Serum- and Glucocorticoid-Regulated Kinase 1Regulates Ubiquitin Ligase Neural Precursor Cell-Expressed, Developmentally Down-RegulatedProtein 4-2 by Inducing Interaction with 14-3-3

Vivek Bhalla, Dorothee Daidie, Hongyan Li, Alan C. Pao, Lila P. LaGrange, Jian Wang,Alain Vandewalle, James D. Stockand, Olivier Staub, and David Pearce

Division of Nephrology, Department of Medicine (V.B., H.L., A.C.P., J.W., D.P.), and Department ofCellular and Molecular Pharmacology (D.P.), University of California, San Francisco, California 94143-0532; Department of Pharmacology and Toxicology, University of Lausanne (D.D., O.S.), CH-1005Lausanne, Switzerland; Department of Physiology, University of Texas Health Science Center (L.P.L.,J.D.S.), San Antonio, Texas 78229-3900; and Institut National de la Sante et de la RechercheMedicale, Unite 478, Faculte de Medecine Xavier Bichat (A.V.), 75870 Paris, France

Serum- and glucocorticoid-regulated kinase 1(SGK1) is an aldosterone-regulated early responsegene product that regulates the activity of severalion transport proteins, most notably that of theepithelial sodium channel (ENaC). Recent evidencehas established that SGK1 phosphorylates and in-hibits Nedd4-2 (neural precursor cell-expressed,developmentally down-regulated protein 4-2), aubiquitin ligase that decreases cell surface expres-sion of the channel and possibly stimulates its deg-radation. The mechanistic basis for this SGK1-in-duced Nedd4-2 inhibition is currently unknown. Inthis study we show that SGK1-mediated phosphor-ylation of Nedd4-2 induces its interaction withmembers of the 14-3-3 family of regulatory pro-

teins. Through functional characterization ofNedd4-2-mutant proteins, we demonstrate thatthis interaction is required for SGK1-mediated in-hibition of Nedd4-2. The concerted action of SGK1and 14-3-3 appears to disrupt Nedd4-2-mediatedubiquitination of ENaC, thus providing a mecha-nism by which SGK1 modulates the ENaC-medi-ated Na� current. Finally, the expression pattern of14-3-3 is also consistent with a functional role indistal nephron Na� transport. These results dem-onstrate a novel, physiologically significant role for14-3-3 proteins in modulating ubiquitin ligase-dependent pathways in the control of epithelialion transport. (Molecular Endocrinology 19:3073–3084, 2005)

THE EPITHELIAL SODIUM channel (ENaC) consti-tutes the rate-limiting step in transepithelial Na�

transport in kidney, colon, lung, and other electricallyhigh resistance epithelia. Disruption of channel func-tion or of its key regulators causes profound diseasestates and in some cases is incompatible with life (1,2). ENaC-mediated sodium transport is under physio-logical control, primarily by the adrenal steroid hor-mones, which act through the mineralocorticoid andglucocorticoid receptors (MR and GR, respectively) tostimulate localization of functional ENaC to the apicalplasma membrane. The ENaC stimulatory effects of

corticosteroid hormones are mediated in part by theserine-threonine kinase, serum- and glucocorticoid-regulated kinase 1 (SGK1), whose gene transcription isstimulated by MR and GR (3, 4). Other hormonal andnonhormonal regulators of ENaC, including insulin andosmotic shock, also appear to act through SGK1 (5, 6)to stimulate Na� current (7–9). One of the cellularsubstrates of SGK1 is Nedd4-2 (neural precursor cell-expressed, developmentally down-regulated protein4-2), a HECT (homologous to E6-AP carboxyl termi-nus) domain-containing ubiquitin ligase that regulatesENaC trafficking and degradation through ubiquitina-tion (10, 11). Several recent reports support the ideathat SGK1 phosphorylates and thereby inhibitsNedd4-2 (7, 12–14), although there are data to suggestthat SGK1 also has effects on ENaC that are indepen-dent of Nedd4 or Nedd4-2 (15–20).

Nedd4-2 contains a series of tryptophan-rich se-quences (WW motifs) that interact with a proline-ty-rosine PY motif within the C-terminal tail of ENaC (2,21–23). Interaction of Nedd4-2 with ENaC is essentialfor its inhibitory effect; notably, the human diseaseLiddle’s syndrome, a form of pseudohyperaldosteron-ism characterized by severe hypertension and hypo-

First Published Online August 11, 2005Abbreviations: ENaC, Epithelial sodium channel; GR, glu-

cocorticoid receptor; GST, glutathione-S-transferase; h, hu-man; HA, hemagglutinin; HEK, human embryonic kidney; MR,mineralocorticoid receptor; Nedd4-2, neural precursor cell-expressed, developmentally down-regulated protein 4-2;PI3-kinase, phosphatidylinositol 3-kinase; PMSF, phenylm-ethylsulfonylfluoride; SGK1, serum- and glucocorticoid-reg-ulated kinase 1; WT, wild type; x, Xenopus.

Molecular Endocrinology is published monthly by TheEndocrine Society (http://www.endo-society.org), theforemost professional society serving the endocrinecommunity.

0888-8809/05/$15.00/0 Molecular Endocrinology 19(12):3073–3084Printed in U.S.A. Copyright © 2005 by The Endocrine Society

doi: 10.1210/me.2005-0193

3073

kalemia, may be caused by mutations in the C-termi-nal tail of ENaC that irreversibly disrupt this interaction(21). In contrast, reversible physiological disruption ofthe Nedd4-2-ENaC interaction is achieved throughSGK1-mediated phosphorylation of Nedd4-2 at keysubstrate residues within the region containing thefour WW domains (7, 12). However, the molecularmechanism by which phosphorylation alters Nedd4-2activity remains unknown.

SGK1 is a member of the phosphatidylinositol 3-kinase (PI3-kinase)-dependent SGK/Akt family (24),which phosphorylates substrates with the general mo-tif RXRXX(S/T) (25, 26). Such phosphorylation can, inprinciple, have a variety of effects on target proteinfunction, including altering conformation or inducinginteraction with other regulatory proteins. Interest-ingly, Akt-mediated phosphorylation of the Forkheadtranscription factor, FKHRL1, induces its interactionwith members of the 14-3-3 family of signaling pro-teins (27). The 14-3-3 interaction motif (28) withinFKHRL1 overlaps an SGK/Akt target sequence,RSX(pS/pT)XP (where pS/pT represents the substratephosphoserine/phosphothreonine). Importantly, in-spection of the functionally critical SGK1 motif withinNedd4-2 revealed that it conforms perfectly to thisconsensus 14-3-3 interaction motif, whereas twoother sites partially match this sequence (14, 29) (Fig.1). Moreover, these SGK1 sites are conserved in ver-tebrate species from Xenopus to human, and SGK1-mediated phosphorylation of Nedd4-2 (7, 12) has beendemonstrated in vivo and in vitro at two of these motifs(7) (Fig. 1). In particular, Ser444 (Xenopus numbering),which is embedded in a perfect 14-3-3 interactionmotif, is critical for SGK1-induced inhibition; Nedd4-2mutants with Ala substitution at this residue inhibitENaC constitutively and do not respond to SGK1.

14-3-3 proteins were initially described as acidicproteins in brain tissue (30). Since then, they havebeen identified in virtually all tissues, and a variety ofphysiological and pathophysiological roles have been

identified in neurodegenerative diseases, carcinogen-esis, and congestive heart failure (31–33). There existseven mammalian isoforms of 14-3-3 (�, �, �, �, �, �,and �), encoded by distinct genes, which can ho-modimerize and heterodimerize to increase bindingaffinity for specific targets (34). Their biology spans avariety of intracellular processes, such as cell cycleregulation, proliferation, metabolism, and ion channelfunction (31, 35, 36). In general, phosphorylation-de-pendent interactions of 14-3-3 proteins with targets ofcellular kinases alter the activity and/or subcellularlocation of these target proteins, thereby propagatingkinase-dependent signaling events.

Together, these observations suggested that SGK1-mediated phosphorylation at Ser444 induces an inhib-itory interaction of Nedd4-2 with 14-3-3 proteinsthereby increasing ENaC-mediated Na� transport. Toexplore this hypothesis, we examined the interactionof 14-3-3 with Nedd4-2 in cultured cells and in vitroand characterized the functional regulation and ubiq-uitination of ENaC by SGK1 together with 14-3-3.

RESULTS

Endogenous Nedd4-2 and 14-3-3 Interact inKidney Tubule Epithelial Cells

To determine whether 14-3-3 proteins physically in-teract with Nedd4-2 in vivo, we performed coimmu-noprecipitation experiments with antibodies directedagainst endogenous Nedd4-2 and 14-3-3 in culturedmpkCCDc14 cells. This mouse collecting duct cell lineforms a monolayer with high electrical resistance whengrown on Transwell filters and expresses several keymediators of Na� transport regulation, including thecorticosteroid receptors (MR and GR), SGK1,Nedd4-2, and ENaC (11, 37–39). In this and othercollecting duct cell lines, corticosteroids induce a ro-bust increase in SGK1 expression, and insulin in-creases SGK1 activity in a PI3-kinase dependent fash-ion. Hence, corticosteroids and insulin togetherstrongly stimulate, whereas PI3-kinase inhibitorsblock, ENaC-mediated Na� currents (38–41) (Fig. 2). Inthis study we compared the interaction of endogenousNedd4-2 and 14-3-3 by coimmunoprecipitation, underbasal conditions, SGK1 activation (dexamethasoneplus insulin), or PI3-kinase inhibition (by LY294002). Asshown in Fig. 2, A and B, the interaction betweenNedd4-2 and 14-3-3 was markedly stimulated bydexamethasone plus insulin, whereas it was attenu-ated by PI3-kinase inhibition with LY294002. Parallelchanges in equivalent current were also observed (Fig.2C). Importantly, total Nedd4-2 and 14-3-3 expressionwere unchanged, consistent with the idea that theeffects are due to changes in interaction per se and notto changes in the level of expression of eithercomponent.

Fig. 1. The Major SGK1 Substrate Sequence in Nedd4-2Overlaps a 14-3-3 Interaction Motif

A, Strip diagram depicting the major functional domains ofNedd4-2, including three SGK1 target motifs (Xenopus num-bering). B, Sequences surrounding Ser338, Thr363, andSer444 of xNedd4-2 are shown. The SGK1 site that includesSer444 is required for SGK1 regulation of Nedd4-2. Aminoacids overlapping this SGK1 site constitute a consensussequence for a high affinity 14-3-3 interaction motif.

3074 Mol Endocrinol, December 2005, 19(12):3073–3084 Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3

SGK1 Modulates Interaction of Nedd4-2with 14-3-3

To examine directly the dependence of the Nedd4-2/14-3-3 interaction on SGK1, we turned to heterolo-

gous expression in human embryonic kidney 293(HEK293) cells. Wild-type (WT) FLAG epitope-taggedNedd4-2 was transfected, with or without a constitu-tively active mutant of SGK1. FLAG-Nedd4-2 was thenimmunoprecipitated, and immunoblots were stainedfor endogenous 14-3-3 (Fig. 3). As shown in Fig. 3, Aand B, SGK1 markedly increased the amount of 14-3-3 coimmunoprecipitated with FLAG-Nedd4-2, evenat low levels of coexpressed SGK1, consistent with anenzymatic effect. This idea is supported by the obser-vations that omitting phosphatase inhibitors from theinteraction buffer markedly attenuated the interactionand that kinase-dead SGK1 failed to stimulate it (Fig.3, C and D, respectively). Interestingly, it appears thatSGK1 itself stably enters the complex with Nedd4-2and 14-3-3 under these conditions, because it wascoimmunoprecipitated with FLAG-Nedd4-2 (Fig. 3A,third panel).

We next examined the role of the functionally im-portant SGK1 substrate sequence within Nedd4-2,and the surrounding amino acids that contribute to the14-3-3 interaction motif (Fig. 1). For these experi-ments, we used two mutants of Nedd4-2, S444A andP446A, which were predicted to block 14-3-3 interac-tion. Nedd4-2/S444A was shown previously to abro-gate SGK1 phosphorylation and disrupt inhibition,hence blocking ENaC disinhibition (7, 12). In contrast,Nedd4-2/P446A was predicted to be incapable of 14-3-3 interaction despite being fully phosphorylated bySGK1 (28, 42, 43). As shown in Fig. 3, D and E, incontrast to WT Nedd4-2, neither mutant was able tospecifically interact with 14-3-3 in an SGK1-depen-dent fashion. Taken together, these results stronglysuggest that the SGK1-dependent phosphorylation ofNedd4-2 at Ser444 induces its interaction with a 14-3-3 family member, and that this interaction is depen-dent on a proline two residues downstream of thesubstrate serine.

SGK1 Mediates Nedd4-2 Phosphorylation andStimulates Its Interaction with 14-3-3� in Vitro

To explore the dependence of the Nedd4-2 interactionwith 14-3-3 on SGK1-induced phosphorylation atSer444, we examined the phosphorylation of Nedd4-2and its interaction with 14-3-3� in vitro. We initiallytested the hypothesis that the P446A, but not theS444A, mutant, would be phosphorylated by SGK1 ata comparable rate to WT Nedd4-2. Nedd4-2 peptidesencompassing the SGK1 and 14-3-3 motifs surround-ing Ser444 were expressed in Escherichia coli as glu-tathione-S-transferase (GST) fusion proteins (Fig. 4A)and incubated with SGK1 and [�-32P]ATP, and SGK1kinase activity was assayed by radioactive phosphateincorporation. As shown in Fig. 4, B and C, the WTpeptide and the P446A mutant were strongly phos-phorylated by SGK1, in contrast to phosphorylation ofthe S444A mutant, which was minimally above back-ground, consistent with previous results (7, 12). Im-portantly, phosphorylation of the P446A mutant pep-

Fig. 2. Endogenous Nedd4-2 and 14-3-3 Interact in KidneyTubule Epithelial Cells

mpkCCDc14 cells were treated with the PI3-kinase inhibitorLY294002, vehicle, or dexamethasone plus insulin for 3 hbefore harvest. A, Immunoblots of cell lysates immunopre-cipitated with anti-Nedd4 antibody and stained with anti-14-3-3 (top panel) and anti-Nedd4-2 (second panel) antibodies.Right panels, Whole cell lysates (WCL) were stained asshown. A representative blot from three independent exper-iments is shown. B, Scanning densitometry of 14-3-3 abun-dance, normalized to the corresponding amount of Nedd4-2,immunoprecipitated from cells treated with LY294002, vehi-cle, or dexamethasone plus insulin in three independent ex-periments. Results are expressed relative to vehicle-treatedcontrols. C, Equivalent current from polarized mpkCCDc14

cells at 3 h after treatment. *, Statistically significant (P �0.05) compared with vehicle treatment.

Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3 Mol Endocrinol, December 2005, 19(12):3073–3084 3075

tide was indistinguishable from that of the WT witheither varying concentrations of SGK1 or varying pe-riods of incubation (Fig. 4, B and C).

We next demonstrated that GST-14-3-3� interactedwith Nedd4-2 in an SGK1-dependent fashion (Fig. 4D).WT or mutant Nedd4-2 was translated in vitro in thepresence of [35S]methionine, incubated with purifiedSGK1 in the presence of (cold) ATP, and then inter-acted with GST-14-3-3�, chosen as a candidate familymember because it is highly expressed in epithelialtissues and cell lines (44). Consistent with the results incultured cells, binding of either Nedd4-2/P446A orS444A to GST-14-3-3� was markedly lower than thatof the WT and was unaffected by SGK1. The depen-dence of the Nedd4-2/14-3-3 interaction on directSGK1 phosphorylation was confirmed by performingthe assay inversely using GST-fused Nedd4-2 pep-

tides incubated with radiolabeled 14-3-3� (data notshown).

Functional Role of Nedd4-2/14-3-3 Interaction inStimulating ENaC-Mediated Na� Transport

To study the functional role of the Nedd4-2/14-3-3interaction, we next examined the ability of SGK1 tostimulate ENaC-mediated Na� currents in the pres-ence of WT Nedd4-2 or either of the two interaction-deficient mutants. Six Xenopus 14-3-3 isoforms havebeen identified, at least three of which are expressedin oocytes (45, 46). Hence, we coexpressed ENaCsubunits (�, �, and �), WT or mutant Nedd4-2, andSGK1 in Xenopus oocytes and used a two-electrodevoltage clamp to determine ENaC-mediated (phena-mil-sensitive) Na� currents. As shown previously, WT

Fig. 3. Nedd4-2 and 14-3-3 Interaction Is Dependent on Phosphorylation of Ser444 by SGK1Coimmunoprecipitation of transfected WT FLAG-tagged Nedd4-2 and SGK1 with endogenous 14-3-3 in HEK293 cells, with

titration of transfected SGK1 (A), titration of transfected Nedd4-2 (B), or removal of phosphatase inhibitors (C). D, Mutation ofNedd4-2 at Pro446 or Ser444 abrogated the SGK1-dependent interaction of Nedd4-2 with 14-3-3. Kinase-dead (K127M) SGK1was unable to stimulate interaction. All immunoblots displayed are representative of between two and seven experiments for eachcondition. E, Scanning densitometry of results from experiments with conditions identical with those depicted in D. Results for14-3-3 abundance are normalized to the amount of Nedd4-2 immunoprecipitated in the corresponding lane. WCL, Whole celllysate. *, Statistically significant (P � 0.05) compared with Nedd4-2 WT without SGK1 (lane 1).

3076 Mol Endocrinol, December 2005, 19(12):3073–3084 Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3

Nedd4-2 markedly inhibited the Na� current, an effectthat was largely reversed by SGK1 (Fig. 5A). The phos-phorylation-deficient S444A Nedd4-2 mutant retainedthe ability to inhibit ENaC-mediated current, similar toWT; however, SGK1 only weakly stimulated Na� cur-rent in the presence of this mutant (7). Similarly, theP446A mutant of Nedd4-2 strongly inhibited ENaC-mediated current, but stimulation of ENaC currents bySGK1 was markedly blunted. The residual stimulationof Na� current by SGK1 in the presence of the P446Aand S444A mutants may occur through secondarySGK1 phosphorylation sites as previously described(7), through endogenous Nedd4-2 (10, 14), or througha Nedd4-2-independent effect.

To directly test the stimulatory effect of 14-3-3 onENaC-mediated Na� current, we also used the Xeno-pus oocyte coexpression assay. We measured phena-mil-sensitive Na� current in oocytes coinjected with

ENaC, Nedd4-2, SGK1, and/or 14-3-3�. ENaC-medi-ated Na� current was modestly, but significantly, in-creased in the presence of 14-3-3 (Fig. 5B, lane 4). Wespeculate that the only modest increase observed with14-3-3 may be due to the abundance of endogenous14-3-3 in oocytes, or perhaps the � isoform, which hasnot been identified in Xenopus laevis, is not optimal forheterologous expression in oocytes. Patch-clamp re-cordings of amiloride-sensitive current density (pA/pF)in transfected Chinese hamster ovary cells were con-sistent with these results (data not shown).

14-3-3 Overexpression Suppresses Nedd4-2-Dependent Ubiquitination of �ENaC

Previous reports have suggested that Nedd4 orNedd4-2 inhibits ENaC through direct ubiquitination ofchannel subunits (10, 47, 48). To examine the mech-

Fig. 4. SGK1 Phosphorylates WT Nedd4-2 and P446A Mutant, But Not the S444A Mutant, in VitroA, GST fusion proteins were expressed in E. coli and incubated with purified SGK1. B, In vitro kinase assay with titrated

amounts of purified SGK1. SGK1 activity is displayed as counts per minute. C, Similar to B, using 25 ng purified SGK1, andaliquots of each reaction were removed at the incubation times indicated. Kinase assay experiments (B and C) were performedin triplicate. D, 14-3-3� interacts with WT, but not mutant, Nedd4-2 in an SGK1-dependent manner in vitro. 35S-Labeled WT or Nedd4-2mutant proteins were incubated with equimolar GST or GST-14-3-3�. Input, 10% of radiolabeled WT protein, was directly loaded. TheCoomassie Blue-stained gel (bottom panel) shows uniform loading. Data are representative of four independent experiments. *,Statistically significant (P � 0.05) lower level of phosphorylation of S444A compared with WT or P446A mutant.

Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3 Mol Endocrinol, December 2005, 19(12):3073–3084 3077

anism of inhibition of ENaC-mediated Na� current bythe Nedd4-2 and 14-3-3 interaction, we assessed theeffect of 14-3-3 overexpression on Nedd4-2-mediatedubiquitination of �ENaC. 293 cells were cotransfectedwith the three ENaC subunits, including hemagglutinin(HA)-tagged �ENaC, Nedd4-2, SGK1, and 14-3-3�.HA-tagged �ENaC was then immunoprecipitated, andimmunoblots were stained for endogenous ubiquitin,and �ENaC. The antiubiquitin antibody detected a se-ries of bands above 100 kDa, the intensities of whichwere markedly increased by Nedd4-2 and attenuatedby 14-3-3 (Fig. 6A, lanes 5 and 6). Scanning densi-tometry of blots from three independent experimentsshowed that 14-3-3 overexpression diminished ubi-quitinated ENaC by approximately 70% (Fig. 6B). In-creased ubiquitination of �ENaC did not appear toalter its total levels, consistent with the idea thatNedd4-2-mediated ubiquitination may modulatechannel trafficking without affecting overall expression(10, 47, 48).

14-3-3 Proteins Are Strongly Expressed inMammalian Cortical Collecting Duct

Finally, to determine whether 14-3-3 expression wasappropriate for a role in mediating the effects of SGK1on the collecting duct, we investigated its expressionin rat kidney tissue using a pan-14-3-3 antibody thatrecognizes all seven mammalian isoforms. Interest-ingly, 14-3-3 immunostaining was not uniform in renalcortex, showing a substantially stronger signal in col-lecting duct principal cells (identified by costainingwith aquaporin 2) than in other nephron segments (Fig.7). Proximal tubule showed modest staining, whereasglomerular 14-3-3 was nearly undetectable in multiplesections from four different animals. 14-3-3 was alsoabundantly expressed in medullary collecting duct,but only at modest levels in medullary thick ascendinglimb of Henle’s loop (not shown). These data demon-strate that 14-3-3 is expressed in the appropriate celltype for SGK1- and Nedd4-2-mediated regulation of

Fig. 5. Interaction of Nedd4-2 and 14-3-3 Augments SGK1-Stimulated Na� TransportA, Mutation of the 14-3-3 interaction motif in Nedd4-2 reduces SGK1-mediated ENaC activation in a Xenopus oocyte

coexpression assay. Stage V–VI oocytes were injected with cRNA for ENaC subunits, Nedd4-2, and SGK1, as indicated.Phenamil-sensitive Na� current was recorded 24 h later by two-electrode voltage clamp at a holding potential of �100 mV. B,14-3-3 stimulates ENaC-mediated Na� current in the presence of SGK1. Currents are expressed compared with ENaC alone. Sixoocytes were measured per condition within each experiment. Results show a representative experiment performed three or fourtimes with similar results, using different batches of oocytes. Statistical significance between the different conditions wascompared as indicated (*, P � 0.01).

3078 Mol Endocrinol, December 2005, 19(12):3073–3084 Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3

ENaC, interestingly at substantially higher levels thanin other regions of the kidney.

DISCUSSION

Role of 14-3-3 in SGK1 Inhibition of Nedd4-2

SGK1 mediates the effects of a variety of hormonesand cellular signals on ion transport proteins, mostnotably the stimulatory effect of aldosterone and in-sulin on ENaC. A large body of evidence supports theidea that SGK1 increases ENaC plasma membranelocalization by increasing its exocytosis, inhibiting en-

docytosis, or both (49), and recent evidence suggeststhat this effect is due in part to its phosphorylation andinhibition of the ubiquitin ligase, Nedd4-2. There arethree SGK1 motifs in Nedd4-2, two of which havebeen clearly demonstrated to be SGK1 substrates(S338 and S444). Interestingly, one of these (S444),which appears to be the principal mediator of SGK1inhibition, conforms to the consensus for a 14-3-3binding motif. Our present data suggest that SGK1-mediated phosphorylation of Nedd4-2 at this motifinduces its inhibitory interaction with a 14-3-3 pro-tein(s), and moreover, that this inhibition is physiolog-ically important; Nedd4-2 interacts with 14-3-3 pro-teins in a PI3-kinase/SGK1-dependent fashion; SGK1stimulation of ENaC currents is markedly blunted inthe presence of Nedd4-2 mutants that cannot interactwith 14-3-3; overexpression of 14-3-3 augmentsENaC-mediated Na� current and diminishes Nedd4-2-mediated ubiquitination of ENaC; and finally, 14-3-3expression in rat kidney is highest within aldosterone-sensitive distal nephron cells.

During the preparation of this manuscript, a publi-cation appeared in which Nedd4-2 was identified as a14-3-3�-interacting partner in a proteomics screen(50) and which made some of the same observationsdescribed here. Our present study confirms 14-3-3interaction and additionally shows that this interactionis physiologically relevant, in that it occurs in a PI3-kinase-dependent fashion in cultured distal nephroncells. Moreover, we demonstrate that 14-3-3 proteinsmodulate Nedd4-2 regulation of Na� current andENaC ubiquitination and show an expression patternin rat kidney consistent with a role in aldosterone-regulated ion transport. Although both 14-3-3� and14-3-3� (50) interact with Nedd4-2 in vitro, which one,if either, is important in vivo remains uncertain. The14-3-3 antibody used in our immunoprecipitation andimmunohistochemistry experiments recognizes allseven mammalian isoforms, and it is not possible toconclude decisively at this time which isoform(s) isphysiologically the most relevant. With this caveat, it isinteresting to note that 14-3-3� is selectively ex-pressed in epithelia and contains a pleckstrin homol-ogy domain, which is known to interact with plasmamembrane phosphoinositides (51).

Mechanism of Nedd4-2 Inhibition by 14-3-3

14-3-3 proteins are highly conserved, are expressed ina variety of eukaryotic cells, and modulate a remark-able variety of cellular processes (52). Most often thetarget motifs have an essential phosphoserine that, inconjunction with other key residues, including a pro-line in the 2� position, mediates interaction. The mod-ulation of target protein function is accomplishedthrough several different classes of activities, includingdisplacement of interacting proteins, exposure of nu-clear export signals, alteration of active site conforma-tion, and bridging of target proteins to induce interac-tion (53). Frequently, target proteins have a high

Fig. 6. 14-3-3 Interferes with ENaC UbiquitinationA, HEK293 cells were transfected as indicated. After 24 h,

cells were lysed, and immunoprecipitations carried out withanti-HA (�ENaC) antibodies. Immunoprecipitates were ana-lyzed by SDS-PAGE, using antiubiquitin (upper panel) or anti-�ENaC (lower panel) antibodies. Ubiquitinated �ENaC is in-dicated with brackets. *, Heavy and light chains of theimmunoprecipitated antibodies. NT, Nontransfected cells. B,Comparison of ubiquitination of �ENaC in cells transfectedwith ENaC, Nedd4-2, and SGK1 with or without 14-3-3.Shown is scanning densitometry from three independent ex-periments showing decreased ubiquitination of ENaC (nor-malized to total ENaC) in the presence 14-3-3. *, P � 0.04.

Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3 Mol Endocrinol, December 2005, 19(12):3073–3084 3079

affinity gatekeeper site that initiates interaction with a14-3-3 dimer, followed by interaction with a loweraffinity motif (54).

With regard to Nedd4-2 regulation, it is notable thatthe critical SGK1 phosphorylation site (Ser444), whichhas the characteristics of a 14-3-3 gatekeeper motif,lies between two of the tryptophan-rich WW domainsimplicated in PY motif interactions. One of these,WW3, is immediately downstream of the gatekeepermotif and has been suggested to be the key site forENaC interaction and inhibition (55, 56). A second WWdomain, WW2, is flanked by the gatekeeper site andtwo of the potential secondary SGK1-regulated motifs(Ser444 and Thr363). In conjunction with previous ob-servations (7, 12), this raises the possibility that 14-3-3binds as a dimer to the gatekeeper site surroundingSer444 and one of the other secondary sites, therebyinterfering with Nedd4-2-ENaC interaction by sterichindrance. It is also possible that the Nedd4-2/14-3-3interaction induces a conformational change that al-ters WW domain conformation or affects ligase activityby altering HECT domain conformation.

It is interesting to speculate that 14-3-3 might act asa bridge to induce interaction with other proteins in-volved in modulating trafficking behavior. In this re-gard, recent evidence suggests that SGK1 induces itsown Nedd4-2-driven, proteasome-mediated, degra-dation (57), and that in a different context, SGK1 hasbeen shown to interact with 14-3-3� (58). We specu-late that SGK1, after recruitment to the Nedd4-2/14-3-3 complex, as demonstrated here, phosphorylatesNedd4-2 and induces its interaction with a 14-3-3homo- and/or heterodimer. This interaction may thenblock the interaction of Nedd4-2 with ENaC (hencereducing channel ubiquitination) and stabilize its inter-action with SGK1 itself (12), possibly through bridging,

hence promoting the ubiquitination of SGK1 and itssubsequent proteasome-mediated degradation.

We cannot rule out that 14-3-3 proteins may stim-ulate ENaC-mediated Na� transport through alterna-tive pathways as well. 14-3-3 proteins have beenreported to interact with PI3-kinase, phosphoinosit-ide-dependent kinase 1, MAPK, and, most recently,Na�,K�-adenosine triphosphatase (33, 59–61), whichall have been implicated in modulation of the ENaC-mediated Na� current.

Role of Ubiquitination in ENaC Regulation

Numerous integral membrane proteins, including ionchannels, are regulated through modulation of traffick-ing, and a growing body of evidence supports the ideathat ubiquitin ligases such as Nedd4-2 act at keypoints in trafficking pathways to alter recycling vs.degradative decisions (62–66). The Nedd4 family of E3ligases appears to act by ubiquitinating their targets aswell as components of the endocytic and sorting ma-chinery, such as endophillin and Hrs (63). Our presentdata strongly support the idea that Nedd4-2 mediatesubiquitination of ENaC, and that 14-3-3 inhibitsNedd4-2 in an SGK1-dependent fashion; however, thepresent experiments do not establish the cellular com-partment(s) in which Nedd4-2-mediated ENaC ubiq-uitination occurs. We favor the idea that a substantialportion is occurring at the plasma membrane or earlyendosome, impacting on endocytosis, sorting, andrecycling to the plasma membrane. In immunoblots,some of the ubiquitinated adducts were detected be-low the size of the major �ENaC band. Additionalstudies will be needed to determine whether the lowermolecular weight bands represent ubiquitinated, en-dopeptidase-modified �ENaC (67), unglycosylated

Fig. 7. 14-3-3 Proteins Are Strongly Expressed in Mammalian Cortical Collecting DuctA, Kidney sections from rats fed a standard chow diet were harvested and stained with antibodies against 14-3-3 and

aquaporin-2 (AQP2). 14-3-3 expression overlaps closely with AQP2 expression in principal cells of the cortical collecting duct(CCD). CCD was also identifiable by tubular epithelial cell morphology and the presence of adjacent glomeruli (G). B, An adjacentsection was stained with antibodies incubated with blocking peptide to demonstrate the specificity of the 14-3-3 expressionpattern. The red signal indicates 14-3-3 expression, and the green signal indicates AQP2 expression. The scale bar is indicatedin micrometers.

3080 Mol Endocrinol, December 2005, 19(12):3073–3084 Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3

forms, or degradation products. Importantly for thepresent context, it is clear that Nedd4-2 increasesENaC ubiquitination, and that this effect is blunted by14-3-3 overexpression.

Physiological Implications of Nedd4-2/14-3-3Interactions

The association of 14-3-3 proteins and the regulationof ENaC and Nedd4-2 impacts on aldosterone-regu-lated fluid and electrolyte homeostasis and may alsoimpact on other cellular functions. In addition to thedistal nephron, ENaC is expressed in a variety of tightepithelia in the lung, salivary ducts, distal colon, eye,and skin. Moreover, two new targets of Nedd4-2-me-diated ubiquitination have been identified, Nav 1.5 andN4WBP5A (66, 68), and 14-3-3 proteins may play arole in their regulation. It appears that 14-3-3 modu-lation of Nedd4-2 and other ubiquitin ligases will be animportant general mechanism in the control of iontransporters, and the role of 14-3-3 proteins in Na�

transport will guide future study of their role in salt-sensitive hypertension and edema-forming states.

MATERIALS AND METHODS

Coimmunoprecipitation in mpkCCDc14 Cells

Murine cortical collecting duct (mpkCCDc14) cells were grownin defined medium, and electrical properties were measuredas described previously (37, 40). Cells were pretreated for 30min with 50 M LY294002 (Calbiochem, La Jolla, CA) orvehicle (dimethylsulfoxide) before addition of 10�7 M dexa-methasone (Sigma-Aldrich Corp., St. Louis, MO) and 100 nM

insulin (Sigma-Aldrich Corp.) or vehicle (ethanol).After 3 h, cells were lysed in 900 l lysis buffer (69) con-

taining protease inhibitors [1 mM phenylmethylsulfonylfluo-ride (PMSF), 1 mM, benzamidine, and 1� Complete ProteaseInhibitor Cocktail (Roche, Indianapolis, IN)] and phosphataseinhibitors [25 mM sodium fluoride, 2 M microcystin-LR, and1� Phosphatase-Inhibitor Cocktail Sets I and II (Calbio-chem)]. Immunoprecipitation was performed overnight withanti-Nedd4 antibody (Upstate Biotechnology, Inc., LakePlacid, NY) and protein A-agarose. Samples were separatedby SDS-PAGE and immunoblotted with anti-Nedd4-2 anti-body (69) or anti-14-3-3� (K-19) antibody (Santa Cruz Bio-technology, Inc., Santa Cruz, CA), which recognizes all sevenmammalian isoforms (70). Densitometry was measured usingImageQuant TL software (Amersham Biosciences, ArlingtonHeights, IL), and values for immunoprecipitated 14-3-3 werenormalized to Nedd4-2 within the same lane.

Coimmunoprecipitation in HEK293 Cells

HEK293 cell cultures were maintained at 37 C in culturemedium containing DMEM, 10% fetal bovine serum, and 1%penicillin/streptomycin. Cells were transfected with FLAG-tagged Xenopus Nedd4-2, mouse SGK1, and/or empty vec-tor using Lipofectamine 2000 (Invitrogen Life Technologies,Inc., Gaithersburg, MD). Twenty-four hours after transfection,cells were harvested in 500 l lysis buffer [50 mM Tris-HCl (pH7.5), 120 mM NaCl, and 1% Nonidet P-40] containing pro-tease inhibitors and phosphatase inhibitors as describedabove. Immunoprecipitation was performed with anti-FLAG

antibody conjugated to resin (Sigma-Aldrich Corp.), and sam-ples were separated by SDS-PAGE and immunoblotted withanti-FLAG, anti-14-3-3, or anti-SGK1 antibodies. When indi-cated, phosphatase inhibitors were omitted from the lysisbuffer. Scanning densitometry was measured as describedabove.

In Vitro Kinase Assay

GST fusion peptides encompassing Ser444 of Xenopus (x)Nedd4-2 were expressed in E. coli and isolated by glutathi-one-Sepharose 4B beads according to the manufacturer’sprotocol (Amersham Biosciences). Uniform expression wasconfirmed by SDS-PAGE and Coomassie Blue staining andwas quantified by Bradford assay. Ten micrograms of pep-tide was used as substrate for a kinase reaction with purified,active SGK1 (Upstate Biotechnology, Inc.) and [�-32P]ATPaccording to the manufacturer’s protocol. The reaction prod-uct was then blotted onto P81 phosphocellulose squares(Whatman, Inc., Clifton, NJ), washed, and counted in a scin-tillation counter.

GST Pull-Down Assay

Full-length cDNA encoding human (h) 14-3-3� (MGC-8516,IMAGE clone ID 2822397) was obtained from American TypeCulture Collection (Manassas, VA) and subcloned into thepGEX-4T1 (Amersham Biosciences) vector using standardtechniques. GST fusion proteins were prepared as describedabove.

WT and mutant xNedd4-2 were synthesized in vitro in thepresence of [35S]methionine and incubated as describedabove with 25 ng SGK1 for 60 min. Phosphorylated WT ormutant xNedd4-2 was incubated with 15 mg GST-h14-3-3�(or equimolar amounts of GST alone) and glutathione-Sepha-rose beads in binding buffer [PBS (pH 7.4), 1% Triton X-100,1 mM phenylmethylsulfonylfluoride, and 1� protease inhibitormixture] at room temperature for 30 min. Samples werewashed in binding buffer and then incubated with 50 l of thekinase reaction described above and 150 l interaction buffer(71) at 4 C overnight. Pelleted beads were then washed threetimes in interaction buffer, boiled in Laemmli’s sample buffer,and separated by SDS-PAGE. The gel was stained with Coo-massie Blue as a loading control and dried, and [35S]-labeledproteins were visualized by autoradiography.

Construction of Expression Vectors and X. laevisOocyte Coexpression Assay

All cDNAs were subcloned into the in vitro expression vectorpSDEasy or pMO. Capped cRNAs were synthesized as de-scribed previously (3). Xenopus �-, �-, and �ENaC subunitcRNAs were coinjected (3 ng each) into stage V–VI oocytestogether with xNedd4-2 or mutant cRNA (0.6–1.5 ng), xSGK1cRNA (5 ng), h14-3-3� cRNA (20 ng), or diethylpyrocarbon-ate-treated water as a control. After injection, the oocyteswere incubated in a low-Na� Barth’s solution, and currentmeasurements were performed as described previously (3).The macroscopic phenamil-sensitive current was defined asthe difference between currents obtained in the presence (2M) and the absence of phenamil.

In Vivo Ubiquitination Assay

HEK293 cells, grown on 75-cm2 dishes, were transfected bythe calcium phosphate technique with 4.5 g of the plasmidsencoding differentially tagged rat ENaC subunits (HA-tagged�ENaC and �ENaC, and FLAG-tagged �ENaC) (47, 72), 2 ghNedd4-2 (68), 1 g mSGK1 S422D, and 13.5 g h14-3-3�.Twenty-four hours after transfection, cells were dissociated

Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3 Mol Endocrinol, December 2005, 19(12):3073–3084 3081

in 1 ml dissociation buffer [5% glycerol, 1 mM EDTA, and 1mM EGTA (pH 8.0)]. Cells were recovered with 10 ml PBS andcentrifuged at 4 C, 1500 � g, for 2 min. Cell pellets werefrozen at �70 C. Pellets were then resuspended in 1 ml lysisbuffer [50 mM HEPES (pH 8.0), 150 mM NaCl, 1 mM EGTA,10% glycerol, 1% Triton X-100, 1 mM dithiothreitol, 100 mM

sodium fluoride, 10 mM disodium pyrophosphate, 1 mM

PMSF, and 1� protease mixture], and lysates were centri-fuged for 15 min at 21,500 � g (4 C). Supernatants wererecovered, and immunoprecipitations were carried out withmonoclonal anti-HA antibodies (Santa Cruz Biotechnology,Inc.) and protein G-Sepharose. Samples were analyzed by5–15% (ubiquitin-blot) or 8% (HA-blot) SDS-PAGE and im-munoblotted with monoclonal antiubiquitin antibodies (FK2,Affinity, Nottingham, UK) or polyclonal anti-rat �ENaC anti-bodies (73). Quantification of the signals on fluorograms wasperformed using a molecular imager FX (Bio-Rad Laborato-ries, Hercules, CA), and the results were normalized to thecontrol values.

Immunofluorescence

Fifty- to 75-gm Sprague-Dawley male rats (Harlan Laborato-ries, Indianapolis, IN) were maintained on a regular chow dietwith normal Na� content. Rats were anesthetized by ip in-jection with pentobarbital (40–60 mg/kg) according to thestandard protocol and perfused with chilled 0.9% NaCl, fol-lowed by chilled PBS (pH 7.4). Kidneys were removed, bi-sected sagittally, and incubated in 4% paraformaldehyde at4 C overnight, then rinsed several times with PBS, embeddedin paraffin, sliced into 4- to 5-m sections, and mounted onglass slides.

Slides were washed in PBS for 5 min to remove theircoverslips and incubated in blocking solution [10% donkeyserum (Chemicon International, Temecula, CA) and 0.1%BSA in PBS] for 1 h at room temperature. Sections were thenincubated with primary antibody [rabbit anti-14-3-3 and goatanti-aquaporin 2 (Santa Cruz Biotechnology, Inc.)] or anti-body incubated with 5� (wt/vol) 14-3-3 blocking peptide(Santa Cruz Biotechnology, Inc.) overnight at 4 C. Sectionswere then rinsed three times with PBS and overlaid with theappropriate secondary antibodies (Texas Red-conjugateddonkey anti-rabbit and fluorescein isothiocyanate-conju-gated donkey anti-goat IgG, Jackson ImmunoResearch Lab-oratories, Inc., West Grove, PA). Slides were rinsed in PBSand mounted with VectaMount (Vector Laboratories, Inc.,Burlingame, CA). Images were obtained at �400 magnifica-tion using an Axioscope epifluorescence microscope (CarlZeiss, Inc., New York, NY) and MetaMorph imaging software(Molecualr Devices, Sunnyvale, CA).

Statistical Analysis

Data are presented as the mean � SEM. Unless otherwisespecified, all statistical comparisons were performed usingStudent’s unpaired t test, and significance was defined asP � 0.05.

Experimental Animal Safety

All animal experimentation described in this manuscript wasconducted in accord with accepted standards of humaneanimal care and was approved by the committee on animalresearch at University of California-San Francisco.

Acknowledgments

We thank Martin McMahon, Anthony Lau, Jeffrey Lau, andFred Cohen for helpful discussions, and Bernard Rossier andLaurent Schild for ENaC plasmids.

Received May 13, 2005. Accepted August 2, 2005.Address all correspondence and requests for reprints to:

Dr. David Pearce, 600 16th Street, N272 Genentech Hall,Mission Bay, Box 2140, University of California, San Fran-cisco, California 94107-214. E-mail: pearce@medicine.ucsf.edu.

This work was supported by grants from the NationalInstitutes of Health, National Institute of Diabetes and Diges-tive and Kidney Diseases (F32-DK-066968 and R01-DK5–6695), and the Swiss National Science Foundation(3100A0-103779).

REFERENCES

1. Bonny O, Rossier BC 2002 Disturbances of Na/Kbalance: pseudohypoaldosteronism revisited. J Am SocNephrol 13:2399–2414

2. Lifton RP, Gharavi AG, Geller DS 2001 Molecular mech-anisms of human hypertension. Cell 104:545–556

3. Chen SY, Bhargava A, Mastroberardino L, Meijer OC,Wang J, Buse P, Firestone GL, Verrey F, Pearce D 1999Epithelial sodium channel regulated by aldosterone-in-duced protein sgk. Proc Natl Acad Sci USA 96:2514–2519

4. Naray-Fejes-Toth A, Canessa C, Cleaveland ES, AldrichG, Fejes-Toth G 1999 Sgk is an aldosterone-inducedkinase in the renal collecting duct. Effects on epithelialNa� channels. J Biol Chem 274:16973–16978

5. Faletti CJ, Perrotti N, Taylor SI, Blazer-Yost BL 2002 Sgk:an essential convergence point for peptide and steroidhormone regulation of ENaC-mediated Na� transport.Am J Physiol 282:C494–C500

6. Rozansky DJ, Wang J, Doan N, Purdy T, Faulk T, Bhar-gava A, Dawson K, Pearce D 2002 Hypotonic inductionof SGK1 and Na� transport in A6 cells. Am J Physiol283:F105–F113

7. Debonneville C, Flores SY, Kamynina E, Plant PJ, TauxeC, Thomas MA, Munster C, Chraibi A, Pratt JH, Horis-berger JD, Pearce D, Loffing J, Staub O 2001 Phosphor-ylation of Nedd4-2 by Sgk1 regulates epithelial Na�

channel cell surface expression. EMBO J 20:7052–70598. Loffing J, Zecevic M, Feraille E, Kaissling B, Asher C,

Rossier BC, Firestone GL, Pearce D, Verrey F 2001 Al-dosterone induces rapid apical translocation of ENaC inearly portion of renal collecting system: possible role ofSGK. Am J Physiol 280:F675–F682

9. Wulff P, Vallon V, Huang DY, Volkl H, Yu F, Richter K,Jansen M, Schlunz M, Klingel K, Loffing J, KauselmannG, Bosl MR, Lang F, Kuhl D 2002 Impaired renal Na�

retention in the sgk1-knockout mouse. J Clin Invest 110:1263–1268

10. Abriel H, Loffing J, Rebhun JF, Pratt JH, Schild L, Horis-berger JD, Rotin D, Staub O 1999 Defective regulation ofthe epithelial Na� channel by Nedd4 in Liddle’s syn-drome. J Clin Invest 103:667–673

11. Kamynina E, Debonneville C, Bens M, Vandewalle A,Staub O 2001 A novel mouse Nedd4 protein suppressesthe activity of the epithelial Na� channel. FASEB J 15:204–214

12. Snyder PM, Olson DR, Thomas BC 2002 Serum andglucocorticoid-regulated kinase modulates Nedd4-2-mediated inhibition of the epithelial Na� channel. J BiolChem 277:5–8

13. Snyder PM, Steines JC, Olson DR 2004 Relative contri-bution of Nedd4 and Nedd4-2 to ENaC regulation inepithelia determined by RNA interference. J Biol Chem279:5042–5046

14. Fouladkou F, Koopaei RA, Vogt B, Flores SY, Malbert-Colas L, Lecomte MC, Loffing J, Frey FJ, Frey BM, StaubO 2004 A naturally occurring human Nedd4-2 variant

3082 Mol Endocrinol, December 2005, 19(12):3073–3084 Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3

displays impaired ENaC regulation due to differentialphosphorylation of Nedd4-2 in Xenopus laevis oocytes.Am J Physiol 287:F550–F561

15. Diakov A, Korbmacher C 2004 A novel pathway of epi-thelial sodium channel activation involves a serum- andglucocorticoid-inducible kinase consensus motif in the Cterminus of the channel’s �-subunit. J Biol Chem 279:38134–38142

16. Alvarez de la Rosa D, Canessa CM 2003 Role of SGK inhormonal regulation of epithelial sodium channel in A6cells. Am J Physiol 284:C404–C414

17. Alvarez de la Rosa D, Li H, Canessa CM 2002 Effects ofaldosterone on biosynthesis, traffic, and functional ex-pression of epithelial sodium channels in A6 cells. J GenPhysiol 119:427–442

18. Alvarez de la Rosa D, Zhang P, Naray-Fejes-Toth A,Fejes-Toth G, Canessa CM 1999 The serum and glu-cocorticoid kinase sgk increases the abundance of epi-thelial sodium channels in the plasma membrane of Xe-nopus oocytes. J Biol Chem 274:37834–37839

19. Auberson M, Hoffmann-Pochon N, Vandewalle A, Kellen-berger S, Schild L 2003 Epithelial Na� channel mutantscausing Liddle’s syndrome retain ability to respond toaldosterone and vasopressin. Am J Physiol 285:F459–F471

20. De La Rosa DA, Paunescu TG, Els WJ, Helman SI, Ca-nessa CM 2004 Mechanisms of regulation of epithelialsodium channel by SGK1 in A6 cells. J Gen Physiol124:395–407

21. Staub O, Dho S, Henry P, Correa J, Ishikawa T, McGladeJ, Rotin D 1996 WW domains of Nedd4 bind to theproline-rich PY motifs in the epithelial Na� channel de-leted in Liddle’s syndrome. EMBO J 15:2371–2380

22. Schild L, Lu Y, Gautschi I, Schneeberger E, Lifton RP,Rossier BC 1996 Identification of a PY motif in the epi-thelial Na channel subunits as a target sequence formutations causing channel activation found in Liddlesyndrome. EMBO J 15:2381–2387

23. Thomas CP, Itani OA 2004 New insights into epithelialsodium channel function in the kidney: site of action,regulation by ubiquitin ligases, serum- and glucocorti-coid-inducible kinase and proteolysis. Curr Opin NephrolHypertens 13:541–548

24. Mora A, Komander D, van Aalten DM, Alessi DR 2004PDK1, the master regulator of AGC kinase signal trans-duction. Semin Cell Dev Biol 15:161–170

25. Kobayashi T, Deak M, Morrice N, Cohen P 1999 Char-acterization of the structure and regulation of two novelisoforms of serum- and glucocorticoid-induced proteinkinase. Biochem J 344:189–197

26. Sakoda H, Gotoh Y, Katagiri H, Kurokawa M, Ono H,Onishi Y, Anai M, Ogihara T, Fujishiro M, Fukushima Y,Abe M, Shojima N, Kikuchi M, Oka Y, Hirai H, Asano T2003 Differing roles of Akt and serum- and glucocorti-coid-regulated kinase in glucose metabolism, DNA syn-thesis, and oncogenic activity. J Biol Chem 278:25802–25807

27. Brunet A, Bonni A, Zigmond MJ, Lin MZ, Juo P, Hu LS,Anderson MJ, Arden KC, Blenis J, Greenberg ME 1999Akt promotes cell survival by phosphorylating and inhib-iting a Forkhead transcription factor. Cell 96:857–868

28. Muslin AJ, Tanner JW, Allen PM, Shaw AS 1996 Interac-tion of 14-3-3 with signaling proteins is mediated by therecognition of phosphoserine. Cell 84:889–897

29. Snyder PM, Olson DR, Kabra R, Zhou R, Steines JC 2004cAMP and SGK regulate ENaC through convergentphosphorylation of Nedd4-2. J Biol Chem 279:45753–45758

30. Moore BW, Perez VJ 1967 Specific acidic proteins of thenervous system. In: Carlson FD, ed. Physiological andbiochemical aspects of nervous integration. EnglewoodCliffs, NJ: Prentice-Hall; 343–359

31. MacKintosh C 2004 Dynamic interactions between 14-3-3� and phosphoproteins regulate diverse cellular pro-cesses. Biochem J 381(Pt 2):329–342

32. Hermeking H 2003 The 14-3-3 cancer connection. NatRev Cancer 3:931–943

33. Xing H, Zhang S, Weinheimer C, Kovacs A, Muslin AJ2000 14-3-3 proteins block apoptosis and differentiallyregulate MAPK cascades. EMBO J 19:349–358

34. Aitken A 2002 Functional specificity in 14-3-3 isoforminteractions through dimer formation and phosphoryla-tion. Chromosome location of mammalian isoforms andvariants. Plant Mol Biol 50:993–1010

35. Aitken A 1996 14-3-3 and its possible role in co-ordinat-ing multiple signalling pathways. Trends Cell Biol6:341–347

36. Bunney TD, van den Wijngaard PW, de Boer AH 200214-3-3 protein regulation of proton pumps and ion chan-nels. Plant Mol Biol 50:1041–1051

37. Bens M, Vallet V, Cluzeaud F, Pascual-Letallec L, KahnA, Rafestin-Oblin ME, Rossier BC, Vandewalle A 1999Corticosteroid-dependent sodium transport in a novelimmortalized mouse collecting duct principal cell line.J Am Soc Nephrol 10:923–934

38. Bhargava A, Wang J, Pearce D 2004 Regulation of epi-thelial ion transport by aldosterone through changes ingene expression. Mol Cell Endocrinol 217:189–196

39. Flores SY, Loffing-Cueni D, Kamynina E, Daidie D, Ger-bex C, Chabanel S, Dudler J, Loffing J, Staub O 2005Aldosterone-induced serum and glucocorticoid-inducedkinase 1 expression is accompanied by Nedd4-2 phos-phorylation and increased Na� transport in cortical col-lecting duct cells. J Am Soc Nephrol 16:2279–2287

40. Wang J, Barbry P, Maiyar AC, Rozansky DJ, Bhargava A,Leong M, Firestone GL, Pearce D 2001 SGK integratesinsulin and mineralocorticoid regulation of epithelial so-dium transport. Am J Physiol 280:F303–F313

41. Shelly C, Herrera R 2002 Activation of SGK1 by HGF,Rac1 and integrin-mediated cell adhesion in MDCK cells:PI-3K-dependent and -independent pathways. J Cell Sci115:1985–1993

42. Yaffe MB, Rittinger K, Volinia S, Caron PR, Aitken A,Leffers H, Gamblin SJ, Smerdon SJ, Cantley LC 1997The structural basis for 14-3-3:phosphopeptide bindingspecificity. Cell 91:961–971

43. Lehoux S, Abe J, Florian JA, Berk BC 2001 14-3-3 Bind-ing to Na�/H� exchanger isoform-1 is associated withserum-dependent activation of Na�/H� exchange. J BiolChem 276:15794–15800

44. Leffers H, Madsen P, Rasmussen HH, Honore B,Andersen AH, Walbum E, Vandekerckhove J, Celis JE1993 Molecular cloning and expression of the transfor-mation sensitive epithelial marker stratifin. A member ofa protein family that has been involved in the proteinkinase C signalling pathway. J Mol Biol 231:982–998

45. Wang B, Yang H, Liu YC, Jelinek T, Zhang L, Ruoslahti E,Fu H 1999 Isolation of high-affinity peptide antagonists of14-3-3 proteins by phage display. Biochemistry 38:12499–12504

46. Wu C, Muslin AJ 2002 Role of 14-3-3 proteins in earlyXenopus development. Mech Dev 119:45–54

47. Staub O, Gautschi I, Ishikawa T, Breitschopf K, Ciech-anover A, Schild L, Rotin D 1997 Regulation of stabilityand function of the epithelial Na� channel (ENaC) byubiquitination. EMBO J 16:6325–6336

48. Malik B, Yue Q, Yue G, Chen X, Price SR, Mitch WE,Eaton DC 2005 Role of Nedd4-2 and polyubiquitinationin epithelial sodium channel degradation in untransfectedrenal A6 cells expressing endogenous ENaC subunits.Am J Physiol 289:F107–F116

49. McCormick JA, Bhalla V, Pao AC, Pearce D 2005 SGK1:a rapid aldosterone-induced regulator of renal sodiumreabsorption. Physiology 20:134–139

Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3 Mol Endocrinol, December 2005, 19(12):3073–3084 3083

50. Ichimura T, Yamamura H, Sasamoto K, Tominaga Y,Taoka M, Kakiuchi K, Shinkawa T, Takahashi N, ShimadaS, Isobe T 2005 14-3-3 proteins modulate the expressionof epithelial Na� channels by phosphorylation-depen-dent interaction with Nedd4-2 ubiquitin ligase. J BiolChem 280:13187–13194

51. Dellambra E, Patrone M, Sparatore B, Negri A, CecilianiF, Bondanza S, Molina F, Cancedda FD, De Luca M 1995Stratifin, a keratinocyte specific 14-3-3 protein, harbors apleckstrin homology (PH) domain and enhances proteinkinase C activity. J Cell Sci 108:3569–3579

52. Rubio MP, Geraghty KM, Wong BH, Wood NT, CampbellDG, Morrice N, Mackintosh C 2004 14-3-3-affinity puri-fication of over 200 human phosphoproteins reveals newlinks to regulation of cellular metabolism, proliferationand trafficking. Biochem J 379:395–408

53. Tzivion G, Avruch J 2002 14-3-3 proteins: active cofac-tors in cellular regulation by serine/threonine phosphor-ylation. J Biol Chem 277:3061–3064

54. Yaffe MB 2002 How do 14-3-3 proteins work? Gate-keeper phosphorylation and the molecular anvil hypoth-esis. FEBS Lett 513:53–57

55. Asher C, Sinha I, Garty H 2003 Characterization of theinteractions between Nedd4-2, ENaC, and sgk-1 usingsurface plasmon resonance. Biochim Biophys Acta1612:59–64

56. Kanelis V, Rotin D, Forman-Kay JD 2001 Solution struc-ture of a Nedd4 WW domain-ENaC peptide complex. NatStruct Biol 8:407–412

57. Zhou R, Snyder PM 2005 Nedd4-2 phosphorylation in-duces serum and glucocorticoid-regulated kinase (SGK)ubiquitination and degradation. J Biol Chem 280:4518–4523

58. Chun J, Kwon T, Lee EJ, Kim CH, Han YS, Hong SK,Hyun S, Kang SS 2004 14-3-3 protein mediates phos-phorylation of microtubule-associated protein byserum- and glucocorticoid-induced protein kinase 1. MolCell 18:360–368

59. Cahill CM, Tzivion G, Nasrin N, Ogg S, Dore J, Ruvkun G,Alexander-Bridges M 2001 Phosphatidylinositol 3-kinasesignaling inhibits DAF-16 DNA binding and function via14-3-3-dependent and 14-3-3-independent pathways.J Biol Chem 276:13402–13410

60. Sato S, Fujita N, Tsuruo T 2002 Regulation of kin-ase activity of 3-phosphoinositide-dependent proteinkinase-1 by binding to 14-3-3. J Biol Chem 277:39360–39367

61. Efendiev R, Chen Z, Krmar RT, Uhles S, Katz AI, Pe-demonte CH, Bertorello AM 2005 The 14-3-3 proteintranslates the Na�r,K�r-ATPase �1-subunit phosphor-ylation signal into binding and activation of phospho-inositide 3-kinase during endocytosis. J Biol Chem280:16272–16277

62. Rotin D, Staub O, Haguenauer-Tsapis R 2000 Ubiquiti-nation and endocytosis of plasma membrane proteins:role of Nedd4/Rsp5p family of ubiquitin-protein ligases. JMembr Biol 176:1–17

63. Hicke L, Dunn R 2003 Regulation of membrane proteintransport by ubiquitin and ubiquitin-binding proteins. Anew ticket for entry into budding vesicles-ubiquitin. AnnuRev Cell Dev Biol 19:141–172

64. Hettema EH, Valdez-Taubas J, Pelham HR 2004 Bsd2binds the ubiquitin ligase Rsp5 and mediates the ubiq-uitination of transmembrane proteins. EMBO J 23:1279–1288

65. Marchese A, Raiborg C, Santini F, Keen JH, Stenmark H,Benovic JL 2003 The E3 ubiquitin ligase AIP4 mediatesubiquitination and sorting of the G protein-coupled re-ceptor CXCR4. Dev Cell 5:709–722

66. Shearwin-Whyatt LM, Brown DL, Wylie FG, Stow JL,Kumar S 2004 N4WBP5A (Ndfip2), a Nedd4-interactingprotein, localizes to multivesicular bodies and the Golgi,and has a potential role in protein trafficking. J Cell Sci117:3679–3689

67. Hughey RP, Bruns JB, Kinlough CL, Harkleroad KL, TongQ, Carattino MD, Johnson JP, Stockand JD, Kleyman TR2004 Epithelial sodium channels are activated by furin-dependent proteolysis. J Biol Chem 279:18111–18114

68. Van Bemmelen MX, Rougier JS, Gavillet B, Apotheloz F,Daidie D, Tateyama M, Rivolta I, Thomas MA, Kass RS,Staub O, Abriel H 2004 Cardiac voltage-gated sodiumchannel Nav1.5 is regulated by Nedd4-2 mediated ubiq-uitination. Circ Res 95:284–291

69. Debonneville C, Staub O 2004 Participation of the ubiq-uitin-conjugating enzyme UBE2E3 in Nedd4-2-depen-dent regulation of the epithelial Na� channel. Mol CellBiol 24:2397–2409

70. Pozuelo Rubio M, Peggie M, Wong BH, Morrice N,MacKintosh C 2003 14-3-3� regulate fructose-2,6-bisphosphate levels by binding to PKB-phosphorylatedcardiac fructose-2,6-bisphosphate kinase/phosphatase.EMBO J 22:3514–3523

71. Onuma H, Osawa H, Yamada K, Ogura T, Tanabe F,Granner DK, Makino H 2002 Identification of the insulin-regulated interaction of phosphodiesterase 3B with 14-3-3 � protein. Diabetes 51:3362–3367

72. Firsov D, Schild L, Gautschi I, Merillat AM, SchneebergerE, Rossier BC 1996 Cell surface expression of the epi-thelial Na channel and a mutant causing Liddlesyndrome: a quantitative approach. Proc Natl Acad SciUSA 93:15370–15375

73. Duc C, Farman N, Canessa CM, Bonvalet JP, Rossier BC1994 Cell-specific expression of epithelial sodium chan-nel �, �, and � subunits in aldosterone-responsive epi-thelia from the rat: localization by in situ hybridization andimmunocytochemistry. J Cell Biol 127:1907–1921

Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremostprofessional society serving the endocrine community.

3084 Mol Endocrinol, December 2005, 19(12):3073–3084 Bhalla et al. • SGK1 Inhibits Nedd4-2 through 14-3-3