyll protoplasts. In (A) to (D), yellow fluorescence indicates an interaction. Representative images of at least three independent experiments are shown. In (A) and (C), one-quarter of an oocyte is shown. In (C) and (D), red represents the autofluorescence of chloroplasts. (E) Instantaneous currents recorded at −100 mV of oocytes expressing SLAH3 alone (None), SLAH3 with CPK21 (CPK21), or SLAH3 with CPK21 and ABI1 (CPK21 ABI1) (n = 3 experiments, mean ± SD).

Because ABI1 and ABI2 (referred to collectively as ABI1/2) interact with SLAC1 and OST1 (23), we coexpressed CPK21 together with ABI1 or ABI2 and detected an interaction between CPK21 and these phosphatases (Fig. 4C and fig. S3F). BiFC experiments with ABI1 and CPK21 performed in Arabidopsis protoplasts revealed that this complex was targeted to the plasma membrane (Fig. 4D). Most likely, the plasma membrane association of the CPK21-ABI1 complex is mediated by S-acylation and myristoylation at the N terminus of the kinase (39, 40). Because ABI1/2 inhibit guard cell responses to nitrate, we coexpressed SLAH3, CPK21, and ABI1; analyzed SLAH3 activation in nitrate buffers; and found that ABI1 abolished SLAH3-mediated anion currents in oocytes (Fig. 4E).

CPK21 phosphorylates SLAH3

To test whether phosphorylation was essential for SLAH3 activity, we impaired the CPK21 kinase activity by exchanging Asp204 for an Ala (CPK21ΔEFD204A) (41). Coexpression of SLAH3::YFPC and CPK21ΔEFD204A::YFPN resulted in YFP fluorescence in the oocyte BiFC assay (fig. S2E), indicating that the two proteins interacted. However, SLAH3-mediated anion currents were not detected in oocytes coexpressing this kinase-inactive CPK21 mutant (Fig. 2B). We performed in vitro kinase assays with recombinant CPK21 and either the N- or the C-terminal domains of SLAH3 (amino acids 1 to 257 and 563 to 635, respectively). Using radiolabeled [γ-32P]ATP, we found that CPK21 exclusively phosphorylated the N terminus of SLAH3 (Fig. 5A). We detected the largest anion currents from SLAH3- and CPK21-expressing oocytes when SLAH3 was coexpressed with the truncated CPK21 variant lacking its junction and calcium-binding domain (CPK21ΔEF) (3133), and a much smaller current in the presence of wild-type CPK21 (Fig. 5D). Because this difference may be due to the Ca2+ dependency of the kinase activity of CPK21, we performed in vitro kinase assays with CPK21 and the SLAH3 N terminus in the presence of various concentrations of Ca2+ (Fig. 5B) and found that CPK21 phosphorylated SLAH3 in a Ca2+-dependent manner. To quantify the CPK21 phosphorylation activity, we counted the radioactive decay of excised phosphorylated N-terminal SLAH3 bands and fitted the calcium-dependent phosphorylation to the Hill equation. This analysis revealed a half-maximal kinase activity of 127.7 ± 7.8 nM Ca2+ and a Hill coefficient of 4 (Fig. 5C), which is consistent with the known properties of the four EF-hands in the C-terminal domains of CPKs (31, 32, 39) and explains the smaller SLAH3 currents detected in the presence of wild-type CPK21 in the oocyte system, where resting calcium concentrations are around 100 nM (42).

Fig. 5

CPK21 phosphorylates the N terminus of SLAH3 in a Ca2+-dependent manner. (A) Phosphorylation of SLAH3 N terminus (NT) or C terminus (CT) by CPK21 in the presence of radiolabeled [γ-32P]ATP. Arrowheads indicate the position of recombinant proteins (left panels: Coomassie-stained SDS-PAGE; right panels: autoradiogram of the gel; the presence of proteins in the reaction assay is indicated by +). A representative of three experiments is shown. (B) Ca2+ dependence of SLAH3 NT phosphorylation by CPK21. Arrowheads indicate the position of recombinant proteins (left panels: Coomassie-stained SDS-PAGE; right panels: autoradiogram of the gel; free Ca2+ concentrations are indicated). In (A) and (B), the SLAH3 NT and CT were partially degraded. (C) In vitro kinase assays like the ones shown in (B) were performed and analyzed by a Hill equation to reveal an apparent Ca2+ affinity of CPK21 of 127 nM and a Hill coefficient of 4, reminiscent of the four EF-hands in the C terminus of CPKs. Data represent means ± SE, n = 5 experiments. Relative phosphorylation activity was normalized to the maximal activity, which was obtained by fitting the original data of each single experiment with the Hill equation. (D) Comparison of SLAH3 activation by full-length CPK21 (Ca2+-dependent) or the constitutively active (Ca2+-independent) truncation mutant CPK21ΔEF in oocytes (n = 3 experiments, mean ± SD). (E) Instantaneous currents (IT) recorded at −150 mV of the indicated phosphorylation site mutants of SLAH3 and wild-type SLAH3 in the absence of coexpressed CPK21 in oocytes. Aspartate mutants should mimic phosphorylation (n = 4 experiments, mean ± SD).

To identify the phosphorylation hot spot within the SLAH3 N terminus, we used peptide arrays of the cytosolic N- and C-terminal parts within the SLAH3 protein. The arrays consisted of 20–amino acid oligomer peptides with an overlap of 10 amino acids (22). After 2 hours of incubation with recombinant CPK21 and radiolabeled [γ-32P]ATP, we identified one region of the SLAH3 N terminus that was phosphorylated by CPK21 (position from Ser184 to Ser202, table S1). Within this region, we identified four putative phosphorylation sites: Ser184, Thr187, Ser189, and Thr197. When the native residues were separately replaced by an aspartate, to mimic phosphorylation, only the SLAH3 mutant T187D was fully active when expressed in oocytes without coexpression of CPK21 (Fig. 5E), which suggests that phosphorylation of Thr187 may be sufficient for SLAH3 activation.

In vitro ABA promotes SLAH3 phosphorylation in the presence of ABI1 and RCAR1

The cytosolic RCAR (also known as PYR or PYL) ABA receptor family inhibits the activity of PP2Cs in an ABA-dependent manner (27, 28). We assayed CPK21-dependent phosphorylation of the SLAH3 N-terminal domain in the presence of RCAR1 and ABI1 to reconstitute in vitro this part of the fast, posttranslational ABA signaling pathway of guard cells. We performed in vitro kinase assays in the presence or absence of ABA with RCAR1, ABI1, CPK21, and the N terminus of SLAH3. Consistent with the inhibition of CPK21- and SLAH3-derived anion currents by ABI1 in oocytes (Fig. 4E), ABI1 inhibited the phosphorylation of the SLAH3 N-terminal domain by CPK21 in vitro, and this inhibitory effect was not alleviated by the addition of ABA (Fig. 6A). ABA had no effect on the phosphorylation of the SLAH3 N terminus in the presence of RCAR1 and CPK21 without ABI1. However, when both ABI1 and RCAR1 were present with CPK21 and the SLAH3 N terminus, the addition of ABA promoted the phosphorylation of the SLAH3 N terminus (Fig. 6A).

Fig. 6

Reconstitution of phosphorylation of the SLAH3 N-terminal domain in response to ABA signaling in vitro. (A) In vitro kinase assays of the N-terminal domain of SLAH3 (NT) in the presence or absence of ABA in the presence of the indicated proteins. (B) Testing for the action of ABI1 on CPK21 or SLAH3 with in vitro kinase assays of SLAH3 NT. In lanes 1 and 2, the phosphorylation reaction was performed in the presence of radiolabeled ATP in the presence or absence of ABI1 for 10 min. In lane 3, the phosphorylation reaction was performed in the presence of an excess of ATP-γ-S or unlabeled ATP or the kinase inhibitor staurosporine also for 10 min. Inhibitor indicates the presence of any of these compounds. In lane 4, the SLAH3 NT was in vitro–phosphorylated in the presence of radiolabeled ATP for 10 min, and then the indicated inhibitor and ABI1 were added and the samples were analyzed after 45-min incubation at room temperature. Note that the SLAH3 NT was partially degraded. Data in (A) and (B) are representative of three or more experiments. (C) Arabidopsis leaf mesophyll protoplasts from cpk21-1 (Salk 029412) mutant plants were transiently transformed to coexpress Strep-tagged SLAH3 and Strep-tagged CPK21, or Strep-tagged CPK21 D204A (kinase-inactive control) and 30 μM ABA (+) or an equal volume of buffer (−) were added for 40 min. After Strep affinity purification, isolated proteins were analyzed by Pro-Q Diamond phosphoprotein staining (upper panel for SLAH3), by Sypro Ruby protein staining (middle panel for SLAH3), or by Western blot with antibodies that recognize the Strep tag (α-strep) (lower panel for CPK21). The results of the experiment were quantified with an imaging system and normalized to the respective amount of each protein and are shown in the graph. (D) The fold increase in SLAH3 phosphorylation in response to CPK21 or the presence of ABA was determined as described in (C). Left, increase in SLAH3 phosphorylation in protoplasts coexpressing active CPK21 relative to that in protoplasts coexpressing CPK21D204A in the presence of ABA. Right, increase in SLAH3 phosphorylation in protoplasts coexpressing CPK21 in the presence of ABA relative to that in the absence of ABA. Data represent means ± SD for three independent experiments.

In vitro kinase assays revealed that phosphorylation of the SLAH3 N-terminal region was inhibited in the presence of ABI1 (Fig. 6, A, lanes 1 to 4, and B, lanes 1 and 2). The inhibitory effect of ABI1 appears weaker in Fig. 6A than in Fig. 6B due to the conditions required for the reconstitution of ABA-dependent regulation. To elucidate whether ABI1 inhibited the activity of CPK21 or dephosphorylated the N terminus of SLAH3, we initially phosphorylated the SLAH3 N-terminal domain with CPK21 in the presence of [γ-32P]ATP, and then added excess adenosine 5′-O-(3-thiotriphosphate) (ATP-γ-S), a nonhydrolyzable ATP analog, or unlabeled ATP. Phosphorylation in the presence of ATP-γ-S produces a bond that is resistant to hydrolysis by phosphatases, and, thus, if the SLAH3 N-terminal domain was directly dephosphorylated by ABI1, the amount of radiolabeled SLAH3 N-terminal domain would be expected to decrease as the labeled ATP was replaced with nonhydrolyzable ATP-γ-S. Addition of unlabeled ATP would result in exchange with radiolabeled [γ-32P]ATP and also decrease the amount of radiolabeled SLAH3 N-terminal domain. However, if ABI1 inhibited CPK21, then the phosphorylated SLAH3 N-terminal domain would remain radiolabeled, because CPK21 would have already phosphorylated the SLAH3 N-terminal domain before the addition of ATP-γ-S or unlabeled ATP together with ABI1. A reduction in the amount of radiolabeled SLAH3 N-terminal domain in the presence of ABI1 required a prolonged reaction time of 45 min (Fig. 6B), suggesting that SLAH3 was not a direct target of ABI1 and that ABI1 more likely inhibited the activity of CPK21. Consistent with ABI1 inhibiting CPK21, the radiolabeled SLAH3 N-terminal domain was also only reduced after prolonged exposure to the kinase inhibitor staurosporine in the presence of ABI1.

CPK21 is responsible for SLAH3 phosphorylation in vivo

To confirm a functional relationship between CPK21 and SLAH3 in planta, we transiently coexpressed the tagged N-terminal domain of SLAH3 with either wild-type CPK21 or the kinase-inactive form (CPK21 D204A). In the absence of ABA or after the addition of ABA to protoplasts for 40 min, proteins were affinity-purified from crude extracts. The extent of SLAH3 phosphorylation normalized to the respective protein amount was determined after analysis by Pro-Q Diamond phosphoprotein versus SYPRO Ruby staining. The abundance of CPK21 was confirmed by Western blot analysis (Fig. 6C). We observed a factor of ~2 increase in the extent of phosphorylation of the SLAH3 N-terminal domain in the presence of ABA when active CPK21 was coexpressed (Fig. 6, C and D). No increase in the phosphorylation of the SLAH3 N-terminal domain was detected in the absence of ABA or upon coexpression of the kinase-inactive CPK21 in the presence of ABA. Thus, in plant tissue, CPK21-mediated phosphorylation of the SLAH3 N-terminal domain was promoted by ABA, which inhibits the activity of the phosphatase ABI.


We reconstituted in vitro the portion of the fast ABA signaling pathway in guard cells from ABA perception to SLAH3 anion channel phosphorylation, which activated the channel. On the basis of this work and studies with SLAC1, we propose that ABA sensed by RCAR1 inhibits the phosphatases ABI1/2 to promote the phosphorylation and activation of the anion channels SLAC1 and SLAH3, finally leading to stomatal closure (Fig. 7).

Fig. 7

Schematic illustration of the fast ABA signaling pathway leading to SLAH3 and SLAC1 activation in guard cells. In the absence of ABA, the protein phosphatases ABI1 and ABI2 inhibit autophosphorylation of OST1 and CPKs, rendering those kinases inactive. Upon perception of ABA by RCAR/PYR/PYL ABA receptors (indicated as RCAR1), ABI1 and ABI2 are bound and inactivated by these cytosolic ABA receptors, relieving the inhibition of OST1 and CPK. SLAH3 is phosphorylated by calcium-dependent CPK21, and SLAC1 is activated by both calcium-independent OST1 and calcium-dependent CPKs. In contrast to SLAC1, SLAH3 activity requires extracellular nitrate in addition to phosphorylation. OST1 also inhibits the activity of the voltage-dependent inward-rectifying K+ channel KAT1 (14). Anion efflux through SLAH3 and SLAC1 leads to depolarization of the guard cell membrane potential (13) and, in turn, outward rectifying potassium (29, 30) channels release K+, leading to loss of turgor pressure in the guard cells and stomatal closure.

Although we did not detect a stomatal phenotype besides the lack of nitrate-activated anion currents under our experimental conditions, it remains to be determined if there are conditions under which SLAH3 has a prominent role in regulating stomatal closure. Additionally, it appears that growth conditions may influence the abundance of SLAH3. We detected SLAH3 expression in guard cells with three independent methods, whereas Negi et al. (25) failed to detect SLAH3 expression in guard cells. We speculate that these differences may be related to both methodological differences and different growing conditions for the plants.

Nitrate concentration, ABA signaling, and water status control stomata-dependent hydraulic properties of the plant shoot (43). Here, we show that SLAH3 is a voltage-dependent, nitrate-activated plasma membrane anion channel present in Arabidopsis guard cells. Together with SLAC1, SLAH3 seems to represent the anion release pathway required for stomatal closure. Both channels, which differ in their Cl/NO3 permeability, are activated by a distinct set of protein kinases and are stimulated by ABA. We found that SLAH3 was activated by ABA through the RCAR1-ABI1 ABA receptor–phosphatase complex and the calcium-dependent kinase CPK21 (Fig. 7). In addition to phosphorylation, SLAH3 required an increase in extracellular nitrate concentrations for full channel opening (Figs. 2, A, C, and D, and 3, B to D). Nitrate, which is a major plant nutrient and indicator of nitrogen status (11, 44, 45), activated SLAH3 by shifting the voltage-dependent gate of SLAH3 closer to the resting potential of guard cells. Thus, we propose that the guard cell nitrate channel SLAH3, which is activated by ABA and NO3 signaling pathways, contributes to closing of stomata, the plant water gates.

Nitrate status also triggers a slow cellular response involving changes in gene expression. CHL1 transports nitrate and senses the extracellular nitrate concentrations through a mechanism that is independent from its transport function (11). Dependent on the availability of nitrate, CHL1 activates signaling cascades that ultimately lead to changes in transcription (11). In the presence of nitrate, CHL1 in guard cells promotes stomatal opening (10). In contrast, SLAH3 represents an effector of a short and fast signaling pathway that mediates stomatal closure. In the presence of nitrate, SLAH3 may be primed for the activation upon drought stress. How CHL1 and SLAH3 coordinate stomatal movement to nitrogen sensing, uptake, and metabolism remains to be investigated.

Materials and Methods

Plant material

Arabidopsis thaliana ecotype Columbia (Col-0), slah3-1 (GABI Kat371G03), slac1-3 (Salk 099139), and cpk21-1 (Salk 029412) mutant plants were grown on soil in a growth chamber at an 8-hour day/16-hour night regime and 22°C/16°C day/night temperature for 6 to 8 weeks. Gene- and T-DNA–specific primers for the slah3-1 (GABI Kat371G03) mutant line were designed to verify the T-DNA insertion within the third exon of SLAH3 and homozygosity of slah3-1 mutant plants. The following primer pairs were used: LB-GK2 (5′-CCC ATT TGG ACG TGA ATG TAG ACA C-3′) in combination with SLAH3-1gsp-s (5′-TCT TAC TTG ATT CCC CTG AAG AAA-3′) and SLAH3 N terminus forward (5′-ACC CCA TTT CCA CCT TCG GTA TG-3′) in combination with SLAH3 C terminus reverse (5′-GGA TAA TGG TGG TCA CGA GCA G-3′).

Detection of SLAH3 transcripts by real-time PCR and quantitative PCR

Quantification of actin2 and 8, SLAC1, SLAH1-4, CPK21, ABI1, and GORK transcripts was performed by real-time PCR as described elsewhere (46). Rosette leaves of 6- to 8-week-old A. thaliana ecotype Columbia (Col-0) plants were harvested for reverse transcription–PCR analysis. Guard cells were isolated by the “blender method” (4750). Leaves from 6- to 8-week-old Arabidopsis Col-0 were harvested and epidermal fragments were isolated. Major veins were removed from three to six midsized leaves; leaf blades were blended for 1 to 2 min in ice-cold deionized water with additional crushed ice and filtered through a 210-μm nylon mash. After two further rounds of blending, the remaining dark green tissue fragments were manually removed from the light green epidermal fraction. By vital staining, ≥90% of the living cells were identified as guard cells. Cell wall–free mesophyll protoplasts were prepared according to Beyhl et al. (51). After removal of the lower epidermis from fully developed young rosette leaves of 6- to 8-week-old plants, the leaves were incubated in 0.5% (w/v) cellulase Onozuka R-10 (Serva), 0.05% (w/v) pectolyase Y23 (Seishin Corp.), 0.5% (w/v) macerozyme R10 (Serva), 1% (w/v) bovine serum albumin (Sigma-Aldrich), 1 mM CaCl2, and 10 mM Hepes/tris (pH 7.4) for 45 min at 23°C and 80 rpm on a rotary shaker. The enzyme solution was adjusted to an osmolality of 400 mosmol/kg with sorbitol. Released protoplasts were filtered through a 50-μm nylon mesh and washed with 400 mM sorbitol and 1 mM CaCl2. After centrifugation at 100g and 4°C for 10 min, protoplasts were used for RNA extraction.

Quantitative PCR (qPCR) experiments were performed with isolated guard cell protoplasts, purified following previously published methods (25, 46). In addition, 60 guard cell protoplasts were manually collected by the CellTram method, and RNA was isolated as previously described (52).

Transcripts were each normalized to 10,000 molecules of actin2/8. The following primers were used: AtACT2/8 forward (fwd), 5′-GGT GAT GGT GTG TCT-3′; AtACT2/8 reverse (rev), 5′-ACT GAG CAC AAT GTT AC-3′; SLAC1 fwd, 5′-CCG GGC TCT AGC ACT CA-3′; SLAC1 rev, 5′-TCA GTG ATG CGA CTC TT-3′; ABI1 fwd, 5′-CTG CAA TAA CCA ATA CTC-3′; ABI1 rev, 5′-TCT TCT TCT CGC TAG TAA-3′; SLAH1 fwd, 5′-TGG CTT AAT GCT TAT CT-3′; SLAH1 rev, 5′-GGT ATG GTT GAC AAG TA-3′; SLAH2 fwd, 5′-CCA GTG CGA CGA TCA AA-3′; SLAH2 rev, 5′-CCG TGG AAC TAT CTA CA-3′; SLAH3 fwd, 5′-GGT CCT ATG TGC CAT TG-3′; SLAH3 rev, 5′-ATC ATT ACT CTG ACT GC-3′; SLAH4 fwd, 5′-ATG TCG CTT CGG TCT TG-3′; SLAH4 rev, 5′-AGT AGT CTG TAG TTG GT-3′; CPK21LC2 fwd, 5′-TAA ATC CCA ACG GTT TAT-3′; CPK21LC2 rev, 5′-CAA GGA GCT AAG ACA TGA-3′.

Detection of SLAH3 in plants with the SLAH3-promoter:GUS construct

SLAH3-promoter:GUS plants (25) were cultivated on sterile Sussman plates [1.2% agar and 1% sucrose (pH 5.7)] under the following conditions for 10 to 12 days: 12-hour photoperiod with 250 μmol of electrons per square meter per second and 22°C/16°C day/night temperature. GUS histochemical staining was performed according to Jefferson et al. (53). Fifteen-micrometer cross sections of GUS-stained leaves were prepared with a microtome (Leica RM2165) from leaves embedded in 2-hydroxyethyl methacrylate [Agar GMA (HEMA) Kit, Agar Scientific Limited], and pictures were taken with an epifluorescence microscope (Axioskop2 Mot, Zeiss) equipped with a spot slider camera (RT3 Slider, Visitron Systems GmbH).

Patch-clamp experiments on guard cell protoplasts

Rosette leaves of 6- to 8-week-old plants were prepared for patch-clamp analysis as described (22, 23). S-type anion currents were measured in the whole-cell patch-clamp configuration essentially as described (54, 55). For experiments under nitrate-based conditions, the pipette solution contained 150 mM NaNO3, 2 mM Mg(NO3)2, 5 mM Mg-ATP, 5 mM tris-GTP (guanosine triphosphate), 6.7 mM EGTA, 2 mM CaCl2, 3.9 mM Ca(NO3)2, and 10 mM Hepes adjusted with tris to pH 7.1. The bath solution was composed of 30 mM NaNO3, 2 mM Mg(NO3)2, 1 mM Ca(NO3)2, 0.5 mM LaCl3, and 10 mM MES adjusted with tris to pH 5.6. For measurements of channel activation under chloride-based conditions in the presence and absence of 3 mM nitrate, the pipette solution contained 150 mM TEA-HCl (tetraethylammonium chloride), 2 mM MgCl2, 5 mM Mg-ATP, 5 mM tris-GTP, 6.7 mM EGTA, 10 mM Hepes (pH 7.1)/tris, and 5.9 CaCl2. The bath solution contained 5 mM CaCl2, 10 mM MES, 0.5 mM LaCl3, 2 mM MgCl2, with or without 3 mM NaNO3 (pH 5.6). Under both nitrate- and chloride-based conditions, the pipette solution had a free Ca2+ concentration of 2 μM. The osmolality of the pipette and bath media was adjusted with d-sorbitol to 440 and 400 mosmol/kg, respectively. For recordings of S-type anion currents with nitrate-based solutions 7 min after whole-cell access, voltage pulses lasting 7.5 s were applied from a holding voltage of +11 mV in the range +46 to −134 mV in 30-mV decrements. For measurements of S-type anion currents with chloride-based media, voltage pulses lasting 5 s were applied from a holding voltage of +10 mV in the range +70 to −110 mV in 20-mV decrements. Anion currents were recorded 7 min after whole-cell access in the standard bath solution (without nitrate). Subsequently, the same protoplast was challenged with the standard bath solution containing 3 mM nitrate, and anion currents were measured 15 min after whole-cell access. The clamped voltages were corrected offline for the liquid junction potential (55).

Cloning and complementary RNA generation

The complementary DNAs (cDNAs) of SLAH3, CPK21, CPK23, OST1, ABI1, and ABI2 were cloned into oocyte (BiFC) expression vectors (based on pGEM vectors), as well as into plant binary vectors (based on pCAMBIA vectors) by an advanced uracil-excision–based cloning technique as described (56). Site-directed mutations were introduced with the QuikChange Site-Directed Mutagenesis kit according to the manufacturer’s instructions (Stratagene). For functional analysis, complementary RNA (cRNA) was prepared with the mMESSAGE mMACHINE T7 transcription kit (Ambion). Oocyte preparation and cRNA injection were performed as described (57). For oocyte BiFC and electrophysiological experiments, 10 ng each of SLAH3, SLAC1, CPK, or PP2C cRNA was injected.

Oocyte recordings

In double-electrode voltage-clamp studies (DEVC), oocytes were perfused with tris/Mes buffers. The standard solution contained 10 mM tris/Mes (pH 7.5), 1 mM Ca(gluconate)2, 1 mM Mg(gluconate)2, 100 mM NaNO3, and 1 mM LaCl3. To balance the ionic strength, we compensated for the nitrate or Cl variations with gluconate. Solutions for selectivity measurements were composed of 50 mM Cl, HCO3, SO42−, NO3, gluconate, or malate sodium salts; 1 mM Ca(gluconate)2; 1 mM Mg(gluconate)2; and 10 mM tris/Mes (pH 7.5). Osmolality was adjusted to 220 mosmol/kg with d-sorbitol. Starting from a holding potential (VH) of 0 mV, single-voltage pulses were applied in 20-mV decrements from +60 to −200 mV. Steady-state currents (ISS) were extracted at the end of the test pulses lasting 20 s. The relative open probability PO was determined from current responses to a constant voltage pulse to −120 mV subsequent to different test pulses lasting 20 s. These currents were normalized to the saturation value of the calculated Boltzmann distribution. The half-maximal activation potential (V1/2) and the apparent gating charge (z) were determined by fitting the experimental data points with a single Boltzmann equation. Instantaneous currents (IT) were extracted right after the voltage jump from the holding potential of 0 mV to 50-ms test pulses ranging from +70 to −150 mV. For calculations of deactivation and activation kinetics, SLAH3 was activated by test pulses in the range −180 to +40 mV in 20-mV increments starting from a holding potential of −180 mV. The relaxation of current deactivation/activation (τdeactact) was determined by fitting the current response at an external NO3 concentration of 10 mM with a single-exponential equation.

BiFC experiments

Transient protoplast expression was performed with a modified polyethylene glycol transformation method (58). Sixteen to 28 hours after transformation, protoplast images were taken. For documentation of the oocyte and protoplast BiFC results, pictures were taken with a confocal laser scanning microscope (LSM 5 Pascal, Carl Zeiss Jena GmbH) equipped with a Zeiss Plan-Neofluar 20×/0.5 objective for oocyte images and a Zeiss Plan-Neofluar 63×/1.25 oil objective for protoplasts. Images were processed (low-pass filtered and sharpened) identically with the image-acquisition software LSM 5 Pascal (Carl Zeiss).

Protein purification and in vitro kinase assays

CPK21, RCAR1, and ABI1, as well as the SLAH3 N- and C-terminal domains were subcloned into the recombinant expression vector pGEX 6P1 (GE Healthcare) and transformed into Escherichia coli (DE3) pLysS strain (Novagen). Bacteria were grown to an optical density (OD) of 0.5 to 0.8 at 600 nm, and production of glutathione S-transferase (GST)–tagged proteins was induced by 0.4 mM isopropylthio-β-galactoside for 16 hours at 18°C. Cells were collected by centrifugation at 10,000g and then lysed in 50 ml of phosphate-buffered saline (PBS) by five 15-s sonication steps. To remove insoluble bacterial material, we centrifuged the lysate for 30 min at 10,000g. Native purification using Glutathione Sepharose 4B beads (Amersham Biosciences) was performed in columns according to the manufacturer’s instructions. Tris-HCl (50 mM) with 10 mM reduced glutathione was used for protein elution. Lysis and wash buffers were composed of 140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, and 1.8 mM KH2PO4 (pH 7.3). Proteins were dialyzed and stored in 150 mM NaCl and 50 mM Hepes. In vitro kinase buffer was composed of 50 mM Hepes (pH 7.5), 10 mM MgCl2, 1× protease inhibitor cocktail (Roche), 2 mM dithiothreitol (DTT), 5 mM EGTA, and 5 μCi of [γ-32P]ATP (3000 Ci/mmol). To obtain defined free Ca2+ concentrations, we added 1.05 mM for 20 nM, 1.95 mM for 50 nM, 2.78 mM for 100 nM, 3.50 mM for 200 nM, 4.26 mM for 500 nM, 4.6 mM for 1000 nM, 4.87 mM for 3000 nM, and 4.92 mM for 5000 nM CaCl2. The concentrations were calculated with the WEBMXC extended Web site ( In vitro kinase assays in Figs. 5A and 6, A and B, were performed in the presence of 1 μM free Ca2+ and the absence or presence of 125 μM ABA. Reactions were carried out for 10 min at room temperature and then stopped by addition of 6× SDS loading buffer and heating to 90°C for 5 min. Proteins were separated by SDS-gel electrophoresis with an 8 to 16% gradient acrylamide gel (Precise protein gel, Thermo Scientific). Gels were stained with Coomassie blue, and phosphorylated proteins were determined by autoradiography with a Fujifilm BAS 2000 scanner. The Ca2+-dependent phosphorylation activity of CPK21 was determined by counting the radioactivity of excised gel slices harboring the SLAH3 N-terminal domain. The affinity of CPK21 for Ca2+ was described by a Hill equation with a Hill factor of 4.

CelluSpots peptide arrays were ordered from Intavis Bioanalytical Instruments. Twenty–amino acid peptides with 10–amino acid overlap at each end of the SLAH3 N and C termini were coupled to microscope slides by acetylation. To prevent nonspecific binding, we blocked the arrays by immersing the slides in bovine serum albumin solution (1 mg/ml) for 2 hours at room temperature. The phosphorylation reaction was carried out for 2 hours at room temperature by the use of 2 μg of kinase, 5 μCi of [γ-32P]ATP (3000 Ci/mmol), and the in vitro kinase buffer described above. Subsequently, slides were washed three to four times with PBS buffer containing 0.05% Tween 20. Phosphorylation was detected by autoradiography.

Cloning of SLAH3 N-terminal domain and CPK21 enzyme variants

The SLAH3 coding sequence from base 1 to 768 (SLAH3 N-terminal domain) was amplified with the primer pairs AtSLAH3 fwd (5′-TAGAATTCATGGAGGAGAAACCAAACTATGTG-3′) and AtSLAH3 rev (5′-CCCGGGGAGAAGAAACGGCCACTTTTTATC-3′) introducing Eco RI and Sma I sites. CPK21 CDS was amplified by PCR with primers introducing an Eco RI (5′-gaattcATGGGTTGCTTCAGCAGTAAACA-3′) and a Sma I (5′-cccgggATGGAATGGAAGCAGTTTCCCC-3′) restriction site at the 5′ and 3′ end, respectively. The products were cloned in pJET1.2/blunt with CloneJET PCR Cloning Kit (Fermentas), and the corresponding Eco RI/Sma I fragments were transferred into the destination vector pXCS-HA-StrepII [AY457636 (59)]. The construction of the CPK21 kinase-inactive variant has already been described (60).

Transient coexpression of SLAH3 N-terminal domain and CPK21 in protoplasts

The transfection of Arabidopsis leaf mesophyll protoplasts from cpk21-1 (Salk 029412) mutant plants was conducted as described (61). To ~4 × 105 Arabidopsis protoplasts, 60 μg of DNA of each expression construct was added and protoplasts were incubated in the dark for 14 hours. For ABA treatment, protoplasts were incubated for 40 min with 30 μM ABA or the same volume of 10 mM MES (pH 7.4) as solvent control. Reactions were stopped by two centrifugation steps (10,000g, 2 s), frozen in liquid nitrogen, and stored at −80°C until use.

Protein extraction and immunoprecipitation

For protein extraction, frozen pellets were resuspended by vortexing in 500 μl of extraction buffer: 100 mM tris-HCl (pH 8.0), 150 mM NaCl, 20 mM DTT, 5 mM EGTA, 5 mM EDTA, 10 mM NaF, 10 mM NaVO4,10 mM β-glycerol phosphate, 1× phosphatase inhibitor cocktail (Sigma), 0.5 mM AEBSF [4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride], AL-mix [antipain (2 μg/ml) and leupeptin (2 μg/ml)], 1× protease inhibitor cocktail 3 (Sigma), avidin (100 μg/ml), and 0.5% (v/v) Triton X-100 in H2O. Crude extracts were then centrifuged at 20,000g for 7 min at 4°C. After removal of insoluble debris, the supernatant was incubated with 20 μl of Strep-Tactin beads (IBA) and gently rotated end over end at 4°C for 40 min. Affinity-bound proteins were washed three times with 1 ml of wash buffer 1 [100 mM tris-HCl (pH 8.0), 100 mM NaCl, 2 mM DTT, 10 mM NaF, 10 mM NaVO4, 10 mM β-glycerol phosphate, 1× phosphatase inhibitor cocktail 3 (Sigma), and 0.05% Triton X-100 in H2O], once with 1 ml of wash buffer 2 [50 mM tris-HCl (pH 8.0) and 20 mM NaCl in H2O], and once with 1 ml of H20. The matrix was then resuspended in 20 μl of 1× SDS sample buffer and incubated for 4 min at 70°C.

Phosphoprotein and total protein staining

Protein samples were separated by 12% SDS–polyacrylamide gel electrophoresis (SDS-PAGE) gel (Bio-Rad) and analyzed by Pro-Q Diamond phosphoprotein gel stain and SYPRO Ruby protein gel stain (both Molecular Probes/Invitrogen) following the manufacturer’s instructions. Phosphoprotein and total protein on wet gels were detected and quantified by imaging (Fluorescent Image Analyzer FLA-2000G; Fuji Photo Film Co.) with 532-nm (Pro-Q Diamond) and 473-nm (SYPRO Ruby) band-pass excitation filters and a 580-nm band-pass emission filter. CPK21 Strep-tagged protein was detected by Western blot with Strep-Tactin AP conjugate (IBA).

Supplementary Materials

Fig. S1. SLAH3 expression in guard cells.

Fig. S2. Nitrate and chloride dependence of SLAC1- and CPK23ΔEF-expressing oocytes.

Fig. S3. BiFC experiments in oocytes.

Table S1. Putative phosphorylation sites in SLAH3.

References and Notes

  1. Acknowledgments: We thank G. Harms for critical reading of the manuscript. We thank O. Matsuda and K. Iba (Department of Biology, Faculty of Sciences, Kyushu University, Fukuoka, Japan) for providing us with seeds for SLAH3-promoter:GUS experiments. Funding: This work was supported by grants of the Deutsche Forschungsgemeinschaft within SFB567 to R.H. and T.M.; FOR 964 to R.H., T.R., A.L., and E.G.; GR938/6 to E.G.; FORPLANTA to E.G. and R.H.; GK1342 to R.H., S.S., and D.G.; and KSU grant to R.H. and K.A.S.A.-R. Author contributions: D.G., K.A.S.A.-R., I.M., P.A., E.G., T.R., and R.H. designed the research; D.G., T.M., S.S., P.M., C.W., and A.L. performed the research; D.G., T.M., S.S., P.M., C.W., A.L., and I.M. analyzed the data; and D.G., I.M., T.R., and R.H. wrote the paper. Competing interests: The authors declare that they have no competing interests.
View Abstract

Stay Connected to Science Signaling

Navigate This Article