Potassium transporters and channels play crucial roles in K+ uptake and translocation in plant cells. These roles are essential for plant growth and development. AKT1 is an important K+ channel in Arabidopsis roots that is involved in K+ uptake. It is known that AKT1 is activated by a protein kinase CIPK23 interacting with two calcineurin B-like proteins CBL1/CBL9. The present study showed that another calcineurin B-like protein (CBL10) may also regulate AKT1 activity. The CBL10-over-expressing lines showed a phenotype as sensitive as that of the akt1 mutant under low-K+ conditions. In addition, the K+ content of both CBL10-over-expressing lines and akt1 mutant plants were significantly reduced compared with wild-type plants. Moreover, CBL10 directly interacted with AKT1, as verified in yeast two-hybrid, BiFC and co-immunoprecipitation experiments. The results of electrophysiological analysis in both Xenopus oocytes and Arabidopsis root cell protoplasts demonstrated that CBL10 impairs AKT1-mediated inward K+ currents. Furthermore, the results from the yeast two-hybrid competition assay indicated that CBL10 may compete with CIPK23 for binding to AKT1 and negatively modulate AKT1 activity. The present study revealed a CBL-interacting protein kinase-independent regulatory mechanism of calcineurin B-like proteins in which CBL10 directly regulates AKT1 activity and affects ion homeostasis in plant cells.
Potassium (K+), as the most abundant cation in living plant cells, plays essential roles in plant growth and development (Leigh and Jones, 1984). Plants absorb K+ from soil into root cells via K+ transporters and channels (Spalding et al., 1999; Mäser et al., 2001; Véry and Sentenac, 2003; Lebaudy et al., 2007). Thus, regulation of K+ transporters and channels across the plasma membrane is essential for proper K+ homeostasis in plants. The inward-rectifying K+ channel AKT1 (Arabidopsis K+ transporter 1) has been identified as an important K+ channel that is expressed in epidermal and cortex cells of Arabidopsis roots, where it mediates K+ uptake from soil into roots (Lagarde et al., 1996; Hirsch et al., 1998).
Previous reports have shown that phosphorylation/dephosphorylation regulates the channel activity of AKT1 (Li et al., 2006; Xu et al., 2006; Lee et al., 2007). The CBL-interacting protein kinase CIPK23 directly phosphorylates and activates the AKT1 channel at the plasma membrane by binding the calcineurin B-like proteins CBL1 or CBL9, significantly increasing the K+ uptake of Arabidopsis (Li et al., 2006; Xu et al., 2006). Conversely, a type 2C protein phosphatase (PP2C) AIP1 was found to interact with AKT1 and inhibit AKT1 activity in oocytes (Lee et al., 2007). In addition, recent reports revealed that heterotetramerization is also a key regulatory mechanism for the AKT1 channel (Duby et al., 2008; Geiger et al., 2009; Wang et al., 2010). AtKC1, a silent K+ channel subunit, forms a functional heterotetrameric K+ channels with AKT1 in vivo, whose voltage dependence is negatively shifted compared with AKT1 channel homotetramers (Geiger et al., 2009; Wang et al., 2010).
CBLs have been identified as plant calcium sensors that are similar to calcineurin B and neuronal calcium sensors in mammals and yeast (Kudla et al., 1999). CBL proteins interact with a group of serine/threonine kinases designated as CBL-interacting protein kinases (CIPKs) (Shi et al., 1999; Kudla et al., 2010). In Arabidopsis, ten CBL-type calcium sensor proteins form a complex signaling network with 26 CIPKs (Weinl and Kudla, 2009). Multiple but specific CBL-CIPK complexes connect upstream Ca2+ signaling with the downstream responses, and appear to regulate numerous target proteins in various physiological processes (Albrecht et al., 2001; Guo et al., 2001). Many studies have shown that the activities of some plant ion channels or transporters (such as AKT1 and SOS1) are regulated by CBL-CIPK signaling pathways, which appear to be a specific regulatory mechanism in plants. In Arabidopsis, the CBL1/9-CIPK23 complex was found to regulate AKT1-mediated K+ uptake under low-K+ stress (Li et al., 2006; Xu et al., 2006), while the SOS3-SOS2 (CBL4-CIPK24) complex was shown to modulate SOS1-mediated salt tolerance (Liu and Zhu, 1998; Qiu et al., 2002).
Previous studies have revealed that the calcium sensor CBL10 (SCABP8) interacts with CIPK24 (SOS2), and that the CBL10-CIPK24 complex regulates the Arabidopsis response to salt stress by modulating the activities of Na+ transporters (Kim et al., 2007; Quan et al., 2007). Here, we report that CBL10 directly interacts with the K+ channel AKT1 and modulates AKT1 activity in maintenance of K+ homeostasis in Arabidopsis under ion stress conditions in a CIPK-independent manner.
CBL10 regulates K+ homeostasis in Arabidopsis
The calcineurin B-like protein CBL10 (SCABP8) has been reported to be a key element in SOS2 (CIPK24) regulation during the Arabidopsis salt-stress response (Kim et al., 2007; Quan et al., 2007). The K+ content of the cbl10 mutant was significantly increased under salt stress compared with wild-type plants (Kim et al., 2007), which suggests that CBL10 may also be involved in the process of K+ homeostasis in Arabidopsis. To verify this hypothesis, a phenotype observation and K+ content measurement of the Arabidopsis cbl10 mutant, complemented cbl10 (line COM29) and CBL10-over-expressing lines (OE3 and OE12) under normal (MS medium) and low-K+ (LK medium) conditions were performed. Because Arabidopsis akt1 mutant seedlings were observed to exhibit a low K+-sensitive phenotype in a previous study (Xu et al., 2006), akt1 plants were used as control in these experiments.
After growth on MS medium for 4 days, the Arabidopsis plants were transferred onto MS medium or LK medium and grown for 10 days. The CBL10-over-expressing lines displayed a similar low-K+ sensitive phenotype to akt1 mutants under low-K+ conditions (Figure 1a). Specifically, we observed chlorotic leaves and continuous root growth in akt1 mutants and CBL10-over-expressing lines, but these were not detected in wild-type plants (Figure 1a). However, all tested plant lines showed a similar dry weight (Figure 1a). Correlated with these phenotypes, K+ content measurement indicated that the shoot K+ content of CBL10-over-expressing lines and akt1 mutants was significantly decreased under low-K+ conditions (Figure 1b). In contrast, the cbl10 mutants and the complementation plants did not show any significant difference compared with wild-type plants in either the phenotype test or K+ content measurement (Figure 1a,b). As there is no difference in dry weight among the tested plant lines, the decreased K+ content in the CBL10-over-expressing lines and akt1 mutants (as shown in Figure 1b) may be due to reduction of K+ uptake. The accumulation of CBL10 transcripts in all the test materials was determined and is shown in Figure 1(c). The accumulation of CBL10 transcripts was also significantly reduced in Arabidopsis roots after the low-K+ treatment (Figure 1d), suggesting that CBL10 is responsive to low-K+ stress and participates in low-K+ signal transduction.
CBL10 interacts directly with the K+ channel AKT1
Because the CBL10-over-expressing lines displayed a low-K+ sensitive phenotype and K+ content similar to akt1 mutant plants (Figure 1a,b), we hypothesized that CBL10 may function in Arabidopsis K+ homeostasis by regulating the activity of the AKT1 channel. It has been reported that CIPK23 activates AKT1 by binding calcium sensors CBL1 or CBL9, but not upon interaction with CBL10 (Xu et al., 2006). To investigate whether CBL10 directly interacts with AKT1 and regulate AKT1-mediated K+ uptake, a potential protein interaction between cAKT1 (C-terminal region) and CBL10 was tested by the yeast two-hybrid assay (Pilot et al., 2003). The results clearly suggest that CBL10 directly interacts with cAKT1 in yeast (Figure 2a).
As recent reports have shown that the potassium channel AtKC1 interacts with AKT1 and regulates AKT1-mediated K+ uptake (Geiger et al., 2009; Wang et al., 2010), the interaction between CBL10 and the C-terminal region of AtKC1 (cAtKC1, 333 amino acids from His330 to Phe662) was also tested. The results showed that CBL10 did not interact with AtKC1 at all (Figure 2a). We also investigated the possible interaction between CBL10 and all other Shaker K+ channels from Arabidopsis. CBL10 exclusively interacted with the AKT1 channel (Figure S1). In addition, we determined the possible interaction between cAKT1 and all other nine Arabidopsis CBL proteins. In addition to CBL10, CBL5 and CBL7 also slightly interacted with cAKT1 (Figure 2b).
To investigate the interaction between CBL10 and full-length AKT1 in living plant cells, a bimolecular fluorescence complementation (BiFC) assay (Walter et al., 2004) was performed. The combination of YFPC-CBL10 (YC-CBL10) and YFPN-AKT1 (YN-AKT1) was co-transformed into tobacco (Nicotiana benthamiana) leaves using the Agrobacterium-mediated infiltration method (Waadt and Kudla, 2008). The YFP fluorescence of the BiFC complex was detected in co-transformed epidermal cells (Figure 2c), indicating in vivo interaction between CBL10 and AKT1 in plant cells. CBL5 and CBL7 displayed slight interaction with cAKT1 in yeast (Figure 2b) but did not interact with full-length AKT1 in plant cells (Figure 2c). The in vivo interaction between CBL10 and AKT1 was further confirmed by a co-immunoprecipitation assay. Constructs containing FLAG-CBL10 and cMyc-cAKT1 driven by the CaMV 35S promoter were introduced into Arabidopsis mesophyll protoplasts. Proteins extracted from the transformed mesophyll protoplasts were used for the co-immunoprecipitation assay. The results further confirmed the interaction between CBL10 and AKT1 in vivo (Figure 2d).
CBL10 impairs AKT1-mediated inward K+ currents in Xenopus oocytes
The similar phenotypes of CBL10-over-expressing lines and the akt1 loss-of-function mutant in the phenotype test and K+ content measurements suggest that CBL10 negative regulates AKT1 activity. We tested this hypothesis in Xenopus oocytes using a two-electrode voltage-clamp technique. When co-expressing AKT1 and CBL10 in oocytes in the presence of CIPK23 and CBL1, the AKT1-mediated K+ inward currents were remarkably impaired by CBL10 under hyper-polarization conditions (Figure 3a,b). Substitution of CBL10 with CBL5 or CBL7 did not affect AKT1 currents at all (Figure 3a,b), indicating that only CBL10 specifically affects AKT1 activity. Furthermore, the inhibition of AKT1-mediated inward K+ currents by CBL10 was significantly enhanced in line with the increase in the amount of injected CBL10 cRNA (AKT1:CBL10 cRNA ratios increased from 1:0.25 to 1:1) (Figure 3a,c). These results are shown as current-voltage (I/V) curves in Figure 3(c).
It has been reported that AtKC1 inhibits AKT1-mediated K+ currents by negatively shifting the voltage dependence of AKT1 channels (Geiger et al., 2009; Wang et al., 2010). We wished to determine whether the regulatory effect of CBL10 on AKT1 was similar to that of AtKC1. Therefore, the biophysical properties (relative open probability, voltage dependence and maximal cord conductance) of AKT1-mediated K+ currents regulated by CBL10 and AtKC1 in Xenopus oocytes were analyzed. The results showed that CBL10 and AtKC1 both inhibited AKT1-mediated K+ currents (Figure 4a). CBL10 did not change the voltage dependence of the AKT1 channel in contrast to AtKC1 (Figure 4b and Table 1), but the maximal cord conductance (Gmax) of oocytes co-injected with AKT1 and CBL10 cRNA was significantly decreased (Figure 4c and Table 1). These results clearly demonstrate that CBL10 has a different regulatory effect on the AKT1 channel compared with AtKC1.
Table 1. Channel property comparison of various AKT1 channel complexes expressed in Xenopus oocytes
40.0 ± 5.0
−106.0 ± 5.0
AKT1 + CBL10
7.8 ± 1.3
−113.2 ± 1.1
AKT1 + AtKC1
38.9 ± 5.2
−169.9 ± 0.1
CBL10 modulates inward K+ currents in Arabidopsis root cells
Having established that CBL10 regulates AKT1 currents in Xenopus ooctyes, we further investigated CBL10-dependent regulation of AKT1 in vivo using the patch-clamp technique in Arabidopsis root cells. Whole-cell inward K+ currents were recorded in root cell protoplasts isolated from wild-type plants, cbl10 mutants and CBL10-over-expressing lines OE3 and OE12. As shown in Figure 5(a,b), the inward K+ currents in root cells of CBL10-over-expressing lines were significantly reduced compared with wild-type plants, while the K+ currents for the cbl10 mutant were unchanged. However, a slight difference in K+ currents between wild-type and cbl10 mutant was observed after the number of protoplasts was increased. As shown in Table 2, the percentage of protoplasts with high current density (>200 pA/pF) in cbl10 mutants (16.9%) was significantly increased compared with wild-type plants (9.1%).
Table 2. Statistical analysis of K+ currents recorded in the root cell protoplasts
Current density (pA/pF)
Col (n =99)
cbl10 (n =65)
OE3 (n =74)
OE12 (n =62)
In addition, the voltage dependence and Gmax of the K+ currents recorded from root cell protoplasts were also analyzed. Although there was no difference observed in the voltage dependence of the K+ currents among various plants (Figure 5c), the normalized Gmax of K+ currents from the two CBL10-over-expressing lines was markedly reduced compared with wild-type plants (Figure 5d). All these results are consistent with the observations in Xenopus oocytes, and suggest that CBL10 impairs AKT1-mediated inward K+ currents, most likely by reducing the number of activated AKT1 channels at the plasma membrane of Arabidopsis root cells.
The CIPK23-AKT1 interaction is repressed by CBL10
The protein kinase CIPK23 is a positive regulator that activates the AKT1 channel by phosphorylation of AKT1 (Xu et al., 2006). As shown in Figure 2(a), CBL10 also directly interacts with AKT1, but does not interact with CIPK23 (Xu et al., 2006). Based on the results from electrophysiological experiments indicating that CBL10 reduced the number of activated AKT1 channels at the plasma membrane (Figures 4 and 5), we hypothesized that CBL10 may compete with CIPK23 for binding to AKT1, and that the number of AKT1 channels activated by CIPK23 may be reduced. To verify this hypothesis, a yeast two-hybrid competition assay was performed. The coding sequence of the AKT1 C-terminal region was cloned into the prey (AD) vector (Figure 6a). A BD vector expressing both CIPK23 and CBL10 was also constructed (Figure 6a). CBL6 was substituted for CBL10 as a control. The expression of CBL10 and CBL6 proteins may be repressed by inclusion of 1 mM Met in the medium, but is induced by reducing the Met concentration to 0.01 mm. As shown in Figure 6(b), the growth of two yeast clones expressing CBL10 or CBL6 showed no difference in the presence of 1 mm Met. However, the growth of the yeast clone expressing CBL10 was repressed compared with the clone expressing CBL6 in the presence of 0.01 mm Met (Figure 6b). The growth differences of yeast clones under various conditions were further confirmed by analyzing the relative β-galactosidase activity (Figure 6c). Protein expression of CBL10 and CBL6 in the yeast two-hybrid competition assay was verified by Western blotting using antibody against hemagglutinin (HA) epitope tag (Figure 6d). All these results demonstrate that CBL10 significantly represses the protein interaction between CIPK23 and AKT1.
The CIPK-independent regulatory mechanism of CBL10
The specific and preferential interaction of CBL proteins with a member of the CIPK family creates a unique calcium signal transduction network in plants (Kudla et al., 2010). Previous investigations have revealed that each CBL interacts with several CIPK targets, and each CIPK associates with one or more CBLs (Kim et al., 2000; Gong et al., 2004; Batistič et al., 2010). Therefore, the CBL and CIPK proteins constitute a complex CBL-CIPK network with functional specificity and redundancy that participates in diverse physiological processes (Halfter et al., 2000; Xu et al., 2006; Ho et al., 2009; Kudla et al., 2010). However, the data presented here reveal a regulatory mode for CBL10. The protein CBL10 may interact with its target protein AKT1 and regulate AKT1 channel activity in a CIPK-independent way. The findings offer an additional perspective on CBL regulatory mechanisms in plant cells. The CBL calcium sensors may sense and conduct the upstream signaling by directly interacting with the downstream target proteins in a CIPK-independent manner. This increases the phoshorylation-independent regulatory repertoire of CBL proteins, as phosphorylation-independent but CIPK-dependent positive regulation of the K+ channel AKT2 by the calcium sensor CBL4 was recently reported (Held et al., 2011).
The diverse regulatory mechanisms of AKT1
AKT1 is the most investigated K+ channel in Arabidopsis, but observation of AKT1-mediated K+ currents was not successful in Xenopus oocytes (Gaymard et al., 1996) until the AKT1 regulators had been identified. The protein kinase CIPK23 was first identified as a key regulatory factor of AKT1 that directly phosphorylates and activates the AKT1 channel (Li et al., 2006; Xu et al., 2006). In contrast, the PP2C protein AIP1 was identified as a negative regulator of AKT1 by interaction in yeast and electrophysiological studies in oocytes (Lee et al., 2007). In addition, another K+ channel α-subunit AtKC1 from the Shaker K+ channel family regulates AKT1 channel properties as well by forming a heteromeric channel with AKT1 (Geiger et al., 2009; Wang et al., 2010). Here, CBL10 is also found to modulate AKT1-mediated inward K+ currents. The diverse modulation and fine-tuning of AKT1 by various regulators may enhance plant adaptability in terms of responses to various environmental changes and stimuli such as low-K+ stress.
We observed that AtKC1 and CBL10 both attenuated AKT1-mediated K+ currents (Figure 4a), but they regulate channel activity by different mechanisms. AtKC1 changed the gating property of the AKT1 channel by shifting the voltage dependence of AKT1 towards a negative membrane potential (Geiger et al., 2009; Wang et al., 2010), but CBL10 only reduced the maximal cord conductance (Gmax) of the plasma membrane, but did not affect the gating property of the AKT1 channel (Figures 4c and 5d). These results demonstrate that CBL10 impairs AKT1-mediated K+ currents probably by reducing the number of activated AKT1 channels at the plasma membrane. In the present study, we demonstrated that the calcium sensor CBL10 directly interacts with AKT1 by competition with CIPK23, and thereby inhibits AKT1-mediated K+ currents (Figure 6). These findings indicate that CBL10 acts as a regulatory factor of K+ channels in plant cells, and also suggest a mechanism of action of CBL-type calcium sensors in ion channel regulation.
The physiological role of CBL10 in Arabidopsis ion homeostasis
The phenotype test showed that the cbl10 mutant did not show significant phenotype differences compared with wild-type plants (Figure 1a). There are two possible explanations for the lack of an obvious phenotype in the cbl10 mutant plants. First, the expression level of CBL10 is relatively low in roots compared with shoots (Kim et al., 2007; Quan et al., 2007), and thus mutation of CBL10 may not result in an obvious phenotype. Second, the transcriptional level of CBL10 was significantly reduced in roots after low-K+ treatment (Figure 1d). When the wild-type seedlings were transferred onto the LK medium, CBL10 transcription was decreased and was close to that of cbl10 mutant plants. Therefore, no obvious phenotype difference was observed between cbl10 mutants and wild-type seedlings. However, a slight difference between them was observed in patch-clamp experiments (Table 2). The results of current density analysis showed that the proportion of root cell protoplasts with a large current density (>200 pA/pF) was increased in cbl10 mutant plants (16.9%) compared with wild-type seedlings (9.1%) (Table 2), which suggests that CBL10 negatively regulates K+ uptake in Arabidopsis.
As shown in Figure 2(b), CBL5, CBL7 and CBL10 all interact with the AKT1 C-terminus in yeast; however, only CBL10 interacts with full-length AKT1 in planta (Figure 2c). Moreover, in Xenopus oocytes, the AKT1-mediated K+ inward currents were impaired only by CBL10, but not by CBL5 or CBL7 (Figure 3a,b). This evidence indicates that, of the CBL proteins tested, only CBL10 interacts with AKT1 and specifically affects AKT1 activity.
It has been reported that CBL10 (SCaBP8) was involved in salt responses by partially participating in the SOS pathway (Kim et al., 2007; Quan et al., 2007). Salt-induced SCaBP8 (CBL10) interacted with SOS2 (CIPK24) and modulated SOS1 activity and vacuolar Na+ accumulation, resulting in the Arabidopsis salt tolerance (Kim et al., 2007; Quan et al., 2007; Lin et al., 2009). Here, we found that CBL10 is also involved in K+ homeostasis and negatively regulates AKT1 channel activity in Arabidopsis roots. It appears that CBL10 may function as an interconnecting regulator to connect SOS1-mediated salt tolerance and AKT1-mediated K+ homeostasis by interacting with different target proteins (SOS2 or AKT1). It is known that plant responses to salt stress are closely related to K+/Na+ discrimination in K+ (and/or Na+) transport and translocation. Thus, CBL10 may play crucial roles in ion homeostasis (K+/Na+ discrimination) under salt stress by regulating both Na+ and K+ uptake/translocation. As shown in Figure S2, both cbl10 mutant and CBL10 OE plants showed a sensitive phenotype to salt stress compared with wild-type and complemented plants when the K+ concentration was decreased. Thus proper regulation of CBL10 transcription is essential in plant responses to salt stress, especially under K+-deficient conditions. Therefore, the physiological roles of CBL10 in coordinating Na+ and K+ homeostasis in plant responses to salt and/or low-K+ stresses should be further investigated.
Plant materials and growth conditions
Arabidopsis thaliana ecotype Columbia was used in this study. The T-DNA insertion lines, including cbl10 (SALK_056042) and akt1 (SALK_071803), were obtained from the Arabidopsis Biological Resource Center (http://www.arabidopsis.org/abrc/). For seed harvest and hybridization, Arabidopsis plants were grown in potting soil mixture (rich soil/vermiculite, 2:1 v/v) and kept in growth chambers at 22°C with illumination at 120 μmol m−2 sec−1 for a 16 h daily light period. The relative humidity was 70 ± 5%.
Vector constructions and Arabidopsis transformation
For generation of CBL10-over-expressing lines, the construct was generated by cloning the coding sequence of CBL10 into the pBIB vector under the control of the SUPER promoter (Li et al., 2001). For generation of complementation lines of cbl10 mutants, CBL10 was cut from the BAC F26P21, and then was cloned into the pCAMBIA1300 vector (http://www.cambia.org). Arabidopsis transformation with Agrobacterium (strain GV3101) was performed by the floral-dip method (Clough and Bent, 1998).
For real-time PCR analysis, total RNA was extracted from 4-week-old plants using Trizol reagent (Invitrogen, http://www.invitrogen.com), and then treated with DNase I (TaKaRa, http://www.takara.com.cn) to eliminate genomic DNA. cDNA was synthesized from treated RNA using SuperScript II RNase H reverse transcriptase (Invitrogen) with random hexamer primers (Promega, www.promega.com). Quantitative real-time PCR was performed using Power SYBR Green PCR Master Mix (Applied Biosystems, http://www.appliedbiosystems.com) on an ABI PRISM 7500 real-time PCR system (Applied Biosystems) according to the manufacturer's protocols. Real-time PCR was performed using CBL10-specific primers (Table S3), and quantification of the relative gene expression was performed by normalization to 18S rRNA.
Yeast two-hybrid assays
The coding sequences of all CBLs (Primers were listed in Table S1) were cloned into the pGBKT7 vector (Clontech, http://www.clontech.com). The coding sequences for the cytosolic C-terminus regions of nine Shaker potassium channels were cloned into the pACT2 vector (Clontech, http://www.clontech.com). The vectors were transformed into yeast strain AH109 for yeast two-hybrid assays according to the manufacturer's instructions (Clontech, http://www.clontech.com). The yeast strain AH109 was transformed with bait and prey vectors, and incubated on SC medium lacking Leu, Trp and 3-AT (3-Amino-1,2,4-triazole) at 28°C for 5 days to obtain positive clones. Then the positive clones were incubated in SC medium lacking Leu, Trp and 3-AT (3-Amino-1,2,4-triazole) at 28°C to an absorbance at 600 nm of 1.0. Aliquots (5 μl) were spotted onto selective medium (lacking Leu, Trp, His and Ade) and non-selective medium (lacking Leu and Trp).
BiFC assays and microscopic analysis
For generation of the BiFC vectors (The primers were listed in Table S1), the coding regions of CBL10, CBL5 and CBL7 were cloned via the BamHI XhoI sites into vector SPYCE(MR) (Waadt and Kudla, 2008). The coding regions of AKT1 were cloned via the XbaI-XhoI sites into vector SPYNE(R)173 (Waadt and Kudla, 2008). All BiFC assays and subcellular localization analysis were performed as described previously (Walter et al., 2004; Waadt and Kudla, 2008). The incubation time was the same in all constructs and controls. All BiFC assay images were taken after 4 days infiltration. Microscopic analyses were performed using a Leica DMIRE2 inverted microscope equipped with a Leica TCS SP2 laser scanning device (Leica, http://www.leica.com).
The coding sequences of CBL10 and cAKT1 fused with the FLAG and cMYC tags (FLAG-CBL10 and cMyc-cAKT1) were cloned into the pCAMBIA1205 vector (Quan et al.,2007). The plasmids were purified by CsCl gradient centrifugation and transformed into Arabidopsis mesophyll protoplasts. Mesophyll protoplasts were isolated from 5-week-old wild-type plants (ecotype Columbia) and transformed with the tested pairs of constructs as described (Quan et al.,2007). The protoplasts were lysed, sonicated and centrifuged at 12000 g for 10 min at 4°C after overnight incubation. The supernatant was incubated with 10 μl anti-cMyc agarose conjugate (Sigma-Aldrich, http://www.sigmaaldrich.com) overnight at 4°C. The co-immunoprecipitation products were washed briefly for five times at 4°C in extraction buffer (1 ml), and then detected via immunoblot analysis. Both anti-cMyc (Sigma-Aldrich) and anti-FLAG (Sigma-Aldrich) antibodies were used at 1:5000 dilutions, and chemiluminescence signals were detected using Kodak film (http://www.kodak.com).
K+ content measurement
Four-day-old Arabidopsis seedlings were transferred from MS medium to LK medium and treated for 10 days. The root and shoot tissues were harvested separately and dried at 80°C for 48 h. The dry plant tissues were treated in a muffle furnace at 575°C for 5 h, and then dissolved in 0.1 N HCl. Potassium concentrations of the samples were measured by atomic absorption spectrophotometry (Hitachi Z-5000, http://www.hitachi.com). The LK medium was prepared by modification of MS medium (Xu et al., 2006).
Expression of AKT1 and related regulatory components in Xenopus oocytes
The coding sequences of AKT1, AtKC1, CIPK23, CBL1, CBL5, CBL7 and CBL10 were cloned into the pGEMHE vector (Liman et al.,1992). The primers were listed in the Table S2. The cRNAs were transcribed in vitro using the T7 RiboMAX™ large-scale RNA production system (Promega). The oocytes were isolated from Xenopus laevis and injected with cRNAs. Oocytes injected with 50 nl distilled water were used as the control. AKT1-expressing oocytes were injected with a mixture of AKT1, CIPK23 and CBL1 cRNAs (6 ng each in 50 nl). Oocytes co-expressing AKT1 and CBL10, CBL5, CBL7 or AtKC1 were injected with a mixture of AKT1, CBL10/CBL5/CBL7/AtKC1, CIPK23 and CBL1 cRNAs (6 ng each in 50 nl). To analyze the CBL10 dependence of AKT1 currents, the amount of injected CBL10 cRNA was adjusted to obtain the various ratios of AKT1 to CBL10, as indicated in Figure 3. Before use for voltage-clamp recordings, the injected oocytes were incubated at 17°C in modified Barth's solution for 2 days (Xu et al., 2006). The two-electrode voltage-clamp technique was applied as described previously (Xu et al., 2006).
Patch-clamp whole-cell recordings from root cell protoplasts
Root cell protoplasts were isolated from the primary roots of 5-day-old Arabidopsis seedlings as described previously (Xu et al., 2006). Standard whole-cell recording techniques (Hamill et al., 1981) were used, utilizing an Axopatch-200A amplifier (Axon Instruments, http://www.moleculardevices.com) at 20°C in dim light. The contents of the bath and pipette solutions for the whole-cell recordings were the same as described previously (Ivashikina et al., 2001).
Yeast two-hybrid competition assay
The pBridge vector (Clontech, http://www.clontech.com) was used as the BD vector to express both CIPK23 and CBL10 (or CBL6). NotI and BglII were used to insert the coding regions of CBL10 and CBL6 into the BD vector. XhoI and SalI were used to insert CIPK23 into the BD vector. The AD vector (pACT2) was used to express the C-terminus of AKT1 (cAKT1). The primers were listed in the Table S4. Both pairs of plasmids (CIPK23-CBL10-BD/cAKT1-AD and CIPK23-CBL6-BD/cAKT1-AD) were transformed into the AH109 yeast strain. The co-transformed colonies were selected on SC medium lacking Trp and Leu. Positive clones were selected and cultivated in SC medium lacking Trp and Leu with 1 mM Met or 0.01 mM Met (plus 1 mM Asp) separately at 30°C, 200 rpm to an absorbance at 600 nm of 1.0. Aliquots (5 μl) were used for spot assay. The relative β-galactosidase activity measurement was performed using a yeast β-galactosidase assay kit (Thermo, www.thermo.com/pierce).
We thank Dr Emily Liman (Department of Biological Sciences, University of Southern California, CA) for providing the pGEMHE vector for the Xenopus oocyte experiments, and Jörg Kudla (Institut für Biologie und Biotechnologie der Pflanzen, Universität Münster, Muenster, Germany) for the helpful discussion and providing the vectors for BiFC assays. We are grateful to Yan Guo (College of Biological Sciences, China Agricultural University, Beijing, China) for helpful discussion, experimental suggestions and kindly providing for the pCAMBIA1205 vector. We thank Hong-Quan Yang (School of Life Science and Biotechnology, Shanghai Jiaotong University, Shanghai, China) for providing the vectors and yeast strain for the yeast two-hybrid competition assay. This work was supported by the ‘973’ Project (grant number 2012CB114203), the National Natural Science Foundation of China (grant number 30830013 and 31121002 to W.H.W.) and the Program of Introducing Talents of Discipline to Universities (grant number B06003 to W.H.W.).