Progesterone Down-regulates the Open Probability of the Amiloride-sensitive Epithelial Sodium Channel via a Nedd4-2-dependent Mechanism*

Activation of the mitogen-activated protein (MAP) kinase cascade by progesterone in Xenopus oocytes leads to a marked down-regulation of activity of the amiloride-sensitive epithelial sodium channel (ENaC). Here we have studied the signaling pathways involved in progesterone effect on ENaC activity. We demonstrate that: (i) the truncation of the C termini of the αβγENaC subunits results in the loss of the progesterone effect on ENaC; (ii) the effect of progesterone was also suppressed by mutating conserved tyrosine residues in the Pro-X-X-Tyr (PY) motif of the C termini of the β and γ ENaC subunits (βY618A and γY628A); (iii) the down-regulation of ENaC activity by progesterone was also suppressed by co-expression ENaC subunits with a catalytically inactive mutant of Nedd4-2, a ubiquitin ligase that has been previously demonstrated to decrease ENaC cell-surface expression via a ubiquitin-dependent internalization/degradation mechanism; (iv) the effect of progesterone was significantly reduced by suppression of consensus sites (βT613A and γT623A) for ENaC phosphorylation by the extracellular-regulated kinase (ERK), a MAP kinase previously shown to facilitate the binding of Nedd4 ubiquitin ligases to ENaC; (v) the quantification of cell-surface-expressed ENaC subunits revealed that progesterone decreases ENaC open probability (whole cell Po, wcPo) and not its cell-surface expression. Collectively, these results demonstrate that the binding of active Nedd4-2 to ENaC is a crucial step in the mechanism of ENaC inhibition by progesterone. Upon activation of ERK, the effect of Nedd4-2 on ENaC open probability can become more important than its effect on ENaC cell-surface expression.

The amiloride-sensitive epithelial sodium channel (ENaC) 2 is a highly selective Na ϩ channel found in the apical membrane of salt-reabsorbing tight epithelia, including the kidney distal nephron, the distal colon, the salivary and sweat glands, and the lung. ENaC activity is essential for maintaining extracellular fluid volume and blood pressure. The activity of this channel is controlled at two levels: the number of active channels on the cell surface (N) and the channel open probability (P o ) (1). However, quantitative estimates of N and P o values remain difficult, mainly because of the low cell-surface expression and high variability of the open probability of ENaC. Aldosterone, the principal hormone control-ling ENaC activity, has been shown to have a major effect on N. One of the molecular mechanisms proposed for the regulation of ENaC cellsurface expression by aldosterone involves Nedd4-2 ubiquitin ligase and serum-and glucocorticoids-induced kinase isoform 1 (sgk1). In a model proposed by Debonneville et al. (2) for the principal cell of the kidney collecting duct, binding of Nedd4-2 to the Pro-X-X-Tyr (PY) motif in the ␤ and ␥ ENaC subunits results in ENaC ubiquitylation and intracellular degradation. Stimulation of sgk1 expression by aldosterone provokes sgk1-dependent Nedd4-2 phosphorylation, resulting in a decreased binding affinity of Nedd4-2 to ENaC and, finally, in an increased ENaC number at the cell surface. However, recent evidence indicates that increased sgk1 expression may also be involved in the regulation of ENaC P o . Diakov et al. (3) have shown that sgk1 can directly phosphorylate the C terminus of the ␣ENaC subunit, and that this phosphorylation significantly increases ENaC activity via an increase in channel P o . Alvarez de la Rosa et al. (4) have recently demonstrated that expression of a constitutively active mutant of sgk1 in A6 cells results in a significant increase of both N and P o . Using measurements of wcP o , a similar dual effect of sgk1 on ENaC N and P o was observed by Vuagniaux et al. (5) in Xenopus oocyte expression system. We have also shown that a naturally occurring mutation of the PY motif (␤ R564stop ), which causes an autosomal dominant form of salt-sensitive hypertension (Liddle syndrome), results in both an increase in N and predominant change in P o (6). Collectively, these data indicate that sgk1-and/or Nedd4-2-dependent mechanisms may control ENaC activity by controlling both the N and P o .
Another mechanism known to play an important role in the control of ENaC activity is the extracellular signal-regulated protein kinases 1 and 2 (ERK1/2) signaling pathway. This signaling pathway is thought to have a dual effect on ENaC. The long term (hours/days) activation of this pathway leads to the down-regulation of ENaC activity via inhibition of the transcription rate of ENaC subunits (7,8). The short term (minutes) activation of ERK1/2 also down-regulates ENaC activity. The molecular mechanism responsible for this acute effect remains poorly understood. Shi et al. (9) have shown that ERK is capable of directly phosphorylating the ␤ and ␥ ENaC subunits on residues Thr-613 and Thr-623, respectively. The phosphorylation of ␤ T613 and ␥ T623 increases the affinity of WW domains of Nedd4 -1 ubiquitin ligase (a closely related homologue of Nedd4-2) for the C termini of the ␤ and ␥ subunits. ␤ T613A and ␥ T623A mutations significantly increased the I Na , but whether changes in N and/or P o were responsible for this increase was not determined (9). We have recently shown that activation of ERK pathway by progesterone in Xenopus oocytes leads to a marked downregulation of ENaC activity (10). However, our data did not fit well with a Nedd4-2-dependent ENaC ubiquitylation/degradation mechanism, because quantification of ENaC expression at the cell surface revealed that the main effect of progesterone is a decrease of channel P o . The aim of the present study was to analyze the molecular mechanisms of the observed P o effect. Detailed analysis of the progesterone effect on ENaC activity demonstrated, contrary to our expectations, that the regulation of ENaC P o by progesterone is PY motif-and Nedd4-2-dependent. These results demonstrate a novel mode of action of Nedd4-2 in the regulation of channel function by P o .

EXPERIMENTAL PROCEDURES
Expression of ENaC and Nedd4-2 in Xenopus oocytes-Complementary RNAs for rat ␣␤␥ ENaC subunits, human Nedd4-2, and mouse NaPi-IIa sodium phosphate co-transporter were synthesized in vitro using SP6 polymerase. The cRNAs were injected into Xenopus oocytes (10 ng of total cRNA/oocyte), and injected oocytes were kept in modified Barth solution (MBS) containing 88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO 3 , 0.8 mM MgSO 4 , 0.3 mM Ca(NO 3 ) 2 , 0.4 mM CaCl 2 , 10 mM Hepes-NaOH (pH 7.2). Electrophysiological measurements and cell surface biotinylation were performed 24 h after injection. Macroscopic amiloride-sensitive Na ϩ current (I Na ) was defined as the difference between Na ϩ current obtained in the presence (5 M) and in the absence of amiloride and was recorded, using the two-electrode voltage clamp method. The mean I Na values in different oocyte batches ranged between 1 and 15 A.
Biotinylation of ENaC Subunits on the Cell Surface-Biotinylation was performed in 48-well plates, using 20 oocytes per experimental condition. All biotinylation steps were performed in a cold room, using ice-cold solutions. After incubation in ice-cold MBS solution for 30 min, the oocytes were washed three times with MBS. After the last wash, the MBS solution was removed and replaced by a biotinylation buffer containing 10 mM triethanolamine, 150 mM NaCl, 2 mM CaCl 2 , 1 mg/ml EZ-link sulfo-NHS-SS-Biotin (Pierce), pH 9.5. The incubation lasted 15 min with gentle agitation. The biotinylation reaction was stopped by washing the oocytes two times with a Quench buffer containing 192 mM glycine, 25 mM Tris-Cl (pH 7.5) added to the MBS solution. After the second wash, oocytes were incubated for 5 min in the Quench buffer with gentle agitation. Oocytes were then washed two times with the MBS solution and transferred to 1.5-ml Eppendorf tubes. Oocytes were lysed by repeated pipetting with Pasteur pipettes in a lysis buffer containing: 1% Triton X-100, 500 mM NaCl, 5 mM EDTA, 50 mM Tris-Cl (20 l/oocyte). The lysates were vortexed for 30 s, and the mixture was centrifuged for 10 min at 12,000 rpm. Supernatants were transferred to the new 1.5-ml Eppendorf tubes containing 50 l of immunopure immobilized streptavidin beads (Pierce) washed with lysis buffer. After overnight incubation at 4°C with shaking, the tubes were centrifuged for 1 min at 5000 rpm. Supernatant was removed, and beads were washed three times with lysis buffer. 60 l of SDS-PAGE sample buffer was added to the beads. All samples were heated for 5 min at 95°C before loading on the 8% SDS-PAGE.
Antibodies-The anti-␣␤␥ENaC antibodies were previously described. The anti-actin antibody was from Sigma.

RESULTS
The Effect of Progesterone on ENaC Activity Is PY Motif-and Nedd4-2-dependent-As shown in Fig. 1, the addition of progesterone (15 M) to ENaC-expressing oocytes leads to a significant down-regulation of amiloride-sensitive Na ϩ current (I Na ). The I Na is inhibited rapidly; ϳ50% of I Na inhibition is reached within 30 min of progesterone treatment. The short-time required for I Na inhibition indicated that ENaC activity and/or cell-surface expression, rather than cRNA stability or protein synthesis, are regulated by progesterone. This was confirmed by the stable expression of ␣␤␥ENaC subunit proteins, over 4 h of treatment with progesterone (see below). To assess which parts of the ENaC subunits are involved in channel responsiveness to progesterone, we first expressed ␣␤␥ENaC subunits with truncated C termini. For the ␣ truncation we used a mutant in which all C-terminal amino acids were eliminated (␣ R618stop ϭ ␣ T ). For the ␤ and ␥ subunits truncation, we used the ␤ R564stop (␤ T ) and ␥ W574stop (␥ T ) mutants that cause Liddle disease (11). As shown in Fig. 2A, the truncation of ␣␤␥ C termini results in a complete loss of the progesterone effect on ENaC. Because the PY motifs in the C termini of ␤ and ␥ subunits have been demonstrated as the important regulatory elements of ENaC activity, we next assessed the effect of progesterone on the ␤ Y618A and ␥ Y628A mutants. As shown in Fig. 2B, the mutation of the PY motifs in both ␤ and ␥ subunits, either alone or together, resulted in the complete suppression of the progesterone effect on ENaC. This result clearly demonstrated that the progesterone effect on ENaC is dependent on PY motif of either ␤ or ␥ (but not ␣) ENaC subunits. The PY motifs have been proposed to bind various WW-domain-containing proteins, including dystrophin, caveolins, ubiquitin ligases, and other proteins (12). Also, Shimkets et al. (13) have proposed that ␤ and ␥ PY motifs are required for clathrin-dependent ENaC endocytosis. Because ENaC interaction with Nedd4-2 has been shown to be an important mechanism controlling ENaC activity, we next tested the effect of the catalytically inactive Nedd4-2 C145S The Effect of Progesterone on ENaC Activity Is Reduced by Mutation of the Consensus Sites for ERK Phosphorylation-Recently, several amino acids lying outside the PY motifs have been proposed to be involved in the interaction between ENaC and Nedd4-2 (9,14). For instance, Shi et al. (9) have shown that phosphorylation of ␤Thr-613 and ␥Thr-623 by ERK facilitates the interaction between ENaC and WW-domains of Nedd4 -1 ubiquitin ligase, a closely related homolog of Nedd4-2. Because the effect of progesterone on ENaC depends on the activation of the ERK pathway (10), we also tested the effect of these two phospho-rylation sites on progesterone inhibition of I Na . As shown in Fig. 3A, the mutation of ␤Thr-613 and/or ␥Thr-623 to alanine results in a partial but significant reduction of the progesterone effect on I Na . These mutations also resulted in a significant increase in basal I Na : 8.8 Ϯ 2.3-fold increase for ␣␤ T613A ␥ T623A channel (p Ͻ 0.01, compared with ␣␤␥ channel); 2.7 Ϯ 0.6-fold increase for ␣␤ T613A ␥ channel (p Ͻ 0.05, compared with ␣␤␥ channel); and 2.5 Ϯ 0.8-fold increase for ␣␤ T613A ␥ T623A channel (p Ͻ 0.05, compared with ␣␤␥ channel). To check whether progesterone induces ERK-dependent ENaC phosphorylation, we attempted 32 P-phosphorylation experiments in living oocytes. However, our first experiments revealed only low, almost undetectable levels of ENaC phosphorylation (data not shown). This was probably due to a low uptake of [ 32 P]orthophosphate across the oocyte plasma membrane under standard conditions. To improve oocyte loading with [ 32 P]orthophosphate, we co-injected ENaC subunits with NaPiIIa sodium phosphate co-transporter. As shown in Fig. 3B, co-injection Oocytes were co-injected with FLAG-tagged ␣␤␥ENaC cRNAs and NaPiIIa cRNA. 24 h after injection, oocytes were loaded for 4 h with [ 32 P]orthophosphate. Immunoprecipitation was performed in non-denaturing conditions, using anti-FLAG antibody. C, progesterone-induced phosphorylation of ␤ENaC is abolished in ␤ Y618A mutant. Oocytes were co-injected with NaPiIIa cRNA and ␣␤␥ENaC or ␣␤ Y618A ␥ENaC FLAG-tagged cRNAs. 24 h after injection, oocytes were loaded for 4 h with [ 32 P]orthophosphate. Immunoprecipitation was performed in non-denaturing conditions, using anti-FLAG antibody.
with NaPiIIa allowed the detection of ␤ENaC phosphorylation after co-immunoprecipitation of all three ␣␤␥ subunits in non-denaturing conditions. Statistical analysis of six independent experiments revealed a 30 Ϯ 8% (p ϭ 0.05) increase in ␤ENaC phosphorylation under progesterone treatment. Mutation of Thr-613 to alanine significantly reduced the basal ␤ENaC phosphorylation and abolished progesteronedependent ␤ENaC phosphorylation (Fig. 3C). The phosphorylation of ␣ and ␥ ENaC subunits was not detectable, potentially due to the complex pattern of ␣ and ␥ ENaC processing by endogenous proteases (see below). Collectively, these results demonstrate that elimination of extra-PY motif binding sites for Nedd4 ubiquitin ligases reduces the effect of progesterone on ENaC. Development of a New Assay for Quantification of Channel N-It has been proposed that the main effect of Nedd4-2 on ENaC activity consists of the reduction of channel N. However, our previous experiments demonstrated that the main effect of progesterone is a decrease of channel wcP o (10). The effect of progesterone on wcP o was previously determined, using an assay based on the binding of an iodinated monoclonal antibody to a FLAG epitope introduced in the extracellular loops of ␣␤␥ENaC subunits (6). To avoid some difficulty inherent to a binding assay, we have developed an independent procedure to quantitate N. This procedure is based on the biotinylation of cell-surface-expressed ENaC subunits (see "Experimental Procedures"). The biotinylated proteins were precipitated with streptavidin-agarose beads and electrophoresed through SDS-PAGE. In parallel, 10% of intracellular proteins recovered after precipitation of biotinylated proteins were loaded on the same gel. The proteins were blotted with anti-␣, anti-␤, and anti-␥ ENaC antibodies. As a control for the biotinylation procedure, the same blots were re-probed with an anti-actin antibody. To validate this procedure, we quantified the cell-surface expression of the wild-type channel and the channel composed of the wild-type ␣ and ␤ subunits and the ␥ subunit with mutated PY motif (␥ Y628A mutant). The use of the ␥ subunit rather than the ␤ subunit mutant was chosen, because the epitope for the anti-␤ subunit antibody encompassed the PY motif. As shown in Fig. 4A, the anti-␣ antibody specifically recognizes several molecular species of the ␣ subunit expressed at the cell surface: a band of ϳ82 kDa corresponding to the glycosylated form of full-length ␣ENaC protein, a band of ϳ65 kDa and the double bands of ϳ25 kDa. Because the epitope for this antibody is located in the N-terminal part of ␣ENaC, the double bands of ϳ25 kDa could correspond to the N-terminal products of furin-dependent ␣ENaC cleavage described by Hughey et al. (15). The ϳ65-kDa band represents an as-yet uncharacterized ␣ENaC protein species, but could correspond to a species of ϳ65 kDa observed in airway cells, using a similar protocol and the same antibody (16,17). The intensity of both the ϳ82-kDa band and the ϳ25-kDa double bands was significantly increased in the ␣␤␥ Y628A channel, whereas the intensity of the ϳ65-kDa band remained unchanged. In the intracellular protein pool, the anti-␣ antibody specifically recognizes two ␣ENaC species; one, which is identical to the ϳ65-kDa band of the cell-surfaceexpressed ␣ subunit and the second one, which is migrating at ϳ78 kDa. A possible explanation for the difference between ␣ENaC migration profile in the biotinylated and non-biotinylated protein pools consists of the presence in the intracellular pool of important amounts of yolk proteins migrating in a range of molecular masses of ϳ80 -100 kDa (see Ponceau red protein staining in Fig. 4D). This large amount of yolk proteins in oocytes could be a confounding factor for the migration of other intracellular proteins in this range of molecular masses, including the 82-kDa band of ␣ENaC.
As shown in Fig. 4B, the anti-␤ antibody specifically recognizes two molecular species of ␤ENaC expressed at the cell surface: a major band of ϳ110 kDa that corresponds to the glycosylated form of full-length ␤ENaC protein and an as-yet uncharacterized minor band of ϳ75 kDa. The intensity of both ϳ95-kDa and 75-kDa bands was significantly increased in ␣␤␥ Y628A channel. In the intracellular protein pool, the ␤ENaC antibody recognizes two bands: one, which is identical to the minor ϳ75-kDa band, and the second one, migrating at ϳ100 kDa. As discussed above, the difference in migration profiles of cell surface and intracellular ␤ENaC potentially comes from the interference between ␤ENaC and yolk proteins migration. As shown in Fig. 4C, probing of biotinylated proteins with an anti-␥ENaC antibody reveals a ϳ85-kDa band corresponding to the glycosylated form of full-length ␥ENaC protein and the double bands of ϳ63 kDa. These double bands could correspond to the furin-cleaved ␥ENaC described by Hughey et al. (15). Cell surface expression of both the ϳ 65-kDa band and the ϳ63-kDa double bands was significantly increased in ␣␤␥ Y628A mutant channel. Probing of intracellular proteins with the anti-␥ antibody revealed that this antibody strongly cross-reacts with oocyte yolk proteins. Importantly, only intracellular protein pools were positive for anti-actin staining, thus demonstrating that only cell-surface-expressed proteins were biotinylated in our experiments (Fig. 4E). The absence of the 25-kDa ␣ENaC species and a low amount of the 63-kDa ␥ENaC species in the intracellular pool of proteins (Fig. 4, A and C, respectively) indicate that only a minor fraction of cell-surface-expressed ␣ and ␥ ENaC subunits remains non-biotinylated.
Collectively, these experiments demonstrated that the anti-␣, anti-␤, and anti-␥ ENaC antibodies can be used for Western blot detection of cell-surface-expressed ENaC subunits. These experiments also indicate that quantitative analysis of Western blots can be used for the quantification of ENaC expression at the cell surface. However, a complex pattern of molecular species for ␣ and ␥ subunits could potentially reduce the precision of estimates for these subunits. Thus, for quantitative analysis of ENaC at the cell surface, we used the ␤ subunit, which is represented by a singe major band of ϳ95 kDa. As shown in Fig. 5, the ␣␤␥ Y628A channel exhibits 3.13-fold increased I Na , as compared with the wild-type channel, whereas cell-surface expression is increased only by 2.18-fold. These results fit well with the previously observed double effect of the Liddle mutation on N and P o (6) and, thus, validate the use of biotinylation for quantification of ENaC expression at the cell surface.
Progesterone Down-regulates ENaC Activity by Decreasing Channel P o -To assess the mechanism of the progesterone effect on ENaC activity, we performed a parallel analysis of I Na and cell-surface expression of ENaC in oocytes treated with progesterone. As shown in Fig. 6A, 4-h treatment with progesterone resulted in a ϳ80% inhibition of I Na . Biotinylation experiments presented in Fig. 6B revealed no significant difference for ␤ENaC expression, neither at the cell surface, nor in the intracellular pool. Paired statistical analysis of nine independent experiments revealed a ϳ20% non-significant reduction in N in oocytes treated with progesterone (Fig. 6C). Interestingly, we have previously observed a ϳ20% non-significant reduction of N under progesterone treatment by the antibody binding assay (10). Collectively, these results demonstrate that the major effect of progesterone is the decrease of channel wcP o .

Nedd4-2 Has Two Distinct Effects on ENaC Function in the Xenopus
Oocyte Expression System-Ubiquitylation is a common post-translational modification, consisting in the covalent transfer of one or multiple copies of ubiquitin, a 76-amino acid protein, to the target proteins. Although ubiquitylation is usually considered as a signal for protein degradation by the proteasome, the attachment of ubiquitin can also regulate protein localization and/or activity, independent of proteolysis. For example, ubiquitylation has been shown to act as a regulated local-ization signal for nuclear or integral membrane proteins (reviewed in Ref. 18). Ubiquitylation has also been demonstrated to control the activity of components of the endocytic machinery, the activity of transcription factors and the kinase activity (reviewed in Ref. 19).
Ubiquitylation is a multiple step reaction in which attachment of ubiquitin to the target proteins is catalyzed by ubiquitin protein ligases, or E3 enzymes. Numerous studies have demonstrated that Nedd4-2, an E3 enzyme, is capable of ubiquitylation of ENaC subunits (20,21). This ubiquitylation has been proposed to down-regulate the activity of ENaC by decreasing the number of active channels (N) at the cell surface (22)(23)(24). This hypothesis implies that binding of Nedd4-2 to C-terminal PY motifs leads to ubiquitylation and degradation of ENaC subunits through the ubiquitin-26 S proteasome pathway. This hypothesis, however, does not take into account the fact that the suppression of ENaC-Nedd4-2 interaction in a Liddle syndrome mutant also results in a parallel increase in channel wcP o (6). To explain this double effect of PY mutation, one might propose either the existence of a Nedd4-2-inde-  pendent, but PY motif-dependent mechanism, regulating channel P o , or alternatively, the involvement of Nedd4-2 in regulating both N and P o .
Our present study favors the latter, showing that Nedd4-2 can be directly involved in the regulation of ENaC P o . In Xenopus oocytes, the activationoftheMAPkinasecascadebyprogesteroneleadstotheNedd4-2dependent decrease in channel wcP o . Importantly, this effect of progesterone requires both Nedd4-2 binding to ENaC via PY motifs and Nedd4-2 catalytic activity. Because Nedd4-2 binding to PY motifs is also required for regulation of channel N, the difference may arise from the amino acids lying outside of the canonical PY motif involved in ENaC-Nedd4-2 binding. Our data demonstrate that ␤Thr-613 and/or ␥Thr-623 are required for the full effect of progesterone on ENaC. Shi et al. (9) have shown that phosphorylation of these amino acids by ERK facilitates ENaC-Nedd4 interaction. One may propose that these additional binding sites may be responsible for the novel effect of Nedd4-2 on channel P o . Alternatively, progesterone treatment may lead to Nedd4-2-dependent ubiquitylation of non-ENaC proteins. Also, we cannot formally rule out the possibility that progesterone affects the channel conductance, but it is, at present, a very unlikely possibility, because the C terminus of either ␤ or ␥ ENaC subunit is located far away from the channel pore region. All mutations of the C termini so far studied have changed either N and/or P o but not channel conductance.
Previously, several lysine residues in N-terminal parts of ␣ and ␥ ENaC subunits have been identified as the targets for Nedd4-2-dependent ubiquitylation (25). The major effect of the mutation of these lysine residues was an increase of channel N. Whether ubiquitylation of other lysines may influence the mode of Nedd4-2 action on ENaC remains to be established. Recently, Rajan et al. (26) have demonstrated that addition of SUMO, a ubiquitin-like protein, to a single lysine residue in the K2P1 (TWIK) potassium channel results in a complete silencing of this channel. Removal of SUMO by a desumoylation enzyme restored channel activity. This study provided the proof of principle that a ubiquitylation-like mechanism can be involved in the regulation of the open probability of ion channels.
Hormonal Regulation of ENaC: N versus P o ?-The N versus P o regulation of ENaC activity in different physiological or pathophysiological situations remains a matter of controversy (reviewed in Ref. 1). We will briefly discuss the mechanisms of ENaC regulation by aldosterone in the context of the Liddle syndrome, because it involves a lack of interaction between the ENaC ␤ and ␥ subunits with Nedd4-2.
Aldosterone-dependent Signaling Cascade-Dahlmann et al. (27) have observed that the mineralocorticoid regulation of ENaC was fully maintained in a mouse model of Liddle syndrome. The I Na through epithelial sodium channels was measured in the principal cell of the cortical collecting duct (CCD) of mice expressing mutated ␤ENaC subunit. The Liddle mice had a much larger I Na measured in the principal cell when the animals were fed with a low sodium diet or when they were infused with aldosterone. The I Na from Liddle mice was also larger when animals were pretreated with a high potassium diet, a maneuver that increases plasma aldosterone levels. Noise analysis of channel activity indicated that the open probability was similar in the two genotypes. Thus the increase in whole cell current was attributed to a difference in the density of conducting channels, but as discussed by the authors, their method cannot distinguish between de novo insertion of channels into the plasma membrane from activation of near silent channels (P o Ͻ 0.05) already expressed at the surface. In addition, our own measurement of wcP o (6) suggests that the number of channels that can be measured by noise or patch clamp analysis represents only a minor fraction of ENaCs expressed at the cell surface. Immunolocalization studies also show a translocation of the ENaC subunit at the apical membrane under salt restriction (28). No channel protein could be detected by immunostaining in salt-repleted animals. Again, the apparent lack of channels in the apical membrane of principal cells is probably due to the difficulty to detect by immunostaining a small number of channels, which are postulated to be expressed under a standard salt diet. Indeed, if there was no functional channels under this physiological condition, one should not observe severe, life-threatening, salt-loosing syndromes in pseudo-hypoaldosteronism type 1 patients or in mouse models in which ENaC activity has been drastically reduced or totally abolished (29,30). The absence of a P o effect by aldosterone may be methodological rather than biological. The apparent controversy will be, however, difficult to solve as long as the measurement of whole cell P o will be restricted to large cells (oocytes) and will remain technically not feasible in small somatic cells (CCD principal cells) Physiological Relevance of ERK-dependent Regulation of ENaC Activity-What could be the physiological relevance of the present data? To our knowledge, under physiological conditions, progesterone does not have any significant effect on wild-type ENaC activity in the distal nephron. However, the ERK-dependent inhibition of ENaC activity is the principal mechanism of the epidermal growth factor-induced inhibition of transepithelial sodium transport in the kidney CCD. Shen and Cotton (7) have shown that acute addition of epidermal growth factor to the basolateral side of the CCD monolayer leads to a rapid ERK1/2 phosphorylation/activation and a parallel inhibition of I Na . This effect was abolished by pretreatment of CCD monolayers with an inhibitor of ERK pathway. Veizis and Cotton (31) have proposed that apical delocalization of the epidermal growth factor receptor, leading to the hyperactivation of ERK1/2 and the sustained inhibition of I Na , can contribute to the pathogenesis of an autosomal recessive form of a polycystic kidney disease. The mechanism of epidermal growth factor action on N and/or P o was not reported in these studies. Our present data indicate that P o inhibition could be the principal mechanism.