Nedd4–2 isoforms differentially associate with ENaC and regulate its activity
http://www.100md.com
《美国生理学杂志》
Department of Internal Medicine, Graduate Program in Molecular Biology, University of Iowa College of Medicine and Veterans Affairs Medical Center, Iowa City, Iowa
ABSTRACT
Mutations that disrupt a PY motif in epithelial Na+ channel (ENaC) subunits increase surface expression of Na+ channels in the collecting duct, resulting in greater Na+ reabsorption. Nedd4 and Nedd4–2 have been identified as ubiquitin ligases that can interact with ENaC via its PY motifs to regulate channel activity. We recently reported that human Nedd4–2 (hNedd4–2) is expressed as many isoforms because of alternative promoter usage and/or variable splicing. To understand the relevance of hNedd4–2 isoforms for collecting duct Na+ transport, we studied the interaction with ENaC and the intracellular localization and function of the following three naturally occurring hNedd4–2 isoforms: full-length Nedd4–2 (Nedd4–2), Nedd4–2 lacking the NH2-terminal C2 domain (Nedd4–2C2), and Nedd4–2 lacking the C2 domain and WW domains 2 and 3 (Nedd4–2WW2,3). Nedd4–2 and Nedd4–2C2 associate with ENaC and robustly reduce Na+ transport in Xenopus oocytes, whereas the interaction with and functional effect of Nedd4–2WW2,3 on ENaC is weak. Nedd4–2 is expressed in the mouse collecting duct, and overexpression of Nedd4–2 reduces endogenous ENaC activity in a collecting duct cell line. This reduction in ENaC activity can be reversed early with exposure to dexamethasone, an effect that is associated with an increase in sgk1 abundance. The C2 domain is required to target Nedd4–2 to the plasma membrane in response to elevation of intracellular Ca2+ concentration ([Ca2+]i) in MDCK cells, although it does not appear to mediate the inhibitory effect of [Ca2+]i on Na+ transport. Our data illustrate that naturally occurring hNedd4–2 isoforms differentially associate with ENaC to regulate its activity.
epithelial sodium channel
NA+ TRANSPORT IN SOME epithelial cells, such as the connecting tubules and collecting ducts of the kidney, airway and alveolar epithelia of the lung, in the colon, and in sweat glands and salivary glands occurs via the epithelial Na+ channel (ENaC). ENaC present at the apical membrane of transporting epithelia is composed of three subunits (, , and ) that form a heteromultimeric protein complex. This channel permits the passive diffusion of Na+ in cells where its concentration is kept low by extrusion of Na+ through the action of Na+-K+-ATPase at the basolateral membrane. The resulting effect is the reabsorption of Na+ from the lumen into the interstitial space (2, 18, 42).
The sgk and Nedd4/Rsp5 protein families are considered to be important physiological regulators of ENaC activity. Sgk1, a kinase with pleiotropic effects on multiple transporters, has been shown to increase ENaC activity in heterologous expression systems by increasing abundance of surface channels, increasing open probability, and/or activating silent channels (3, 11, 12, 47, 48, 54). Knock down of sgk1 activity, by a dominant-negative sgk1 form or by antisense-mediated decay, results in a substantial reduction of ENaC activity (15, 21, 38). Sgk1-null mice are somewhat inefficient at reabsorbing Na+, leading to a reduction of blood pressure when kept on a low-salt diet (55).
Nedd4/Rsp5 proteins are WW domain-containing E3-type ubiquitin ligases that regulate surface expression of ENaC in heterologous expression systems (50). WW domains of Nedd4/Rsp5 bind to the PY motifs of ENaC, and the COOH-terminal HECT domain catalyzes the transfer of ubiquitin residues to ENaC, leading to channel internalization and reduced Na+ transport. ENaC mutations that disrupt a PY motif abolish the effect of Nedd4/Rsp5. It is these mutations that result in a dominantly inherited form of severe hypertension called Liddle's syndrome, a consequence of increased Na+ reabsorption in the connecting tubules and collecting duct (19, 46).
Nedd4–2, a member of the Nedd4/Rsp5 family, is very effective in downregulating ENaC activity in heterologous expression systems (27, 29). More recently, an RNA interference approach was used to show that knocking down Nedd4–2 expression increases endogenous ENaC activity in a lung epithelial cell line (H441) and increases transfected ENaC activity in a rat thyroid epithelial cell line (FRT; see Ref. 48). Sgk1 phosphorylates Nedd4–2 at three distinct serine/threonine residues, thus reducing the affinity between Nedd4–2 and ENaC (11, 47). Because Nedd4–2 reduces ENaC surface expression, phosphorylation of Nedd4–2 may increase ENaC abundance at the plasma membrane and increase ENaC activity. This indirect effect of sgk1 is thought to be one of the mechanisms by which sgk1 stimulates Na+ transport.
We have previously reported that human Nedd4–2 (hNedd4–2) exists as many isoforms that arise from alternate transcription and translation start sites and from variable splicing of some internal exons (27). Some forms of hNedd4–2 have an NH2-terminal C2 domain, a domain that in other proteins is a Ca2+ and phospholipid-binding domain and may dictate membrane localization (41). Other hNedd4–2 isoforms differ in the number of WW domains and sgk1 phosphorylation sites that are present (27). In an in vitro binding assay, WW domains 3 and 4 appear to have strong affinity for the PY motifs of ENaC, whereas domains 1 and 2 have little or no affinity (27). The varying affinities of the WW domains of Nedd4–2 for ENaC have been described in a number of laboratories (4, 16, 23).
In this report, we examined the expression of Nedd4–2 in the collecting duct of the kidney and studied the interaction with ENaC and the intracellular localization and function of three naturally occurring hNedd4–2 isoforms.
EXPERIMENTAL PROCEDURES
Cell culture. COS-7 cells (gift from M. J. Welsh, University of Iowa) were cultured in DMEM containing 10% FBS and 1% penicillin-streptomycin. M-1 cells were (gift from Géza Fejes-Tóth, Dartmouth Medical School) cultured in DMEM-F-12 containing 10% FBS and 1% penicillin-streptomycin unless otherwise specified. Madin-Darby canine kidney (MDCK)-C7 cells were (gift from B. Blazer-Yost and H. Oberleithner) cultured in MEM containing 10% FBS and 1% penicillin-streptomycin.
Transient transfection. The coding regions of human -, -, and -ENaC tagged at the COOH terminus with Flag, hemagglutinin (HA), and Myc epitopes, respectively, were cloned into pcDNA3 (Invitrogen, Carlsbad, CA), whereas the coding regions of all three hNedd4–2 isoforms where cloned into pcDNA4/HisMax C in frame with the NH2-terminal His and Xpress epitopes (Invitrogen) and into EGFP-N1 in frame and upstream of the enhanced green fluorescent protein (EGFP) open reading frame (Clontech, Palo Alto, CA). COS-7 cells (107 cells) were spun down, resuspended in 400 μl cytomix solution (in mM: 120 KCl, 0.15 CaCl2, 10 K2HPO4, 10 KH2PO4, 25 HEPES, 2 EGTA, 5 MgCl2, 2 Na-ATP, and 5 glutathione), and then mixed with 15 μg plasmids expressing human ENaC (hENaC; 5 μg/subunit) and 20 μg pcDNA4/HisMax C plasmids containing hNedd4–2 isoforms or EGFP-N1 plasmid containing hNedd4–2 isoforms. The cell-DNA mixture was electroporated with Gene Pulser II (Bio-Rad, Hercules, CA) and then seeded on 100-mm culture plates for immunoprecipitation experiments and on chamber slides coated with poly-L-Lysine for immunofluorescence experiments.
MDCK-C7 cells (107 cells) were prepared as above in cytomix solution, transfected by electroporation at 2,850 μF and 220 volts with 20 μg EGFP-N1 plasmids expressing Nedd4–2 isoforms, and then seeded on Millicell PCF filters pretreated with human placental collagen at a density of 2.5 x 104 cells/filter. M-1 cells were also suspended in cytomix solution and transfected by electroporation at 500 μF and 280 volts with 20 μg pcDNA4/HisMax C expressing hNedd4–2C2 or the control plasmid pRLSV40 that expresses Renilla luciferase (Promega, Madison, WI). Transfected cells were then seeded on Millicell PCF filters pretreated with human placental collagen.
Immunofluorescence. Some transfected COS-7 cells were grown on chamber slides (Nunc, Rochester, NY). After transfection (72 cells), cells were fixed with 4% paraformaldehyde and blocked with 5% mouse serum and 1% BSA in PBS followed by incubation with rhodamine-conjugated anti-myc antibody (Santa Cruz, Santa Cruz, CA). Cells were washed with PBS, and nuclei were stained with TO-PRO3 (Molecular Probes, Eugene, OR). Slides were then mounted with VectaShield (Vector Laboratories, Burlingame, CA), and fluorescent signals were detected using a Bio-Rad 1040 confocal microscope (Bio-Rad).
Transfected MDCK cells were grown on chamber slides for 72 h. Cells were then washed with 140 mM NaCl, 6 mM KCl, 1 mM MgCl2, 0.1 mM EDTA, 20 mM glucose, and 20 mM HEPES, pH 7.3, and then incubated in the same media for 5 min in the presence of 2 mM EGTA or 1.1 mM CaCl2 and 1 μM ionomycin (Sigma, St. Louis, MO). Cells were then fixed with 4% paraformaldehyde and washed with PBS. PCF filters were cut out and mounted on slides and visualized as described above.
Western blot analysis and immunoprecipitation. Transfected COS-7 cells and M-1 cells grown in 100-mm plates were lysed using M-PER lysis buffer (Pierce, Rockford, IL), and protein concentration was determined using the Bradford method. Protein lysates were boiled, run on a 7% SDS-PAGE, and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA). Membrane was blocked with Tris-buffered saline-Tween 20 (TBST) containing 5% milk then incubated in 5% milk-TBST with an anti-Nedd4–2 antibody (gift from Dr. Howard Pratt), an anti--ENaC antibody (gift from Dr. Mark Knepper), or an anti-SGK1 antibody (Upstate, Charlottesville, VA). Membranes were then washed with TBST and incubated with secondary horseradish peroxidase-conjugated antibody followed by chemiluminescence using Supersignal West Pico chemiluminescent substrate (Pierce). In some instances, membranes were stripped with Restore stripping buffer (Pierce) for 15 min, incubated with 5% milk-TBST, and then reprobed with anti--tubulin antibody (Santa Cruz). Autoradiograms were scanned, and the density of individual bands was measured using Kodak Digital Science Image Analysis software (Rochester, NY) for sgk1 immunoblots. The sgk1 band was normalized for the density of the -tubulin band.
For immunoprecipitation, the ProFound Mammalian HA-Tag IP kit (Pierce) was used according to the supplier's protocol. In brief, transfected COS-7 lysates were incubated with anti-HA antibodies conjugated to agarose beads in the provided columns. Columns were then spun down and washed three times. HA-bound protein complexes were retrieved by elution with loading buffer, boiled, and run on SDS-PAGE followed by Western analysis using the anti-Nedd4–2 antibody.
Xenopus oocyte expression. The hENaC -, -, and -subunits and hNedd4–2 isofroms (Nedd4–2, Nedd4–2C2, and Nedd4–2WW2,3) were subcloned into the pGEM-HE plasmid (53). Linearized plasmids were subjected to in vitro transcription using the mMessage Machine (Ambion, Austin, TX) kit to produce capped cRNA from each construct. The integrity of the cRNAs was evaluated by agarose gel electrophoresis and quantified by densitometry. The cRNAs were diluted in water so that 50-nl injections with a Drummond Nanoject oocyte injector carried 1 ng of each cRNA, unless otherwise indicated.
Mature female Xenopus laevis were purchased from Xenopus I (Dexter, MI) or Nasco (Fort Atkinson, WI). X. laevis were housed in the University of Iowa animal facility in dechlorinated tap water at 18–20°C. Stage V and VI oocytes were removed from toads that were anesthetized in an ice-cold 2 mg/ml tricaine solution. The oocytes were defolliculated with collagenase and stored overnight at 18°C in frog Ringer solution (NFR) consisting of 115 mM NaCl, 2.5 mM KCl, 1.8 mM CaCl2, 10 mM HEPES, 5 mM sodium pyruvate, and 100 U/ml penicillin-streptomycin (pH 7.35). After 12–24 h of recovery from the collagenase treatment, healthy oocytes were injected with cRNA. Ringer solution was changed daily. Whole cell hENaC currents were measured 48–72 h after cRNA injections. Measurements were made in NFR using an OC-725C oocyte voltage-clamp amplifier (Warner Instruments, Hamden, CT). The pCLAMP software suite (Axon Instruments, Union City, CA) was used for amplifier control and data collection/analyses. All recordings were performed at room temperature. Amiloride-sensitive currents were derived by subtracting currents recorded in 10 μM amiloride from preamiloride currents.
After injection (48 h), 20 oocytes/injection group were chilled to 4°C in 200 μl NFR. The oocytes were ruptured with 15 strokes in a small Teflon-glass dounce homogenizer. The resulting slurry was centrifuged at 1,000 g for 5 min to separate yolk, lipids, and other large cellular debris. The lipid layer at the top of the tube was wicked off using a cotton swab, and the supernatant was centrifuged again at 1,000 g for 5 min. An equal volume of NFR + 2% SDS + protease inhibitor cocktail (Roche, Indianapolis, IN) was added to the final supernatant, and the mixture was sheared with 20 passes through a 20-gauge needle. Each sample (10 μl) was then boiled and loaded on an SDS-PAGE gel and transferred to PVDF membranes for Nedd4–2 immunoblotting.
Microdissection and RT-PCR. C57BL/6J mice housed in the University of Iowa animal facility were killed, and kidneys were isolated. Animal experiments were approved by the University of Iowa Institutional Animal Care and Use Committee. Kidney cortex was dissected with a blade and cut into fine pieces that were then incubated in MEM medium containing 0.5 mg/ml type 2 collagenase, 5 mM glycine, 50 U/ml DNase, and 48 μg/ml soybean trypsin inhibitor. Kidney cortex pieces were incubated in the MEM collagenase solution until the medium became cloudy. Collecting duct segments were then picked with a needle with the aid of Olympus stereoscope (Olympus, Melville, NY), as previously described (45). RNA was extracted using TRI reagent (Molecular Research center, Cincinnati, OH) following the supplier's protocol.
RNA samples were then digested with DNase (Promega) and reverse transcribed using SuperScript II (Invitrogen) according to the supplier's protocol. The resulting cDNA was used to perform PCR with AmpliTaq DNA polymerase (Roche) using the following primers: mVPR2, 5'-TTTCGTCCCCTAGCTCTCC-3' and 5'-ATACCCCACTGCCATTTCC-3'; mSGLT1, 5'-GACATCCCAGAGGACTCCAA-3' and 5'-ACCACTGTCCTCCACAAAGG-3'; and mNedd4–2, 5'-TCTCAACTGGTTGCCGTGTA-3' and 5'-AAAGCTGAAAATCAGTCTAAATCA-3'.
RNase protection assay. RNA was extracted from mouse kidney cortex and medulla and from M-1 cells using TRI reagent. The hNedd4–2 template has been previously described (27). The two mouse RNase protection assay (RPA) templates were generated by RT-PCR from M-1 cells. In brief, RNA from M-1 cells was extracted and reverse transcribed. The resulting cDNA was used to amplify two PCR fragments using each of the following primer pairs: mN4–2F2, 5'-TCTAGACCAGCCTTCCTCTCC-3'; mN4–2R2, 5'-CAGGAAGTCGTCTCGTGTCA-3'; and mN4–2F4, 5'-AGTTCATGTGGGGGACAAAG-3'; and mN4–2R4, 5'-TAGGGTCTCTCCATGGTTGG-3'. The resulting PCR products were gel purified and cloned into PCRXl-Topo (Invitrogen). Templates containing the cloned hNedd4–2 and mouse Nedd4–2 (mNedd4–2) cDNA fragments and an 18S rRNA cDNA fragment (pTR1 RNA 18S; Ambion) were used to generate radiolabeled antisense cRNA probes in the presence of [-32P]UTP (PerkinElmer, Wellesley, MA), as previously described (28). RNA samples were cohybridized overnight with hNedd4–2 or mNedd4–2 and 18S cRNAs and then digested with RNase A and T1 and nuclease-protected fragments resolved by PAGE.
Generation of stable cell lines. M-1 cells were transfected by electroporation with sLacR.hyg that expresses the Lac repressor (gift from Dr. Bishop, University of Iowa). Cells were then grown in DMEM-F-12 medium containing 200 μg/ml hygromycin B (Invitrogen). Individual hygromycin-resistant colonies were grown and screened for expression of LacR. The highest expressing clone "S25" was further transfected with popRSV5.neo plasmid that expresses Nedd4–2 in the presence of LacR and isopropyl thiogalactopyranoside (IPTG). Cells were then grown in DMEM-F-12-hygro in the presence of 250 μg/ml geneticin (Invitrogen). Individual geneticin-resistant colonies were isolated, grown, and screened for Nedd4–2 expression.
Short-circuit current measurements. M-1 cells seeded on Millicell PCF filters were grown for 2 days in DMEM-Ham's F-12 supplemented with insulin (5 μg/ml), transferrin (5 μg/ml), triiodothyronine (5 nM), hydrocortisone (50 nM), sodium selenite (10 nM), gentamicin (50 μg/ml), BSA (10 g/l), and dexamethasone (5 nM). The monolayers were then switched to the same medium without albumin and steroids for 1 day. Measurements of transepithelial voltage and resistance and short-circuit current (Isc) were conducted at 37°C in Ussing chambers as described (44).
The short-term time-course measurements were conducted in a separate set of chambers designed to accommodate Millicell filter-bottom cups bathed in a Krebs-Ringer bicarbonate solution containing (in mM) 115 NaCl, 25 NaHCO3, 5 KCl, 5 Na-HEPES, 5 H-HEPES, 1.5 CaCl2, 1 MgCl2, 1 Na2HPO4, and 5 D-glucose. The chambers were water-jacketed to 37°C, and electrical measurements were made with a University of Iowa voltage clamp, as previously described (24). HCO3–-containing solutions were gassed with 5% CO2 in air to maintain pH at 7.4.
Adenoviral transduction. Nedd4–2 and Nedd4–2C2 with NH2-terminal Xpress tags were cloned into a shuttle vector, pacAd5CMV K-N pA, downstream of the CMV promoter. The resulting plasmids were used to generate Nedd4–2- and Nedd4–2C2-expressing adenoviral constructs by the Vector core at the University of Iowa. For adenoviral transduction, M-1 cells seeded on Millicell PCF filters were grown for 3 days followed by the addition of Nedd4–2, Nedd4–2C2, or empty virus (8 x 107 plaque-forming units/filter) to the apical medium for 4 h. Isc measurements were taken 48 h after transduction with cell cultures switched to medium containing 100 nM dexamethasone for the last 24 h. To study the effect of elevation of intracellular Ca2+ concentration ([Ca2+]i) on Isc, short-term time-course experiments were performed as described earlier, adding ionomycin (200 nM) to the basolateral medium, and the effect was followed for 10 min. Benzamil (10 μM) was then added to the apical medium to obtain benzamil-sensitive Isc.
RESULTS
Nedd4–2 exists as many isoforms. We have previously shown that, because of alternative exon usage, hNedd4–2 exists as many transcripts that are predicted to be translated into proteins that either have or lack an NH2-terminal C2 domain (27). Further diversity arises from alternative splicing of exons 12–15 that are predicted to lead to Nedd4–2 proteins with a variable number of WW domains and sgk1 phosphorylation sites (Fig. 1). We have also shown, by RT-PCR and RPA, that Nedd4–2 isoforms are expressed in a tissue-specific manner (27). We selected three different Nedd4–2 isoforms that are abundantly expressed [full-length Nedd4–2, Nedd4–2 lacking the C2 domain (Nedd4–2C2), and Nedd4–2 lacking WW2, WW3, and the C2 domain (Nedd4–2WW2,3)], tested the effect of these isoforms on ENaC activity, and studied their cytosolic localization.
Nedd4–2 and ENaC show cytoplasmic distribution. To study the intracellular distribution of Nedd4–2 isoforms and ENaC, we transfected COS-7 cells with plasmids expressing all three ENaC subunits (, , and ) together with EGFP, or Nedd4–2, Nedd4–2C2 or Nedd4–2WW2,3. -ENaC is myc-tagged, allowing detection with a Rhodamine-conjugated anti-myc antibody, whereas all Nedd4–2 isoforms are EGFP-tagged. Our results show that EGFP is predominantly localized to the nucleus, whereas ENaC exhibits an exclusive diffuse cytoplasmic distribution (Fig. 2A). Expression of Nedd4–2 isoforms as an EGFP fusion protein results in a cytoplasmic EGFP signal, suggesting that Nedd4–2 contains a nuclear export signal. Alternatively, EGFP may have a nuclear localization signal that is masked when expressed as a COOH-terminal fusion protein (Fig. 2, B-D). We could not detect preferential expression of the Nedd4–2 isoforms at the cell membrane, nor did we detect differences in expression patterns between Nedd4–2 isoforms, suggesting that, at least in COS-7 cells, these proteins are similarly distributed. We also noted that, despite its known cell membrane expression and function, most of the detectable ENaC is cytosolic in this expression system. Cytosolic distribution of transfected and native ENaC has been previously reported in native tissues and cultured cells, suggesting that only a small fraction of total ENaC is present at the plasma membrane (20, 33). Interestingly, cells expressing ENaC appear more spherical with less cytoplasmic extensions, suggesting that some fraction of these channels are functional with increased Na+ absorption and subsequent change in cell shape (Fig. 2B).
Nedd4–2 interacts with ENaC. To determine if Nedd4–2 associates with ENaC, we performed immunoprecipitation experiments in COS-7 cells transfected with three ENaC subunits (, , and ) and with Xpress-tagged Nedd4–2, Nedd4–2C2, or Nedd4–2WW2,3. The anti-Nedd4–2 antibody used for immunoblotting detects transfected Nedd4–2 and appears to detect endogenous Nedd4. The amount of expressed Nedd4–2 varied somewhat between each isoform and varied significantly when ENaC was coexpressed. The COS-7 cell lysates were normalized to contain equal amounts of the transfected Nedd4–2 isoforms before immunoprecipitation. On Western blot analysis of normalized cell lysates, each of the transfected tagged Nedd4–2 isoforms migrate at the predicted size as follows: Nedd4–2 and Nedd4–2WW2,3 migrate at 116 and 83 kDa and Nedd4–2C2 migrates at 101 kDa, almost superimposed on endogenous Nedd4 at 104 kDa (Fig. 3A). Cell lysates were immunoprecipitated with an anti-HA antibody that recognizes the tagged -ENaC. When expressed together with -, -, and -ENaC, each of the Nedd4–2 isoforms coimmunoprecipitate with ENaC (Fig. 3B). Although we could not detect a difference in levels of immunoprecipitatable Nedd4–2 and Nedd4–2C2, Nedd4–2WW2,3 is immunoprecipitated to a much lesser extent with ENaC. These results suggest that WW2 and/or WW3 domains are important for the association of Nedd4–2 with ENaC, whereas the C2 domain of Nedd4–2 may not influence this association.
Nedd4–2WW2,3 is not effective in reducing ENaC-dependent Na+ transport. To study the effect of the three Nedd4–2 isoforms on Na+ transport, we expressed ENaC with or without Nedd4–2 in Xenopus oocytes followed by whole cell voltage clamp and amiloride-sensitive current measurements. We also studied the level of Nedd4–2 proteins in injected oocytes by Western analysis. We show that both Nedd4–2 and Nedd4–2C2 robustly reduce ENaC current. Importantly, the effect of Nedd4–2WW2,3 is small and did not reach statistical significance (Fig. 4A). The Western blot results show that the difference in effectiveness of Nedd4–2WW2,3 protein cannot be explained by differences in abundance of expressed isoforms, since Nedd4–2 is expressed at a relatively lower level than Nedd4–2WW2,3 (Fig. 4B). The expression of Nedd4–2C2 is more than Nedd4–2 and may explain the small difference between them in the inhibition of Na+ transport. We then asked if the effect of Nedd4–2WW2,3 on ENaC could be enhanced by expressing it in excess. Our results show that, despite expression at up to 10-fold excess, there was no statistically significant effect of Nedd4–2WW2,3 on ENaC-mediated currents (Fig. 4, C and D).
Because Nedd4–2WW2,3 binds to ENaC (albeit with lower affinity than Nedd4–2) and because it has a small effect on ENaC, we asked whether overexpression of Nedd4–2WW2,3 could compete with Nedd4–2C2 for its effect on ENaC, thus acting in a dominant-negative fashion. To address this issue, we expressed Nedd4–2C2 with 20-fold excess of Nedd4–2WW2,3, followed by current measurements. Our results show that Nedd4–2WW2,3 does not interfere with the ability of Nedd4–2C2 to reduce ENaC-dependent current (Fig. 5). Collectively, our results show that both Nedd4–2 and Nedd4–2C2 can robustly interact with and downregulate ENaC activity in a reconstituted system, whereas the effect of Nedd4–2WW2,3 is weak. These results suggest that Nedd4–2WW2,3 may have a marginal role in the regulation of ENaC activity.
Nedd4–2 is expressed in the kidney collecting duct. Our studies so far indicate that ENaC and Nedd4–2 can interact in heterologous expression systems. We next wanted to determine if Nedd4–2 is expressed and able to regulate ENaC in the collecting duct. We performed RT-PCR on microdissected collecting duct samples of mouse kidney. To confirm collecting duct lineage and to monitor for contaminating proximal tubular epithelium in the microdissected samples, we also studied the expression of the vasopressin 2 receptor (VPR2), a collecting duct-specific transcript, and the sodium glucose transporter (SGLT1), a proximal tubule-specific transcript. Our results indicate that Nedd4–2 and VPR2 are expressed in the mouse collecting duct (Fig. 6A).
We have previously shown that, because of alternative promoter usage, hNedd4–2 exists as isoforms that either have or lack a C2 domain (27). To study if a C2-containing Nedd4–2 is also present in the mouse transcriptome, we first examined all known cDNA sequences that map to the mNedd4–2 locus in the Ensembl and EST databases. As shown in Fig. 6B, one form of mNedd4–2 (accession no. AF277232) initiates transcription in exon 1a, with the open reading frame beginning in exon 6, which leads to translation of a protein without a C2 domain. Interestingly, we found a short cDNA sequence (accession no. AK040106) that extends to exon 4 and is predicted to arise from transcription that initiates in a new exon, 1c. Exon 1c contains a translation initiation codon that is in frame with the downstream initiation codon in exon 6, resulting in the translation of a protein that contains a C2 domain (Fig. 6B).
To study the relative abundance of these two mNedd4–2 isoforms in the mouse kidney and in a collecting duct epithelial cell line, we performed RPA on samples from mouse kidney cortex and medulla and from M-1 cells using two different cRNA probes. The first probe is antisense to sequences in exons 1c through exon 6 and will detect two bands, with the larger corresponding to transcripts that contain exons 1c through 6 (plus C2 form) and the smaller band corresponding to transcripts that contain exons 2 through 6, but not exon 1c (minus C2 form; Fig. 6C). The second probe is antisense to sequences in exons 1a through exon 5 and will also detect two bands. The upper band corresponds to transcripts that contain exons 1a through 5 (minus C2 form), and the lower band corresponds to transcripts that contain exons 2 through 5 (plus C2 form; Fig. 6D). Although expression of the two Nedd4–2 isoforms is equivalent in the kidney cortex and medulla, the C2-containing form is expressed at a higher level in M-1 cells. This suggests that Nedd4–2 isoforms may be differentially regulated in the mouse collecting duct.
Nedd4–2C2 reduces Na+ transport in collecting duct epithelia. To study the effect of Nedd4–2 on ENaC activity in the collecting duct, we transiently expressed Nedd4–2C2, Nedd4–2WW2,3, or an irrelevant protein, Renilla luciferase, in M-1 cells. Based on analysis of transfected green fluorescent protein, we routinely achieve a transfection efficiency of >50% (data not shown). We show that Nedd4–2C2 significantly reduces ENaC-dependent Na+ transport in M-1 cells 72 h after transfection (Fig. 7A), whereas Nedd4–2WW2,3 has no effect (Fig. 7B). We also show that the effect of Nedd4–2C2 is dose-dependent and correlates with the level of expression of transfected Nedd4–2C2 (Fig. 7, C and D). Our results indicate that Nedd4–2C2 can downregulate endogenous ENaC activity in a collecting duct cell line. Because Nedd4–2C2 is localized throughout the cytoplasm, it may regulate ENaC activity by reducing the insertion rate of newly synthesized or recycling channels. Alternatively, enough Nedd4–2C2 is present at the apical membrane, leading to increased retrieval rate of active channels.
Short-term glucocorticoid treatment abolishes the Nedd4–2 effect on Na+ transport. To study the effect of glucocorticoid treatment on Nedd4–2 in the collecting duct, we generated a stable M-1 cell clone that overexpresses Nedd4–2. We transfected M-1 cells with LacR plasmid that constitutively expresses the Lac repressor. One clone, S25, that expressed the highest level of LacR was selected for the second transfection and served as a control for subsequent studies. S25 cells were transfected with a plasmid that expresses Nedd4–2 under the control of an IPTG-inducible promoter. Clones with inducible Nedd4–2 expression were isolated and exhibited a Nedd4–2-dependent decrease in Na+ transport (data not shown). One clone, N1n-4, that expressed the highest levels of Nedd4–2 even in the absence of IPTG (Fig. 8A) was selected for further studies. Basal Isc in N1n-4 is very low compared with S25 cells (0.66 ± 0.19 vs. 2.76 ± 0.19 μA/cm2). The level of immunoblottable -ENaC is not affected in response to overexpression of Nedd4–2 in M-1 cells, suggesting that Nedd4–2 may interact with a small pool of total ENaC (Fig. 8B).
To study the effect of glucocorticoids on Na+ transport, we tested dexamethasone in N1n-4 and S25 cells. Dexamethasone appeared to stimulate Na+ transport in N1n-4 by 1 h, such that both cell lines have matching current over the next few hours. At 6 h, the cell lines diverge again, with N1n-4 now showing reduced Na+ transport compared with S25 (Fig. 8C). These results are consistent with a model where dexamethasone stimulates sgk1 to abolish the early but not the late inhibitory effect of Nedd4–2 on ENaC. To understand the biphasic effect of dexamethasone on Isc in N1n-4, we performed Western blot analysis for sgk1 in M-1 samples treated with dexamethasone for varying time periods (Fig. 8D). Our results show that dexamethasone increases sgk1 levels within 1 h of treatment, with a continued increase for 4 h. The levels of sgk1 then decline and reach basal levels by 24 h of dexamethasone treatment (Fig. 8E). The transient effect of dexamethasone on sgk1 abundance provides an attractive model for the dexamethasone effect on Na+ transport in N1n-4 cells. In this model, sgk1 levels are elevated after dexamethasone treatment (1–4 h), thus inhibiting the effect of Nedd4–2 on ENaC activity; however, sgk1 elevation is transient, thus restoring the effect of Nedd4–2 on ENaC after 6 h. Presumably, the increased Isc at later time points in N1n-4 is the result of increased synthesis of ENaC subunits in response to dexamethasone treatment (14).
The C2 domain is a [Ca2+]i-regulated membrane-targeting sequence. To determine if the C2 domain affects trafficking of Nedd4–2, we transfected MDCK cells grown on permeable supports with EGFP-tagged Nedd4–2 with and without a C2 domain. Our results indicate that Nedd4–2 isoforms have a cytoplasmic distribution in the absence of extracellular Ca2+ (Fig. 9, A and C). Treating the cells with ionomycin, to elevate [Ca2+]i, targets the C2 domain-containing form but not the C2-less form to the plasma membrane (Fig. 9, B and D). The Z-section image series spanning the whole height of cells shows that Nedd4–2 is distributed throughout the plasma membrane (data not shown) and not specifically localized to the apical membrane. This study indicates that the C2 domain of Nedd4–2, like the C2 domain of Nedd4, phospholipase A2, and synaptotagmin, directs Nedd4–2 to the plasma membrane in response to elevation in [Ca2+]i (6, 35, 41).
Elevation of [Ca2+]i reduced Na+ transport in collecting duct epithelia. To determine if elevation of [Ca2+]i affects Isc in M-1 cells, we measured Isc in M-1 cells after treatment with 200 nM ionomycin. Ionomycin stimulates a transient peak of Isc followed by a new plateau, both of which are benzamil-insensitive, suggesting that the rise in Isc is the result of Cl– secretion (Fig. 10A). Indeed, removal of extracellular Cl– inhibited the rise in Isc (data not shown). To determine the effect of ionomycin on ENaC-dependent current, we added benzamil 10 min after ionomycin treatment when the current had reached a new stable baseline. The benzamil-sensitive Isc, after ionomycin, was then compared with the benzamil-sensitive Isc of control M-1 cells not treated with ionomycin (Fig. 10, A and B). Our results show that elevation of [Ca2+]i by ionomycin results in a significant reduction in ENaC activity (Fig. 10B).
To study the effect of ionomycin in cells overexpressing Nedd4–2, we transduced M-1 cells with adenoviral constructs that express Nedd4–2 and Nedd4–2C2. As expected from our data in Figs. 7 and 8, benzamil-sensitive Isc is reduced in Nedd4–2-transduced cells compared with cells transduced with an empty virus. As observed in Fig. 4, Nedd4–2C2 is at least as effective as Nedd4–2 in reducing current. A further decrease in Isc occurs after treatment with ionomycin (Fig. 10B). However, the fractional reduction in Isc was not different between control and each of the Nedd4–2 transduced cells, suggesting that overexpression of Nedd4–2 has no apparent effect on [Ca2+]i-inhibited Isc, at least in a collecting duct cell line with native ENaC (Fig. 10C). Collectively, these data would suggest that Nedd4–2 regulates ENaC within the cytoplasmic pool, rather than at the cell surface, to reduce Na+ transport. Alternatively, elevation of [Ca2+]i leads to the transfer of native Nedd4–2 to the plasma membrane, resulting in inhibition of ENaC at the cell surface that cannot be amplified by adding more Nedd4–2.
DISCUSSION
Nedd4–2 is a recently identified protein that belongs to the Nedd4/Rsp5 family of E3-type ubiquitin ligases. These proteins are present from yeast to human, are widely expressed in mammalian tissues and cells, and are involved in a wide variety of cellular functions. These functions include regulation of the cell cycle, intracellular trafficking, and surface expression of many channels and transporters (25, 43). All members of this family share the same WW and HECT domains, with some members having an NH2-terminal C2 domain as well. The C2 domain is a Ca2+ and phospholipid-binding domain, and the catalytic HECT domain in the COOH-terminus transfers ubiquitin to lysine residues on target proteins. The C2 and HECT domains are separated by a number of WW domains, which are 40 amino acids in length and contain two conserved tryptophan (W) residues. WW domains are well-characterized binding partners for PY motifs (usually consisting of PPXY). Such PY motifs are present at the COOH-terminus of all three ENaC subunits and are important for the regulatory effect of Nedd4–2 on ENaC activity (1, 31, 49).
We have previously demonstrated that hNedd4–2 exists in different isoforms that are, in part, the result of transcription initiation in different exons (27). Some of these exons have an alternate translation start codon leading to a protein with a C2 domain, whereas other exons lack this upstream initiation codon, resulting in an hNedd4–2 isoform lacking the NH2-terminal C2 domain. Other hNedd4–2 variants are the result of internal splicing, which leads to isoforms missing at least one of four WW domains. When tested individually in vitro, WW domains 1 to 4 of Nedd4–2 have differing affinities for the PY motifs of ENaC. We and others have shown that WW3 has the highest affinity followed by WW4, whereas WW1 and WW2 have very weak affinity (4, 16, 23, 27).
We have previously shown that an hNedd4–2 variant lacking WW2 and WW3 (Nedd4–2WW2,3) is a naturally occurring isoform that arises when exons 12 through 15 are spliced out (27). Here we show that Nedd4–2WW2,3 has a substantially lower affinity for ENaC than full-length Nedd4–2 in transfected COS-7 cells. We also report that Nedd4–2WW2,3 does not significantly regulate ENaC activity when reconstituted in Xenopus oocytes or in M-1 cells, suggesting that Nedd4–2WW2,3 may not have a role in regulating ENaC activity in epithelia where it is normally expressed. We speculate that Nedd4–2WW2,3 may regulate the activity of other target proteins where the affinity of individual WW domains for the target PY domain may be different than that for ENaC. A number of new Nedd4–2 targets have been recently described, including the cardiac voltage-gated Na+ channel Nav1.5 (52), the glucose transporter SGLT1 (13), the neuronal voltage-gated Na+ channels (Nav1.2, Nav1.7, and Nav1.8; see Ref. 17), intestinal phosphate cotransporter NaPi IIb (39), the voltage gated K+ channel Kv1.3 (22), and the glutamate transporter EAAT1 (5). Some of these channels (Nav1.2, Nav1.5, Nav1.7, and Nav1.8) have PY motifs, and preliminary studies show that the PY motif of Nav1.5 has a high affinity for WW4 of Nedd4–2 and no measurable affinity for WW2 and WW3 (52), making Nav1.5 a candidate that may be regulated by Nedd4–2WW2,3. Many of the other Nedd4–2-regulated transporters do not have PY motifs, and the effect of Nedd4–2 may be mediated by PY-containing intermediates.
Previous reports demonstrate that Nedd4–2 regulates ENaC activity in heterologous expression systems and in a lung epithelial cell line (27, 30, 47, 48). In this paper, we demonstrate that Nedd4–2 is expressed in the mouse collecting duct and in a mouse collecting duct cell line and confirm that, similar to hNedd4–2, mNedd4–2 has 5'-variants with transcription starting from at least two different exons, leading to proteins with and without a C2 domain. Using three independent approaches (transient transfection, stable transfection, and viral transduction), we show that Nedd4–2 overexpression in M-1 cells inhibits ENaC activity.
The mineralocorticoid aldosterone stimulates ENaC activity in the collecting duct, and this regulation is associated with the transcription of aldosterone-responsive genes sgk1 and the -subunit of ENaC. Sgk1 synthesis begins early (<1 h) after exposure to aldosterone or glucocorticoids; typically, this sgk1 synthesis level is not maintained and returns to baseline after prolonged (>4 h) aldosterone treatment (8, 36, 37). -ENaC exhibits a different response pattern, where increased synthesis does not occur until later (>4 h; see Ref. 14). This suggests that at least a portion of the early effect of aldosterone on increasing ENaC activity may be mediated by sgk1, whereas the later effect is the result of increased channel synthesis. Sgk1 is thought to increase ENaC activity by many mechanisms. One mechanism involves the phosphorylation of Nedd4–2, thereby reducing its affinity to ENaC (11, 47). Consistent with this model, we show that treatment with dexamethasone for 1 h abolishes the Nedd4–2 effect on ENaC activity in M-1 cells overexpressing Nedd4–2. Moreover, the effect of Nedd4–2 on ENaC activity in those cells correlates with sgk1 expression where Nedd4–2 is inactivated when sgk1 levels are elevated and reactivated when sgk1 levels are reduced after prolonged dexamethasone treatment.
We have previously demonstrated that hNedd4–2 forms with and without the C2 domain are both efficient at downregulating ENaC activity when reconstituted in FRT epithelia (27). Earlier studies using Nedd4 protein demonstrated that the C2 domain is important for targeting transfected Nedd4 to the plasma membrane of MDCK cells in the presence of elevated [Ca2+]i (41). Yet Nedd4 without a C2 domain is more efficient at inhibiting ENaC in heterologous expression systems (30). In the current study, Nedd4–2C2 is at least as efficient as Nedd4–2 (if not more) in regulating ENaC activity in Xenopus oocytes and in M-1 cells (Figs. 4 and 10). Like Nedd4, we show that, in transfected cells, Nedd4–2 has a predominant cytoplasmic distribution and that the C2 domain is necessary for targeting Nedd4–2 to the plasma membrane when [Ca2+]i is elevated. This suggests that Nedd4–2 might interact with and regulate the cytoplasmic pool of ENaC rather than a membrane pool of ENaC, thus reducing the number of channels available to be inserted in the apical membrane.
We used ionomycin to investigate the effect of [Ca2+]i elevation on ion transport in M-1 cells. Ionomycin resulted in a biphasic increase in Cl– secretion, with the first phase characterized by an immediate and transient peak of Isc; during the second phase, Cl– secretion reached a plateau that was higher than basal Cl– secretion (Fig. 10). cAMP, ATP, and norepinephrine have been reported to induce a similar Cl– secretion profile in M-1 cells, and at least the ATP effect is mediated by an increase in [Ca2+]i (9, 10, 32, 51). In our experiments, overexpression of Nedd4–2 did not affect Cl– secretion, suggesting that Nedd4–2 does not regulate the [Ca2+]i-stimulated Cl– secretion in M-1 cells (data not shown).
Earlier studies in other models of the collecting duct demonstrated that elevation of [Ca2+]i reduces ENaC activity, although at least one prior study was unable to confirm this effect in M-1 cells (7, 26, 34, 40). The mechanism for the inhibitory effect on Na+ transport is unknown, but an attractive hypothesis is that elevated [Ca2+]i targets the C2-containing form of Nedd4–2 to the plasma membrane where it can interact with ENaC, leading to channel internalization. Our studies confirmed that elevation of [Ca2+]i reduces ENaC activity in M-1 cells; however, overexpression of Nedd4–2 did not result in an enhancement of the Ca2+ effect. There are several possible explanations for this. The first is that the maximal effect of [Ca2+]i on ENaC activity is achieved by targeting endogenous Nedd4–2 to the apical membrane and that this cannot be enhanced by overexpression. Alternatively, the [Ca2+]i effect on Na+ transport may be independent of Nedd4–2.
In conclusion, we demonstrate that Nedd4–2 isoforms are expressed in the kidney collecting duct, and Nedd4–2 overexpression reduces native ENaC activity in a collecting duct cell line. This effect of Nedd4–2 is transiently abolished by exposure to glucocorticoids, which may be mediated by stimulation of sgk1. We also demonstrate that Nedd4–2 isoforms that contain the C2 domain are localized to the plasma membrane in response to elevated [Ca2+]i, although [Ca2+]i-induced reductions in ENaC activity do not seem to be dependent on having a Nedd4–2 with a C2 domain. Nedd4–2 isoforms missing WW2 and WW3 do not appear to regulate ENaC activity, and they may function to regulate other, as yet unknown, target proteins.
GRANTS
This work was supported in part by National Institutes of Health Grants DK-54348 and HL-71664 (to C. P. Thomas) and DK-52617 (to J. B. Stokes) and by Veterans Affairs Merit Review Awards to C. P. Thomas and J. B. Stokes. C. P. Thomas is an Established Investigator of the American Heart Association. O. A. Itani is the recipient of a Predoctoral fellowship from the American Heart Association, Heartland Affiliate.
ACKNOWLEDGMENTS
We thank Russell Husted, Rita Sigmund, Kenneth Volk, Kang Liu, Patrick Wright, and Thomas Moninger for excellent technical support and the University of Iowa DNA core, vector core and microscopy facility for services provided. We thank Dr. Gail Bishop for the LacR.hyg and popRSV5.neo plasmids, Dr. Mark Knepper for anti--ENaC antibody, Dr. Howard Pratt for the anti-Nedd4–2 antibody, and Dr. Jim Schafer for help with microdissection experiments.
The nucleotide sequence reported in this paper will appear in DNA Data Bank of Japan, European Molecular Biology Laboratory, GenBank, and Genome Sequence Database Nucleotide Sequence Databases with the accession number AY751751.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Abriel H, Loffing J, Rebhun JF, Pratt JH, Schild L, Horisberger JD, Rotin D, and Staub O. Defective regulation of the epithelial Na+ channel by Nedd4 in Liddle's syndrome. J Clin Invest 103: 667–673, 1999.
Alvarez de la Rosa D, Canessa CM, Fyfe GK, and Zhang P. Structure and regulation of amiloride-sensitive sodium channels. Annu Rev Physiol 62: 573–594, 2000.
Alvarez de la Rosa D, Zhang P, Naray-Fejes-Toth A, Fejes-Toth G, and Canessa CM. The serum and glucocorticoid kinase sgk increases the abundance of epithelial sodium channels in the plasma membrane of Xenopus oocytes. J Biol Chem 274: 37834–37839, 1999.
Asher C, Sinha I, and Garty H. Characterization of the interactions between Nedd4–2, ENaC, and sgk-1 using surface plasmon resonance. Biochim Biophys Acta 1612: 59–64, 2003.
Boehmer C, Henke G, Schniepp R, Palmada M, Rothstein JD, Broer S, and Lang F. Regulation of the glutamate transporter EAAT1 by the ubiquitin ligase Nedd4–2 and the serum and glucocorticoid-inducible kinase isoforms SGK1/3 and protein kinase B. J Neurochem 86: 1181–1188, 2003.
Chapman ER and Jahn R. Calcium-dependent interaction of the cytoplasmic region of synaptotagmin with membranes. Autonomous function of a single C2-homologous domain. J Biol Chem 269: 5735–5741, 1994.
Chase HS Jr and Al-Awqati Q. Regulation of the sodium permeability of the luminal border of toad bladder by intracellular sodium and calcium: role of sodium-calcium exchange in the basolateral membrane. J Gen Physiol 77: 693–712, 1981.
Chen SY, Bhargava A, Mastroberardino L, Meijer OC, Wang J, Buse P, Firestone GL, Verrey F, and Pearce D. Epithelial sodium channel regulated by aldosterone-induced protein sgk. Proc Natl Acad Sci USA 96: 2514–2519, 1999.
Cuffe JE, Bielfeld-Ackermann A, Thomas J, Leipziger J, and Korbmacher C. ATP stimulates Cl– secretion and reduces amiloride-sensitive Na+ absorption in M-1 mouse cortical collecting duct cells. J Physiol 524: 77–90, 2000.
Cuffe JE, Howard DP, Bertog M, and Korbmacher C. Basolateral adrenoceptor activation mediates noradrenaline-induced Cl– secretion in M-1 mouse cortical collecting duct cells. Pflügers Arch 445: 381–389, 2002.
Debonneville C, Flores SY, Kamynina E, Plant PJ, Tauxe C, Thomas MA, Munster C, Chraibi A, Pratt JH, Horisberger JD, Pearce D, Loffing J, and Staub O. Phosphorylation of Nedd4–2 by Sgk1 regulates epithelial Na+ channel cell surface expression. EMBO J 20: 7052–7059, 2001.
Diakov A and Korbmacher C. A novel pathway of ENaC activation involves an SGK1 consensus motif in the C-terminus of the channel's -subunit. J Biol Chem 279: 38134-38142, 2004.
Dieter M, Palmada M, Rajamanickam J, Aydin A, Busjahn A, Boehmer C, Luft FC, and Lang F. Regulation of glucose transporter SGLT1 by ubiquitin ligase Nedd4–2 and kinases SGK1, SGK3, and PKB. Obes Res 12: 862–870, 2004.
Dijkink L, Hartog A, Deen PM, van Os CH, and Bindels RJ. Time-dependent regulation by aldosterone of the amiloride-sensitive Na+ channel in rabbit kidney. Pflügers Arch 438: 354–360, 1999.
Faletti CJ, Perrotti N, Taylor SI, and Blazer-Yost BL. Sgk: an essential convergence point for peptide and steroid hormone regulation of ENaC-mediated Na+ transport. Am J Physiol Cell Physiol 282: C494–C500, 2002.
Fotia AB, Dinudom A, Shearwin KE, Koch JP, Korbmacher C, Cook DI, and Kumar S. The role of individual Nedd4–2 (KIAA0439) WW domains in binding and regulating epithelial sodium channels. FASEB J 17: 70–72, 2003.
Fotia AB, Ekberg J, Adams DJ, Cook DI, Poronnik P, and Kumar S. Regulation of neuronal voltage-gated sodium channels by the ubiquitin-protein ligases Nedd4 and Nedd4–2. J Biol Chem 279: 28930–28935, 2004.
Garty H and Palmer LG. Epithelial sodium channels: function, structure, and regulation. Physiol Rev 77: 359–396, 1997.
Hansson JH, Nelson-Williams C, Suzuki H, Schild L, Shimkets R, Lu Y, Canessa C, Iwasaki T, Rossier B, and Lifton RP. Hypertension caused by a truncated epithelial sodium channel gamma subunit: genetic heterogeneity of Liddle syndrome. Nat Genet 11: 76–82, 1995.
Hanwell D, Ishikawa T, Saleki R, and Rotin D. Trafficking and cell surface stability of the epithelial Na+ channel expressed in epithelial Madin-Darby canine kidney cells. J Biol Chem 277: 9772–9779, 2002.
Helms MN, Fejes-Toth G, and Naray-Fejes-Toth A. Hormone-regulated transepithelial Na+ transport in mammalian CCD cells requires SGK1 expression. Am J Physiol Renal Physiol 284: F480–F487, 2003.
Henke G, Maier G, Wallisch S, Boehmer C, and Lang F. Regulation of the voltage gated K+ channel Kv1.3 by the ubiquitin ligase Nedd4–2 and the serum and glucocorticoid inducible kinase SGK1. J Cell Physiol 199: 194–199, 2004.
Henry PC, Kanelis V, O'Brien MC, Kim B, Gautschi I, Forman-Kay J, Schild L, and Rotin D. Affinity and specificity of interactions between Nedd4 isoforms and the epithelial Na+ channel. J Biol Chem 278: 20019–20028, 2003.
Husted RF, Zhang C, and Stokes JB. Concerted actions of IL-1beta inhibit Na+ absorption and stimulate anion secretion by IMCD cells. Am J Physiol Renal Physiol 275: F946–F954, 1998.
Ingham RJ, Gish G, and Pawson T. The Nedd4 family of E3 ubiquitin ligases: functional diversity within a common modular architecture. Oncogene 23: 1972–1984, 2004.
Ishikawa T, Marunaka Y, and Rotin D. Electrophysiological characterization of the rat epithelial Na+ channel (rENaC) expressed in MDCK cells. Effects of Na+ and Ca2+. J Gen Physiol 111: 825–846, 1998.
Itani OA, Campbell JR, Herrero J, Snyder PM, and Thomas CP. Alternate promoters and variable splicing lead to hNedd4–2 isoforms with a C2 domain and varying number of WW domains. Am J Physiol Renal Physiol 285: F916–F929, 2003.
Itani OA, Cornish KL, Liu KZ, and Thomas CP. Cycloheximide increases glucocorticoid-stimulated -ENaC mRNA in collecting duct cells by p38 MAPK-dependent pathway. Am J Physiol Renal Physiol 284: F778–F787, 2003.
Kamynina E, Debonneville C, Bens M, Vandewalle A, and Staub O. A novel mouse Nedd4 protein suppresses the activity of the epithelial Na+ channel. FASEB J 15: 204–214, 2001.
Kamynina E, Tauxe C, and Staub O. Distinct characteristics of two human Nedd4 proteins with respect to epithelial Na+ channel regulation. Am J Physiol Renal Physiol 281: F469–F477, 2001.
Kanelis V, Rotin D, and Forman-Kay JD. Solution structure of a Nedd4 WW domain-ENaC peptide complex. Nat Struct Biol 8: 407–412, 2001.
Letz B and Korbmacher C. cAMP stimulates CFTR-like Cl– channels and inhibits amiloride-sensitive Na+ channels in mouse CCD cells. Am J Physiol Cell Physiol 272: C657–C666, 1997.
Loffing J, Loffing-Cueni D, Macher A, Hebert SC, Olson B, Knepper MA, Rossier BC, and Kaissling B. Localization of epithelial sodium channel and aquaporin-2 in rabbit kidney cortex. Am J Physiol Renal Physiol 278: F530–F539, 2000.
Nakhoul NL, Hering-Smith KS, Gambala CT, and Hamm LL. Regulation of sodium transport in M-1 cells. Am J Physiol Renal Physiol 275: F998–F1007, 1998.
Nalefski EA, Sultzman LA, Martin DM, Kriz RW, Towler PS, Knopf JL, and Clark JD. Delineation of two functionally distinct domains of cytosolic phospholipase A2, a regulatory Ca2+-dependent lipid-binding domain and a Ca2+-independent catalytic domain. J Biol Chem 269: 18239–18249, 1994.
Naray-Fejes-Toth A, Canessa C, Cleaveland ES, Aldrich G, and Fejes-Toth G. Sgk is an aldosterone-induced kinase in the renal collecting duct: effects on epithelial Na+ channels. J Biol Chem 274: 16973–16978, 1999.
Naray-Fejes-Toth A and Fejes-Toth G. The sgk, an aldosterone-induced gene in mineralocorticoid target cells, regulates the epithelial sodium channel. Kidney Int 57: 1290–1294, 2000.
Naray-Fejes-Toth A, Helms MN, Stokes JB, and Fejes-Toth G. Regulation of sodium transport in mammalian collecting duct cells by aldosterone-induced kinase, SGK1: structure/function studies. Mol Cell Endocrinol 217: 197–202, 2004.
Palmada M, Dieter M, Speil A, Bohmer C, Mack AF, Wagner HJ, Klingel K, Kandolf R, Murer H, Biber J, Closs EI, and Lang F. Regulation of intestinal phosphate cotransporter NaPi IIb by ubiquitin ligase Nedd4–2 and by serum- and glucocorticoid-dependent kinase 1. Am J Physiol Gastrointest Liver Physiol 287: G143–G150, 2004.
Palmer LG and Frindt G. Effects of cell Ca and pH on Na channels from rat cortical collecting tubule. Am J Physiol Renal Fluid Electrolyte Physiol 253: F333–F339, 1987.
Plant PJ, Yeger H, Staub O, Howard P, and Rotin D. The C2 domain of the ubiquitin protein ligase Nedd4 mediates Ca2+-dependent plasma membrane localization. J Biol Chem 272: 32329–32336, 1997.
Rossier BC, Pradervand S, Schild L, and Hummler E. Epithelial sodium channel and the control of sodium balance: interaction between genetic and environmental factors. Annu Rev Physiol 64: 877–897, 2002.
Rotin D, Staub O, and Haguenauer-Tsapis R. Ubiquitination and endocytosis of plasma membrane proteins: role of Nedd4/Rsp5p family of ubiquitin-protein ligases. J Membr Biol 176: 1–17, 2000.
Sayegh R, Auerbach SD, Li X, Loftus RW, Husted RF, Stokes JB, and Thomas CP. Glucocorticoid induction of epithelial sodium channel expression in lung and renal epithelia occurs via trans-activation of a hormone response element in the 5'-flanking region of the human epithelial sodium channel subunit gene. J Biol Chem 274: 12431–12437, 1999.
Schafer JA, Watkins ML, Li L, Herter P, Haxelmans S, and Schlatter E. A simplified method for isolation of large numbers of defined nephron segments. Am J Physiol Renal Physiol 273: F650–F657, 1997.
Shimkets RA, Warnock DG, Bositis CM, Nelson-Williams C, Hansson JH, Schambelan M, Gill JR Jr, Ulick S, Milora RV, Findling JW, Canessa CM, Rossier BC, and Lifton RB.Liddle's syndrome: heritable human hypertension caused by mutations in the subunit of the epithelial sodium channel. Cell 79: 407–414, 1994.
Snyder PM, Olson DR, and Thomas BC. Serum and glucocorticoid-regulated kinase modulates Nedd4–2-mediated inhibition of the epithelial Na+ channel. J Biol Chem 277: 5–8, 2002.
Snyder PM, Steines JC, and Olson DR. Relative contribution of Nedd4 and Nedd4–2 to ENaC regulation in epithelia determined by RNA interference. J Biol Chem 279: 5042–5046, 2004.
Staub O, Dho S, Henry P, Correa J, Ishikawa T, McGlade J, and Rotin D. 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, 1996.
Staub O, Gautschi I, Ishikawa T, Breitschopf K, Ciechanover A, Schild L, and Rotin D. Regulation of stability and function of the epithelial Na+ channel (ENaC) by ubiquitination. EMBO J 16: 6325–6336, 1997.
Thomas J, Deetjen P, Ko WH, Jacobi C, and Leipziger J. P2Y(2) receptor-mediated inhibition of amiloride-sensitive short circuit current in M-1 mouse cortical collecting duct cells. J Membr Biol 183: 115–124, 2001.
van Bemmelen MX, Rougier JS, Gavillet B, Apotheloz F, Daidie D, Tateyama M, Rivolta I, Thomas MA, Kass RS, Staub O, and Abriel H. Cardiac voltage-gated sodium channel Nav1.5 is regulated by Nedd4–2 mediated ubiquitination. Circ Res 95: 284–291, 2004.
Volk KA, Husted RF, Snyder PM, and Stokes JB. Kinase regulation of hENaC mediated through a region in the COOH-terminal portion of the -subunit. Am J Physiol Cell Physiol 278: C1047–C1054, 2000.
Vuagniaux G, Vallet V, Jaeger NF, Hummler E, and Rossier BC. Synergistic activation of ENaC by three membrane-bound channel-activating serine proteases (mCAP1, mCAP2, and mCAP3) and serum- and glucocorticoid-regulated kinase (Sgk1) in Xenopus oocytes. J Gen Physiol 120: 191–201, 2002.
Wulff P, Vallon V, Huang DY, Volkl H, Yu F, Richter K, Jansen M, Schlunz M, Klingel K, Loffing J, Kauselmann G, Bosl MR, Lang F, and Kuhl D. Impaired renal Na+ retention in the sgk1-knockout mouse. J Clin Invest 110: 1263–1268, 2002.(Omar A. Itani, John B. St)
ABSTRACT
Mutations that disrupt a PY motif in epithelial Na+ channel (ENaC) subunits increase surface expression of Na+ channels in the collecting duct, resulting in greater Na+ reabsorption. Nedd4 and Nedd4–2 have been identified as ubiquitin ligases that can interact with ENaC via its PY motifs to regulate channel activity. We recently reported that human Nedd4–2 (hNedd4–2) is expressed as many isoforms because of alternative promoter usage and/or variable splicing. To understand the relevance of hNedd4–2 isoforms for collecting duct Na+ transport, we studied the interaction with ENaC and the intracellular localization and function of the following three naturally occurring hNedd4–2 isoforms: full-length Nedd4–2 (Nedd4–2), Nedd4–2 lacking the NH2-terminal C2 domain (Nedd4–2C2), and Nedd4–2 lacking the C2 domain and WW domains 2 and 3 (Nedd4–2WW2,3). Nedd4–2 and Nedd4–2C2 associate with ENaC and robustly reduce Na+ transport in Xenopus oocytes, whereas the interaction with and functional effect of Nedd4–2WW2,3 on ENaC is weak. Nedd4–2 is expressed in the mouse collecting duct, and overexpression of Nedd4–2 reduces endogenous ENaC activity in a collecting duct cell line. This reduction in ENaC activity can be reversed early with exposure to dexamethasone, an effect that is associated with an increase in sgk1 abundance. The C2 domain is required to target Nedd4–2 to the plasma membrane in response to elevation of intracellular Ca2+ concentration ([Ca2+]i) in MDCK cells, although it does not appear to mediate the inhibitory effect of [Ca2+]i on Na+ transport. Our data illustrate that naturally occurring hNedd4–2 isoforms differentially associate with ENaC to regulate its activity.
epithelial sodium channel
NA+ TRANSPORT IN SOME epithelial cells, such as the connecting tubules and collecting ducts of the kidney, airway and alveolar epithelia of the lung, in the colon, and in sweat glands and salivary glands occurs via the epithelial Na+ channel (ENaC). ENaC present at the apical membrane of transporting epithelia is composed of three subunits (, , and ) that form a heteromultimeric protein complex. This channel permits the passive diffusion of Na+ in cells where its concentration is kept low by extrusion of Na+ through the action of Na+-K+-ATPase at the basolateral membrane. The resulting effect is the reabsorption of Na+ from the lumen into the interstitial space (2, 18, 42).
The sgk and Nedd4/Rsp5 protein families are considered to be important physiological regulators of ENaC activity. Sgk1, a kinase with pleiotropic effects on multiple transporters, has been shown to increase ENaC activity in heterologous expression systems by increasing abundance of surface channels, increasing open probability, and/or activating silent channels (3, 11, 12, 47, 48, 54). Knock down of sgk1 activity, by a dominant-negative sgk1 form or by antisense-mediated decay, results in a substantial reduction of ENaC activity (15, 21, 38). Sgk1-null mice are somewhat inefficient at reabsorbing Na+, leading to a reduction of blood pressure when kept on a low-salt diet (55).
Nedd4/Rsp5 proteins are WW domain-containing E3-type ubiquitin ligases that regulate surface expression of ENaC in heterologous expression systems (50). WW domains of Nedd4/Rsp5 bind to the PY motifs of ENaC, and the COOH-terminal HECT domain catalyzes the transfer of ubiquitin residues to ENaC, leading to channel internalization and reduced Na+ transport. ENaC mutations that disrupt a PY motif abolish the effect of Nedd4/Rsp5. It is these mutations that result in a dominantly inherited form of severe hypertension called Liddle's syndrome, a consequence of increased Na+ reabsorption in the connecting tubules and collecting duct (19, 46).
Nedd4–2, a member of the Nedd4/Rsp5 family, is very effective in downregulating ENaC activity in heterologous expression systems (27, 29). More recently, an RNA interference approach was used to show that knocking down Nedd4–2 expression increases endogenous ENaC activity in a lung epithelial cell line (H441) and increases transfected ENaC activity in a rat thyroid epithelial cell line (FRT; see Ref. 48). Sgk1 phosphorylates Nedd4–2 at three distinct serine/threonine residues, thus reducing the affinity between Nedd4–2 and ENaC (11, 47). Because Nedd4–2 reduces ENaC surface expression, phosphorylation of Nedd4–2 may increase ENaC abundance at the plasma membrane and increase ENaC activity. This indirect effect of sgk1 is thought to be one of the mechanisms by which sgk1 stimulates Na+ transport.
We have previously reported that human Nedd4–2 (hNedd4–2) exists as many isoforms that arise from alternate transcription and translation start sites and from variable splicing of some internal exons (27). Some forms of hNedd4–2 have an NH2-terminal C2 domain, a domain that in other proteins is a Ca2+ and phospholipid-binding domain and may dictate membrane localization (41). Other hNedd4–2 isoforms differ in the number of WW domains and sgk1 phosphorylation sites that are present (27). In an in vitro binding assay, WW domains 3 and 4 appear to have strong affinity for the PY motifs of ENaC, whereas domains 1 and 2 have little or no affinity (27). The varying affinities of the WW domains of Nedd4–2 for ENaC have been described in a number of laboratories (4, 16, 23).
In this report, we examined the expression of Nedd4–2 in the collecting duct of the kidney and studied the interaction with ENaC and the intracellular localization and function of three naturally occurring hNedd4–2 isoforms.
EXPERIMENTAL PROCEDURES
Cell culture. COS-7 cells (gift from M. J. Welsh, University of Iowa) were cultured in DMEM containing 10% FBS and 1% penicillin-streptomycin. M-1 cells were (gift from Géza Fejes-Tóth, Dartmouth Medical School) cultured in DMEM-F-12 containing 10% FBS and 1% penicillin-streptomycin unless otherwise specified. Madin-Darby canine kidney (MDCK)-C7 cells were (gift from B. Blazer-Yost and H. Oberleithner) cultured in MEM containing 10% FBS and 1% penicillin-streptomycin.
Transient transfection. The coding regions of human -, -, and -ENaC tagged at the COOH terminus with Flag, hemagglutinin (HA), and Myc epitopes, respectively, were cloned into pcDNA3 (Invitrogen, Carlsbad, CA), whereas the coding regions of all three hNedd4–2 isoforms where cloned into pcDNA4/HisMax C in frame with the NH2-terminal His and Xpress epitopes (Invitrogen) and into EGFP-N1 in frame and upstream of the enhanced green fluorescent protein (EGFP) open reading frame (Clontech, Palo Alto, CA). COS-7 cells (107 cells) were spun down, resuspended in 400 μl cytomix solution (in mM: 120 KCl, 0.15 CaCl2, 10 K2HPO4, 10 KH2PO4, 25 HEPES, 2 EGTA, 5 MgCl2, 2 Na-ATP, and 5 glutathione), and then mixed with 15 μg plasmids expressing human ENaC (hENaC; 5 μg/subunit) and 20 μg pcDNA4/HisMax C plasmids containing hNedd4–2 isoforms or EGFP-N1 plasmid containing hNedd4–2 isoforms. The cell-DNA mixture was electroporated with Gene Pulser II (Bio-Rad, Hercules, CA) and then seeded on 100-mm culture plates for immunoprecipitation experiments and on chamber slides coated with poly-L-Lysine for immunofluorescence experiments.
MDCK-C7 cells (107 cells) were prepared as above in cytomix solution, transfected by electroporation at 2,850 μF and 220 volts with 20 μg EGFP-N1 plasmids expressing Nedd4–2 isoforms, and then seeded on Millicell PCF filters pretreated with human placental collagen at a density of 2.5 x 104 cells/filter. M-1 cells were also suspended in cytomix solution and transfected by electroporation at 500 μF and 280 volts with 20 μg pcDNA4/HisMax C expressing hNedd4–2C2 or the control plasmid pRLSV40 that expresses Renilla luciferase (Promega, Madison, WI). Transfected cells were then seeded on Millicell PCF filters pretreated with human placental collagen.
Immunofluorescence. Some transfected COS-7 cells were grown on chamber slides (Nunc, Rochester, NY). After transfection (72 cells), cells were fixed with 4% paraformaldehyde and blocked with 5% mouse serum and 1% BSA in PBS followed by incubation with rhodamine-conjugated anti-myc antibody (Santa Cruz, Santa Cruz, CA). Cells were washed with PBS, and nuclei were stained with TO-PRO3 (Molecular Probes, Eugene, OR). Slides were then mounted with VectaShield (Vector Laboratories, Burlingame, CA), and fluorescent signals were detected using a Bio-Rad 1040 confocal microscope (Bio-Rad).
Transfected MDCK cells were grown on chamber slides for 72 h. Cells were then washed with 140 mM NaCl, 6 mM KCl, 1 mM MgCl2, 0.1 mM EDTA, 20 mM glucose, and 20 mM HEPES, pH 7.3, and then incubated in the same media for 5 min in the presence of 2 mM EGTA or 1.1 mM CaCl2 and 1 μM ionomycin (Sigma, St. Louis, MO). Cells were then fixed with 4% paraformaldehyde and washed with PBS. PCF filters were cut out and mounted on slides and visualized as described above.
Western blot analysis and immunoprecipitation. Transfected COS-7 cells and M-1 cells grown in 100-mm plates were lysed using M-PER lysis buffer (Pierce, Rockford, IL), and protein concentration was determined using the Bradford method. Protein lysates were boiled, run on a 7% SDS-PAGE, and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore, Billerica, MA). Membrane was blocked with Tris-buffered saline-Tween 20 (TBST) containing 5% milk then incubated in 5% milk-TBST with an anti-Nedd4–2 antibody (gift from Dr. Howard Pratt), an anti--ENaC antibody (gift from Dr. Mark Knepper), or an anti-SGK1 antibody (Upstate, Charlottesville, VA). Membranes were then washed with TBST and incubated with secondary horseradish peroxidase-conjugated antibody followed by chemiluminescence using Supersignal West Pico chemiluminescent substrate (Pierce). In some instances, membranes were stripped with Restore stripping buffer (Pierce) for 15 min, incubated with 5% milk-TBST, and then reprobed with anti--tubulin antibody (Santa Cruz). Autoradiograms were scanned, and the density of individual bands was measured using Kodak Digital Science Image Analysis software (Rochester, NY) for sgk1 immunoblots. The sgk1 band was normalized for the density of the -tubulin band.
For immunoprecipitation, the ProFound Mammalian HA-Tag IP kit (Pierce) was used according to the supplier's protocol. In brief, transfected COS-7 lysates were incubated with anti-HA antibodies conjugated to agarose beads in the provided columns. Columns were then spun down and washed three times. HA-bound protein complexes were retrieved by elution with loading buffer, boiled, and run on SDS-PAGE followed by Western analysis using the anti-Nedd4–2 antibody.
Xenopus oocyte expression. The hENaC -, -, and -subunits and hNedd4–2 isofroms (Nedd4–2, Nedd4–2C2, and Nedd4–2WW2,3) were subcloned into the pGEM-HE plasmid (53). Linearized plasmids were subjected to in vitro transcription using the mMessage Machine (Ambion, Austin, TX) kit to produce capped cRNA from each construct. The integrity of the cRNAs was evaluated by agarose gel electrophoresis and quantified by densitometry. The cRNAs were diluted in water so that 50-nl injections with a Drummond Nanoject oocyte injector carried 1 ng of each cRNA, unless otherwise indicated.
Mature female Xenopus laevis were purchased from Xenopus I (Dexter, MI) or Nasco (Fort Atkinson, WI). X. laevis were housed in the University of Iowa animal facility in dechlorinated tap water at 18–20°C. Stage V and VI oocytes were removed from toads that were anesthetized in an ice-cold 2 mg/ml tricaine solution. The oocytes were defolliculated with collagenase and stored overnight at 18°C in frog Ringer solution (NFR) consisting of 115 mM NaCl, 2.5 mM KCl, 1.8 mM CaCl2, 10 mM HEPES, 5 mM sodium pyruvate, and 100 U/ml penicillin-streptomycin (pH 7.35). After 12–24 h of recovery from the collagenase treatment, healthy oocytes were injected with cRNA. Ringer solution was changed daily. Whole cell hENaC currents were measured 48–72 h after cRNA injections. Measurements were made in NFR using an OC-725C oocyte voltage-clamp amplifier (Warner Instruments, Hamden, CT). The pCLAMP software suite (Axon Instruments, Union City, CA) was used for amplifier control and data collection/analyses. All recordings were performed at room temperature. Amiloride-sensitive currents were derived by subtracting currents recorded in 10 μM amiloride from preamiloride currents.
After injection (48 h), 20 oocytes/injection group were chilled to 4°C in 200 μl NFR. The oocytes were ruptured with 15 strokes in a small Teflon-glass dounce homogenizer. The resulting slurry was centrifuged at 1,000 g for 5 min to separate yolk, lipids, and other large cellular debris. The lipid layer at the top of the tube was wicked off using a cotton swab, and the supernatant was centrifuged again at 1,000 g for 5 min. An equal volume of NFR + 2% SDS + protease inhibitor cocktail (Roche, Indianapolis, IN) was added to the final supernatant, and the mixture was sheared with 20 passes through a 20-gauge needle. Each sample (10 μl) was then boiled and loaded on an SDS-PAGE gel and transferred to PVDF membranes for Nedd4–2 immunoblotting.
Microdissection and RT-PCR. C57BL/6J mice housed in the University of Iowa animal facility were killed, and kidneys were isolated. Animal experiments were approved by the University of Iowa Institutional Animal Care and Use Committee. Kidney cortex was dissected with a blade and cut into fine pieces that were then incubated in MEM medium containing 0.5 mg/ml type 2 collagenase, 5 mM glycine, 50 U/ml DNase, and 48 μg/ml soybean trypsin inhibitor. Kidney cortex pieces were incubated in the MEM collagenase solution until the medium became cloudy. Collecting duct segments were then picked with a needle with the aid of Olympus stereoscope (Olympus, Melville, NY), as previously described (45). RNA was extracted using TRI reagent (Molecular Research center, Cincinnati, OH) following the supplier's protocol.
RNA samples were then digested with DNase (Promega) and reverse transcribed using SuperScript II (Invitrogen) according to the supplier's protocol. The resulting cDNA was used to perform PCR with AmpliTaq DNA polymerase (Roche) using the following primers: mVPR2, 5'-TTTCGTCCCCTAGCTCTCC-3' and 5'-ATACCCCACTGCCATTTCC-3'; mSGLT1, 5'-GACATCCCAGAGGACTCCAA-3' and 5'-ACCACTGTCCTCCACAAAGG-3'; and mNedd4–2, 5'-TCTCAACTGGTTGCCGTGTA-3' and 5'-AAAGCTGAAAATCAGTCTAAATCA-3'.
RNase protection assay. RNA was extracted from mouse kidney cortex and medulla and from M-1 cells using TRI reagent. The hNedd4–2 template has been previously described (27). The two mouse RNase protection assay (RPA) templates were generated by RT-PCR from M-1 cells. In brief, RNA from M-1 cells was extracted and reverse transcribed. The resulting cDNA was used to amplify two PCR fragments using each of the following primer pairs: mN4–2F2, 5'-TCTAGACCAGCCTTCCTCTCC-3'; mN4–2R2, 5'-CAGGAAGTCGTCTCGTGTCA-3'; and mN4–2F4, 5'-AGTTCATGTGGGGGACAAAG-3'; and mN4–2R4, 5'-TAGGGTCTCTCCATGGTTGG-3'. The resulting PCR products were gel purified and cloned into PCRXl-Topo (Invitrogen). Templates containing the cloned hNedd4–2 and mouse Nedd4–2 (mNedd4–2) cDNA fragments and an 18S rRNA cDNA fragment (pTR1 RNA 18S; Ambion) were used to generate radiolabeled antisense cRNA probes in the presence of [-32P]UTP (PerkinElmer, Wellesley, MA), as previously described (28). RNA samples were cohybridized overnight with hNedd4–2 or mNedd4–2 and 18S cRNAs and then digested with RNase A and T1 and nuclease-protected fragments resolved by PAGE.
Generation of stable cell lines. M-1 cells were transfected by electroporation with sLacR.hyg that expresses the Lac repressor (gift from Dr. Bishop, University of Iowa). Cells were then grown in DMEM-F-12 medium containing 200 μg/ml hygromycin B (Invitrogen). Individual hygromycin-resistant colonies were grown and screened for expression of LacR. The highest expressing clone "S25" was further transfected with popRSV5.neo plasmid that expresses Nedd4–2 in the presence of LacR and isopropyl thiogalactopyranoside (IPTG). Cells were then grown in DMEM-F-12-hygro in the presence of 250 μg/ml geneticin (Invitrogen). Individual geneticin-resistant colonies were isolated, grown, and screened for Nedd4–2 expression.
Short-circuit current measurements. M-1 cells seeded on Millicell PCF filters were grown for 2 days in DMEM-Ham's F-12 supplemented with insulin (5 μg/ml), transferrin (5 μg/ml), triiodothyronine (5 nM), hydrocortisone (50 nM), sodium selenite (10 nM), gentamicin (50 μg/ml), BSA (10 g/l), and dexamethasone (5 nM). The monolayers were then switched to the same medium without albumin and steroids for 1 day. Measurements of transepithelial voltage and resistance and short-circuit current (Isc) were conducted at 37°C in Ussing chambers as described (44).
The short-term time-course measurements were conducted in a separate set of chambers designed to accommodate Millicell filter-bottom cups bathed in a Krebs-Ringer bicarbonate solution containing (in mM) 115 NaCl, 25 NaHCO3, 5 KCl, 5 Na-HEPES, 5 H-HEPES, 1.5 CaCl2, 1 MgCl2, 1 Na2HPO4, and 5 D-glucose. The chambers were water-jacketed to 37°C, and electrical measurements were made with a University of Iowa voltage clamp, as previously described (24). HCO3–-containing solutions were gassed with 5% CO2 in air to maintain pH at 7.4.
Adenoviral transduction. Nedd4–2 and Nedd4–2C2 with NH2-terminal Xpress tags were cloned into a shuttle vector, pacAd5CMV K-N pA, downstream of the CMV promoter. The resulting plasmids were used to generate Nedd4–2- and Nedd4–2C2-expressing adenoviral constructs by the Vector core at the University of Iowa. For adenoviral transduction, M-1 cells seeded on Millicell PCF filters were grown for 3 days followed by the addition of Nedd4–2, Nedd4–2C2, or empty virus (8 x 107 plaque-forming units/filter) to the apical medium for 4 h. Isc measurements were taken 48 h after transduction with cell cultures switched to medium containing 100 nM dexamethasone for the last 24 h. To study the effect of elevation of intracellular Ca2+ concentration ([Ca2+]i) on Isc, short-term time-course experiments were performed as described earlier, adding ionomycin (200 nM) to the basolateral medium, and the effect was followed for 10 min. Benzamil (10 μM) was then added to the apical medium to obtain benzamil-sensitive Isc.
RESULTS
Nedd4–2 exists as many isoforms. We have previously shown that, because of alternative exon usage, hNedd4–2 exists as many transcripts that are predicted to be translated into proteins that either have or lack an NH2-terminal C2 domain (27). Further diversity arises from alternative splicing of exons 12–15 that are predicted to lead to Nedd4–2 proteins with a variable number of WW domains and sgk1 phosphorylation sites (Fig. 1). We have also shown, by RT-PCR and RPA, that Nedd4–2 isoforms are expressed in a tissue-specific manner (27). We selected three different Nedd4–2 isoforms that are abundantly expressed [full-length Nedd4–2, Nedd4–2 lacking the C2 domain (Nedd4–2C2), and Nedd4–2 lacking WW2, WW3, and the C2 domain (Nedd4–2WW2,3)], tested the effect of these isoforms on ENaC activity, and studied their cytosolic localization.
Nedd4–2 and ENaC show cytoplasmic distribution. To study the intracellular distribution of Nedd4–2 isoforms and ENaC, we transfected COS-7 cells with plasmids expressing all three ENaC subunits (, , and ) together with EGFP, or Nedd4–2, Nedd4–2C2 or Nedd4–2WW2,3. -ENaC is myc-tagged, allowing detection with a Rhodamine-conjugated anti-myc antibody, whereas all Nedd4–2 isoforms are EGFP-tagged. Our results show that EGFP is predominantly localized to the nucleus, whereas ENaC exhibits an exclusive diffuse cytoplasmic distribution (Fig. 2A). Expression of Nedd4–2 isoforms as an EGFP fusion protein results in a cytoplasmic EGFP signal, suggesting that Nedd4–2 contains a nuclear export signal. Alternatively, EGFP may have a nuclear localization signal that is masked when expressed as a COOH-terminal fusion protein (Fig. 2, B-D). We could not detect preferential expression of the Nedd4–2 isoforms at the cell membrane, nor did we detect differences in expression patterns between Nedd4–2 isoforms, suggesting that, at least in COS-7 cells, these proteins are similarly distributed. We also noted that, despite its known cell membrane expression and function, most of the detectable ENaC is cytosolic in this expression system. Cytosolic distribution of transfected and native ENaC has been previously reported in native tissues and cultured cells, suggesting that only a small fraction of total ENaC is present at the plasma membrane (20, 33). Interestingly, cells expressing ENaC appear more spherical with less cytoplasmic extensions, suggesting that some fraction of these channels are functional with increased Na+ absorption and subsequent change in cell shape (Fig. 2B).
Nedd4–2 interacts with ENaC. To determine if Nedd4–2 associates with ENaC, we performed immunoprecipitation experiments in COS-7 cells transfected with three ENaC subunits (, , and ) and with Xpress-tagged Nedd4–2, Nedd4–2C2, or Nedd4–2WW2,3. The anti-Nedd4–2 antibody used for immunoblotting detects transfected Nedd4–2 and appears to detect endogenous Nedd4. The amount of expressed Nedd4–2 varied somewhat between each isoform and varied significantly when ENaC was coexpressed. The COS-7 cell lysates were normalized to contain equal amounts of the transfected Nedd4–2 isoforms before immunoprecipitation. On Western blot analysis of normalized cell lysates, each of the transfected tagged Nedd4–2 isoforms migrate at the predicted size as follows: Nedd4–2 and Nedd4–2WW2,3 migrate at 116 and 83 kDa and Nedd4–2C2 migrates at 101 kDa, almost superimposed on endogenous Nedd4 at 104 kDa (Fig. 3A). Cell lysates were immunoprecipitated with an anti-HA antibody that recognizes the tagged -ENaC. When expressed together with -, -, and -ENaC, each of the Nedd4–2 isoforms coimmunoprecipitate with ENaC (Fig. 3B). Although we could not detect a difference in levels of immunoprecipitatable Nedd4–2 and Nedd4–2C2, Nedd4–2WW2,3 is immunoprecipitated to a much lesser extent with ENaC. These results suggest that WW2 and/or WW3 domains are important for the association of Nedd4–2 with ENaC, whereas the C2 domain of Nedd4–2 may not influence this association.
Nedd4–2WW2,3 is not effective in reducing ENaC-dependent Na+ transport. To study the effect of the three Nedd4–2 isoforms on Na+ transport, we expressed ENaC with or without Nedd4–2 in Xenopus oocytes followed by whole cell voltage clamp and amiloride-sensitive current measurements. We also studied the level of Nedd4–2 proteins in injected oocytes by Western analysis. We show that both Nedd4–2 and Nedd4–2C2 robustly reduce ENaC current. Importantly, the effect of Nedd4–2WW2,3 is small and did not reach statistical significance (Fig. 4A). The Western blot results show that the difference in effectiveness of Nedd4–2WW2,3 protein cannot be explained by differences in abundance of expressed isoforms, since Nedd4–2 is expressed at a relatively lower level than Nedd4–2WW2,3 (Fig. 4B). The expression of Nedd4–2C2 is more than Nedd4–2 and may explain the small difference between them in the inhibition of Na+ transport. We then asked if the effect of Nedd4–2WW2,3 on ENaC could be enhanced by expressing it in excess. Our results show that, despite expression at up to 10-fold excess, there was no statistically significant effect of Nedd4–2WW2,3 on ENaC-mediated currents (Fig. 4, C and D).
Because Nedd4–2WW2,3 binds to ENaC (albeit with lower affinity than Nedd4–2) and because it has a small effect on ENaC, we asked whether overexpression of Nedd4–2WW2,3 could compete with Nedd4–2C2 for its effect on ENaC, thus acting in a dominant-negative fashion. To address this issue, we expressed Nedd4–2C2 with 20-fold excess of Nedd4–2WW2,3, followed by current measurements. Our results show that Nedd4–2WW2,3 does not interfere with the ability of Nedd4–2C2 to reduce ENaC-dependent current (Fig. 5). Collectively, our results show that both Nedd4–2 and Nedd4–2C2 can robustly interact with and downregulate ENaC activity in a reconstituted system, whereas the effect of Nedd4–2WW2,3 is weak. These results suggest that Nedd4–2WW2,3 may have a marginal role in the regulation of ENaC activity.
Nedd4–2 is expressed in the kidney collecting duct. Our studies so far indicate that ENaC and Nedd4–2 can interact in heterologous expression systems. We next wanted to determine if Nedd4–2 is expressed and able to regulate ENaC in the collecting duct. We performed RT-PCR on microdissected collecting duct samples of mouse kidney. To confirm collecting duct lineage and to monitor for contaminating proximal tubular epithelium in the microdissected samples, we also studied the expression of the vasopressin 2 receptor (VPR2), a collecting duct-specific transcript, and the sodium glucose transporter (SGLT1), a proximal tubule-specific transcript. Our results indicate that Nedd4–2 and VPR2 are expressed in the mouse collecting duct (Fig. 6A).
We have previously shown that, because of alternative promoter usage, hNedd4–2 exists as isoforms that either have or lack a C2 domain (27). To study if a C2-containing Nedd4–2 is also present in the mouse transcriptome, we first examined all known cDNA sequences that map to the mNedd4–2 locus in the Ensembl and EST databases. As shown in Fig. 6B, one form of mNedd4–2 (accession no. AF277232) initiates transcription in exon 1a, with the open reading frame beginning in exon 6, which leads to translation of a protein without a C2 domain. Interestingly, we found a short cDNA sequence (accession no. AK040106) that extends to exon 4 and is predicted to arise from transcription that initiates in a new exon, 1c. Exon 1c contains a translation initiation codon that is in frame with the downstream initiation codon in exon 6, resulting in the translation of a protein that contains a C2 domain (Fig. 6B).
To study the relative abundance of these two mNedd4–2 isoforms in the mouse kidney and in a collecting duct epithelial cell line, we performed RPA on samples from mouse kidney cortex and medulla and from M-1 cells using two different cRNA probes. The first probe is antisense to sequences in exons 1c through exon 6 and will detect two bands, with the larger corresponding to transcripts that contain exons 1c through 6 (plus C2 form) and the smaller band corresponding to transcripts that contain exons 2 through 6, but not exon 1c (minus C2 form; Fig. 6C). The second probe is antisense to sequences in exons 1a through exon 5 and will also detect two bands. The upper band corresponds to transcripts that contain exons 1a through 5 (minus C2 form), and the lower band corresponds to transcripts that contain exons 2 through 5 (plus C2 form; Fig. 6D). Although expression of the two Nedd4–2 isoforms is equivalent in the kidney cortex and medulla, the C2-containing form is expressed at a higher level in M-1 cells. This suggests that Nedd4–2 isoforms may be differentially regulated in the mouse collecting duct.
Nedd4–2C2 reduces Na+ transport in collecting duct epithelia. To study the effect of Nedd4–2 on ENaC activity in the collecting duct, we transiently expressed Nedd4–2C2, Nedd4–2WW2,3, or an irrelevant protein, Renilla luciferase, in M-1 cells. Based on analysis of transfected green fluorescent protein, we routinely achieve a transfection efficiency of >50% (data not shown). We show that Nedd4–2C2 significantly reduces ENaC-dependent Na+ transport in M-1 cells 72 h after transfection (Fig. 7A), whereas Nedd4–2WW2,3 has no effect (Fig. 7B). We also show that the effect of Nedd4–2C2 is dose-dependent and correlates with the level of expression of transfected Nedd4–2C2 (Fig. 7, C and D). Our results indicate that Nedd4–2C2 can downregulate endogenous ENaC activity in a collecting duct cell line. Because Nedd4–2C2 is localized throughout the cytoplasm, it may regulate ENaC activity by reducing the insertion rate of newly synthesized or recycling channels. Alternatively, enough Nedd4–2C2 is present at the apical membrane, leading to increased retrieval rate of active channels.
Short-term glucocorticoid treatment abolishes the Nedd4–2 effect on Na+ transport. To study the effect of glucocorticoid treatment on Nedd4–2 in the collecting duct, we generated a stable M-1 cell clone that overexpresses Nedd4–2. We transfected M-1 cells with LacR plasmid that constitutively expresses the Lac repressor. One clone, S25, that expressed the highest level of LacR was selected for the second transfection and served as a control for subsequent studies. S25 cells were transfected with a plasmid that expresses Nedd4–2 under the control of an IPTG-inducible promoter. Clones with inducible Nedd4–2 expression were isolated and exhibited a Nedd4–2-dependent decrease in Na+ transport (data not shown). One clone, N1n-4, that expressed the highest levels of Nedd4–2 even in the absence of IPTG (Fig. 8A) was selected for further studies. Basal Isc in N1n-4 is very low compared with S25 cells (0.66 ± 0.19 vs. 2.76 ± 0.19 μA/cm2). The level of immunoblottable -ENaC is not affected in response to overexpression of Nedd4–2 in M-1 cells, suggesting that Nedd4–2 may interact with a small pool of total ENaC (Fig. 8B).
To study the effect of glucocorticoids on Na+ transport, we tested dexamethasone in N1n-4 and S25 cells. Dexamethasone appeared to stimulate Na+ transport in N1n-4 by 1 h, such that both cell lines have matching current over the next few hours. At 6 h, the cell lines diverge again, with N1n-4 now showing reduced Na+ transport compared with S25 (Fig. 8C). These results are consistent with a model where dexamethasone stimulates sgk1 to abolish the early but not the late inhibitory effect of Nedd4–2 on ENaC. To understand the biphasic effect of dexamethasone on Isc in N1n-4, we performed Western blot analysis for sgk1 in M-1 samples treated with dexamethasone for varying time periods (Fig. 8D). Our results show that dexamethasone increases sgk1 levels within 1 h of treatment, with a continued increase for 4 h. The levels of sgk1 then decline and reach basal levels by 24 h of dexamethasone treatment (Fig. 8E). The transient effect of dexamethasone on sgk1 abundance provides an attractive model for the dexamethasone effect on Na+ transport in N1n-4 cells. In this model, sgk1 levels are elevated after dexamethasone treatment (1–4 h), thus inhibiting the effect of Nedd4–2 on ENaC activity; however, sgk1 elevation is transient, thus restoring the effect of Nedd4–2 on ENaC after 6 h. Presumably, the increased Isc at later time points in N1n-4 is the result of increased synthesis of ENaC subunits in response to dexamethasone treatment (14).
The C2 domain is a [Ca2+]i-regulated membrane-targeting sequence. To determine if the C2 domain affects trafficking of Nedd4–2, we transfected MDCK cells grown on permeable supports with EGFP-tagged Nedd4–2 with and without a C2 domain. Our results indicate that Nedd4–2 isoforms have a cytoplasmic distribution in the absence of extracellular Ca2+ (Fig. 9, A and C). Treating the cells with ionomycin, to elevate [Ca2+]i, targets the C2 domain-containing form but not the C2-less form to the plasma membrane (Fig. 9, B and D). The Z-section image series spanning the whole height of cells shows that Nedd4–2 is distributed throughout the plasma membrane (data not shown) and not specifically localized to the apical membrane. This study indicates that the C2 domain of Nedd4–2, like the C2 domain of Nedd4, phospholipase A2, and synaptotagmin, directs Nedd4–2 to the plasma membrane in response to elevation in [Ca2+]i (6, 35, 41).
Elevation of [Ca2+]i reduced Na+ transport in collecting duct epithelia. To determine if elevation of [Ca2+]i affects Isc in M-1 cells, we measured Isc in M-1 cells after treatment with 200 nM ionomycin. Ionomycin stimulates a transient peak of Isc followed by a new plateau, both of which are benzamil-insensitive, suggesting that the rise in Isc is the result of Cl– secretion (Fig. 10A). Indeed, removal of extracellular Cl– inhibited the rise in Isc (data not shown). To determine the effect of ionomycin on ENaC-dependent current, we added benzamil 10 min after ionomycin treatment when the current had reached a new stable baseline. The benzamil-sensitive Isc, after ionomycin, was then compared with the benzamil-sensitive Isc of control M-1 cells not treated with ionomycin (Fig. 10, A and B). Our results show that elevation of [Ca2+]i by ionomycin results in a significant reduction in ENaC activity (Fig. 10B).
To study the effect of ionomycin in cells overexpressing Nedd4–2, we transduced M-1 cells with adenoviral constructs that express Nedd4–2 and Nedd4–2C2. As expected from our data in Figs. 7 and 8, benzamil-sensitive Isc is reduced in Nedd4–2-transduced cells compared with cells transduced with an empty virus. As observed in Fig. 4, Nedd4–2C2 is at least as effective as Nedd4–2 in reducing current. A further decrease in Isc occurs after treatment with ionomycin (Fig. 10B). However, the fractional reduction in Isc was not different between control and each of the Nedd4–2 transduced cells, suggesting that overexpression of Nedd4–2 has no apparent effect on [Ca2+]i-inhibited Isc, at least in a collecting duct cell line with native ENaC (Fig. 10C). Collectively, these data would suggest that Nedd4–2 regulates ENaC within the cytoplasmic pool, rather than at the cell surface, to reduce Na+ transport. Alternatively, elevation of [Ca2+]i leads to the transfer of native Nedd4–2 to the plasma membrane, resulting in inhibition of ENaC at the cell surface that cannot be amplified by adding more Nedd4–2.
DISCUSSION
Nedd4–2 is a recently identified protein that belongs to the Nedd4/Rsp5 family of E3-type ubiquitin ligases. These proteins are present from yeast to human, are widely expressed in mammalian tissues and cells, and are involved in a wide variety of cellular functions. These functions include regulation of the cell cycle, intracellular trafficking, and surface expression of many channels and transporters (25, 43). All members of this family share the same WW and HECT domains, with some members having an NH2-terminal C2 domain as well. The C2 domain is a Ca2+ and phospholipid-binding domain, and the catalytic HECT domain in the COOH-terminus transfers ubiquitin to lysine residues on target proteins. The C2 and HECT domains are separated by a number of WW domains, which are 40 amino acids in length and contain two conserved tryptophan (W) residues. WW domains are well-characterized binding partners for PY motifs (usually consisting of PPXY). Such PY motifs are present at the COOH-terminus of all three ENaC subunits and are important for the regulatory effect of Nedd4–2 on ENaC activity (1, 31, 49).
We have previously demonstrated that hNedd4–2 exists in different isoforms that are, in part, the result of transcription initiation in different exons (27). Some of these exons have an alternate translation start codon leading to a protein with a C2 domain, whereas other exons lack this upstream initiation codon, resulting in an hNedd4–2 isoform lacking the NH2-terminal C2 domain. Other hNedd4–2 variants are the result of internal splicing, which leads to isoforms missing at least one of four WW domains. When tested individually in vitro, WW domains 1 to 4 of Nedd4–2 have differing affinities for the PY motifs of ENaC. We and others have shown that WW3 has the highest affinity followed by WW4, whereas WW1 and WW2 have very weak affinity (4, 16, 23, 27).
We have previously shown that an hNedd4–2 variant lacking WW2 and WW3 (Nedd4–2WW2,3) is a naturally occurring isoform that arises when exons 12 through 15 are spliced out (27). Here we show that Nedd4–2WW2,3 has a substantially lower affinity for ENaC than full-length Nedd4–2 in transfected COS-7 cells. We also report that Nedd4–2WW2,3 does not significantly regulate ENaC activity when reconstituted in Xenopus oocytes or in M-1 cells, suggesting that Nedd4–2WW2,3 may not have a role in regulating ENaC activity in epithelia where it is normally expressed. We speculate that Nedd4–2WW2,3 may regulate the activity of other target proteins where the affinity of individual WW domains for the target PY domain may be different than that for ENaC. A number of new Nedd4–2 targets have been recently described, including the cardiac voltage-gated Na+ channel Nav1.5 (52), the glucose transporter SGLT1 (13), the neuronal voltage-gated Na+ channels (Nav1.2, Nav1.7, and Nav1.8; see Ref. 17), intestinal phosphate cotransporter NaPi IIb (39), the voltage gated K+ channel Kv1.3 (22), and the glutamate transporter EAAT1 (5). Some of these channels (Nav1.2, Nav1.5, Nav1.7, and Nav1.8) have PY motifs, and preliminary studies show that the PY motif of Nav1.5 has a high affinity for WW4 of Nedd4–2 and no measurable affinity for WW2 and WW3 (52), making Nav1.5 a candidate that may be regulated by Nedd4–2WW2,3. Many of the other Nedd4–2-regulated transporters do not have PY motifs, and the effect of Nedd4–2 may be mediated by PY-containing intermediates.
Previous reports demonstrate that Nedd4–2 regulates ENaC activity in heterologous expression systems and in a lung epithelial cell line (27, 30, 47, 48). In this paper, we demonstrate that Nedd4–2 is expressed in the mouse collecting duct and in a mouse collecting duct cell line and confirm that, similar to hNedd4–2, mNedd4–2 has 5'-variants with transcription starting from at least two different exons, leading to proteins with and without a C2 domain. Using three independent approaches (transient transfection, stable transfection, and viral transduction), we show that Nedd4–2 overexpression in M-1 cells inhibits ENaC activity.
The mineralocorticoid aldosterone stimulates ENaC activity in the collecting duct, and this regulation is associated with the transcription of aldosterone-responsive genes sgk1 and the -subunit of ENaC. Sgk1 synthesis begins early (<1 h) after exposure to aldosterone or glucocorticoids; typically, this sgk1 synthesis level is not maintained and returns to baseline after prolonged (>4 h) aldosterone treatment (8, 36, 37). -ENaC exhibits a different response pattern, where increased synthesis does not occur until later (>4 h; see Ref. 14). This suggests that at least a portion of the early effect of aldosterone on increasing ENaC activity may be mediated by sgk1, whereas the later effect is the result of increased channel synthesis. Sgk1 is thought to increase ENaC activity by many mechanisms. One mechanism involves the phosphorylation of Nedd4–2, thereby reducing its affinity to ENaC (11, 47). Consistent with this model, we show that treatment with dexamethasone for 1 h abolishes the Nedd4–2 effect on ENaC activity in M-1 cells overexpressing Nedd4–2. Moreover, the effect of Nedd4–2 on ENaC activity in those cells correlates with sgk1 expression where Nedd4–2 is inactivated when sgk1 levels are elevated and reactivated when sgk1 levels are reduced after prolonged dexamethasone treatment.
We have previously demonstrated that hNedd4–2 forms with and without the C2 domain are both efficient at downregulating ENaC activity when reconstituted in FRT epithelia (27). Earlier studies using Nedd4 protein demonstrated that the C2 domain is important for targeting transfected Nedd4 to the plasma membrane of MDCK cells in the presence of elevated [Ca2+]i (41). Yet Nedd4 without a C2 domain is more efficient at inhibiting ENaC in heterologous expression systems (30). In the current study, Nedd4–2C2 is at least as efficient as Nedd4–2 (if not more) in regulating ENaC activity in Xenopus oocytes and in M-1 cells (Figs. 4 and 10). Like Nedd4, we show that, in transfected cells, Nedd4–2 has a predominant cytoplasmic distribution and that the C2 domain is necessary for targeting Nedd4–2 to the plasma membrane when [Ca2+]i is elevated. This suggests that Nedd4–2 might interact with and regulate the cytoplasmic pool of ENaC rather than a membrane pool of ENaC, thus reducing the number of channels available to be inserted in the apical membrane.
We used ionomycin to investigate the effect of [Ca2+]i elevation on ion transport in M-1 cells. Ionomycin resulted in a biphasic increase in Cl– secretion, with the first phase characterized by an immediate and transient peak of Isc; during the second phase, Cl– secretion reached a plateau that was higher than basal Cl– secretion (Fig. 10). cAMP, ATP, and norepinephrine have been reported to induce a similar Cl– secretion profile in M-1 cells, and at least the ATP effect is mediated by an increase in [Ca2+]i (9, 10, 32, 51). In our experiments, overexpression of Nedd4–2 did not affect Cl– secretion, suggesting that Nedd4–2 does not regulate the [Ca2+]i-stimulated Cl– secretion in M-1 cells (data not shown).
Earlier studies in other models of the collecting duct demonstrated that elevation of [Ca2+]i reduces ENaC activity, although at least one prior study was unable to confirm this effect in M-1 cells (7, 26, 34, 40). The mechanism for the inhibitory effect on Na+ transport is unknown, but an attractive hypothesis is that elevated [Ca2+]i targets the C2-containing form of Nedd4–2 to the plasma membrane where it can interact with ENaC, leading to channel internalization. Our studies confirmed that elevation of [Ca2+]i reduces ENaC activity in M-1 cells; however, overexpression of Nedd4–2 did not result in an enhancement of the Ca2+ effect. There are several possible explanations for this. The first is that the maximal effect of [Ca2+]i on ENaC activity is achieved by targeting endogenous Nedd4–2 to the apical membrane and that this cannot be enhanced by overexpression. Alternatively, the [Ca2+]i effect on Na+ transport may be independent of Nedd4–2.
In conclusion, we demonstrate that Nedd4–2 isoforms are expressed in the kidney collecting duct, and Nedd4–2 overexpression reduces native ENaC activity in a collecting duct cell line. This effect of Nedd4–2 is transiently abolished by exposure to glucocorticoids, which may be mediated by stimulation of sgk1. We also demonstrate that Nedd4–2 isoforms that contain the C2 domain are localized to the plasma membrane in response to elevated [Ca2+]i, although [Ca2+]i-induced reductions in ENaC activity do not seem to be dependent on having a Nedd4–2 with a C2 domain. Nedd4–2 isoforms missing WW2 and WW3 do not appear to regulate ENaC activity, and they may function to regulate other, as yet unknown, target proteins.
GRANTS
This work was supported in part by National Institutes of Health Grants DK-54348 and HL-71664 (to C. P. Thomas) and DK-52617 (to J. B. Stokes) and by Veterans Affairs Merit Review Awards to C. P. Thomas and J. B. Stokes. C. P. Thomas is an Established Investigator of the American Heart Association. O. A. Itani is the recipient of a Predoctoral fellowship from the American Heart Association, Heartland Affiliate.
ACKNOWLEDGMENTS
We thank Russell Husted, Rita Sigmund, Kenneth Volk, Kang Liu, Patrick Wright, and Thomas Moninger for excellent technical support and the University of Iowa DNA core, vector core and microscopy facility for services provided. We thank Dr. Gail Bishop for the LacR.hyg and popRSV5.neo plasmids, Dr. Mark Knepper for anti--ENaC antibody, Dr. Howard Pratt for the anti-Nedd4–2 antibody, and Dr. Jim Schafer for help with microdissection experiments.
The nucleotide sequence reported in this paper will appear in DNA Data Bank of Japan, European Molecular Biology Laboratory, GenBank, and Genome Sequence Database Nucleotide Sequence Databases with the accession number AY751751.
FOOTNOTES
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
REFERENCES
Abriel H, Loffing J, Rebhun JF, Pratt JH, Schild L, Horisberger JD, Rotin D, and Staub O. Defective regulation of the epithelial Na+ channel by Nedd4 in Liddle's syndrome. J Clin Invest 103: 667–673, 1999.
Alvarez de la Rosa D, Canessa CM, Fyfe GK, and Zhang P. Structure and regulation of amiloride-sensitive sodium channels. Annu Rev Physiol 62: 573–594, 2000.
Alvarez de la Rosa D, Zhang P, Naray-Fejes-Toth A, Fejes-Toth G, and Canessa CM. The serum and glucocorticoid kinase sgk increases the abundance of epithelial sodium channels in the plasma membrane of Xenopus oocytes. J Biol Chem 274: 37834–37839, 1999.
Asher C, Sinha I, and Garty H. Characterization of the interactions between Nedd4–2, ENaC, and sgk-1 using surface plasmon resonance. Biochim Biophys Acta 1612: 59–64, 2003.
Boehmer C, Henke G, Schniepp R, Palmada M, Rothstein JD, Broer S, and Lang F. Regulation of the glutamate transporter EAAT1 by the ubiquitin ligase Nedd4–2 and the serum and glucocorticoid-inducible kinase isoforms SGK1/3 and protein kinase B. J Neurochem 86: 1181–1188, 2003.
Chapman ER and Jahn R. Calcium-dependent interaction of the cytoplasmic region of synaptotagmin with membranes. Autonomous function of a single C2-homologous domain. J Biol Chem 269: 5735–5741, 1994.
Chase HS Jr and Al-Awqati Q. Regulation of the sodium permeability of the luminal border of toad bladder by intracellular sodium and calcium: role of sodium-calcium exchange in the basolateral membrane. J Gen Physiol 77: 693–712, 1981.
Chen SY, Bhargava A, Mastroberardino L, Meijer OC, Wang J, Buse P, Firestone GL, Verrey F, and Pearce D. Epithelial sodium channel regulated by aldosterone-induced protein sgk. Proc Natl Acad Sci USA 96: 2514–2519, 1999.
Cuffe JE, Bielfeld-Ackermann A, Thomas J, Leipziger J, and Korbmacher C. ATP stimulates Cl– secretion and reduces amiloride-sensitive Na+ absorption in M-1 mouse cortical collecting duct cells. J Physiol 524: 77–90, 2000.
Cuffe JE, Howard DP, Bertog M, and Korbmacher C. Basolateral adrenoceptor activation mediates noradrenaline-induced Cl– secretion in M-1 mouse cortical collecting duct cells. Pflügers Arch 445: 381–389, 2002.
Debonneville C, Flores SY, Kamynina E, Plant PJ, Tauxe C, Thomas MA, Munster C, Chraibi A, Pratt JH, Horisberger JD, Pearce D, Loffing J, and Staub O. Phosphorylation of Nedd4–2 by Sgk1 regulates epithelial Na+ channel cell surface expression. EMBO J 20: 7052–7059, 2001.
Diakov A and Korbmacher C. A novel pathway of ENaC activation involves an SGK1 consensus motif in the C-terminus of the channel's -subunit. J Biol Chem 279: 38134-38142, 2004.
Dieter M, Palmada M, Rajamanickam J, Aydin A, Busjahn A, Boehmer C, Luft FC, and Lang F. Regulation of glucose transporter SGLT1 by ubiquitin ligase Nedd4–2 and kinases SGK1, SGK3, and PKB. Obes Res 12: 862–870, 2004.
Dijkink L, Hartog A, Deen PM, van Os CH, and Bindels RJ. Time-dependent regulation by aldosterone of the amiloride-sensitive Na+ channel in rabbit kidney. Pflügers Arch 438: 354–360, 1999.
Faletti CJ, Perrotti N, Taylor SI, and Blazer-Yost BL. Sgk: an essential convergence point for peptide and steroid hormone regulation of ENaC-mediated Na+ transport. Am J Physiol Cell Physiol 282: C494–C500, 2002.
Fotia AB, Dinudom A, Shearwin KE, Koch JP, Korbmacher C, Cook DI, and Kumar S. The role of individual Nedd4–2 (KIAA0439) WW domains in binding and regulating epithelial sodium channels. FASEB J 17: 70–72, 2003.
Fotia AB, Ekberg J, Adams DJ, Cook DI, Poronnik P, and Kumar S. Regulation of neuronal voltage-gated sodium channels by the ubiquitin-protein ligases Nedd4 and Nedd4–2. J Biol Chem 279: 28930–28935, 2004.
Garty H and Palmer LG. Epithelial sodium channels: function, structure, and regulation. Physiol Rev 77: 359–396, 1997.
Hansson JH, Nelson-Williams C, Suzuki H, Schild L, Shimkets R, Lu Y, Canessa C, Iwasaki T, Rossier B, and Lifton RP. Hypertension caused by a truncated epithelial sodium channel gamma subunit: genetic heterogeneity of Liddle syndrome. Nat Genet 11: 76–82, 1995.
Hanwell D, Ishikawa T, Saleki R, and Rotin D. Trafficking and cell surface stability of the epithelial Na+ channel expressed in epithelial Madin-Darby canine kidney cells. J Biol Chem 277: 9772–9779, 2002.
Helms MN, Fejes-Toth G, and Naray-Fejes-Toth A. Hormone-regulated transepithelial Na+ transport in mammalian CCD cells requires SGK1 expression. Am J Physiol Renal Physiol 284: F480–F487, 2003.
Henke G, Maier G, Wallisch S, Boehmer C, and Lang F. Regulation of the voltage gated K+ channel Kv1.3 by the ubiquitin ligase Nedd4–2 and the serum and glucocorticoid inducible kinase SGK1. J Cell Physiol 199: 194–199, 2004.
Henry PC, Kanelis V, O'Brien MC, Kim B, Gautschi I, Forman-Kay J, Schild L, and Rotin D. Affinity and specificity of interactions between Nedd4 isoforms and the epithelial Na+ channel. J Biol Chem 278: 20019–20028, 2003.
Husted RF, Zhang C, and Stokes JB. Concerted actions of IL-1beta inhibit Na+ absorption and stimulate anion secretion by IMCD cells. Am J Physiol Renal Physiol 275: F946–F954, 1998.
Ingham RJ, Gish G, and Pawson T. The Nedd4 family of E3 ubiquitin ligases: functional diversity within a common modular architecture. Oncogene 23: 1972–1984, 2004.
Ishikawa T, Marunaka Y, and Rotin D. Electrophysiological characterization of the rat epithelial Na+ channel (rENaC) expressed in MDCK cells. Effects of Na+ and Ca2+. J Gen Physiol 111: 825–846, 1998.
Itani OA, Campbell JR, Herrero J, Snyder PM, and Thomas CP. Alternate promoters and variable splicing lead to hNedd4–2 isoforms with a C2 domain and varying number of WW domains. Am J Physiol Renal Physiol 285: F916–F929, 2003.
Itani OA, Cornish KL, Liu KZ, and Thomas CP. Cycloheximide increases glucocorticoid-stimulated -ENaC mRNA in collecting duct cells by p38 MAPK-dependent pathway. Am J Physiol Renal Physiol 284: F778–F787, 2003.
Kamynina E, Debonneville C, Bens M, Vandewalle A, and Staub O. A novel mouse Nedd4 protein suppresses the activity of the epithelial Na+ channel. FASEB J 15: 204–214, 2001.
Kamynina E, Tauxe C, and Staub O. Distinct characteristics of two human Nedd4 proteins with respect to epithelial Na+ channel regulation. Am J Physiol Renal Physiol 281: F469–F477, 2001.
Kanelis V, Rotin D, and Forman-Kay JD. Solution structure of a Nedd4 WW domain-ENaC peptide complex. Nat Struct Biol 8: 407–412, 2001.
Letz B and Korbmacher C. cAMP stimulates CFTR-like Cl– channels and inhibits amiloride-sensitive Na+ channels in mouse CCD cells. Am J Physiol Cell Physiol 272: C657–C666, 1997.
Loffing J, Loffing-Cueni D, Macher A, Hebert SC, Olson B, Knepper MA, Rossier BC, and Kaissling B. Localization of epithelial sodium channel and aquaporin-2 in rabbit kidney cortex. Am J Physiol Renal Physiol 278: F530–F539, 2000.
Nakhoul NL, Hering-Smith KS, Gambala CT, and Hamm LL. Regulation of sodium transport in M-1 cells. Am J Physiol Renal Physiol 275: F998–F1007, 1998.
Nalefski EA, Sultzman LA, Martin DM, Kriz RW, Towler PS, Knopf JL, and Clark JD. Delineation of two functionally distinct domains of cytosolic phospholipase A2, a regulatory Ca2+-dependent lipid-binding domain and a Ca2+-independent catalytic domain. J Biol Chem 269: 18239–18249, 1994.
Naray-Fejes-Toth A, Canessa C, Cleaveland ES, Aldrich G, and Fejes-Toth G. Sgk is an aldosterone-induced kinase in the renal collecting duct: effects on epithelial Na+ channels. J Biol Chem 274: 16973–16978, 1999.
Naray-Fejes-Toth A and Fejes-Toth G. The sgk, an aldosterone-induced gene in mineralocorticoid target cells, regulates the epithelial sodium channel. Kidney Int 57: 1290–1294, 2000.
Naray-Fejes-Toth A, Helms MN, Stokes JB, and Fejes-Toth G. Regulation of sodium transport in mammalian collecting duct cells by aldosterone-induced kinase, SGK1: structure/function studies. Mol Cell Endocrinol 217: 197–202, 2004.
Palmada M, Dieter M, Speil A, Bohmer C, Mack AF, Wagner HJ, Klingel K, Kandolf R, Murer H, Biber J, Closs EI, and Lang F. Regulation of intestinal phosphate cotransporter NaPi IIb by ubiquitin ligase Nedd4–2 and by serum- and glucocorticoid-dependent kinase 1. Am J Physiol Gastrointest Liver Physiol 287: G143–G150, 2004.
Palmer LG and Frindt G. Effects of cell Ca and pH on Na channels from rat cortical collecting tubule. Am J Physiol Renal Fluid Electrolyte Physiol 253: F333–F339, 1987.
Plant PJ, Yeger H, Staub O, Howard P, and Rotin D. The C2 domain of the ubiquitin protein ligase Nedd4 mediates Ca2+-dependent plasma membrane localization. J Biol Chem 272: 32329–32336, 1997.
Rossier BC, Pradervand S, Schild L, and Hummler E. Epithelial sodium channel and the control of sodium balance: interaction between genetic and environmental factors. Annu Rev Physiol 64: 877–897, 2002.
Rotin D, Staub O, and Haguenauer-Tsapis R. Ubiquitination and endocytosis of plasma membrane proteins: role of Nedd4/Rsp5p family of ubiquitin-protein ligases. J Membr Biol 176: 1–17, 2000.
Sayegh R, Auerbach SD, Li X, Loftus RW, Husted RF, Stokes JB, and Thomas CP. Glucocorticoid induction of epithelial sodium channel expression in lung and renal epithelia occurs via trans-activation of a hormone response element in the 5'-flanking region of the human epithelial sodium channel subunit gene. J Biol Chem 274: 12431–12437, 1999.
Schafer JA, Watkins ML, Li L, Herter P, Haxelmans S, and Schlatter E. A simplified method for isolation of large numbers of defined nephron segments. Am J Physiol Renal Physiol 273: F650–F657, 1997.
Shimkets RA, Warnock DG, Bositis CM, Nelson-Williams C, Hansson JH, Schambelan M, Gill JR Jr, Ulick S, Milora RV, Findling JW, Canessa CM, Rossier BC, and Lifton RB.Liddle's syndrome: heritable human hypertension caused by mutations in the subunit of the epithelial sodium channel. Cell 79: 407–414, 1994.
Snyder PM, Olson DR, and Thomas BC. Serum and glucocorticoid-regulated kinase modulates Nedd4–2-mediated inhibition of the epithelial Na+ channel. J Biol Chem 277: 5–8, 2002.
Snyder PM, Steines JC, and Olson DR. Relative contribution of Nedd4 and Nedd4–2 to ENaC regulation in epithelia determined by RNA interference. J Biol Chem 279: 5042–5046, 2004.
Staub O, Dho S, Henry P, Correa J, Ishikawa T, McGlade J, and Rotin D. 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, 1996.
Staub O, Gautschi I, Ishikawa T, Breitschopf K, Ciechanover A, Schild L, and Rotin D. Regulation of stability and function of the epithelial Na+ channel (ENaC) by ubiquitination. EMBO J 16: 6325–6336, 1997.
Thomas J, Deetjen P, Ko WH, Jacobi C, and Leipziger J. P2Y(2) receptor-mediated inhibition of amiloride-sensitive short circuit current in M-1 mouse cortical collecting duct cells. J Membr Biol 183: 115–124, 2001.
van Bemmelen MX, Rougier JS, Gavillet B, Apotheloz F, Daidie D, Tateyama M, Rivolta I, Thomas MA, Kass RS, Staub O, and Abriel H. Cardiac voltage-gated sodium channel Nav1.5 is regulated by Nedd4–2 mediated ubiquitination. Circ Res 95: 284–291, 2004.
Volk KA, Husted RF, Snyder PM, and Stokes JB. Kinase regulation of hENaC mediated through a region in the COOH-terminal portion of the -subunit. Am J Physiol Cell Physiol 278: C1047–C1054, 2000.
Vuagniaux G, Vallet V, Jaeger NF, Hummler E, and Rossier BC. Synergistic activation of ENaC by three membrane-bound channel-activating serine proteases (mCAP1, mCAP2, and mCAP3) and serum- and glucocorticoid-regulated kinase (Sgk1) in Xenopus oocytes. J Gen Physiol 120: 191–201, 2002.
Wulff P, Vallon V, Huang DY, Volkl H, Yu F, Richter K, Jansen M, Schlunz M, Klingel K, Loffing J, Kauselmann G, Bosl MR, Lang F, and Kuhl D. Impaired renal Na+ retention in the sgk1-knockout mouse. J Clin Invest 110: 1263–1268, 2002.(Omar A. Itani, John B. St)