Regulation of Epithelial Sodium Channel Trafficking by Proprotein Convertase Subtilisin/Kexin Type 9 (PCSK9)*

Background: The epithelial Na+ channel ENaC functions as a pathway for Na+ absorption across epithelia. Results: PCSK9 reduced ENaC expression at the cell surface by enhancing its proteasomal degradation. Conclusion: PCSK9 inhibits ENaC-mediated Na+ absorption. Significance: These findings provide new insights into mechanisms that regulate Na+ homeostasis and blood pressure. The epithelial Na+ channel (ENaC) is critical for Na+ homeostasis and blood pressure control. Defects in its regulation cause inherited forms of hypertension and hypotension. Previous work found that ENaC gating is regulated by proteases through cleavage of the extracellular domains of the α and γ subunits. Here we tested the hypothesis that ENaC is regulated by proprotein convertase subtilisin/kexin type 9 (PCSK9), a protease that modulates the risk of cardiovascular disease. PCSK9 reduced ENaC current in Xenopus oocytes and in epithelia. This occurred through a decrease in ENaC protein at the cell surface and in the total cellular pool, an effect that did not require the catalytic activity of PCSK9. PCSK9 interacted with all three ENaC subunits and decreased their trafficking to the cell surface by increasing proteasomal degradation. In contrast to its previously reported effects on the LDL receptor, PCSK9 did not alter ENaC endocytosis or degradation of the pool of ENaC at the cell surface. These results support a role for PCSK9 in the regulation of ENaC trafficking in the biosynthetic pathway, likely by increasing endoplasmic reticulum-associated degradation. By reducing ENaC channel number, PCSK9 could modulate epithelial Na+ absorption, a major contributor to blood pressure control.

The epithelial Na ϩ channel (ENaC) 2 plays an important role in absorption of Na ϩ across epithelia, including the kidney, collecting duct and connecting tubule, lung, distal colon, and sweat duct (reviewed in Refs. 1,2). A heterotrimer composed of three homologous subunits (␣, ␤, and ␥), ENaC is expressed at the apical membrane where it forms a pathway for Na ϩ to enter the cell. Na ϩ leaves the cell at the basolateral membrane via the Na ϩ -K ϩ -ATPase, which completes the pathway for Na ϩ absorption. This process is critical to control extracellular volume and to maintain the composition and quantity of epithelial surface liquid. This is illustrated by several diseases. For example, ENaC mutations that slow its retrieval from the cell surface cause an inherited form of hypertension (Liddle's syndrome), resulting from excessive renal Na ϩ absorption (3)(4)(5). Defects in ENaC regulation are responsible for most of the known genetic forms of hypertension (6). Conversely, loss of function mutations cause pseudohypoaldosteronism type I, a disorder of renal Na ϩ wasting (7). In the lung, defects in ENaC activity cause pulmonary edema and may contribute to the pathogenesis of cystic fibrosis (8).
Previous work indicates that ENaC is regulated by serine proteases (reviewed in Ref. 9). Furin, a member of the proprotein convertase family, cleaves the extracellular domain of ␣ENaC at basic motifs, removing a 26-amino acid fragment (10). In ␥ENaC, furin cleaves the extracellular domain at a single site and a second, more distal site is cleaved by additional proteases (e.g. CAP1/prostasin, plasmin, elastase), releasing a fragment of ϳ43 amino acids (11)(12)(13)(14)(15). Proteolytic cleavage of ␣and ␥ENaC converts quiescent channels into active Na ϩ -conducting channels. This activation occurs by relieving the channel from inhibition by extracellular Na ϩ ("Na ϩ self-inhibition") (16). Proteolytic cleavage of ENaC is a regulated process. For example, cleavage is inhibited by increased intracellular Na ϩ , providing a negative feedback mechanism to regulate Na ϩ absorption (17). Conversely, cleavage is enhanced by Na ϩ depletion and aldosterone infusion (18,19). Cleavage is also disrupted in pathological states. In Liddle's syndrome, cleavage is increased, likely through prolonged exposure of ENaC to proteases present at the cell surface (5). There is also evidence to suggest that ENaC cleavage is increased in nephrotic syndrome (15,20) and cystic fibrosis (21,22).
Because proteolytic cleavage modulates ENaC gating, there has been considerable interest in identifying additional proteases that regulate ENaC. The proprotein convertase family has nine members, including furin (23). In this work, we investigated a potential role for another member of this family, proprotein convertase subtilisin/kexin type 9 (PCSK9) (24). Consistent with a potential role in ENaC regulation, PCSK9 is expressed in the kidney and lung (24). It is synthesized as a 72-kDa immature precursor that undergoes autocatalytic cleavage in the endoplasmic reticulum to generate a 63-kDa mature protein (25). The cleaved N-terminal fragment remains associated with the mature protein and is necessary for its secretion, allowing it to circulate in the blood (26).
Previous work has focused on the role of PCSK9 in the regulation of the LDL receptor (LDLR). By reducing expression of the LDLR at the cell surface, PCSK9 increases serum levels of LDL cholesterol (25,27,28). Rare gain-of-function PCSK9 mutations cause hypercholesterolemia and increase the risk of coronary heart disease, whereas loss-of-function mutations cause hypocholesterolemia and protect against heart disease (29 -33). The mechanisms by which PCSK9 alters LDLR surface expression are not completely understood. Secreted PCSK9 (or recombinant PCSK9 added to the extracellular medium) binds to the LDLR and undergoes endocytosis (26, 27, 34 -36). In the endocytic pathway, PCSK9 increases lysosomal degradation of the LDLR. Although secreted PCSK9 regulates LDLR trafficking, additional evidence suggests that PCSK9 may also induce LDLR degradation through an intracellular route (37). Interestingly, although PCSK9 induces degradation of the LDLR, its protease activity is not required (33,38). Thus, it has been proposed that PCSK9 regulates the LDLR through a chaperone mechanism, rather than through its function as a protease. Although it seems clear that the PCSK9 regulates the LDLR and two closely related receptors (very low density lipoprotein receptor and apolipoprotein E receptor 2 (38)), additional substrates for PCSK9 have not been identified. Here we show that PCSK9 regulates ENaC and we explore the mechanisms that underlie this regulation.
Oocytes were voltage-clamped at Ϫ60 mV, and currents were recorded by two-electrode voltage clamp using an oocyte clamp (OC-725C, Warner Instruments), digitized with a Powerlab interface (ADInstruments), and recorded and analyzed with Chart software (ADInstruments). The cells were bathed in 116 mM NaCl, 2 mM KCl, 0.4 mM CaCl 2 , 1 mM MgCl 2 , and 5 mM HEPES (pH 7.4 or 5). The amiloride-sensitive ENaC current was measured by adding 10 M amiloride to the bathing solution. ASIC1 currents were detected by addition of pH 5 to the bathing solution.
For exocytosis experiments, FRT cells were transfected with ␣ENaC, ␤ENaC, and ␥ 536C ENaC (0.167 g each subunit) with PCSK9 or GFP cDNA (0.5 g) (43,49). Channels at the cell surface were irreversibly blocked by covalent modification of the introduced cysteine with 1 mM [2-(trimethylammonium)ethyl]methanethiosulfonate bromide (MTSET). Following removal of MTSET, we measured the rate of current increase to quantitate exocytosis of unblocked channels. Time constants () were determined by fitting the data to single-exponential equations using IGOR Pro 6.01 software.
Degradation-To measure the rate of ENaC degradation, HEK 293 cells transfected with ␣ENaC-FLAG, ␤ENaC, and ␥ENaC with PCSK9 or GFP were incubated with cycloheximide (10 g/ml) for 0 -120 min. Remaining ␣ENaC-FLAG at each time point was detected by immunoblot (anti-FLAG M2-peroxidase-conjugated antibody) and quantitated by densitometry. To identify the location of degradation, cells were treated with 10 M N-acetyl-Leu-Leu-norleucinal or 5 mM NH 4 Cl.
To measure the rate of degradation of the cell surface fraction of ENaC, HEK 293 cells transfected with ␣ENaC-FLAG, ␤ENaC, and ␥ENaC with PCSK9 or GFP were biotinylated on ice and then incubated at 37°C for 0 -120 min (5,41). Biotinylated ␣ENaC-FLAG was isolated using NeutrAvidin-agarose, detected by immunoblot (anti-FLAG M2-peroxidase-conjugated antibody), and quantitated by densitometry.
Endocytosis-To measure the rate of ENaC endocytosis, we used a previously described ␣ENaC construct (␣ Cl-2 ) in which multiple arginines were simultaneously mutated to prevent proteolytic cleavage by furin but to retain the ability to be cleaved by trypsin (R175A, R177A, R178A, R181A, R190A, R192A, R201A, and R204A) (44). HEK 293 cells were transfected with ␣ Cl-2 ENaC-FLAG, ␤ENaC, and ␥ENaC with PCSK9 or GFP were incubated with trypsin (5 g/ml) for 5 min at 37°C to generate a pool of cleaved channels at the cell surface (9)(44). The cells were washed three times with cold PBS-CM to remove trypsin, incubated at 37°C for 0 -60 min to allow endocytosis of cleaved channels, and then placed on ice. Cleaved channels remaining at the cell surface were labeled with biotin, isolated with NeutrAvidin-agarose, detected by immunoblot analysis (anti-FLAG M2-peroxidase-conjugated antibody), and quantitated by densitometry.

RESULTS
PCSK9 Inhibits ENaC-We tested the effect of PCSK9 on ENaC current utilizing two expression systems. First, we injected Xenopus oocytes with ␣-, ␤-, and ␥ENaC cDNA to generate amiloride-sensitive Na ϩ currents (Fig. 1A). We found that coexpression of PCSK9 decreased the Na ϩ current in a dose-dependent manner (Fig. 1, A and B).
As a second strategy, we tested the effect of PCSK9 on the ENaC current in epithelia. Transfection of FRT epithelia with ␣-, ␤-, and ␥ENaC resulted in amiloride-sensitive short-circuit currents (Fig. 1C). Cotransfection with PCSK9 produced a dose-dependent decrease in ENaC current (Fig. 1, C and D), similar to our results in oocytes. Thus, PCSK9 inhibited ENaC in two independent experimental systems.
We also tested the effect of PCSK9 on a related DEG/ENaC channel, ASIC1. PCSK9 reduced the proton-activated ASIC1 current by 24% in Xenopus oocytes (Fig. 1, E and F), less than its effect on ENaC.
PCSK9 Interacts with ENaC-To begin to investigate the mechanism by which PCSK9 inhibits ENaC current, we tested whether PCSK9 and ENaC interact with one another. In Fig.  2A, we transfected HEK 293 cells with ␣-, ␤-, and ␥ENaC (one of the subunits contained a FLAG epitope) along with PCSK9 (V5 epitope) and examined protein interactions using a coimmunoprecipitation assay. When we immunoprecipitated ␣ENaC, we detected coprecipitated PCSK9 in cells cotransfected with ENaC and PCSK9 but not in cells transfected individually with either ENaC or PCSK9 ( Fig. 2A, first panel). Likewise, we detected PCSK9 when we immunoprecipitated ␤or (0 -0.9 g). C, representative short-circuit current traces. 10 M amiloride was added to the apical bathing solution as indicated by the black bar. D, summary plot of amiloride-sensitive current versus the amount of transfected PCSK9 (mean Ϯ S.E. relative to 0 PCSK9 group, n ϭ 9 -12). E and F, Xenopus oocytes were nuclear-injected with cDNAs encoding human ASIC1 (0.6 ng) and PCSK9 or control plasmid (0.8 ng). E, representative current traces. The bath was perfused with pH 5 solution as indicated by the black bar. F, summary plot of proton-activated current (mean Ϯ S.E.; n ϭ 9; *, p Ͻ 0.03). ␥ENaC ( Fig. 2A, first panel). There are two forms of PCSK9, full-length pro-PCSK9 (72 kDa) and autocatalytically cleaved PCSK9 (63 kDa) ( Fig. 2A, third panel) (25). ENaC selectively coprecipitated the pro-PCSK9 form ( Fig. 2A, first panel). Using a reciprocal strategy we found that ␣-, ␤-, and ␥ENaC each coprecipitated when we immunoprecipitated PCSK9 ( Fig. 2A,  second panel). In the immunoblot analysis, we observed two bands for ␣and ␥ENaC, which correspond to the full-length (immature) and proteolytically cleaved (mature) forms, respectively (␤ENaC does not undergo cleavage). The bands that coprecipitated with PCSK9 correspond to the full-length forms of ␣and ␥ENaC.
Because ␣-, ␤-, and ␥ENaC form a complex, we asked if each of the individual subunits could also bind to PCSK9. We cotransfected HEK 293 cells with one of the ENaC subunits, with or without PCSK9. When we immunoprecipitated each of the ENaC subunits, we detected pro-PCSK9 by immunoblot (Fig. 2B, top panel). Thus, the data indicate that PCSK9 interacts with each of the three ENaC subunits. Moreover, the inter-actions occur selectively between the uncleaved immature forms of PCSK9 and ENaC.
PCSK9 Reduces ENaC Cell Surface Expression-We asked whether PCSK9 inhibits ENaC current through a change in ENaC surface expression. In Fig. 3A, we used a biotinylation assay to detect the cell surface fraction of ␣ENaC (coexpressed with ␤and ␥ENaC) in HEK 293 cells. Fig. 3B shows quantitative summary data. PCSK9 decreased both the full-length and proteolytically cleaved forms of ␣ENaC at the cell surface. This decrease in surface expression corresponded to a decrease in ␣ENaC in the total cellular pool, as detected by immunoblot analysis of cell lysates (Fig. 3A, bottom panel, and B; also see Fig.  2A). PCSK9 produced a similar decrease in expression of ␤and ␥ENaC at the cell surface and in ␤ENaC in the total cellular pool (Fig. 3, A and B). As negative controls, PCSK9 had no effect on the abundance of heterologously expressed Nedd4-2 or endogenous ␤-actin (Fig. 3C). These results indicate that PCSK9 inhibits ENaC current by reducing the number of channels at the cell surface.
To determine whether PCSK9 also regulates ENaC gating, we took advantage of a mutation that locks ENaC in the open state ("DEG" mutation, ␤ S520K ) (50). If PCSK9 inhibits ENaC in part through a change in gating, this mutation should blunt the effect. However, we found that PCSK9 inhibited mutant ENaC to the same extent as wild-type ENaC (Fig. 3D). This finding indicates that the changes we observed in ENaC surface expression are sufficient to explain PCSK9 inhibition of ENaC.
Prior work has shown that the PY motifs located in the C termini of ENaC subunits play an important role in trafficking (51). The PY motifs function as binding sites for Nedd4-2, an E3 ubiquitin ligase that catalyzes ENaC ubiquitination. This functions as a signal to induce ENaC endocytosis and degradation in lysosomes. Importantly, mutations in the PY motifs cause Liddle's syndrome, an inherited form of hypertension. To test whether the PY motifs are required for ENaC regulation by PCSK9, we mutated the conserved tyrosine residue within the motif of each ENaC subunit (␣ Y644A , ␤ Y620A , and ␥ Y627A ). In Fig.  3E, we found that PCSK9 reduced surface expression of the mutant ENaC. This finding suggests that PCSK9 regulates ENaC surface expression through a pathway that is independent of the PY motifs and Nedd4-2.
To test whether protease activity is needed for PCSK9 to reduce ENaC surface expression, we introduced a mutation that abolishes proteolytic activity (S386A) (33). PCSK9 S386A decreased ␣ENaC expressed at the cell surface and in the total cellular pool similar to wild-type PCSK9 (Fig. 3F). Thus, the catalytic activity is not required for PCSK9 to regulate ENaC, similar to its regulation of the LDLR.
PCSK9 Increases ENaC Degradation-To further investigate the mechanism by which PCSK9 reduced ENaC cell surface expression, we asked whether PCSK9 alters ENaC degradation using a cycloheximide chase assay. HEK 293 cells expressing ENaC and PCSK9 or GFP (control) were treated with cycloheximide for 0 -120 min to inhibit protein synthesis. In Fig. 4, A and B, we detected and quantitated the remaining ␣ENaC-FLAG at each time point by immunoblot analysis. In the absence of PCSK9, there was no significant decrease in ␣ENaC over the 120-min time course of the experiment. In contrast, in cells transfected with PCSK9, there was a time-dependent decrease in ␣ENaC. This finding indicates that PCSK9 accelerates the rate of ENaC degradation.
To localize the site of the PCSK9-induced ENaC degradation, we incubated cells with inhibitors of the proteasome (N-acetyl-Leu-Leu-norleucinal) or lysosomes (NH 4 Cl). We found that N-acetyl-Leu-Leu-norleucinal partially reversed the effect of PCSK9 on ␣ENaC expression, whereas NH 4 Cl had no effect (Fig. 4C). Together, the data indicate that PCSK9 reduces ENaC surface expression in part by enhancing its degradation in the proteasome.
Effect of PCSK9 on ENaC Exocytosis-ENaC surface expression is controlled through a balance between exocytosis of newly formed channels, endocytosis of cell surface channels, and recycling of channels in the endocytic pathway. Because PCSK9 increased ENaC degradation, it seemed likely that PCSK9 would reduce the pool of ENaC available for exocytosis. To test this possibility, we used a functional strategy we reported previously (49). We covalently modified the cell surface pool of ENaC and then measured the rate of appearance of unmodified channels at the cell surface. For these experiments, we placed a cysteine in the pore of ␥ENaC (G536C) (43,49). When the mutant ␥ subunit was coexpressed in FRT epithelia with wild-type ␣and ␤ENaC, MTSET irreversibly blocked the channel by modifying the introduced cysteine, as shown in the representative current traces in Fig. 5, A and B. Because MTSET is not membrane-permeable, the intracellular ENaC pool was protected from modification. Following removal of MTSET from the bathing solution, we measured the increase in ENaC current over time as an assay of exocytosis of unmodified/unblocked channels. Fig. 5C shows the averaged time courses of current recovery, and the time constants are shown in D. PCSK9 reduced the maximal increase in current recovery over time but did not significantly alter the time course of the increase, as reflected by the lack of difference in the rate constant. These results are consistent with a reduction in the size of the ENaC pool available for exocytosis. However, the observed decrease in current recovery could also be explained by an increase in the rate of ENaC endocytosis. We therefore tested the effect of PCSK9 on ENaC endocytosis.  Fig. 6, we quantitated endocytosis of mature proteolytically cleaved ENaC using a method that we described previously (44). We mutated the two furin consensus sites in the extracellular domain of ␣ENaC to prevent cleavage by furin (␣ Cl-2 ) and then treated the cells briefly with trypsin to generate a pool of proteolytically cleaved channels (65-kDa band). After removal of trypsin, we incubated the cells at 37°C for 0 -60 min to allow endocytosis of the cleaved channels and then biotinylated and detected channels remaining at the cell surface. The disappearance of the 65-kDa band reflects the rate of ENaC endocytosis. Because newly synthesized mutant channels reaching the cell surface are uncleaved, they can be distinguished from the cleaved channels undergoing endocytosis. As shown in Fig. 6, A and B, the 65-kDa cleaved band was rapidly removed from the cell surface with a half-life of ϳ15 min, consistent with our prior work (44). However, PCSK9 did not alter the rate of ENaC disappearance from the cell surface. Thus, PCSK9 does not alter ENaC surface expression through a change in endocytosis.

PCSK9 Does Not Alter Trafficking of the Cell Surface ENaC Pool-In
Following endocytosis, ENaC can either recycle back to the cell surface or traffic to lysosomes for degradation. To test whether PCSK9 alters ENaC surface expression in part by regulating this sorting step, we measured the effect of PCSK9 on degradation of the cell surface pool of ENaC. In HEK 293 cells expressing ENaC, we pulse-labeled the cell surface fraction of channels with biotin, incubated the cells at 37°C for 0 -120 min, and then quantitated the remaining (non-degraded) biotinylated channels. PCSK9 did not increase the rate of degradation of biotinylated ␣ENaC (Fig. 7, A and B). Rather, it slightly delayed degradation, which may partially counter the effect of PCSK9 on ENaC degradation in the biosynthetic pathway.

DISCUSSION
Recent work has focused on the regulation of epithelial Na ϩ transport by proteases, including furin, a member of the proprotein convertase family (9). Here we found that ENaC is regulated by PCSK9, another member of this protease family. However, furin and PCSK9 have opposite effects on ENaC current and they regulate ENaC through different mechanisms. In contrast to furin, which activates ENaC by proteolytic cleavage of the extracellular domains of the ␣ and ␥ subunits, PCSK9 inhibits ENaC by reducing its cell surface expression. Moreover, unlike furin, PCSK9 regulates ENaC independent of its protease activity.
The data indicate that PCSK9 reduces ENaC surface expression primarily by increasing its degradation in the biosynthetic pathway, which reduces the pool of ENaC available for exocytosis. Consistent with this concept, we found that PCSK9 decreased ENaC exocytosis and increased the rate of ENaC degradation in the proteasome. Moreover, PCSK9 had no effect on the rate of ENaC endocytosis or its degradation in the endocytic pathway. Coimmunoprecipitation studies suggest that ENaC and PCSK9 interact with one another in their immature (uncleaved) states, likely prior to ENaC cleavage in the Golgi apparatus (although we cannot exclude the possibility that proteolytic cleavage induces conformation changes that prohibit binding). Together, these findings are most consistent with a model in which PCSK9 enhances endoplasmic reticulum associated degradation of ENaC.
The mechanism by which PCSK9 regulates ENaC shares some similarities to its regulation of the LDLR. In both cases, PCSK9 reduces surface expression through a change in trafficking, culminating in increased degradation. Likewise, both occur independent of PCSK9 protease activity. However, there are also important differences. Contrary to regulation of ENaC in the biosynthetic pathway, PCSK9 predominately regulates the LDLR in the endocytic pathway (34,35). This suggests that PCSK9 can function in multiple cellular compartments. There are also differences in the mechanisms of binding. PCSK9 interacts with the LDLR through two interfaces. Crystallization revealed that the catalytic domain of PCSK9 binds to the epidermal growth factor-like A domain in the LDLR (36). A recent report found a second interaction between the PCSK9 C-terminal domain and the LDLR ligand binding domain, which was required for regulation (52). Importantly, ENaC lacks homologous motifs, indicating that it interacts with PCSK9 through a novel binding mechanism. This interaction could be direct or could occur indirectly through an adaptor protein.
The proprotein convertase furin regulates ENaC by proteolytic cleavage of the extracellular domains of ␣and ␥ENaC, which releases inhibitory peptides (9). In this manner, furin regulates ENaC gating, converting near-silent channels into active channels. However, there may be an additional level of complexity. Furin also proteolytically cleaves PCSK9, which inactivates it (53). Thus, through a decrease in PCSK9 activity, furin could increase ENaC cell surface expression. This raises the interesting possibility that furin regulates ENaC through dual effects on channel trafficking and gating. Defects in ENaC regulation are responsible for the majority of the known genetic forms of hypertension, which is an important risk factor for coronary heart disease and other cardiovascular diseases. Thus, PCSK9 could modulate cardiovascular risk in part through its regulation of ENaC. We speculate that a decrease in PCSK9 activity would increase renal Na ϩ absorption and, therefore, raise the risk of hypertension and associated cardiovascular disease. However, such a mechanism would counter the previously reported effect of PCSK9 mutations on cardiovascular risk. Activating mutations were found to increase the risk, whereas loss-of-function mutations reduced the risk (29 -33). These effects are thought to occur through changes in expression of the LDLR, which produce changes in serum levels of cholesterol. Thus, it is possible that PCSK9 reg-ulation of ENaC and the LDLR have opposing effects on cardiovascular risk. On the other hand, our data suggest that PCSK9 regulates ENaC and the LDLR through different binding sites and different mechanisms. Thus, the PCSK9 mutations that disrupt LDLR regulation may have dissimilar effects on ENaC. Additional work will be required to test whether naturally occurring mutations in PCSK9 alter ENaC trafficking, renal Na ϩ homeostasis, and blood pressure. . PCSK9 decreases ENaC exocytosis. Short-circuit currents were recorded in FRT epithelia transfected with ␣-, ␤-, and ␥ G536C ENaC subunits (0.16 g each) with or without PCSK9 (0.5 g). A and B, representative current traces. MTSET (1 mM) was added to the apical membrane as indicated by the black bars. C, summary data for the amiloride-sensitive current in the absence and presence of PCSK9 (mean Ϯ S.E., n ϭ 7). D, time constants for single exponential fit of the data in C (mean Ϯ S.E.; n ϭ 7; n.s., p Ն 0.05). FIGURE 6. PCSK9 does not alter ENaC endocytosis. A, immunoblot analysis (anti-FLAG) of biotinylated ␣ Cl-2 ENaC-FLAG in HEK 293 cells cotransfected with ␤and ␥ENaC (1 g each) with or without PCSK9 (3 g). The cells were treated with 5 g/ml trypsin for 5 min, incubated at 37°C for 0 -60 min, and then biotinylated. B, quantification of the cleaved ␣ENaC band relative to 0 min (mean Ϯ S.E., n ϭ 5). Cell surface proteins were pulse-labeled with biotin and then the cells were incubated at 37°C for 0 -120 min. B, biotinylated ␣ENaC at each time was quantified relative to 0 min (mean Ϯ S.E.; n ϭ 3; *, p Ͻ 0.05).