Casein kinase 2 specifically binds to and phosphorylates the carboxy termini of ENaC subunits Haikun Shi 1 , Carol Asher 1 , Yuval Yung 2 , Luba Kligman 1 , Eitan Reuveny 1 , Rony Seger 2 and Haim Garty 1 1 Department of Biological Chemistry, and 2 Department of Biological Regulation, The Weizmann Institute of Science, Rehovot, Israel A number of findings have suggested the involvement of protein phosphorylation in the regulation of the epithelial Na + channel (ENaC). A recent study has demonstrated that the C tails of the b and c subunits of ENaC are subject to phosphorylation by at least three protein kinases [Shi, H., Asher, C., Chigaev, A., Yung, Y., Reuveny, E., Seger, R. & Garty, H. (2002) J. Biol. Chem. 277, 13539–13547]. One of them was identified as ERK which phosphorylates bT613 and cT623 and affects the channel interaction with Nedd4. The current study identifies a second protein kinase as casein kinase 2 (CK2), or CK-2-like kinase. It phosphorylates bS631, a well-conserved serine on the b subunit. Such phosphorylation is observed both in vitro using glutathi- one-S-transferase–ENaC fusion proteins and in vivo in ENaC-expressing Xenopus oocytes. The c subunit is weakly phosphorylated by this protein kinase on another residue (cT599), and the C tail of a is not significantly phosphory- lated by this kinase. Thus, CK2 may be involved in the regulation of the epithelial Na + channel. Keywords: casein kinase 2; ENaC; epithelial Na + channel; phosphorylation. Active Na + reabsorption in kidney collecting duct, distal colon, lung, and exocrine glands is mediated by an apical Na + specific channel, termed ENaC (epithelial Na + - channel) [1–3]. The channel is a major target of the action of several hormones such as the mineralocorticoid aldosterone, the anti-diuretic peptide vasopressin, and insulin [1,4]. It is composed of three homologous subunits (a, b,andc)which transverse the membrane twice so that both the C and N termini are intracellular [5–8]. Cell surface expression of ENaC is determined by an interaction between the C tails of b and c and the ubiquitin ligase Nedd4. The WW domains of Nedd4 bind to the proline-rich PY motifs of b and cENaC, leading to channel ubiquitination, internalization and degradation [9,10]. Recently it was demonstrated that the aldosterone-induced kinase sgk (serum and glucocorti- coid dependent kinase) phosphorylates Nedd4-2 on two serines and thereby reduces its interaction with the channel [11,12]. In addition, aldosterone and insulin, as well as intracellular signalling components such as protein kinase C and protein kinase A, were found to increase the in vivo phosphorylation of the C termini of both b and cENaC [13]. We have previously demonstrated phosphorylation of the C termini of ENaC subunits, expressed as glutathione-S- transferase (GST) fusion proteins by crude epithelial cytosolic fractions [14]. Fractionating rat colon cytosol by ion exchange chromatography revealed at least three kinases phosphorylating b and cENaC [15]. One of these was identified as ERK (extracellular regulated kinase) which phosphorylates two conserved threonines in the immediate vicinity of the PY motif bT613 and cT623. Phosphorylation of these residues increases the channel’s affinity towards WW sequences and down-regulates the channel activity [15]. A second peak corresponded to an as yet unidentified protein kinase which phosphorylates bS623 and cT630 [14,15]. The third, major peak of protein kinase activity is reported here. In the current paper we provide evidence that this protein kinase is likely to be casein kinase 2 (CK2) and demonstrate that the residues phosphorylated by it are bS631 and cT599. EXPERIMENTAL PROCEDURES Materials 32 P orthophosphate (10 mCiÆmL )1 ), [c- 32 P] ATP (10 mCiÆmL )1 , 3000 Ci mmol )1 )and[c- 32 P] GTP (10 mCiÆmL )1 , 5000 CiÆmmol )1 )werefromAmersham Pharmacia Biotech; glutathione–agarose beads, dephospho- rylated casein, heparin, 2,3-diphosphoglycerate and rat liver CK2 (a mixture of a 2 b 2 and aa¢b 2 )werefromSigma- Aldrich Fine Chemicals. Human recombinant CK2 (Escherichia coli) was from Calbiochem. An antibody directed against the a and a¢ subunits of CK2 was kindly provided by D. W. Litchfield, University of Western Ontario, Canada. Methods Construction, expression and purification of GST fusion proteins containing the cytoplasmic C-tail domains of ENaC subunits, were carried out as described before [14]. Point mutations were introduced by using the Quik- Change TM site-directed mutagenesis kit (Stratagene) and verified by sequencing. Extraction of rat colon cytosol, its fractionation by ion exchange chromatography and phos- phorylation of GST–ENaC fusion proteins by either Correspondence to H. Garty, Department of Biological Chemistry, The Weizmann Institute of Science, Rehovot 76100 Israel. Fax: +972 8 9344177, Tel.: +972 8 9342706, E-mail: h.garty@weizmann.ac.il Abbreviations: ENaC, epithelial Na + channel; CK2, casein kinase 2; GST, glutathione S-transferase; DPG, 2,3-diphosphoglycerate; HA, hemagglutinin A. (Received 15 March 2002, revised 7 July 2002, accepted 30 July 2002) Eur. J. Biochem. 269, 4551–4558 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03154.x fractionated cytosol or purified enzyme are detailed in Shi et al. [15]. Kinetic parameters were determined by phos- phorylating GST–bENaC with human recombinant CK2 at different substrate concentrations or for different time periods. Stochiometry was determined by excising the phosphorylated band from the gel and determining its radioactivity and protein content. Phosphoamino-acid analysis was performed as described previously [14]. Specific binding of kinases to GST fusion proteins was determined by coprecipitation. A 200-lL aliquot of the desired cytosolic fraction was diluted with 70 lL water and 30 lL10· binding buffer composed of 1.5 M NaCl, 220 m M Hepes pH 7.7, 20 m M MgCl 2 , 0.75% Triton X-100, 200 m M b-glycerolphosphate, 1 m M EDTA, 1 m M sodium orthovanadate and protease inhibitors (1 m M phenylmethanesulfonyl fluoride, 10 lgÆmL )1 aprotinin, 10 lgÆmL )1 leupeptin, 2 lgÆmL )1 pepstatin A). The solution was mixed with 5–20 lg GST-fusion protein immobilized on glutathione-agarose beads, and mixtures were rotated at 4 °C for 2 h. Beads were sedimented by centrifugation and washed six times using the following protocol: (a) two washes in 0.5 M LiCl, 100 m M Tris/HCl pH 8.0; (b) two more washes in 1 M NaCl plus HB1B buffer (20 m M Hepes pH 7.7, 50 m M NaCl, 0.1 m M EDTA, 25 m M MgCl 2 , 0.05% Triton X-100); (c) one wash in buffer A [50 m M b-glycerophosphate pH 7.3, 1.5 m M EGTA, 1.0 m M EDTA, 1.0 m M dithiothreitol, and 0.1 m M de-aerated sodium orthovanadate]; (d) A final wash in a kinase assay buffer composed of 20 m M Hepes pH 7.7, 20 m M MgCl 2 , 20 m M 2-glycerolphosphate, 2 m M dithiothreitol, 0.1 m M sodium orthovanadate. Beads were suspended in 30 lLof the above kinase assay buffer, a second substrate was added, and phosphorylation was initiated by the addition of 2 l M ATP plus 2 lCi [c- 32 P]ATP. The suspension was incubated for 30 min at 30 °C and the reaction was stopped by washing. Beads were suspended in 30 lL Laemmli sample buffer, boiled for 5 min, resolved by 12% acrylamide SDS/ PAGE and visualized by autoradiography. ÔIn-gelÕ assays were performed as described in [16]. In brief, SDS/polyacrylmide gels were cast with 0.5 mgÆmL )1 GST-b,GST-c, or casein in the gel polymerization solution. Cytosolic fractions (50 lL) and purified CK2 (0.05 U) were applied to different wells of the gel and resolved by electrophoresis. The gel was then soaked in 20% 2-propra- nol in 50m M Tris/HCl, pH 8.0, and then in 50 m M Tris/HCl pH 8.0 plus 5 m M 2-mercaptoethanol. Proteins were dena- turated by soaking the gel in 6 M urea and then renaturated in 50 m M Tris/HCl, pH 8.0, 0.04% Tween-20 and 5 m M 2-mercaptoethanol. For in-gel phosphorylation, the gel was first preincubated for 30 min at 30 °Cin20m M Hepes pH 8.0 plus 20 m M MgCl 2 . ATP (100 lCi [c- 32 P]ATP plus 20 l M nonradioactive ATP) and 2 m M dithiothreitol were added and the phosphorylation was allowed to proceed for 2hat30°C. The gel was washed in 5% (w/v) trichloro- acetic acid plus 1% (w/v) sodium pyrophosphate, dried and subjected to autoradiography. Western blotting of CK2 was carried out as follows: 20 lL aliquots of different MonoQ fractions were resolved by SDS/PAGE on a 10% acrylamide gel. Proteins were blotted onto a nitrocellulose membrane, blocked with 5% low fat milk, and probed with polyclonal antibodies against the a and a¢ subunits of CK2 (dilution 1 : 1000). Blots were overlaid with horse radish peroxidase conjugated goat antirabbit antibody (1 : 10 000), and binding was detected by enhanced chemiluminesence. For functional expression in Xenopus oocytes, stage V–VI oocytes were injected with cRNA mixtures containing 2.5 ng each ENaC subunit. Oocytes were incubated at 17 °C in a medium that contained 96 m M NaCl and 10 l M amiloride. Electrophysiological measurements were per- formed 48–72 h after the injection by means of the two- electrode voltage clamp technique. Channel activity was determined as the amiloride-sensitive current amplitudes monitored at )100 mV. To study phosphorylation of bENaC, oocytes were injected with a cRNA mixture containing a, c, and a hemagglutinin A (HA)-tagged bENaC. The HA epitope was introduced in the ecto domain at a position shown before not to affect channel activity [17], and the construct was kindly provided by B. Schwappach (Zentrum fu ¨ r Molekulare Biologie, Uni- versita ¨ t Heidelberg, Germany). Injected oocytes were divi- ded into two groups each containing  40 oocytes and incubated with either 100 lCi [ 35 S]methionine (2 days) or 3.3 mCi 32 Pi (4 h). 32 P- and 35 S-labelled oocytes were washed and homogenized in buffer A containing protease inhibitors and membranes were isolated by centrifugation through a sucrose cushion. Membranes were solubilized in 1% Triton X-100 in buffer A and centrifuged for 5 min at 11 000 g to remove insoluble material. Aliquots of  400 lL detergent soluble membrane protein extracts were incubated for 12–16 h at 4 °Cwith2lg of a mouse mAb anti-HA antibody (clone 12CA5, Roche Molecular Bio- chemicals) and then for another 2 h with Protein A Sepharose beads. The beads were sedimented, washed twice in buffer A + 0.1% Triton-X-100, and a third time in buffer A+0.5 M LiCl. Immunopellets were suspended in SDS sample buffer, resolved by SDS/PAGE (8% acrylamide gel) and assayed for 35 Sand 32 P radioactivity by phosphor- imaging. RESULTS Previous studies have demonstrated phosphorylation of the C termini of ENaC subunits expressed as GST fusion proteins by various cytosolic fractions extracted from rat distal colon [14,15]. In particular, three peaks of kinase activity which phosphorylate the C termini of b and cENaC were identified by ion exchange chromatography. Two of them have been studied before [14,15]. The third, a major peak eluted at  0.25–0.27 M NaCl, has not been charac- terized yet. Fig. 1A depicts phosphorylation of GST-b and -c by fractions 66–84 eluted from the monoQ column. In both cases a peak was observed around fraction 76 ( 0.26 M NaCl, Fig. 1B). This kinase phosphorylated GST-b better than GST-c, while GST and GST-a did not incorporate a significant amount of 32 P (Fig. 1C). (Usually, the GST–ENaC fusion proteins were partly degraded, resulting in multiple bands by Coomassie blue staining, not all of which are phosphorylated. This however, does not pose a problem as phosphorylation could be quantified by phosphorimaging of the band corresponding to the full- length protein, and normalizing to the Coomassie blue staining of the same band.) It was next demonstrated that the kinase eluted in fraction 76 tightly binds to both b and c C tails. In this assay, GST, GST-b or GST-c immobilized on glutathione 4552 H. Shi et al. (Eur. J. Biochem. 269) Ó FEBS 2002 beads was first incubated with the above cytosolic fraction in the absence of ATP. The beads containing the fusion proteins (and other proteins associated with them) were precipitated, washed stringently, and incubated with [ 32 P]ATP with no added kinases. Phosphorylation of the fusion protein under these conditions is possible only if the phosphorylating kinase is tightly bound to its substrate. As shown in Fig. 2 both b and c could be effectively phosphorylated under these conditions (second and sixth lanes). Moreover, it was found that the kinase precipitated by one subunit could also phosphorylate the other. In this case, the kinase was precipitated by one of the two fusion proteins (1st) and the other subunit (2nd) was added only to the final phosphorylation mixture. Both fusion proteins could be phosphorylated irrespective of the order of addition, indicating that the kinase bound to GST-b can phosphorylate GST-c and vice versa. This protocol has also been used to demonstrate that binding requires the ENaC domain of the fusion protein and much less kinase activity is precipitated by GST alone. Next, b and c residues phosphorylated by this kinase were identified. Phospho amino acid analysis demonstrated that in b, all of the radioactivity is incorporated into serines, while in c the phosphorylated residues are threonines (Fig. 3A). No tyrosine phosphorylation could be detected for any of the subunits. Identity of the phosphorylated residues was further determined by assessing effects of various point mutations on 32 P incorporation into GST-b and -c. Accordingly, several serines and threonines on the C termini of b and c weremutatedtoalanineandexamined for phosphorylation by fraction 76. In b the major phosphorylated residue was S631 and mutating it to alanine blocked almost all 32 P incorporation (Fig. 3B). bS631 is located past the PY motif and is well conserved among known b sequences (Fig. 3D). Partial inhibition of b Fig. 1. Phosphorylation of different fusion proteins by kinase enriched fractions. (A) Matched comparison of the phosphorylation of GST-b and GST-c by monoQ fractions 64–84. (B) Quantification of the phosphorylation of b and c by fractions 66–86 (a different experiment from the one shown in A). (C) GST fusion proteins containing the C termini of a, b,andcENaC were phosphorylated by fraction 76. Autoradiogram and Coomassie blue staining are shown. Fig. 2. Binding of kinases to GST-b and -c. GST, GST-b or GST-c (1st) were first incubated with or without cytosolic fraction 76, and then precipitated. The pellets were washed stringently as described in Experimental procedures and phosphorylation was induced by the addition of [c- 32 P]ATP in the presence or absence of a second substrate (2nd). Autoradiogram and Coomassie blue staining are shown. Ó FEBS 2002 CK2-mediated phosphorylation of ENaC (Eur. J. Biochem. 269) 4553 phosphorylation was seen also upon the mutation of a neighbouring residue, S633. This residue may have an indirect effect on the incorporation of 32 P into S631. The analogous residue on the c subunit (cT644) did not incorporate 32 P and this subunit appeared to be phospho- rylated mainly on T599, a nonconserved threonine that precedes the PY motif (Fig. 3C and D). Mutating other conserved serine/threonines such as bS620 or cT623 and cT630hadnoeffecton 32 P incorporation by fraction 76 (data not shown). bS631 and S633 have recently been shown to be involved in the regulation of ENaC in mandibular duct cells [18]. As above, the homologous c residue was not involved in any such mechanism. To determine whether bS631 is phosphorylated also in vivo the a-, c-, and HA-tagged bENaC were expressed in Xenopus oocytes and metabolically labelled with either [ 35 S]methionine or 32 Pi. The b subunit could be specifically immunoprecipitated from oocytes by an anti-HA antibody (Fig. 4A). It was endogenously phosphorylated, and the amount of phosphorylation was considerably lower if S631 was mutated into alanine (Fig. 4B,C). Incorporation of 32 P into bS631A was 0.55 ± 0.07 of the value obtained for wild type b (mean ± SEM of three independent experiments). Thus, S631 accounts for  50% of the endogenous phos- phorylation of bENaC. Possible effects of this phosphory- lation on channel activity were examined by recording the amiloride-sensitive Na + current amplitudes in Xenopus oocytes expressing wild-type and mutated ENaC. bS631 was mutated into either glutamic acid or alanine. It was found that neither mutation evokes a significant effect on the macroscopic amiloride blockable current measured in this system (Fig. 5). Both bS631 and cT599 are located in a cluster of acidic residues and in particular have an acidic group at position n + 3. This suggests that the phosphorylating protein kinase might be CK2 [19]. The next set of experiments provided further evidence that the cytosolic kinase phosphorylating bS631 and cT599 is indeed likely to be CK2. First, Western blot analysis of the active cytosolic fractions was performed using polyclonal antibodies against the a and a¢ subunits of CK2. The catalytic subunit of the enzyme could indeed be detected in fractions 70–80 and the predominant isoform was a (Fig. 6). (Protein samples from column fractions have high concentration of NaCl resulting in somewhat diffused bands.) Next we showed that purified CK2 can effectively phosphorylate GST-b and -c and that this phosphorylation occurs primarily on bS631 and cT599 (Fig. 7A). Phospho- rylation of bS631 was further investigated using human recombinant CK2. The phosphorylation reaction was linear for at least 40 min The fraction of protein phosphorylated during a 3-h incubation corresponded to  10% of the total protein, i.e. incorporation of 32 PintobS631 is not residual. The K m for phosphorylation of bS631 was estimated to be Fig. 3. Identification of the phosphorylated residues. (A) Amino acid analysis of phosphorylated GST-b and -c. The circles indicate positions of marker phospho serine, threonine and tyrosine. (B and C) Phosphorylation of wild-type and mutated b and c subunits by fraction 76. Auto- radiograms (top) and Coomassie blue stained gels (bottom) are shown. (D) Sequence alignments of the C termini of b and cENaC from different species. 4554 H. Shi et al. (Eur. J. Biochem. 269) Ó FEBS 2002 1.4 l M (Fig. 7B). This value compares well to the K m measured for b-casein (36.5 l M ) or specific substrate peptides [20]. A unique feature of CK2 is its ability to use both ATP and GTP as phosphate donors [21]. Other criteria for CK2- mediated phosphorylation are inhibition by heparin and 2,3-diphosphoglycerate as well as the ability to phosphory- late casein. Experiments summarized in Fig. 8 indicate that the ENaC phosphorylating kinase eluted in fraction 76 does indeed have CK2 characteristics. The two fusion proteins could be phosphorylated by GTP, and the GTP-dependent phosphorylation took place at bS631 and cT599 (Fig. 8A). Both heparin and DPG inhibited phosphorylation by fraction 76 of either GST-b or GST-c. Finally, we used an Ôin-gelÕ assay to determine the electrophoretic mobility of the kinase eluted in fraction 76, and compare it to that of CK2. In this assay, GST, GST-b, GST-c and casein were copolymerized with polyacrylamide to form substrate-containing SDS gels. Fraction 76 and purified CK2 were resolved electrophoretically, gels were re-naturated, and phosphorylation reactions were carried out within the gels. The radioactive bands detected in such an assay correspond to the electrophoretic mobility of the kinase acting on the in-gel substrate. The kinase eluted in fraction 76 appeared as a 40–42 kDa doublet which phosphorylated b or cENaC but not GST (Fig. 9A). The same bands were visualized using b, c or casein, indicating that the same protein kinase phosphorylates all three substrates (Fig. 9A and B). These bands were also visualized using purified CK2 instead of fraction 76 (Fig. 9C). Taken together, the data summarized in Figs 6–9 strongly suggest that CK2 is the cytosolic kinase phosphorylating bS631 and cT599 and that this phosphorylation is of high affinity. DISCUSSION The current study identifies CK2 as one of the kinases that specifically phosphorylates bENaC. It was motivated by accumulating data suggesting the involvement of phosphate transfer reactions in the cellular regulation of this channel, Fig. 4. Phosphorylation of bENaC in Xenopus oocytes. Oocytes were injected with cRNA mixtures coding for a-, c- and HA-tagged b.Theywere metabolically labelled with either [ 35 S]methionine or 32 Pi. bENaC was immunoprecipitated with an anti-HA antibody and resolved by SDS/PAGE. (A) Autoradiogram of 35 S-labelled proteins immunoprecipitated from ENaC injected (b) and noninjected (–) oocytes. (B) HA-tagged b and b S631A immunoprecipitated from 35 S- and 32 P-labelled oocytes. (C) Quantification of the incorporation of 32 Pintowild-typeandmutatedb. 32 P/ 35 Sratios were calculated for three independent experiments each averaging  40 oocytes. Means ± SEM are shown. Fig. 5. Functional expression of ENaC in Xenopus oocytes. Oocytes were injected with cRNA mixtures corresponding to a, c, and either b or b S631A or b S631E . Amiloride blockable current amplitudes at )100 mV were measured 3 days later, as described in Experimental procedures. Data were normalized to the average current in oocytes injected with the wild-type constructs (11.38 ± 0.94 lA). Means ± SEM of 16–20 oocytes from two different frogs are shown. Fig. 6. Detection of CK2 in ENaC phosphorylating cytosolic fractions. Western blot hybridization of cytosolic fractions with anti-a and -a¢ CK2 antibody was performed as described in Experimental proce- dures. The last lane contains 20 lg of rat brain CK2 (aa’b 2 ). Ó FEBS 2002 CK2-mediated phosphorylation of ENaC (Eur. J. Biochem. 269) 4555 and the identification of cytosolic fractions that incorporate 32 P into the C termini of b and cENaC [13,15,22–24]. Detailed fractionation of rat colon cytosol by ion exchange chromatography has identified several kinase enriched fractions acting on b and c [15]. Two of them have been studied in recent publications. The first is an as yet unidentified kinase which incorporates 32 PintobS620 and cT630 [14]. The second is ERK which phosphorylates bT613 and cT623 and facilitates interactions between the channel and Nedd4 [15]. However, the strongest phospho- rylation activity was evoked by a kinase eluted with  0.25– 0.27 M NaCl, which had not been characterized so far. The current study demonstrates that this protein kinase is likely to be CK2 and that it phosphorylates bS631 and cT599. This identification is based on the ability of GTP to act as a phosphate donor (Fig. 8A), the inhibitory effects of heparin and DPG (Fig. 8B), the fact that purified CK2 can phosphorylate bS631 and cT599 (Fig. 7A), the presence of the a subunit of CK2 in the active cytosolic fractions (Fig. 6), and the finding that the electrophoretic mobility of the phosphorylating kinase corresponds to that of the a subunit of CK2 (Fig. 9). However, other experiments aiming to detect the b subunit of CK2 in the active cytosolic fraction were inconclusive (data not shown). Thus the possibility remains that the protein kinase characterized is another, CK2-like enzyme. CK2 is a ubiquitously expressed serine/threonine kinase, composed of two catalytic (a and/or a¢) and two regulatory (b) subunits [25–28]. It has a large number of substrates which include components of signalling pathways, cytoskel- etal elements, transcription factors and others [25]. Relat- ively little is known about its function, and evidence has been provided that the enzyme plays a role in cell survival, division and proliferation, as well as synaptic development and transmission [25,26]. It is generally believed that the enzyme is constitutively active [26–28]. However, a number of studies have reported activation of CK2 by extracellular signals such as serum, steroid hormones and growth factors [25]. Fig. 7. Phosphorylation of ENaC by CK2. (A) Wild-type and mutated fusion proteins were phosphorylated as described in Experimental procedures using 0.01 U of purified rat liver CK2. (B) Different amounts of GST-b were phosphorylated for 40 min by human recombinant CK2. The amount of phosphate incorporated was determined by phosphorimaging (V) and the protein content of the radioactive band (S) was estimated. A Lineweaver–Burk presentation of the data could be fitted to straight line (R >0.98)withaK m of 1.4 l M . Fig. 8. GTP-mediated phosphorylation of ENaC. (A) Phosphorylation of wild-type and mutated GST constructs was carried out using equal amounts [c- 32 P]ATP or [c- 32 P]GTP. (B) Phosphorylation of GST-b and -c was performed with and without 10 l M heparin or 10 m M DPG. Autoradiograms and Coomassie blue stained gels are shown. Fig. 9. ‘In-gel’ assay identifying the kinase in fraction 76. Various substrates (GST, GST-b,GST-c and casein) were copolymerized with the gel, and fraction 76 (A, B) or purified CK2 (C) were resolved electrophoretically. ÔIn-gelÕ phosphorylation was performed as des- cribed in Experimental procedures. In all cases the same 40–42 kDa doublet was identified. 4556 H. Shi et al. (Eur. J. Biochem. 269) Ó FEBS 2002 Phosphorylation of bENaC by CK2 occurs on residue S631. This residue and the acidic amino acids surrounding it are well conserved in evolution (Fig. 3D). Because mutating S631 fully blocked 32 P incorporation into GST-b we assume that this is the only residue phosphorylated. However S633 too is a conserved serine with a consensus CK2 site, mutation of which inhibits phosphorylation. Phosphorylation of such neighbouring serines may also be interrelated [29]. Taken together with the fact that the phosphorylation does take place also in whole cells, is characterized by a K m of 1.4 l M , and that the kinase tightly binds to its substrate, it is likely that the above phosphorylation plays a role in the regulation of Na + transport. A role of the region that includes bS631 in determining the Na + conductance of ENaC was inferred from a number of studies. First it was shown that truncating the last eight amino acids of b elevates channel activity by  50% and a similar activation can be seen by mutating the four acidic residues in this region [30]. A peptide corresponding to the last 10 amino acids of b inhibited the channel in planer bilayer [31] and in mandibular duct cells [18]. In both cases, the analogous c peptide had no or a much smaller effect. The b motif involved in this interaction has been studied by Dinudom et al. [18] and found to involve S631, D632 and S633. The analogous c residues did not participate in such interac- tion, and in this case another nonconserved serine mediated channel inhibition. It was suggested that the above serines participate in protein–protein interactions which may involve their phosphorylation. To further assess the above possibility we have deter- mined the activity of ENaC in Xenopus oocytes expressing wild-type and mutated b subunits. Substituting bS631 by either alanine or glutamic acid had no effect on the mac- roscopic Na + current (Fig. 5). This is in spite of the fact that bS631 was endogenously phosphorylated in the oocytes. It is however, possible that CK2 plays a role in one of several ENaC regulatory processes that can be measured in mammalian cells but not in the oocyte expression system. In this respect it is interesting to note that CK2 has been reported to be activated by both insulin and dexamethasone [25]. The two hormones have well-established effects on ENaC which may be mediated by protein phosphorylation and also cannot be mimicked in Xenopus oocytes [1,24,32]. An in vitro aldosterone- and insulin-dependent phosphorylation of several residues in the C tail of b expressedinMDCKcellshasbeen reported previously [13]. These residues have not been fully identified but one of them is a serine located between amino acids 619 and 638. The only serines in this range are S620, S631 and S633. Also interesting is the fact that several studies have documented a GTP-dependent regulation of ENaC [1]. The underlying mechanism has not yet been determined and in particular no apical, G-protein coupled receptor is known to play a role in the regulation of Na + transport. The current observations raise the possibility that GTP acts on the channel directly by promoting a CK2-mediated phospho- rylation, rather than by activating a G-protein. In summary, this study demonstrates tight binding of aCK2 to the C tail of ENaC and phosphorylation of a conserved serine in the b subunit. The cellular role of this event awaits further studies. ACKNOWLEDGEMENTS We thank D. W. Litchfield of the University of Western Ontario, Canada for the anti-aCK2 antibody. This study was supported by research grants from the Israel Science Foundation and the US-Israel Binational Science foundation to H. G and E. R. REFERENCES 1. Garty, H. & Palmer, L.G. (1997) Epithelial Na + channels: func- tion, structure, and regulation. Physiol. Rev. 77, 359–396. 2. Horisberger, J.D. (1998) Amiloride-sensitive Na channels, Curr. Opin. Cell Biol. 10, 443–449. 3. Alvarez de la Rosa, D., Canessa, C.M., Fyfe, G.K. & Zhang, P. (2000) Structure and regulation of amiloride-sensitive sodium channels. Annu. Rev. Physiol. 62, 573–594. 4. Palmer, L.G. & Garty, H. (2000) Epithelial Na + channels. In The Kidney, Physiology and Pathophysiology (Seldin, D.W. & Giebisch, G., eds), pp. 251–276, Lippincott Williams & Wilkins, Philadelphia. 5. Canessa, C.M., Schild, L., Buell, G., Thorens, B., Gautschi, I., Horisberger, J D. & Rossier, B.C. (1994) Amiloride-sensitive epithelial Na + channel is made of three homologous subunits. Nature 367, 463–467. 6. Renard, S., Lingueglia, E., Voilley, N., Lazdunski, M. & Barbry, P. (1994) Biochemical analysis of the membrane topology of the amiloride-sensitive Na+ channel. J. Biol. Chem. 269, 12981– 12986. 7. Snyder, P.M., Mcdonald, F.J., Stokes, J.B. & Welsh, M.J. (1994) Membrane topology of the amiloride-sensitive epithelial sodium channel. J. Biol. Chem. 269, 24379–24383. 8. Canessa, C.M., Me ´ rillat, A M. & Rossier, B.C. (1994) Membrane topology of the epithelial sodium channel. Am. J. Physiol. Cell Physiol. 267, C1682–C1690. 9. Staub, O., Gautschi, I., Ishikawa, T., Breitschopf, K., Ciechanover, A., Schild, L. & Rotin, D. (1997) Regulation of stability and function of the epithelial Na + channel (ENaC) by ubiquitination. EMBO J. 16, 6325–6336. 10. Staub, O., Dho, S., Henry, P.C., Correa, J., Ishikawa, T., Mcglade, J. & Rotin, D. (1996) WW domains of Nedd4 bind to the proline-rich PY motifs in the epithelial Na + channel deleted in Liddle’s syndrome. EMBO J. 15, 2371–2380. 11. Debonneville, C., Flores, S.Y., Kamynina, E., Plant, P.J., Tauxe, C.,Thomas,M.A.,Munster,C.,Chraibi,A.,Pratt,J.H., Horisberger, J.D., Pearce, D., Loffing, J. & Staub, O. (2001) Phosphorylation of Nedd4-2 by Sgk1 regulates epithelial Na+ channel cell surface expression. EMBO J. 20, 7052–7059. 12. Snyder,P.M.,Olson,D.R.&Thomas,B.C.(2002)Serumand glucocorticoid-regulated kinase modulates Nedd4-2-mediated inhibition of the epithelial Na + Channel. J. Biol. Chem. 277, 5–8. 13. Shimkets, R.A., Lifton, R. & Canessa, C.M. (1998) In vivo phosphorylation of the epithelial sodium channel. Proc. Natl Acad.Sci.USA95, 3301–3305. 14. Chigaev, A., Lu, G., Shi, H K., Asher, C., Xu, R., Latter, H., Seger, R., Garty, H. & Reuveny, E. (2001) In vitro phosphoryla- tion of the carboxy termini of the epithelial Na + channel (ENaC) and its effects on the channel activity in Xenopus oocytes. Am. J. Physiol. Renal Fluid Electrolyte Physiol. 280, F1030–F1036. 15. Shi, H., Asher, C., Chigaev, A., Yung, Y., Reuveny, E., Seger, R. & Garty, H. (2002) Interactions of beta and gamma ENaC with Nedd4 can be facilitated by an ERK-mediated phosphorylation. J. Biol. Chem. 277, 13539–13547. 16. Rubinfeld, H. & Seger, R. (1998) Detection of MAP kinase sig- naling. In Current Protocols in Cell Biology. John Wiley & Sons, Inc., New York. 17. Firsov, D., Schild, L., Gautschi, I., Merillat, A.M., Schneeberger, E. & Rossier, B.C. (1996) Cell surface expression of the epithelial Ó FEBS 2002 CK2-mediated phosphorylation of ENaC (Eur. J. Biochem. 269) 4557 Na channel and a mutant causing Liddle syndrome: a quantitative approach. Proc. Natl Acad. Sci. USA 93, 15370–15375. 18. Dinudom, A., Harvey, K.F., Komwatana, P., Jolliffe, C.N., Young, J.A., Kumar, S. & Cook, D.I. (2001) The roles of the carboxyl termini of a, b and c subunits of ENaC in regulating epithelial Na + channels and in mediating their inhibition by cytosolic Na + . J. Biol. Chem. 276, 13744–14749. 19. Meggio, F., Marin, O. & Pinna, L.A. (1994) Substrate specificity of protein kinase CK2. Cell. Mol. Biol. Res. 40, 401–409. 20. Marin, O., Meggio, F. & Pinna, L.A. (1994) Design and synthesis of two new peptide substrates for the specific and sensitive monitoring of casein kinases-1 and -2. Biochem. Biophys. Res. Commun. 198, 898–905. 21. Hathaway, G.M. & Traugh, J.A. (1982) Casein kinases – multi- potential protein kinases. Curr. Top. Cell Regul. 21, 101–127. 22. Shigaev, A., Asher, C., Latter, H., Garty, H. & Reuveny, E. (2000) Regulation of sgk by aldosterone and its effects on the epithelial Na + channel. Am. J. Physiol Renal Fluid Electrolyte Physiol. 278, F613–F619. 23. Blazer-Yost, B.L., Punescu, T.G., Helman, S.I., Lee, K.D. & Vlahos, C.J. (1999) Phosphoinositide 3-kinase is required for aldosterone-regulated sodium reabsorption. Am. J. Physiol Cell Physiol. 277, C531–C536. 24. Record, R.D., Froelich, L.L., Vlahos, C.J. & Blazer-Yost, B.L. (1998) Phosphatidylinositol 3-kinase activation is required for insulin- stimulated sodium transport in A6 cells. Am.J.Physiol. 274, E611–E617. 25. Blanquet, P.R. (2000) Casein kinase 2 as a potentially important enzyme in the nervous system. Prog. Neurobiol. 60, 211–246. 26. Pinna, L.A. & Meggio, F. (1997) Protein kinase CK2 (Ôcasein kinase-2Õ) and its implication in cell devision and proliferation. Prog.CellCycleRes.3, 77–97. 27. Guerra,B.,Boldyreff,B.,Sarno,S.,Cesaro,L.,Issinger,O.G.& Pinna, L.A. (1999) CK2: a protein kinase in need of control. Pharmacol. Ther. 82, 303–313. 28. Pinna, L.A. (1997) Protein kinase CK2. Int. J. Biochem. Cell. Biol. 29, 551–554. 29. Perich, J.W., Meggio, F., Reynolds, E.C., Marin, O. & Pinna, L.A. (1992) Role of phosphorylated aminoacyl residues in gen- erating atypical consensus sequences which are recognized by casein kinase-2 but not by casein kinase-1. Biochemistry. 31, 5893– 5897. 30. Snyder, P.M., Price, M.P., Mcdonald, F.J., Adams, C.M., Volk, K.A.,Zeiher,B.G.,Stokes,J.B.&Welsh,M.J.(1995)Mechanism by which Liddle’s syndrome mutations increase activity of a human epithelial Na + channel. Cell. 83, 969–978. 31. Ismailov, I.I., Berdiev, B.K., Fuller, C.M., Bradford, A.L., Lifton, R.P., Warnock, D.G., Bubien, J.K. & Benos, D.J. (1996) Peptide block of constitutively activated Na + channels in Liddle’s disease. Am.JPhysiolCellPhysiol.270, C214–C223. 32. Garty, H. (2000) Regulation of the epithelial Na + channel by aldosterone: open questions and emerging answers. Kidney Int. 57, 1270–1276. 4558 H. Shi et al. (Eur. J. Biochem. 269) Ó FEBS 2002 . Casein kinase 2 specifically binds to and phosphorylates the carboxy termini of ENaC subunits Haikun Shi 1 , Carol Asher 1 , Yuval Yung 2 , Luba. (20 01) The roles of the carboxyl termini of a, b and c subunits of ENaC in regulating epithelial Na + channels and in mediating their inhibition by cytosolic