Plant Physiol. (1998) 116: 785-795

Guard Cells Possess a Calcium-Dependent Protein Kinase That Phosphorylates the KAT1 Potassium Channel¹

Jiaxu Li, Yuh-Ru Julie Lee, and Sarah M. Assmann^*

Department of Biology and Plant Physiology Program, The Pennsylvania State University, University Park, Pennsylvania 16802

	ABSTRACT

Top Abstract Introduction Methods Results Discussion References

Increasing evidence suggests that changes in cytosolic Ca²⁺ levels and phosphorylation play important roles in the regulation of stomatal aperture and as ion transporters of guard cells. However, protein kinases responsible for Ca²⁺ signaling in guard cells remain to be identified. Using biochemical approaches, we have identified a Ca²⁺-dependent protein kinase with a calmodulin-like domain (CDPK) in guard cell protoplasts of Vicia faba. Both autophosphorylation and catalytic activity of CDPK are Ca²⁺ dependent. CDPK exhibits a Ca²⁺-induced electrophoretic mobility shift and its Ca²⁺-dependent catalytic activity can be inhibited by the calmodulin antagonists trifluoperazine and N-(6-aminohexyl)-5-chloro-1-naphthalenesulfonamide. Antibodies to soybean CDPK cross-react with CDPK. Micromolar Ca²⁺ concentrations stimulate phosphorylation of several proteins from guard cells; cyclosporin A, a specific inhibitor of the Ca²⁺-dependent protein phosphatase calcineurin enhances the Ca²⁺-dependent phosphorylation of several soluble proteins. CDPK from guard cells phosphorylates the K⁺ channel KAT1 protein in a Ca²⁺-dependent manner. These results suggest that CDPK may be an important component of Ca²⁺ signaling in guard cells.

	INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

Guard cells define and control stomatal aperture by osmotic swelling and shrinking. Stomatal opening involves hyperpolarization of the plasma membrane by a H⁺-ATPase, uptake of K⁺ and Cl, and production of organic solutes (Assmann, 1993). Stomatal closure requires depolarization of the plasma membrane and efflux of anions and K⁺ (Assmann, 1993). Increases in cytosolic Ca²⁺ regulate several ion transporters that are essential in the control of stomatal aperture (MacRobbie, 1997; McAinsh et al., 1997). The plasma membrane proton pump of Vicia faba guard cells, which hyperpolarizes the plasma membrane and thus provides the driving force for K⁺ and Cl uptake, is inhibited by increased cytosolic Ca²⁺ (Kinoshita et al., 1995). The inward K⁺ channels in the plasma membrane of V. faba guard cells, which are responsible for K⁺ influx, are also inhibited by elevated cytosolic Ca²⁺ (Schroeder and Hagiwara, 1989). Conversely, a type of K⁺ channel in the tonoplast of V. faba guard cells, which may control K⁺ efflux from guard cell vacuoles during stomatal closure, is activated when cytosolic Ca²⁺ is increased to approximately 1 µm (Ward and Schroeder, 1994). Both S-type (slow) and R-type (rapid) plasma membrane anion channels, which allow Cl and malate efflux during stomatal closure, are also activated by elevated cytosolic Ca²⁺ concentrations (Schroeder and Hagiwara, 1989; Hedrich et al., 1990). Consistent with these electrophysiological data, exogenous application of Ca²⁺ inhibits opening of closed stomata and stimulates closure of open stomata in isolated epidermal peels of Commelina communis (De Silva et al., 1985; Schwartz, 1985; Schwartz et al., 1988); such Ca²⁺ application is known to increase cytosolic Ca²⁺ levels (Gilroy et al., 1991). In addition, a variety of stimuli such as ABA, CO₂, and oxidative stress can rapidly induce increases in cytosolic Ca²⁺ concentrations in guard cells (McAinsh et al., 1997, and refs. therein). These data all point to increases in cytosolic Ca²⁺ concentrations as being critical in the inhibition of stomatal opening and promotion of stomatal closure. On the other hand, calmodulin antagonists inhibit stomatal opening and H⁺ pumping in V. faba, suggesting that Ca²⁺ may also play a role in mediating stomatal opening (Shimazaki et al., 1992).

Despite a growing body of evidence that physiological signals increase cytosolic Ca²⁺ levels in guard cells and that this in turn affects ion transporters, the biochemical steps between elevation of cytosolic Ca²⁺ concentrations and electrophysiological response are incompletely understood. Recently, elegant work from Pei et al. (1996) showed that a CDPK activated a tonoplast Cl channel in isolated vacuoles from V. faba guard cells. However, it is still unclear whether CDPK directly phosphorylates the ion channel or phosphorylates an intermediary regulatory protein(s). In addition, the study of Pei et al. (1996) utilized recombinant Arabidopsis CDPK protein purified from Escherichia coli. There is no direct evidence that CDPK exists in guard cells, which have not only unique morphology and highly specialized metabolism but also unique responses to environmental signals.

Electrophysiological studies using general protein kinase inhibitors indicate that inward K⁺ channels, outward K⁺ channels, and anion channels of the guard cell plasma membrane may be modulated by phosphorylation (Armstrong et al., 1995; Schmidt et al., 1995). However, the kinases involved in regulation of these ion channels remain to be identified. Inward K⁺ channels provide a major pathway for K⁺ uptake into plant cells including guard cells (Schroeder et al., 1994). KAT1, a plant K⁺ channel gene, was initially cloned from Arabidopsis by complementation of a K⁺-transport-deficient strain of Saccharomyces cerevisiae (Anderson et al., 1992). Electrophysiological studies utilizing heterologous expression of KAT1 in Xenopus oocytes, S. cerevisiae, or the insect cell line Sf9 indicate that KAT1 encodes a voltage-gated inward K⁺ channel (Schachtman et al., 1992; Bertl et al., 1995; Hoshi, 1995; Marten et al., 1996). Moreover, the KAT1 gene is primarily expressed in guard cells of transgenic Arabidopsis plants (Nakamura et al., 1995). These results indicate that the KAT1 protein is very likely to be the guard cell inward K⁺ channel in which activity is inhibited by elevated cytosolic Ca²⁺ levels (Schroeder and Hagiwara, 1989; Blatt et al., 1990; Lemtiri-Chlieh and MacRobbie, 1994). Sequence analysis of KAT1 suggests that the deduced protein from the KAT1 gene is rich in potential phosphorylation sites for protein kinases. However, whether the KAT1 protein can be phosphorylated by protein kinases remains to be determined.

In the present study we biochemically identify and characterize a CDPK from guard cells of V. faba. We also demonstrate that this guard cell CDPK can phosphorylate the KAT1 protein in a Ca²⁺-dependent manner. Our data suggest that CDPK may be involved in Ca²⁺-regulated modulation of plasma membrane ion channels in guard cells.

	MATERIALS AND METHODS

Top Abstract Introduction Methods Results Discussion References

Chemicals

Plant Material

2

1

Preparation of Proteins

Gel Electrophoresis

In-Gel Autophosphorylation and Kinase Assays

1

1

1

In Vitro Protein Kinase Activity Assay

1

1

Immunoblotting

Phosphorylation of Guard Cell Proteins

Transcription and Translation of KAT1

Phosphorylation of KAT1 Protein

1

	RESULTS

Top Abstract Introduction Methods Results Discussion References

Identification of a 57-kD Kinase from Guard Cells as a CDPK

View larger version (61K): [in this window] [in a new window]   Figure 1. Detection of Ca2+-dependent autophosphorylation activity in guard cells. Soluble fractions (lanes S, 40 µg of protein) and membrane fractions (lanes M, 20 µg of protein) from GCPs were separated on a 12% SDS-polyacrylamide gel. After proteins in the gels were denatured and renatured, autophosphorylation activity was detected in gel in the presence of 100 µm Ca2+ (A) or 450 µm EGTA (B). The molecular masses of protein standards (10-kD protein ladder) are shown at the left of each panel in kilodaltons. The arrows and asterisks at the right of each panel indicate the positions of the 57- and 38-kD radiolabeled proteins, respectively.

We next examined the catalytic activity of the 57-kD protein using histone III-S as a substrate. The 57-kD protein with Ca²⁺-dependent autophosphorylation was gel purified (see ``Materials and Methods'') and then incubated with histone III-S and [-³²P]ATP in the presence of Ca²⁺ or EGTA. The proteins were then separated by SDS-PAGE and ³²P-labeled polypeptides were detected by autoradiography. In the presence of 450 µm EGTA, only minor phosphorylation of histone III-S was detected (Fig. 2, lane 3). In contrast, histone III-S was strongly phosphorylated in the presence of 20 µm Ca²⁺ (Fig. 2, lane 4). When the 57-kD protein was omitted from the incubation system (histone III-S and [-³²P]ATP only), no phosphorylation of histone III-S was observed (Fig. 2, lanes 1 and 2). These results show that the catalytic activity of the 57-kD kinase is also Ca²⁺ dependent. In addition, the Ca²⁺-dependent phosphorylation of histone III-S by the 57-kD kinase could be inhibited by the calmodulin antagonists TFP and W7 (Fig. 2, lanes 5 and 6). By contrast, W-5, an inactive analog of W-7, had no apparent effect on histone III-S phosphorylation catalyzed by the 57-kD kinase in the presence of Ca²⁺ (Fig. 2, lane 7).

View larger version (31K): [in this window] [in a new window]   Figure 2. Ca2+-dependent catalytic activity of the 57-kD kinase and its inhibition by calmodulin (CaM) antagonists. The catalytic activity of the gel-purified 57-kD kinase was assayed using histone type III-S as a substrate in the presence of 20 µm Ca2+ or 450 µm EGTA as described in ``Materials and Methods''. TFP, W-7, or W-5 each was used at 250 µm. The phosphorylated histone III-S was resolved on a 12% polyacrylamide gel. , Absent; +, present.

Ca²⁺-binding proteins such as CDPK migrate in gels at different rates in the Ca²⁺-bound versus Ca²⁺-free state (Roberts and Harmon, 1992). To investigate this phenomenon for the 57-kD kinase, Ca²⁺ or EGTA was added to guard cell protein samples just before electrophoresis and then the proteins in the gel were assayed for autophosphorylation (Fig. 3, A and B) or kinase activity (Fig. 3, C and D). When the autophosphorylation assay was performed in the presence of Ca²⁺, a ³²P-labeled polypeptide with different mobilities was observed (Fig. 3A, arrows). The apparent molecular mass of this polypeptide was 52 kD when the sample buffer contained Ca²⁺ and 57 kD when the sample buffer contained EGTA. When the autophosphorylation assay was performed in the presence of EGTA, the ³²P-labeled polypeptide with a Ca²⁺-induced mobility shift was no longer detected (Fig. 3B). In contrast, the 38-kD band that showed autophosphorylation that was not Ca²⁺ dependent (Fig. 1, asterisks) did not exhibit a Ca²⁺-dependent electrophoretic mobility shift (Fig. 3, A and B, asterisks). In accord with the autophosphorylation assay, a major ³²P-labeled band with a Ca²⁺-induced mobility shift (52 or 57 kD in the presence of Ca²⁺ or EGTA, respectively) was also observed by an in-gel kinase assay in which histone III-S was included in the gel as a substrate (Fig. 3, C and D, arrows). The activity of this kinase as determined by the in-gel assay was strongly enhanced by Ca²⁺ (Fig. 3, C and D, arrows), which is consistent with the in vitro kinase activity assay (Fig. 2, lanes 3 and 4). In addition to the 57-kD band with a Ca²⁺-induced mobility shift, several faint ³²P-labeled bands were found, but neither the intensities nor the mobilities of these bands seemed to be affected by Ca²⁺ (Fig. 3, C and D). These results demonstrate that the 57-kD kinase that has Ca²⁺-dependent autophosphorylation and catalytic activities exhibits a Ca²⁺-induced electrophoretic mobility shift.

View larger version (61K): [in this window] [in a new window]   Figure 3. The 57-kD kinase exhibits a Ca2+-dependent electrophoretic mobility shift in both autophosphorylation and in-gel activity assays. Ca2+ or EGTA to a final concentration of 2 mm was added to the GCP-soluble proteins dissolved in SDS-PAGE sample buffer. The Ca2+- and EGTA-treated samples were loaded (A and B, 40 µg of protein per lane; C and D, 20 µg of protein per lane) with a blank lane between the two samples and resolved on 12% polyacrylamide gels. Autophosphorylation (A and B) and in-gel kinase activity (C and D) assays were performed by incubating the renatured gels with [-32P]ATP in the presence of 100 µm free Ca2+ (A and C) or 450 µm EGTA (B and D). The presence of histone III-S (0.5 mg mL1) in the polyacrylamide separating gel precludes protein staining; therefore, prestained protein molecular mass standards are used in the kinase activity assay (C and D). The arrows indicate the positions of the 57-kD kinase with a Ca2+-induced electrophoretic mobility shift. The asterisks indicate the positions of the 38-kD kinase that does not exhibit a Ca2+-induced electrophoretic mobility shift.

We further assessed the relatedness of the 57-kD Ca²⁺-dependent kinase from GCPs to CDPKs. As shown in Figure 4A, affinity-purified antibodies to the calmodulin-like domain of soybean CDPK cross-reacted with a single band at 57 or 52 kD from EGTA- or Ca²⁺-treated protein samples, respectively. When the CDPK antibodies were replaced by nonimmune serum, no bands were detected (Fig. 4B). These results demonstrated that the 57-kD kinase that exhibited a Ca²⁺-induced electrophoretic mobility shift can be recognized by CDPK antibodies. The identity of the 57-kD kinase and the 57-kD band recognized by affinity-purified CDPK antibodies was further confirmed by two-dimensional electrophoretic analysis. The autophosphorylation assay was performed on a portion of the two-dimensional gel between the 30- and 70-kD molecular mass standards (Fig. 5, A and B). Although the background in the two-dimensional autophosphorylation assay was higher than that of the one-dimensional autophosphorylation assay, as also observed by other researchers (Keen et al., 1987), a ³²P-labeled 57-kD protein spot was consistently found when the autophosphorylation assay was carried out in the presence of 100 µm Ca²⁺ (n = 5, Fig. 5B). However, this ³²P-labeled 57-kD protein spot was no longer detected when the autophosphorylation assay was performed in the presence of 450 µm EGTA (n = 5, Fig. 5A), indicating that the autophosphorylation of the 57-kD protein spot is Ca²⁺ dependent. Furthermore, on two-dimensional immunoblots, the affinity-purified CDPK antibodies detected a 57-kD protein spot with a position identical to that of the 57-kD protein spot with Ca²⁺-dependent autophosphorylation (n = 2, Fig. 5C), confirming that they are the same protein. Taken together, all of the results consistently indicate that the 57-kD kinase from guard cells is a CDPK.

View larger version (70K): [in this window] [in a new window]   Figure 4. Affinity-purified CDPK antibodies cross-react with the 57-kD kinase. The soluble (S) or membrane (M) proteins (80 µg per lane) were treated with Ca2+ or EGTA as described in Figure 3 and resolved on a 12% polyacrylamide gel. The proteins were then transferred to nitrocellulose membranes and subjected to immunostaining with affinity-purified CDPK antibodies (A) or nonimmune rabbit serum (B) as described in ``Materials and Methods''.

View larger version (89K): [in this window] [in a new window]   Figure 5. Detection of the 57-kD kinase from GCPs on two-dimensional gels. Soluble proteins (50 µg) from GCPs were resolved by IEF in the horizontal dimension and then by SDS-PAGE in the vertical dimension. Proteins on the two-dimensional gels were subjected to the autophosphorylation assay in the presence of 450 µm EGTA (n = 5, A) or 100 µm Ca2+ (n = 5, B) or transferred to a nitrocellulose membrane that was then probed with affinity-purified CDPK antibodies (n = 2, C). The arrows indicate the position of the protein spot showing Ca2+-dependent autophosphorylation (B) or the identical position of the protein spot recognized by the CDPK antibody (C).

Micromolar Levels of Ca²⁺ Stimulate Protein Phosphorylation

View larger version (47K): [in this window] [in a new window]   Figure 6. Ca2+ stimulates phosphorylation of certain guard cell proteins. Protein phosphorylation was performed by incubating 30 µg of soluble (S) proteins (A) or membrane (M) proteins (B) with [-32P]ATP in the presence of various concentrations of free Ca2+ ([Ca2+]f). Phosphorylation of soluble proteins (30 µg) in the presence of 10 µm CsA (lanes +) or 0.2% ethanol (lanes ) at nominally 0 or 1 µm free Ca2+ was performed as described in ``Materials and Methods''. The phosphoproteins were resolved on 5 to 20% gradient polyacrylamide gels. The arrows indicate the positions of proteins with Ca2+-stimulated phosphorylation. The asterisk indicates the position of a protein with CsA-enhanced phosphorylation. The 120- and 52-kD proteins that exhibited Ca2+-stimulated phosphorylation (indicated by the highest and the lowest arrows) also exhibited CsA-enhanced phosphorylation.

Since the phosphorylation status of a given protein is determined by the balance of activities of protein kinases and protein phosphatases, we further examined the effect of CsA (a specific inhibitor of the Ca²⁺-dependent protein phosphatase calcineurin) on phosphorylation of guard cell proteins. We did not see any effect of CsA on phosphorylation of membrane proteins from GCPs (data not shown). However, in the presence of 1 µm Ca²⁺, 10 µm CsA did increase the phosphorylation of the 120-, 52-, and 35-kD polypeptides from the soluble fraction of GCPs (Fig. 6C, the highest and the lowest arrows and the asterisk). In the absence of Ca²⁺, CsA had no apparent effect on the phosphorylation of these proteins (Fig. 6C). Although the CsA-induced changes in protein phosphorylation were not as marked as those induced by Ca²⁺, they were consistently observed in four replicate experiments. These results suggest that a Ca²⁺-dependent protein phosphatase may be involved in regulating the Ca²⁺-dependent phosphorylation status of guard cell proteins.

Guard Cell CDPK Phosphorylates KAT1 Protein in a Ca²⁺-Dependent Manner

View larger version (77K): [in this window] [in a new window]   Figure 7. CDPK phosphorylates KAT1 protein in a Ca2+-dependent manner. The membrane (Mem; A, lanes 1-6, and 10) or supernatant (Sup; B, lanes 1-3) fractions of the product translated in the presence of microsome membranes were subjected to phosphorylation by CDPK in the presence of [-32P]ATP and Ca2+ (A, lanes 4-6, 9 and 10; B, lanes 1-3) or EGTA (A, lanes 1-3) as described in ``Materials and Methods''. A, Lane 8, and B, lane 5, [35S]Met-labeled translation product from the KAT1 cDNA template. A, Lane 7 and B, lane 4, [35S]Met-labeled translation product without DNA template. In lane 9, the microsome membranes were added to the translation system just after translation and then phosphorylated by CDPK in the presence of Ca2+. The arrows indicate the position of the 35S-labeled KAT1 protein; note the corresponding phosphorylated band in lanes 6 and 10. The protein samples (20 µg of protein per lane) were resolved on 9% polyacrylamide gels. , Absent; +, present.

View larger version (83K): [in this window] [in a new window]   Figure 8. AAPK does not phosphorylate the KAT1 protein. The membrane (Mem, A, lanes 3 and 4) or supernatant (Sup, B, lanes 3 and 4) fractions of the product translated in the presence of microsome membranes were subjected to phosphorylation by AAPK purified from GCPs treated with ABA (lanes 4) or without ABA (lanes 3). Lanes 1, [35S]Met-labeled translation product without KAT1 cDNA template. Lanes 2, [35S]Met-labeled translation product from the KAT1 cDNA template. The arrows indicate the position of the 35S-labeled KAT1 protein. The protein samples (20 µg of protein per lane) were resolved on 9% polyacrylamide gels. , Absent; +, present.

To examine phosphorylation of the KAT1 protein by the guard cell CDPK, the translated proteins were incubated with CDPK in the presence of [-³²P]ATP. In the presence of Ca²⁺ and in the absence of CDPK, a number of phosphoproteins were found in the translation product in the membrane fraction (Fig. 7A, lane 4), indicating the presence of endogenous kinase activities in the in vitro transcription/translation system. However, in the presence of CDPK and Ca²⁺, a new phosphoprotein at 79 kD was detected in the translated product when the KAT1 cDNA was used as the template (Fig. 7A, lanes 6 and 10), but was not detected when the KAT1 DNA template was omitted (Fig. 7A, lane 5). In addition, when canine pancreatic microsomes were added to the translation system just after translation rather than concomitantly, the 79-kD phosphoprotein was not found in the membrane fraction of the translation product (Fig. 7A, lane 9), suggesting that the 79-kD phosphoprotein found in the membrane fraction (Fig. 7A, lanes 6 and 10) is not an endogenous membrane protein but rather the translation product that translocated or inserted into the membrane during translation. In contrast, in the presence of EGTA, no 79-kD phosphoprotein was observed in any of the three treatments (Fig. 7A, lanes 1-3). These data show that CDPK can, in a Ca²⁺-dependent fashion, phosphorylate the translated KAT1 protein in the membrane fraction. When the translated products in the supernatant fraction were analyzed for phosphorylation, no 79-kD phosphoprotein was found despite the fact that ³⁵S labeling confirmed the presence of KAT1 protein in the supernatant as well as the membrane fraction (Fig. 7B). In sum, these results suggest that KAT1 can be phosphorylated by CDPK only when it is membrane localized.

We next examined whether the KAT1 protein can be phosphorylated by another guard cell protein kinase, AAPK, which is Ca²⁺ independent and ABA activated (Li and Assmann, 1996). As shown in Figure 8, no 79-kD phosphoprotein was found in either the membrane or the supernatant fractions of the translated products, suggesting that AAPK could not phosphorylate the KAT1 protein.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

A 57-kD protein kinase has been identified in V. faba GCPs. Autophosphorylation of this kinase, like CDPKs (Harmon et al., 1987; Binder et al., 1994), is Ca²⁺ dependent (Figs. 1 and 3, A and B). Both in solution and in gel, activity assays show that the catalytic activity of this kinase is Ca²⁺ dependent (Fig. 2 and Fig. 3, C and D). Furthermore, this kinase, like soybean CDPK (one of the best-characterized CDPKs), exhibits a Ca²⁺-induced electrophoretic mobility shift on SDS-PAGE gels (Figs. 3 and 4). The Ca²⁺-induced electrophoretic mobility shift of this kinase suggests that it may contain a calmodulin-like domain (Harper et al., 1991; Roberts and Harmon, 1992). Evidence further supporting this notion comes from the following observations: (a) this kinase can be specifically recognized by affinity-purified antibodies to the calmodulin-like domain of soybean CDPK on both one- and two-dimensional blots (Figs. 3 and 5); (b) the Ca²⁺-dependent catalytic activity can be inhibited by the calmodulin antagonists TFP and W-7 but not by the inactive analog W-5 (Fig. 2). Taken together, all of the data indicate that the 57-kD protein kinase from V. faba GCPs is a Ca²⁺-dependent protein kinase with a calmodulin-like domain (CDPK).

It has been well documented from physiological studies of stomata that Ca²⁺ plays very important roles in mediating stomatal closure and opening (for review, see McAinsh et al., 1997). However, signaling intermediaries between Ca²⁺ and stomatal responses have not been identified. CDPKs are encoded by a large gene family and members of the CDPK family expressed in different cell types may have distinct roles in signal transduction (Estruch et al., 1994; Abo-El-Saad and Wu, 1995; Hrabak et al., 1996). Biochemical identification of a CDPK in guard cells, one of the most specialized cell types in plants, strengthens the notion that CDPK mediates at least a subset of Ca²⁺-regulated stomatal responses (Assmann, 1993).

To examine potential targets of CDPK in guard cells, phosphorylation of guard cell proteins was performed at various Ca²⁺ concentrations. Physiological levels of free Ca²⁺ (1 µm) promote the phosphorylation of a number of soluble and membrane proteins from GCPs (Fig. 6, A and B). A similar experiment was carried out by Kinoshita and Shimazaki (1995). However, in contrast to their observations, we did not see Ca²⁺-stimulated phosphorylation of proteins at 41, 31, and 25 kD in either the soluble or membrane fractions (Fig. 6, A and B). Instead, we found that 1 µm free Ca²⁺ stimulated phosphorylation of several proteins with molecular masses different from those that they observed (Fig. 6, A and B). The differences may be due to different plant growth environments (Kinoshita and Shimazaki [1995] used greenhouse grown plants) or different methods used for isolating GCPs. Differences in guard cell Cl content from V. faba plants grown in a growth chamber versus a greenhouse have been reported (Talbott and Zeiger, 1996), suggesting that different growth environments may also affect other cellular processes, such as protein phosphorylation.

The level of phosphorylation is determined by the relative activities of protein kinases and protein phosphatases. In this light, it is interesting to note that the Ca²⁺-stimulated phosphorylation of the 120- and 52-kD soluble polypeptides from GCPs is enhanced by CsA (Fig. 6C), a specific inhibitor of the Ca²⁺-dependent protein phosphatase calcineurin (Liu et al., 1992). A Ca²⁺-dependent phosphatase activity that is inhibited by CsA has been detected in epidermal peels of V. faba (Luan et al., 1993). In addition, CsA has been recently shown to antagonize ABA-induced stomatal closure and ABA-inhibited stomatal opening (Hey et al., 1997). Our biochemical data suggest that CsA-sensitive Ca²⁺-dependent protein phosphatase(s) as well as CDPK are present not only in the epidermis but specifically in guard cells, and that they are involved in the Ca²⁺-regulated phosphorylation of guard cell proteins and therefore Ca²⁺-regulated stomatal responses.

The phosphorylation of several proteins was enhanced by micromolar levels of Ca²⁺ (Fig. 6), suggesting that these proteins are potential targets of the guard cell CDPK. However, without other information, the identities of these proteins are very difficult to determine because of the limited amount of proteins that can be obtained from GCPs. In particular, ion channel proteins, which are key players in osmotic regulation of stomatal aperture, are present at very low concentrations (Sussman and Harper, 1989). The low abundance and water-insoluble nature of channel proteins make biochemical analysis of ion channels very difficult. Sequence analysis of KAT1, the inward K⁺ channel gene primarily expressed in guard cells (Nakamura et al., 1995), suggests that the KAT1 protein contains a number of potential phosphorylation sites for protein kinases. Therefore, it would be reasonable to examine whether the KAT1 protein can be phosphorylated by the CDPK from guard cells.

Although electrophysiological techniques have been widely used to study ion channel regulation, including phosphorylation, these approaches cannot distinguish whether regulation of an ion channel by phosphorylation is due to direct phosphorylation of the channel protein or whether the regulation is due to phosphorylation of some intermediate protein(s), which in turn affects the channel activity. To circumvent these problems, we translated the cloned KAT1 gene in vitro in the presence of canine pancreatic microsomal membranes. The inclusion of microsome membranes in the in vitro translation system has been widely used to study membrane insertion and translocation of channel proteins (Miao et al., 1992; Dunlop et al., 1995). The translated ³⁵S-labeled-KAT1 protein was found in both membrane and supernatant fractions (Figs. 7 and 8), as was also true for the expression of a rat brain K⁺ channel gene RCK1 in Xenopus oocytes (Ivanina et al., 1994). However, guard cell CDPK only phosphorylated the translated KAT1 protein in the membrane fraction and only if microsome membranes were provided during the translation step (Fig. 7). Since it is known that, upon insertion into the membrane, integral proteins fold into a conformation that exposes the hydrophobic residues to the lipid bilayer (Singer, 1990), our observation implies that CDPK may phosphorylate the KAT1 protein only when it is correctly configured in the membrane. The fact that phosphorylation of KAT1 by CDPK is not indiscriminate suggests that this phosphorylation will be of physiological relevance for channel regulation. Moreover, the phosphorylation of the KAT1 protein seems to be specific to CDPK, since another guard cell protein kinase, AAPK, which is Ca²⁺ independent and ABA activated (Li and Assmann, 1996), was not able to phosphorylate the KAT1 protein (Fig. 8). Because of the very low abundance of channel proteins in cells (Sussmann and Harper, 1989), phosphorylation of channel proteins such as the KAT1 protein homolog is not expected to be detected in the microsome membrane fraction of GCPs (Fig. 6B). Furthermore, the guard cell inward K⁺ channel from V. faba has not yet been purified or cloned, so it cannot be identified by molecular mass.

Preliminary studies involving expression of KAT1 in Xenopus oocytes have shown that inward K⁺ currents were in fact greatly reduced when CDPK was co-expressed with KAT1 (Kamasani et al., 1997). Electrophysiological studies of guard cells have shown that the inward K⁺ channel can be inhibited by increased cytosolic Ca²⁺ (Schroeder and Hagiwara, 1989; MacRobbie, 1997). However, it is not clear from the study of Kamasani et al. (1997) whether CDPK directly phosphorylates the KAT1 protein or whether it phosphorylates other proteins. Such proteins could be regulators of KAT1 activity or other ion transporters such as the H⁺ or Ca²⁺ ATPase, modulation of which could result in altered intracellular ion concentrations that might in turn affect K⁺ channel activity.

Our data showing that CDPK phosphorylates the KAT1 protein itself in a Ca²⁺-dependent manner suggests that the inhibition of the inward K⁺ channel in guard cells by Ca²⁺ could be mediated by direct Ca²⁺-dependent phosphorylation of the inward K⁺ channel by CDPK. At first sight, this result seems contradictory to the results from the study by Luan et al. (1993), whose electrophysiological data suggested that the inhibition of the inward K⁺ channel in guard cells by Ca²⁺ was mediated by a Ca²⁺-dependent dephosphorylation mechanism. However, it is still not known whether Ca²⁺ stimulates dephosphorylation of the inward K⁺ channel itself or whether regulatory proteins are the dephosphorylation target. Even if the channel were to be directly dephosphorylated by a Ca²⁺-dependent phosphatase, Ca²⁺-dependent dephosphorylation and Ca²⁺-dependent phosphorylation would not necessarily occur at the same site(s), since the KAT1 K⁺ channel protein is rich in potential phosphorylation sites. In addition, in vitro phosphorylation data showed that Ca²⁺-dependent phosphorylation of certain soluble proteins from GCPs could be enhanced by CsA, a specific inhibitor of Ca²⁺-dependent protein phosphatase (Fig. 6C).

Taken together, our results and those of Luan et al. (1993) imply that the effect of Ca²⁺ on the inward K⁺ channel may involve phosphorylation and dephosphorylation on multiple sites of the K⁺ channel protein or multiple routes of phosphorylation and dephosphorylation to achieve a fine modulation of the K⁺ channel in response to a variety of environmental stimuli. Further studies of KAT1 incorporated into lipid bilayers (Rosenberg and East, 1992) or expressed in Xenopus oocytes should shed new light on the regulation of this inward K⁺ channel.

	FOOTNOTES

   Received August 29, 1997; accepted November 3, 1997.

	ABBREVIATIONS

Abbreviations: AAPK, ABA-activated protein kinase. CDPK, calcium-dependent protein kinase containing a calmodulin-like domain. CsA, cyclosporin A. GCP, guard cell protoplast. KAT1, a potassium channel cDNA from Arabidopsis. TFP, trifluoperazine. W-5, N-(6-aminohexyl)-1-naphthalenesulfonamide. W-7, N-(6-aminohexyl)-5-chloro-1-naphthalenesulfonamide.

	ACKNOWLEDGMENTS

We would like to thank Dr. Alice Harmon for providing CDPK antibodies, Drs. Leon Kochian, Gerald Berkowitz, and Julian Schroeder for providing KAT1 cDNA plasmids, and Dr. Simon Gilroy for helpful discussion concerning initial experiments.

	LITERATURE  CITED

Top Abstract Introduction Methods Results Discussion References

Abo-El-Saad M, Wu R (1995) A rice membrane calcium-dependent protein kinase is induced by gibberellin. Plant Physiol 108: 787-793 [Abstract]

Anderson JA, Huprikar SS, Kochian LV, Lucas WJ, Gaber RF (1992) Functional expression of a probable Arabidopsis thaliana potassium channel in Saccharomyces cerevisiae. Proc Natl Acad Sci USA 89: 3736-3740 [Abstract/Free Full Text]

Armstrong F, Leung J, Grabov A, Brearley J, Giraudat J, Blatt MR (1995) Sensitivity to abscisic acid of guard cell K⁺ channels is suppressed by abi-1, a mutant Arabidopsis gene encoding a putative protein phosphatase. Proc Natl Acad Sci USA 92: 9520-9524 [Abstract/Free Full Text]

Assmann SM (1993) Signal transduction in guard cells. Annu Rev Cell Biol 9: 345-375 [CrossRef][Web of Science]

Bachmann M, Shiraishi N, Campbell WH, Yoo B-C, Harmon AC, Huber SC (1996) Identification of Ser-543 as the major regulatory phosphorylation site in spinach leaf nitrate reductase. Plant Cell 8: 505-517 [Abstract]

Bertl A, Anderson JA, Slayman CL, Gaber RF (1995) Use of Saccharomyces cerevisiae for patch-clamp analysis of heterologous membrane proteins: characterization of Kat1, an inward-rectifying K⁺ channel from Arabidopsis thaliana, and comparison with endogenous yeast channels and carriers. Proc Natl Acad Sci USA 92: 2701-2705 [Abstract/Free Full Text]

Binder BM, Harper JF, Sussman MR (1994) Characterization of an Arabidopsis calmodulin-like domain protein kinase purified from Escherichia coli using an affinity sandwich technique. Biochemistry 33: 2033-2041 [CrossRef][Medline]

Blatt MR, Thiel G, Trentham DR (1990) Reversible inactivation of K⁺ channels of Vicia stomatal guard cells following the photolysis of caged inositol 1,4,5-trisphosphate. Nature 346: 766-769 [CrossRef][Medline]

Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72: 248-254 [CrossRef][Web of Science][Medline]

Chang D, Hsieh PS, Dawson DC (1988) Calcium: a program in BASIC for calculating the composition of solutions with specified free concentrations of calcium, magnesium and other divalent cations. Comput Biol Med 18: 351-366 [CrossRef][Web of Science][Medline]

De Silva DLR, Hetherington AM, Mansfield TA (1985) Synergism between calcium ions and abscisic acid in preventing stomatal opening. New Phytol 100: 473-482 [CrossRef][Web of Science]

Dunlop J, Jones PC, Finbow ME (1995) Membrane insertion and assembly of ductin: a polytopic channel with dual orientations. EMBO J 14: 3609-3616 [Web of Science][Medline]

Estruch JJ, Kadwell S, Merlin E, Crossland L (1994) Cloning and characterization of a maize pollen-specific calcium-dependent calmodulin-independent protein kinase. Proc Natl Acad Sci USA 91: 8837-8841 [Abstract/Free Full Text]

Gilroy S, Fricker MD, Read ND, Trewavas AJ (1991) Role of calcium in signal transduction of Commelina guard cells. Plant Cell 3: 333-344 [Abstract/Free Full Text]

Harmon AC, Putnam-Evans C, Cormier MJ (1987) A calcium-dependent but calmodulin-independent protein kinase from soybean. Plant Physiol 83: 830-837 [Abstract/Free Full Text]

Harper JF, Sussman MR, Schaller GE, Putnam-Evans C, Charbonneau H, Harmon AC (1991) A calcium-dependent protein kinase with a regulatory domain similar to calmodulin. Science 252: 951-954 [Abstract/Free Full Text]

Hedrich R, Busch H, Raschke K (1990) Ca²⁺ and nucleotide dependent regulation of voltage dependent anion channels in the plasma membrane of guard cells. EMBO J 9: 3889-3892 [Web of Science][Medline]

Hey SJ, Bacon A, Burnett E, Neill SJ (1997) Abscisic acid signal tranduction in epidermal cells of Pisum sativum L. Argenteum: both dehydrin mRNA accumulation and stomatal responses require protein phosphorylation and dephosphorylation. Planta 202: 85-92 [CrossRef][Web of Science]

Hochstrasser DF, Harrington MG, Hochstrasser A-C, Miller MJ, Merril CR (1988) Methods for increasing the resolution of two-dimensional protein electrophoresis. Anal Biochem 173: 424-435 [CrossRef][Web of Science][Medline]

Hoshi T (1995) Regulation of voltage dependence of the KAT1 channel by intracellular factors. J Gen Physiol 105: 309-328 [Abstract/Free Full Text]

Hrabak EM, Dickmann LJ, Satterlee JS, Sussmann MR (1996) Characterization of eight new members of the calmodulin-like domain protein kinase gene family of Arabidopsis thaliana. Plant Mol Biol 31: 405-412 [CrossRef][Web of Science][Medline]

Ivanina T, Perets T, Thornhill WB, Levin G, Dascal N, Lotan I (1994) Phosphorylation by protein kinase A of RCK1 K⁺ channels expressed in Xenopus oocytes. Biochemistry 33: 8786-8792 [CrossRef][Medline]

Kamasani U, Zhang X, Lawton M, Berkowitz GA (1997) Ca²⁺-dependent protein kinase modulates activity of the K⁺ channel KAT1 (abstract no. 980). Plant Physiol 114: S-197

Kameshita I, Fujisawa H (1989) A sensitive method for detection of calmodulin-dependent protein kinase II activity in sodium dodecyl sulfate-polyacrylamide gel. Anal Biochem 183: 139-143 [CrossRef][Web of Science][Medline]

Keen JH, Chestnut MH, Beck KA (1987) The clathrin coat assembly polypeptide complex: autophosphorylation and assembly activities. J Biol Chem 262: 3864-3871 [Abstract/Free Full Text]

Kim JS, Raines RT (1993) Ribonuclease S-peptide as a carrier in fusion proteins. Protein Sci 2: 348-356 [Web of Science][Medline]

Kinoshita T, Nishimura M, Shimazaki K (1995) Cytosolic concentration of Ca²⁺ regulates the plasma membrane H⁺-ATPase in guard cells of fava bean. Plant Cell 7: 1333-1342 [Abstract]

Kinoshita T, Shimazaki K (1995) Evidence for Ca²⁺-dependent protein phosphorylation in vitro in guard cells from Vicia faba L. Plant Sci 110: 173-180 [CrossRef]

Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 277: 680-685

Lemtiri-Chlieh F, MacRobbie EAC (1994) Role of calcium in the modulation of Vicia guard cell potassium channels by abscisic acid: a patch clamp study. J Membr Biol 137: 99-107 [Web of Science][Medline]

Li B, Chollet R (1993) Resolution and identification of C₄ phosphoenolpyruvate-carboxylase protein-kinase polypeptides and their reversible light activation in maize leaves. Arch Biochem Biophys 307: 416-419 [CrossRef][Web of Science][Medline]

Li J, Assmann SM (1996) An abscisic acid-activated and calcium-independent protein kinase from guard cells of fava bean. Plant Cell 8: 2359-2368 [Abstract]

Ling V, Assmann SM (1992) Cellular distribution of calmodulin and calmodulin-binding proteins in Vicia faba L. Plant Physiol 100: 970-978 [Abstract/Free Full Text]

Liu J, Albers MW, Wandless, TJ, Luan S, Alberg DG, Belsshaw PJ, Cohen P, MacKintosh C, Klee CB, Schreiber SL (1992) Inhibition of T cell signaling by immunophilin-ligand complexes correlates with loss of calcineurin phosphatase activity. Biochemistry 31: 3896-3901 [CrossRef][Medline]

Luan S, Li W, Rusnak, F, Assmann SM, Schreiber SL (1993) Immunosuppressants implicate protein phosphatase regulation of K⁺ channels in guard cells. Proc Natl Acad Sci USA 90: 2202-2206 [Abstract/Free Full Text]

MacRobbie EAC (1997) Signalling in guard cells and regulation of ion channel activity. J Exp Bot 48: 515-528

Marten I, Gaymard F, Lemaillet G, Thibaud J-B, Sentenac H, Hedrich R (1996) Functional expression of the plant K⁺ channel KAT1 in insect cells. FEBS Lett 380: 229-232 [Medline]

McAinsh MR, Brownlee C, Hetherington AM (1997) Calcium ions as second messengers in guard cell signal transduction. Physiol Plant 100: 16-29 [CrossRef]

Miao GH, Hong Z, Verma DPS (1992) Topology and phosphorylation of soybean nodulin-26, an intrinsic protein of the peribacteroid membrane. J Cell Biol 118: 481-490 [Abstract/Free Full Text]

Nakamura RL, McKendree WL, Hirsch RE, Sedbrook JC, Gaber RF, Sussman MR (1995) Expression of an Arabidopsis potassium channel gene in guard cells. Plant Physiol 109: 371-374 [Abstract]

Pei Z-M, Ward JM, Harper JF, Schroeder JI (1996) A novel chloride channel in Vicia faba guard cell vacuoles activated by the serine/threonine kinase, CDPK. EMBO J 15: 6564-6574 [Web of Science][Medline]

Roberts DM, Harmon AC (1992) Calcium-modulated proteins: targets of intracellular calcium signals in higher plants. Annu Rev Plant Physiol Plant Mol Biol 43: 375-414 [CrossRef][Web of Science]

Rosenberg RL, East JE (1992) Cell-free expression of functional Shaker potassium channels. Nature 360: 166-169 [CrossRef][Medline]

Schachtman DP, Schroeder JI, Lucas WJ, Anderson JA, Gaber RF (1992) Expression of an inward-rectifying potassium channel by the Arabidopsis KAT1 cDNA. Science 258: 1654-1658 [Abstract/Free Full Text]

Schmidt C, Schelle I, Liao Y, Schroeder JI (1995) Strong regulation of slow anion channels and abscisic acid signaling in guard cells by phosphorylation and dephosphorylation events. Proc Natl Acad Sci USA 92: 9535-9539 [Abstract/Free Full Text]

Schroeder JI, Hagiwara S (1989) Cytosolic calcium regulates ion channels in the plasma membrane of Vicia faba guard cells. Nature 338: 427-430 [CrossRef][Web of Science]

Schroeder JI, Ward JM, Gassmann W (1994) Perspectives on the physiology and structure of inward-rectifying K⁺ channels in higher plants: biophysical implications for K⁺ uptake. Annu Rev Biophys Biomol Struct 23: 441-471 [Web of Science][Medline]

Schwartz A (1985) Role of Ca²⁺ and EGTA on stomatal movements in Commelina communis L. Plant Physiol 79: 1003-1005 [Abstract/Free Full Text]

Schwartz A, Ilan N, Grantz DA (1988) Calcium effects on stomatal movements in Commelina communis. Plant Physiol 87: 583-587 [Abstract/Free Full Text]

Shimazaki K, Kinoshita T, Nishimura M (1992) Involvement of Ca²⁺/calmodulin-dependent myosin light chain kinase in blue light-dependent H⁺ pumping of guard cell protoplasts from Vicia faba L. Plant Physiol 99: 1416-1421 [Abstract/Free Full Text]

Short TW, Porst M, Briggs WR (1992) A photoreceptor system regulating in vivo and in vitro phosphorylation of a pea plasma membrane protein. Photochem Photobiol 55: 773-781 [Web of Science]

Singer SJ (1990) The structure and insertion of integral proteins in membranes. Annu Rev Cell Biol 6: 247-296 [CrossRef][Web of Science]

Smith JA, Francis SH, Corbin JD (1993) Autophosphorylation: a salient feature of protein kinases. Mol Cell Biochem 127: 51-73

Sussman MR, Harper JF (1989) Molecular biology of the plasma membrane of higher plants. Plant Cell 1: 953-960 [Free Full Text]

Talbott LD, Zeiger E (1996) Central roles for potassium and sucrose in guard-cell osmoregulation. Plant Physiol 111: 1051-1057 [Abstract]

Towbin H, Staehelin T, Gordon T (1979) Electrophoretic transfer of proteins from polyacrylamide gel to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 76: 4350-4354 [Abstract/Free Full Text]

Ward JM, Schroeder JI (1994) Calcium-activated K⁺ channels and calcium-induced calcium release by slow vacuolar ion channels in guard cell vacuoles implicated in the control of stomatal closure. Plant Cell 6: 669-683 [Abstract/Free Full Text]

Yao B, Zhang Y, Delikat S, Mathias S, Basu S, Kolesnick R (1995) Phosphorylation of Raf by ceramide-activated protein kinase. Nature 378: 307-310 [CrossRef][Medline]

Copyright Clearance Center:   0032-0889/98/116/0785/11
 
© 1998 American Society of Plant Physiologists