The goal of the present study was to provide a direct demonstration of the regulation of ENaC by Nedd4 and the implications of this regulation for Liddle's syndrome. The Xenopus laevis
oocyte represents a convenient experimental system to study such questions, as we and others have previously shown that elevated Na+
channel (ENaC) activity can be observed when channels containing Liddle's syndrome mutations are expressed in these oocytes (22
). Moreover, because Xenopus
oocytes express endogenous Nedd4 (27
), they are likely to possess all the necessary machinery for proper Nedd4 function.
In our previous work, we demonstrated that rat Nedd4 is expressed in the same cells that express ENaC within the distal nephron and lung epithelia (27
), that Nedd4-WW domains bind to the PY motifs of ENaC (27
), that point mutations within the PY motifs of ENaC identified in Liddle's syndrome patients (19
) also abrogate binding to Nedd4-WW domains (27
), and finally that ENaC stability and function are regulated by ubiquitination (32
). In this study, we present several lines of evidence suggesting that it is indeed Nedd4 that is involved in the ubiquitination and the control of ENaC activity. First, overexpression of xNedd4-wt with ENaC in Xenopus
oocytes caused a strong inhibition of INa
when compared with oocytes expressing ENaC alone in the presence of endogenous xNedd4. Second, a catalytically inactive form of xNedd4 stimulates these currents, probably by competing with endogenous xNedd4. Third, the effect of xNedd4 is dependent on the presence of intact PY motifs within the ENaC subunits, suggesting an interaction between xNedd4 and the PY motifs, similar to our previous demonstration with rat Nedd4 (27
). Moreover, FaNaCh, another channel of the ENaC/degenerin superfamily, which does not contain PY motifs (38
), is not regulated by Nedd4.
The current findings provide some interesting insights into the mechanism by which Nedd4 mediates ENaC regulation. The observation that the PY motifs are necessary for Nedd4-dependent regulation strongly suggests that Nedd4 is indeed exerting its effect by binding via its WW domains to the ENaC PY motif. The completely opposite effects on channel function of Nedd4-wt versus the xN4C938S mutant support the hypothesis that Nedd4 is acting via ubiquitination, most likely by ubiquitinating ENaC directly, and thus leading to its retrieval from the plasma membrane and to degradation of the ENaC complex, as is clearly demonstrated by both quantitative surface labeling and immunostaining (Figs. and ). However, an effect of Nedd4 on ENaC during biosynthesis and recycling processes may also play a role. The inhibition of ENaC internalization in the catalytically impaired xNedd4 mutant is in agreement with our p revious findings demonstrating increased channel numbers and increased retention at the cell surface of ENaC bearing Lys→Arg mutations that render the channel ubiquitination defective (32
). Recently, Shimkets et al
) have shown that ENaC is internalized via clathrin-coated pits and that this internalization is dependent on the presence of COOH-terminal endocytosis signals that encompass the PY motifs. It is interesting that a growing number of studies have now demonstrated a link between ubiquitination of transmembrane proteins at the cell surface and their subsequent endocytosis and degradation by the lysosomes or vacuoles in yeast (reviewed in ref. 31
). Although it is not yet known how ubiquitination (which normally directs cytosolic or ER proteins to proteasomal degradation) provides a signal for endocytosis of transmembrane proteins, it is quite likely that ENaC also belongs to this class of cell-surface proteins in which ubiquitination and endocytosis and lysosomal degradation are tightly linked.
Our earlier demonstration of binding of Nedd4 to the regions (PY motifs) in ENaC that are deleted/mutated in patients with Liddle's syndrome (27
) and our current demonstration of impaired regulation by Nedd4 of ENaC channels lacking one (as in Liddle's syndrome) of its PY motifs clearly implicate Nedd4 in the pathophysiology of this inherited form of hypertension. However, the increased retention of ENaC at the plasma membrane caused by loss of the PY motif, which leads to inhibition of internalization (39
) and impaired binding to Nedd4 (thus likely to impaired ubiquitination), cannot provide a full explanation for the defective function of ENaC in Liddle's syndrome. As shown earlier by Firsov et al
), at least half of the increase in ENaC activity associated with the Liddle's mutations can be attributed to increased open probability of the channel. Whether or not Nedd4 can also affect channel gating is currently unknown. Interestingly, our recent work has demonstrated that ENaC is downregulated by elevated intracellular Na+
), and that this feedback inhibition is impaired in ENaC chains carrying the Liddle's mutations in their PY motifs (26
). Because Nedd4 binds to these PY motifs, it is possible that it is involved in the negative regulation by intracellular Na+
. In support of this notion, a recent report has suggested that Nedd4 mediates the control by intracellular Na+
of an epithelial Na+
channel in salivary ducts (41). The molecular identity of the latter channel, however, is unknown; hence deciphering its putative biochemical interactions with Nedd4 is currently not possible.
Despite the strong effect of wild-type or catalytically inactive xNedd4 on ENaC function, a much smaller effect of rat Nedd4 on ENaC activity was observed (data not shown) even though the protein was expressed at high levels and despite the fact that we used rat ENaC in our experiments. This suggests that either the presence of an additional WW domain in xNedd4 (which contains four WW domains versus three in rNedd4), which may allow better binding to the ENaC tetramer, or the presence of nonconserved NH2-terminal sequences in xNedd4, which may play a role in localization or functioning of this ubiquitin–protein ligase, could contribute to these differences.
Although thus far there have not been any mutations in Nedd4 identified in patients with Liddle's syndrome, the search for such mutations in inherited forms of hypertension is nevertheless worthwhile in view of our clear demonstration of regulation of the channel by Nedd4. On the other hand, because Nedd4 is likely to have numerous cellular substrates, it is possible that it is an essential gene, as seen in yeast (42
), and that mutations that severely interfere with its function may be lethal in higher organisms as well. A gene knockout of Nedd4 in mice, not yet available, may help shed some light on this issue.
In summary, our work provides strong evidence that Nedd4 is a suppressor of ENaC activity that regulates the number of ENaC channels at the plasma membrane. Moreover, it shows that the regulatory effect of Nedd4 requires the interaction between its WW domains and the PY motifs of ENaC, the same motifs that, when mutated, cause Liddle's syndrome. Accordingly, we have demonstrated that this suppressive effect is attenuated in Liddle's syndrome. Thus, this work has implications for our comprehension of the molecular and biochemical mechanisms underlying this disease, as well as for our understanding of ENaC function in general.