Regulation of the epithelial Na+ channel (ENaC) by ubiquitylation is controlled by the activity of two counteracting enzymes, the E3 ubiquitin-protein ligase Nedd4-2 (mouse ortholog of human Nedd4L) and the ubiquitin-specific protease Usp2-45. Previously, Usp2-45 was shown to decrease ubiquitylation and to increase surface function of ENaC in Xenopus laevis oocytes, whereas the splice variant Usp2-69, which has a different N-terminal domain, was inactive toward ENaC. It is shown here that the catalytic core of Usp2 lacking the N-terminal domain has a reduced ability relative to Usp2-45 to enhance ENaC activity in Xenopus oocytes. In contrast, its catalytic activity toward the artificial substrate ubiquitin-AMC is fully maintained. The interaction of Usp2-45 with ENaC exogenously expressed in HEK293 cells was tested by coimmunoprecipitation. The data indicate that different combinations of ENaC subunits, as well as the α-ENaC cytoplasmic N-terminal but not C-terminal domain, coprecipitate with Usp2-45. This interaction is decreased but not abolished when the cytoplasmic ubiquitylation sites of ENaC are mutated. Importantly, coimmunoprecipitation in HEK293 cells and GST pull-down of purified recombinant proteins show that both the catalytic domain and the N-terminal tail of Usp2-45 physically interact with the HECT domain of Nedd4-2. Taken together, the data support the conclusion that Usp2-45 action on ENaC is promoted by various interactions, including through binding to Nedd4-2 that is suggested to position Usp2-45 favorably for ENaC deubiquitylation.
- epithelial sodium channel
body na+ homeostasis and blood pressure control involve the precise regulation of the epithelial Na+ channel (ENaC) in aldosterone-sensitive distal nephron cells (ASDN) (12). ENaC is ubiquitylated by the E3 ligase Nedd4-2 (Nedd4L in humans), leading to a reduction in active ENaC at the cell surface (10, 15, 22, 26). Aldosterone interferes with this effect notably by upregulating two proteins, serum- and glucocorticoid-regulated kinase 1 (Sgk1) and ubiquitin-specific protease Usp2-45, and thereby stimulating Na+ reabsorption (2, 4, 6, 25). Phosphorylation of Nedd4-2 by the early aldosterone-induced Sgk1 interferes negatively with Nedd4-2 function (2, 3). The ubiquitin-specific protease Usp2-45, also rapidly induced by aldosterone at the mRNA and protein levels, was shown to deubiquitylate ENaC and stimulate ENaC-mediated Na+ transport (4, 20, 25).
Considering the diversity of ubiquitylated proteins within a cell and the remarkable number of deubiquitylating enzymes (∼100 in the human genome) (16), it is expected that Usp2-45 shows specificity toward its target, ENaC. We hypothesized that this specificity is at least partly mediated by direct or indirect protein-protein interactions between Usp2-45 and ENaC. Consistent with this hypothesis, Usp2-45 coimmunoprecipitates with ENaC when coexpressed in HEK293 cells (20). However, it was not shown whether this interaction is direct or indirect. Since Nedd4-2 binds via its WW domains to ENaC (9, 10), binding of Usp2-45 to Nedd4-2 may also play a role in the selectivity of Usp2-45 action on ENaC. There are at least two isoforms of Usp2, Usp2-45 (Usp2b) and Usp2-69 (Usp2a) (7). Usp2-69 has a longer N-terminal domain and is found in the testis, heart, and skeletal muscle, and at a very low level in the kidney (7, 17). Additionally, it was shown to be specifically overexpressed in human prostate cancer and to prolong the half-life of fatty acid synthase (8). Since the catalytic core of these two Usp2 splice variants is identical, their different N-terminal domains might be involved in their target specificity. Target specificity may be conferred either by favoring interactions with target protein(s) (e.g., of Usp2-45 with ENaC) and/or by preventing other interactions.
To elucidate the possible role of the N-terminal domain of Usp2-45, we tested possible protein-protein interactions between Usp2-45 and ENaC or Nedd4-2 by coimmunoprecipitation of proteins expressed in HEK293 cells and by GST pull-down assays of purified recombinant proteins.
Constructs for Xenopus laevis ENaC subunits (αβγX-ENaC) and mouse Usp2-45 were described previously (4, 18). The construct for Usp2cc (catalytic core) was made by deleting bases encoding for amino acids (AA) 2–49 of Usp2-45 using an adapted QuickChange protocol (Stratagene). Constructs for expression in mammalian cells were described previously (20). Briefly, Usp2-45 wild-type (WT) and the inactive mutant (C67A) were cloned into pcDNA3. Usp2-45 N-terminal AA 1–53 (Usp2N) and Usp2-45 catalytic core (Usp2cc) were labeled either with an S-tag or a c-myc tag at the N-terminal side and generated by PCR. Usp2N was subcloned into a pCMV-GST vector, and Usp2cc was subcloned into pcDNA3.1(−). Human Nedd4-2 (KIAA0439) and its inactive mutant (C962S) were cloned into pCDNA3. Myc-tagged human Nedd4-2 and myc-tagged fragment Nedd4H2 (HECT domain of Nedd4-2) were generated by PCR and cloned into pcDNA3, the tag being located at the N terminus. All ENaC constructs were based on rat ENaC, and most of them have been previously described (20). Briefly, α-ENaC-(HA)3 was subcloned into pCMV4 (24), β-ENaC into pCDNA3, and γ-ENaC into pcDNA3.1(+)Zeo. β-ENaC was tagged with a C-terminal myc tag, and γ-ENaC with a C-terminal flag-tag. The ENaC lysine-to-arginine mutants were generated by PCR as described (4, 21).
Coexpression of mUsp2-45 or mUsp2cc and αβγ-XENaC in X. laevis oocytes.
Synthesis of cRNA was performed using a MEGAscipt SP6 kit as described (4). Stage V and VI oocytes were injected with 10 ng of Usp2-45, Usp2cc, and 0.05 ng each of α-, β-, and γ-XENaC subunit cRNA in 50 nl of nuclease-free water and incubated for 2 days before use (4).
Two-electrode voltage clamp on X. laevis oocytes.
The two-electrode voltage-clamp technique was used for recording whole-cell currents from X. laevis oocytes 42–50 h after injection as described (4). Data were obtained from 7 batches of oocytes with 1–10 oocytes/group and batch. Oocytes showing high amiloride-insensitive currents (current at −100 mV below −0.2 μA and <50% amiloride sensitivity) were considered to be leaky and excluded from data analysis. Significance was tested using the Kruskal-Wallis test and Dunn's posttest using GraphPad Prism 4.0 software.
Western blotting of Usp2 expressed in X. laevis oocytes.
Western blotting of X. laevis oocytes expressing ENaC and Usp2-45 or Usp2cc was performed as described (4) using an antibody against a C-terminal region of human Usp2 (no. AP2131c, Abgent). Oocytes (8/group) were lysed as described, and an equivalent of one oocyte was loaded per lane.
DNA constructs for expression in Escherichia coli.
For expression as a glutathione S-transferase (GST) fusion protein, full-length mouse Usp2-45 (GenBank accession no. NP_932759) was cloned into vector pGEX-6P-1 (GE Healthcare), resulting in plasmid pGEX6Pusp2Fl. A construct for the N-terminal domain of Usp2-45 (Usp2N, AA 1–49) was made by introducing two stop codons downstream of AA 49 in plasmid pGEX6Pusp2Fl using an adapted QuickChange protocol (Stratagene). A construct for the catalytic core (Usp2cc, AA 50–396) was made by deleting AA 2–49 in plasmid pGEX6Pusp2Fl. For all three constructs, the GST tag can be removed using the PreScission protease cleavage site. The cleaved proteins start with the N-terminal AA sequence GPLGSPEF (Usp2-45, Usp2N) or GPLGSPEFM (Usp2cc), followed by the Usp2 sequence.
Fragments Nedd4N (AA 1–256) and Nedd4D (AA 218–419) of mouse Nedd4-2 were cloned into vector pET-16b (Novagen) for expression as N-terminal His10-tagged proteins in E. coli. Fragment Nedd4H1 (AA 466–855) of mouse Nedd4-2 was cloned into vector pET-28a (Novagen) for expression as N- and C-terminal His6-tagged protein in E. coli. The plasmid for expression of Nedd4H2 (AA 474–847) with an N-terminal His6-tag was a kind gift from Sirano Dhe-Paganon (Structural Genomic Consortium, Univ. of Toronto). Fragment Nedd4H2 was cloned from human Nedd4-2 but differs only at one AA position from the corresponding mouse sequence (AA 698: N in humans, D in mice). The AA numbers of all Nedd4-2 fragments are based on GenBank accession no. AAK00809.
The N-terminal cytoplasmic part (AA 1–110) of mouse α-ENaC (GenBank accession no. NP_035454) was cloned into vector pET24b (Novagen) for expression as a C-terminal His6-tagged protein in E. coli and named α-ENAC-Nt.
Protein expression in E. coli.
Proteins were expressed in E. coli strain BL21(DE3). Bacteria were grown in Luria Broth medium supplemented with the appropriate antibiotic at 30 or 37°C, induced with 0.5 mM isopropyl-1-thio-d-galactopyranoside at an OD600 of 0.6, incubated for 3–4 h at 30°C (Nedd4N, Nedd4H1, α-ENAC-Nt, GST) or overnight at 15–23°C (other proteins). Cells were harvested by centrifugation, washed with 10 mM Tris·HCl, pH 8.0, frozen in liquid nitrogen, and stored at −80°C.
Purification of recombinant proteins.
GST-Usp2-45, GST-Usp2N, GST-Usp2cc, and GST were purified by affinity chromatography using glutathione-Sepharose 4FF (GE Healthcare), followed by size-exclusion chromatography on a Superdex 200, 10/300 GL column (GE Healthcare). For protease activity assays, the GST tag of GST-Usp2-45 and GST-Usp2cc was removed on the glutathione column with PreScission protease (GE Healthcare). Nedd4D, Nedd4N, and Nedd4H1 were purified by affinity chromatography using Ni-NTA agarose (Qiagen). Imidazol was removed on a HiTrap 5-ml desalting column (GE Healthcare). α-ENAC-Nt and Nedd4H2 were purified by affinity chromatography using Ni-NTA agarose, followed by size-exclusion chromatography on a Superdex 200, 10/300 GL column. A detailed description of all purification steps is given in supplementary data (all supplementary material for this article is available online at the journal web site).
Usp2 activity assay.
Proteolytic activity of Usp2 constructs was determined essentially as described (19) using ubiquitin-7-amido-4-methylcoumarin (ubiquitin-AMC, Boston Biochem) as substrate. Final concentration was 0.5 nM for the enzymes and 0–2.5 μM for ubiquitin-AMC. Kinetic analysis was performed at 28°C. A detailed description of all purification steps is given in supplementary data.
Cell culture and transfection.
HEK293 (human embryonic kidney) cells were cultured in Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 10% of fetal bovine serum (Invitrogen) and 0.05 U/ml of penicillin/streptomycin (Invitrogen) and incubated at 37°C/5% CO2. Cells were transiently transfected in 10-cm dishes at 60% confluence, using the Ca2+ phosphate method.
Cells were dissociated 24 h after transfection in 1 ml of cell dissociation buffer (5% glycerol, 1 mM EDTA, and 1 mM EGTA in PBS). Cells were recovered with 10 ml PBS (1×) in 15-ml polypropylene tubes. After 2-min centrifugation at 3,000 g, supernatants were discarded and pellets were frozen at −70°C for at least 20 min. They were lysed in 1 ml lysis buffer [50 mM HEPES, 150 mM NaCl, 1 mM EGTA, 10% glycerol, and 1% Triton X-100; containing protease inhibitors (protease inhibitor cocktail complete; Roche), 0.2 mM N-ethylmaleimide (Fluka), 1 mM DTT (Sigma), and phosphatase inhibitors (100 mM NaF and 10 mM Di-Na-pyrophosphate)]. Cells were solubilized for 30 min at 4°C with rotation. The lysates were centrifuged for 15 min at 20,000 g. Supernatants were recovered, and protein concentrations were quantified by the Bradford method. Immunoprecipitations were performed with at least 400 μg total protein. Antibody [1/1250, heme agglutinin (HA; Santa Cruz Biotechnology) for α-ENaC, c-Myc (Sigma) for β-ENaC, flag M2-agarose (Sigma) for γ-ENaC, biotinylated S-protein (Novagen) for Usp2-45] was added to the lysate and incubated for 2.5 h at 4°C with rotation. Twenty microliters of protein G-Sepharose (for HA, c-myc, GE Healthcare) or streptavidin-Sepharose (for S-protein, GE Healthcare) beads were added to the protein-antibody mix and incubated for 1.25 h at 4° with rotation. Beads were pelleted at 3,000 g, washed 4× at 4°C with 1 ml wash buffer (50 mM HEPES, pH 7.4, 150 mM NaCl, 1 mM EGTA, 10% glycerol, 0.2% Triton X-100, and 1.5 mM MgCl2), and resuspended in 40 μl of 2× sample buffer. Samples were fractionated by SDS-PAGE, transferred to nitrocellulose membranes, and analyzed by Western blotting as described (20). Anti S-tag monoclonal antibody (Novagen) was used at a dilution of 1:6,000.
Purified GST fusion proteins (GST-Usp2-45, GST-Usp2cc, and GST-Usp2-45N) and GST were diluted in buffer A [25 mM Tris·HCl, pH 7.5, 150 mM NaCl, 0.03% n-dodecyl-β-d-maltoside (DDM), 1 mM DTT]. Purified fragments of Nedd4-2, α-ENaC, and ovalbumin (GE Healthcare) were diluted in buffer B (25 mM Tris·HCl pH 7.5, 150 mM NaCl, 20% glycerol, 0.1% Triton X-100). The GST fusion protein (30 μl) was combined with prey protein (97 μl), and 1 mM MgCl2, 0.1 mM CaCl2, 0.1 μM ZnCl2 and 2–10 mM DTT were added. Final protein concentrations were 0.7–5 μM. The GST dimer concentration was calculated. The combined proteins were incubated for 20 min at room temperature. Then, 115-μl aliquots were added to 5 μl of washed glutathione S- Sepharose 4 Fast Flow beads (GE Healthcare) and incubated for 130–165 min at 20°C with rotation. Samples were centrifuged for 1 min at 0.5 g, the supernatants were removed, and the beads were washed 3× with 100 μl buffer B supplemented with 1 mM MgCl2, 0.1 mM CaCl2, 0.1 μM ZnCl2, and 1 mM DTT. Elutions were performed with 45 μl buffer C (50 mM Tris·HCl, 10 mM glutathione, 0.1% Triton X-100, 2 mM DTT). The concentration of individual proteins, of the supernatant, and eluated proteins was determined by lab-on-a-chip electrophoresis on a Bioanalyzer 2300 using Protein 80 chips (Agilent) according to the manufacturer's instructions.
Role of Usp2-45 N-terminal domain in ENaC regulation.
The impact of full-length Usp2-45 and of its catalytic core lacking the 49 N-terminal residues (Usp2cc) on ENaC-mediated, amiloride-sensitive Na+ current was tested in X. laevis oocytes. As previously shown, coexpression of full-length Usp2-45 increased ENaC activity several-fold (Fig. 1A) (4). In contrast, Usp2cc showed a reduced activation of ENaC relative to full-length Usp2-45 (2.0 vs. 4.3 times) when the same amounts of Usp2cc or Usp2-45 cRNA where injected. As shown by Western blotting, this difference was not due to a lower expression of Usp2cc (Fig. 1B). The activation of ENaC by Usp2cc was even relatively more reduced compared with similar amounts of expressed Usp2-45 protein (similar Western blotting signal intensity obtained by varying the amount of injected cRNA; data not shown). These results indicate that the N-terminal domain of Usp2-45 is involved in modulation of ENaC activity. Interestingly, it was previously shown in X. laevis oocytes that Usp2-69, which differs from Usp2-45 only in its N-terminal domain, does not increase ENaC activity (4). To test whether the N-terminal region of Usp2-45 increases intrinsic activity of Usp2-45, we compared the catalytic activity of Usp2-45 with Usp2cc using the artificial substrate ubiquitin-AMC (Fig. 2). The calculated Km value for mouse Usp2-45 was 1.7 μM and for Usp2cc 2.0 μM and kcat values of 0.92 s−1 and 0.91 s−1, respectively, suggesting that the N-terminal domain does not influence protease activity. These values are in the same range as the published values for human Usp2cc (Km = 0.554 μM, kcat = 0.14 s−1) (19). We conclude that the effect of the N-terminal domain of Usp2-45 on ENaC activity observed in X. laevis oocytes is independent of proteolytic activity, but rather may be due to protein-protein interactions. Therefore, we performed a series of experiments to test possible interactions of Usp2-45 with its known partners.
Interactions of Usp2-45 with ENaC.
We have previously shown that full-length Usp2-45 and Usp2cc but not the N-terminal domain of Usp2-45 coimmunoprecipitate with ENaC (20), suggesting that the C-terminal enzymatic domain interacts with ENaC. Furthermore, coprecipitation did not depend on the PY motifs of ENaC and therefore binding of Nedd4-2 to ENaC (10). To address the question of whether the Usp2-45-ENaC interaction involves a specific ENaC subunit(s), we tested coprecipitation of various combinations of ENaC subunits with S-tagged Usp2-45 in HEK293 cells (Supplemental Fig. S1). Since α-ENaC and/or γ-ENaC coprecipitated with Usp2-45, the direct or indirect interaction of Usp2-45 with ENaC can involve either subunit (β-ENaC was not tested).
To test whether the N- and/or the C-terminal domain of α-ENaC mediate(s) this interaction with Usp2-45, both cytoplasmic domains were separately expressed as GST fusion proteins in HEK293 cells with Usp2-45. The N-terminal-, but not the C-terminal domain of α-ENaC coprecipitated with Usp2-45 (Fig. 3), indicating that the Usp2-45-ENaC interaction requires the N- but not the C-terminal domain of α-ENaC. The interaction observed by coimmunoprecipitation in HEK293 cells is not necessarily direct, but rather may depend on posttranslational modifications. For instance, Usp2-45 directly interacts with protein-conjugated or free ubiquitin (19). Therefore, we tested the possibility that Usp2-45 binds the ubiquitin chains rather than directly to ENaC using ubiquitin-deficient ENaC mutants (ENaC K-R) in which every cytoplasmic lysine in each of the three ENaC subunits was mutated to arginine (4). ENaC K-R mutants thus lack all putative cytoplasmic ubiquitylation sites (4). Although the interaction was not completely abolished, each of the three ENaC K-R mutant subunits (particularly α-ENaC) coprecipitated with Usp2-45 to a lesser extent than the wild-type ENaC, consistent with ubiquitylation strongly contributing but not being essential for the interaction of ENaC with Usp2-45 (Fig. 4). Furthermore, mutation of the ENaC C-terminal PY motifs reduced the interaction, likely due to reduced ubiquitylation by Nedd4-2, since mutation of the PY motif did not additionally prevent coprecipitation of ENaC K-R with Usp2-45 (Fig. 4, compare the K-R mutant with K-R, PY-double mutant).
Usp2-45 interacts with ubiquitin-protein ligase Nedd4-2.
Our previous findings suggest that Usp2-45 counteracts Nedd4-2 by deubiquitylating ENaC (4, 20). Furthermore, since some deubiquitylating enzymes have been reported to complex with ubiquitin-protein ligases (11, 13, 23), we tested whether Usp2-45 interacts with Nedd4-2. Coimmunoprecipitation experiments using transfected HEK293 cells indicated an interaction which does not depend on the catalytic activity of Usp2-45 nor Nedd4-2 as shown by the data from catalytically inactive mutants (Usp2-45-C67A and Nedd4-2-C962S) (Fig. 5A). Testing the interaction of full-length Nedd4-2 with truncated Usp2-45, coprecipitation was observed with the Usp2-45 catalytic domain but not the N-terminal domain (Fig. 5B). In contrast, when interaction of the Nedd4-2 HECT domain was tested, coprecipitation was observed with both the catalytic and the N-terminal domain of Usp2-45 (Fig. 5C).
GST pull-down of affinity purified recombinant proteins expressed in E. coli was used to probe for a direct interaction between Usp2-45 and Nedd4-2. The Nedd4-2 fragment Nedd4H1 containing the catalytic HECT domain was pulled down by GST-Usp2-45, but not by GST alone (Fig. 6). In contrast, Nedd4N and Nedd4D fragments containing three of the four WW domains of Nedd4-2 did not bind to Usp2-45. Due to technical difficulties, a region of 46 AA comprising WW4 of Nedd4-2 could not be tested. These results indicate that the HECT domain of Nedd4-2, but not the WW domains or the regions between WW domains, directly interact with Usp2-45. Since proteins were expressed in bacterial cells and hence lacked specific eukaryotic posttranslational modifications, e.g., ubiquitylation, phosphorylation or glycosylation, such modifications are not required for the physical interaction of Usp2-45 with Nedd4-2.
The two domains of Usp2-45 [N-terminal domain (Usp2N) and catalytic core (Usp2cc)] were produced separately as purified recombinant proteins, and their interaction with Nedd4-2 was compared with full-length Usp2-45. To remove aggregates that might lead to nonspecific interactions, all recombinant proteins used for the GST pull-down assay (except ovalbumin) were purified by size-exclusion chromatography. Consistent with coprecipitation using transfected HEK293 cells (Fig. 5C), the purified recombinant N-terminal and catalytic core domains of Usp2-45 interacted with the HECT domain of Nedd4-2, similar to full-length Usp2-45 (Fig. 7).
In ASDN, Usp2-45 controls the surface expression and function of ENaC by deubiquitylating it (4, 20, 25). Due to the large diversity of ubiquitylated proteins within cells, this implies Usp2-45 specificity for ENaC. The N-terminal domain of Usp2-45, which is unique to this Usp2 isoform, is likely involved in target specificity. Consistent with this conclusion, the deletion of the N-terminal domain of Usp2-45 reduces its stimulation of ENaC function without affecting catalytic activity. This finding supports the hypothesis that the N-terminal domain of Usp2-45 is involved in target specificity, presumably by mediating or interfering with protein-protein interactions. Our previous observation that the Usp2–69 isoform does not affect ENaC activity in X. laevis oocytes (4) suggests that the N-terminal domain of Usp2–69 results in a subcellular localization distinct from ENaC and/or inhibits Usp2–69 catalytic activity toward ubiquitylated ENaC.
To identify possible protein-protein interactions of Usp2-45 that could mediate ENaC target specificity, we previously performed coexpression experiments in HEK293 cells and demonstrated that ENaC is coprecipitated with Usp2-45 (20). In the present study, we coexpressed ENaC mutants devoid of cytoplasmic ubiquitylation sites (N-terminal lysines) and/or lacking the cytoplasmic binding sites for Nedd4-2 (C-terminal PY motifs) (Fig. 4). Surprisingly, the interaction of Usp2-45 with ENaC was maintained in the absence of these cytoplasmic structures/modifications, although at a clearly lower level. This indicates additional direct or indirect interactions between Usp2-45 and ENaC, beyond the direct interaction of Usp2-45 with ENaC via ubiquitin moieties and the indirect binding via Nedd4-2 docked on C-terminal ENaC PY motifs. However, since HEK293 cells can express high levels of exogenous proteins, ubiquitylation of misfolded and improperly inserted ENaC on extracellular lysine residues could generate targets for Usp2-45. Therefore, the question of whether there is a direct physical interaction between Usp2-45 and nonubiquitylated ENaC cannot be definitely answered based on these coprecipitation experiments. To test for direct interaction of Usp2-45 in the absence of other cellular components, we attempted, with partial success, to produce recombinant cytoplasmic domains of the ENaC subunits in bacteria. Intriguingly, the α-ENaC N-terminal domain was efficiently pulled down as large MW multimers. However, the size exclusion-purified fragment was not significantly pulled down (Fig. 6), indicating that any possible binding of this ENaC cytoplasmic N-terminal fragment to Usp2-45 in the absence of cellular context, and thus of ubiquitylation, is of insufficient affinity to be observed. This result together with the coprecipitation results discussed above suggest the possibility that the targeting of Usp2-45 to ENaC could be facilitated by interactions involving ubiquitin moieties in the context of the N-terminal ENaC sequence.
In contrast, a clear interaction of Usp2-45 with the HECT domain of Nedd4-2 was observed by both coimmunoprecipitation of proteins expressed in HEK293 cells and GST pull-down of purified proteins expressed in E. coli, which reveal direct interactions in the absence of posttranslational modifications. Since WW domains of Nedd4-2 interact with ENaC C-terminal cytoplasmic PY motifs (9, 10), Usp2-45 binding to Nedd4-2 situates Usp2-45 next to ENaC for deubiquitylation and protection from Nedd4-2 activity. This model, as shown in Fig. 8, includes the interaction of both the catalytic and the N-terminal domains of Usp2-45 with the HECT domain of Nedd4-2 (Figs. 5C and 7) and explains our original observations that the Usp2-45 N-terminal domain increases the efficacy of Usp2-45 in stimulating ENaC activity in X. laevis oocytes (Fig. 1). Interestingly, the other Usp2 isoform, Usp2a (Usp2-69), can interact with the ubiquitin ligase Mdm2 RING E3 that ubiquitylates p53 (23). In this latter case, however, the interaction appears to deubiquitylate and stabilize the E3 ligase rather than its substrate. However, analogous interactions as postulated for Usp2-45 and Nedd4-2-counteracting E3 ligase activity on its substrate have also been described. For example, the pVHL-interacting deubiquitylating enzyme 2 (VDU2) binds to the von Hippel-Lindau protein (pVHL)-E3 ubiquitin-protein ligase and deubiquitylates its substrate, HIF-1α (14). Interestingly, the isolated HECT domain, but not full-length Nedd4-2 was coprecipitated with the N-terminal domain of Usp2-45 (Fig. 5, B and C). This suggests that within Nedd4-2 the N-terminal WW domain region prevents the HECT domain, possibly by binding to it (1), from interacting with the-N-terminal domain of Usp2-45.
Taken together, our results support a model in which multiple interactions participate in the formation of a transient multiprotein complex including the E3 ubiquitin ligase Nedd4-2 and in which Usp2-45 is localized close to its target, ubiquitylated cytoplasmic ENaC domains.
This work was supported by the Swiss National Science Foundation (31003A_125422/1 and IZK0Z3-125255 to O. Staub and B. Oberfeld, respectively), by the Leducq Foundation, Paris (Transatlantic Network on Hypertension; to O. Staub), by the Swiss Kidney Foundation (to O. Staub), and by the Transregio Sonderforschungsbereich Konstanz-Zurich (TR SFB 11; to F. Verrey).
No conflicts of interest, financial or otherwise, are declared by the authors.
We thank Gabriele Adam for help with electrophysiology, Eva Hänsenberger for preparing oocytes, Katja Huggel for technical assistance, and Victoria Makrides for proofreading the manuscript.
Current address of K. M. Pos: Institute of Biochemistry, Goethe University, Frankfurt, Germany.
- Copyright © 2011 the American Physiological Society