Processing of ENaC in mouse kidney: Effects of aldosterone and a Liddle syndrome mutation

Abstract

We investigated the interplay between the mineralocorticoid aldosterone and a mutation mimicking Liddle syndrome in the control of the processing of the epithelia Na⁺ channel (ENaC) in mouse kidneys. Rates of processing were assessed by the appearance of the cleaved form of the γENaC subunit. Cleaved γENaC increased with decreasing dietary Na intake and with administration of aldosterone. Measurements taken from isolated tubules indicated that enhanced processing was similar in connecting tubules and in late distal convoluted tubules. In a mouse model with a truncated βENaC subunit (Liddle mice), levels of cleaved γENaC were similar in wild-type (WT) and Liddle animals. The amounts of the full-length form of the subunit were lower in the Liddle mice on control and high-Na diets. Infusion of a low dose of aldosterone produced similar increases in cleaved γENaC in WT and Liddle mice, while with maximal doses, levels in Liddle animals were 35% higher than in WT. Acute Na repletion of Na-depleted mice decreased cleaved γENaC with a time constant of 5 hours. Rates of decrease were similar in WT and Liddle genotypes. The Liddle’s mutation produces modest changes in ENaC processing, and a major effect of the mutation is on the activation of processed channels.

Graphical Abstract

INTRODUCTION

The epithelial Na channel (ENaC) controls the amounts of Na⁺ and K⁺ excreted in the urine (1, 2). The adrenal corticosteroid aldosterone increases ENaC activity, at least in part through increased surface expression of the channel protein (3). These increases result mainly from increased forward trafficking of ENaC to the apical membrane and are strongly correlated with channel protein processing including proteolytic cleavage and mature glycosylation (4–6).

Liddle syndrome is a monogenic form of hypertension caused by gain-of-function mutations in ENaC (7–9). Increased ENaC activity has been demonstrated directly in renal tubules isolated from mouse model of the disease (10, 11). Effects of the mutation have been ascribed to decreased channel ubiquitination and turnover expression (12, 13). However, in a heterologous expression system, enhanced Na⁺ currents through mutant channels were only partly explained by increased ENaC surface expression (14).

Aldosterone and the Liddle mutation had synergistic effects on ENaC activity in renal tubules as well as the distal colon (10, 15). These findings suggest that in vivo, lowered levels of aldosterone that occur in both the human disease and in the Liddle mouse strain (9, 16) might compensate for the hyperactivity of the channels. Nesterov et al (11) proposed that Na⁺ retention and hypertension could reflect the effects of the mutation in the late distal convoluted tubule (DCT2) in which ENaC activity was aldosterone- independent.

We have further examined the interactions between aldosterone and Liddle mutations in controlling ENaC processing in mouse kidney. Our results suggest that, at least in this model, activation of processed channels may be the most important factor leading to ENaC hyperactivity.

METHODS

Animals

All procedures using animals were approved by the Institutional Animal Care and Use Committee of Weill Cornell Medicine. For experiments using only wild-type animals we used commercial C57BL/6N mice (Charles River). We also studied a strain of mice with a truncating mutation in the C-terminal of βENaC generated by a stop codon at the R566 position, originally described by Pradervand and colleagues (16). For experiments comparing wild-type and Liddle genotypes we used homozygous wild-type (WT) or Liddle mutation (LL) littermates obtained from breeding heterozygotes. Breeding pairs were a kind gift of Christoph Korbacher of Erlangen University. All the commercial mice used were male. With the Liddle mouse strain, we used both males and females with both homozygote WT and Liddle genotypes, with sexes matched within each experiment. Females tended to have a slightly higher expression of γENaC but we noted no significant differences between the sexes in the ratio of cleaved to full-length species or the responses to dietary or hormonal challenges. When mice of both sexes were analyzed in the same experiment, data were normalized separately for males and females.

Mice were fed either normal lab chow (0.4 % Na), a synthetic diet containing < 0.02% Na (Harlan-Teklad; TD07904), or this same diet with 1% or 5% NaCl added. In some experiments animals were given drinking water containing 150 mM NaCl together with the high-salt diet. In some mice, osmotic minipumps (model 1007D, Alzet Corporation) were implanted subcutaneously. The minipumps contained aldosterone dissolved in PEG300 (Spectrum Chemical) at concentrations up to 4 mg/ml. This stock was diluted further with PEG300 to produce the target rate of infusion of the hormone. Concentrations of aldosterone in plasma were measured using a competitive ELISA kit (Enzo ADI-900–173) according to the manufacturer’s instructions.

For mouse tubule suspensions, kidneys were removed and pieces of cortex from each were finely minced with a razor blade and incubated for 20 min at 37 °C with gentle rocking in 2 ml of solution containing (in mM) 120 NaCl, 25 NaHCO₃, 3.3 KH₂PO₄, 0.8 K₂HPO₄, 1.2 MgCl₂, 1.2 CaCl₂, 10 glucose and 2 mg/ml collagenase type II (Worthington, catalogue # LS004176). Suspensions were washed twice with the same solution without collagenase, triturated gently with a Pasteur pipette, and suspended in 1 ml of the above solution. Distal convoluted and connecting tubules were identified and isolated by hand under a dissecting microscope. They were transferred to microcentrifuge tubes, sedimented by centrifugation for 5 min at 1000 rpm and dissolved in ~50 μl sample buffer.

Western blots

For Western blots, a single kidney from an individual mouse was homogenized with a tight-fitting Dounce in 5 ml of lysis buffer containing (in mM) 250 sucrose, 10 triethanolamine HCl, 4 NaF, 3 Na orthovanadate, 10 Na pyrophosphate at pH 7.40 and 60 μl protease-inhibitor cocktail (APExBio #K1007). Homogenates were sieved with a nylon mesh (70 μm porosity) to separate intact tissue and then centrifuged at 100,000 × g for 100 min to sediment a total membrane pellet. This was resuspended in 2 ml of lysis buffer, aliquoted, and frozen at −70 °C for later analysis. The pellet was resuspended in ~50 μl of sample buffer and stored at −70 °C for later analysis. This pellet was used for all Western blots of kidney microsomes.

Protein in the membrane fractions was measured (BCA Kit, Pierce Biotechnology, Rockford, IL). Samples containing 0.2 mg protein were prepared for electrophoresis in 100 μl of sample buffer with sample reducing agent (Invitrogen) and denatured at 70 °C for 10 min.

For SDS-PAGE, 4–12% bis-Tris gels (Invitrogen, Carlsbad, CA) were loaded with 40–80 μg protein/lane. Proteins were transferred electrophoretically from the gels to PVDF membranes. After blocking, membranes were incubated overnight at 4 °C with primary antibodies. Polyclonal antibodies against a C-terminal epitope of the γ subunit of the rat ENaC were described previously (3, 4, 17). They were diluted 1:500 for Western blots. A polyclonal antibody against NCC (18) was a generous gift of Dr. Alicia McDonough of the U. Southern California. This was diluted 1:5000 for Western blots

Anti-rabbit IgG conjugated with alkaline phosphatase was used as a secondary antibody. Bound antibody was visualized with a Syngene Pxi imager using a chemiluminescence substrate (Western Breeze, Invitrogen). Band densities were quantified using Adobe Photoshop. Blots were inverted to show intensity as an increase in brightness. Boxes were drawn by eye delimiting the strongest band of each species, and the mean intensity was calculated using the measurement function. Background was measured in an identical box placed below the bands being analyzed. For the cleaved γENaC species, the boxes were drawn to include only the band of lowest apparent molecular mass, shown to be the fully cleaved species (19). Densities were within the linear range of the detection system (4).

Data analysis

Kinetic data from Western blots during Na⁺ repletion were fit with exponential functions using Igor (WaveMetrics, Portland, OR). Decreases in cleaved γENaC and increases in the full-length form in response to increased Na⁺ intake were described with single exponential decay processes reflecting first-order kinetics. For the cleaved form:

$$\mathrm{A}\left(\mathrm{t}\right)=1-\mathrm{B}•(1-\mathrm{e}\mathrm{x}\mathrm{p}(-\mathrm{t}/\mathrm{\tau }\left)\right)$$

where A(t) is the abundance as a function of time t after initiation of Na⁺ repletion and B and τ are constants. For the full-length form:

$$\mathrm{A}\left(\mathrm{t}\right)=1+\left({\mathrm{A}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-1\right)•(1-\mathrm{e}\mathrm{x}\mathrm{p}(\mathrm{t}/\mathrm{\tau }\left)\right)$$

where A(t) is the abundance as a function of time and A_max and τ are constants. The same equations were used previously to describe the effects of rapid reduction in aldosterone (20).

In experiments with mice receiving different rates of aldosterone infusion, Western-blot data were fit by a hyperbolic relationship, consistent with a simple first-order binding reaction:

$$\mathrm{A}\left(\mathrm{r}\right)={\mathrm{A}}_0+\left(\left({\mathrm{A}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{A}}_0\right)/\left(1+{\mathrm{r}}_{0.5}/\mathrm{r}\right)\right)$$

Where A(r) is the abundance as a function of the infusion rate r, A₀ is the abundance with no infusion, and A_max and r_0.5 are constants representing respectively the abundance at maximal rates of infusion and the infusion rate at which the effect of the hormone is half-maximal.

Statistical analysis was done with unpaired Student t-tests using Prism (GraphPad Software Inc., San Diego, CA, USA).

RESULTS

We investigated the processing of ENaC in mice under three levels of dietary Na⁺ intake. We have focused on the γENaC subunit because the processed form is most easily and reliably detected by the appearance of a lower molecular mass cleavage product. Processing of βENaC requires assessment of an EndoH-insensitive species, while that of αENaC is complicated by simultaneous increases in the total subunit abundance (4). We assume that the α and β subunits are processed simultaneously with γ (5), but this was not rigorously tested here.

As shown in fig.1, in wild-type (WT) mice the amounts of cleaved and full-length γENaC showed the expected pattern (4, 6), with the cleaved species increasing and the full-length form decreasing with diminishing dietary Na⁺. In some blots (e.g. fig 1A) a band with an intermediate apparent mass was observed. This presumably represents subunits that are cleaved only by furin (19). Since this band could not always be clearly resolved we have not included it in our analyses.

Figure 1.

Effects of dietary Na⁺ and a Liddle’s mutation on γENaC processing. A. Western blots of mouse kidney microsomes from mice on low- Na⁺, control- Na⁺ and high- Na⁺ diets. Each lane from a single blot was loaded with equal amounts of protein (low Na 80 μg; control Na 40 μg; high Na 50 μg) from individual WT or Liddle mouse kidneys. Sexes of the mice are marked on each lane (m = male, f = female) B. Band densities for full-length and cleaved forms were normalized to those of the cleaved form of WT animals. The ratios of cleaved/full-length are plotted as absolute values. Data are shown as means ± SEM for n = 8 (low-Na), or 4 (control Na, high Na). p-values from Student t-tests are shown above the bars.

When fed normal chow with 0.4% Na, Liddle’s mice had a slightly lower amount of cleaved γENaC (30%, p = 0.04), and a more pronounced decrease in the full-length (unprocessed) form (50%, p = 0.001) compared with WT mice (fig. 1). There was no significant difference in the ratio of cleaved/full-length subunits. In animals fed a high-Na diet (2% Na), there was no significant difference in cleaved γENaC but again a significant decrease in the full-length species (50%, p = 0.003), which together lead to a decrease in the cleaved/full-length ratio in the Liddle mice. Finally, in animals on a low-Na diet, no differences in either cleaved or uncleaved species or their ratio could be demonstrated. These results suggest that the effects of this Liddle mutation to increase channel activity during Na restriction (10) do not result from changes in ENaC processing or the abundance of the cleaved, presumably active form of the channel protein. We did not routinely measure plasma electrolytes under these conditions. Previous studies (16) revealed no differences with normal dietary Na⁺ intake but a hypokalemia in Liddle mice when Na⁺ intake was high.

One caveat to this conclusion is that ENaC processing may be heterogeneous along the nephron. In particular, Nesterov et al showed that ENaC activity in the late DCT/early CNT is independent of aldosterone, and that this segment may be an important site of increased Na⁺ retention in Liddle syndrome (11, 21). To test if ENaC processing is different in this part of the nephron we isolated 25–40 tubule fragments from both the late DCT and the CNT (fig 2) and analyzed protein by Western blot. In these preparations CNTs were recognized from their branching pattern, while the distal end of the DCT was located at the point of decreased tubule diameter at the proximal end of the CNT (fig 2A) as described previously (22). Western blots confirmed modest NCC expression in the isolated late DCT fragments, but the cotransporter was not detected in the CNTs (fig 2B). We then compared tubules isolated from mice on either control-Na (0.4%) or low-Na diets. Because individual lanes were loaded with slightly different numbers and lengths of tubules, we analyzed the ratios of cleaved to full-length forms of γENaC. Similar changes in this ratio were observed in the DCT2 and CNT populations (fig. 2D). These measurements were made with two pairs each of WT, heterozygotes and Liddle mice. There was no clear dependence on genotype, although given the limited number of observations small changes cannot be ruled out.

Figure 2.

γENaC processing in late DCT and CNT. A. Photomicrograph showing an isolated nephron fragment with both a proximal portion of a CNT and a distal portion of the upstream DCT. The black line indicates the location of a cut to separate the two nephron segments. B. Western blot of samples from WT mice on control diet containing 25–40 tubules identified as either DCT2 (late DCT) or CNT, together with a sample (40 μg) of total kidney cortex. The blot was cut at ~100 kDa (dashed line) and stained with antibodies against NCC (upper section) or γENaC (lower section). Lanes containing kidney cortex and DCT2 expressed NCC and γENaC. Lanes containing CNT expressed γENaC but no appreciable NCC. All mice were male. C. Western blot of late DCT and CNT fragments from mice on control-Na or low-Na diets. 25–40 tubules were loaded onto individual lanes and stained for γENaC. D. Ratio of band densities of cleaved/full length γENaC. Data were obtained from WT (black symbols), heterozygotes (red symbols) and Liddle mice (blue symbols). Data represent means ± SEM for n = 6 animals on each diet, pooling results from the 3 genotypes.

A second caveat regarding the lack of increased cleaved γENaC in the Liddle’s mice is that plasma aldosterone levels are reduced in these animals (16), as they are in human patients with Liddle syndrome (9), likely reducing ENaC processing. It is possible that the Liddle mutation makes ENaC processing more sensitive to aldosterone. To examine this, we treated mice fed control-Na (0.4%) diet with aldosterone using osmotic minipumps. This a less physiological preparation than that of dietary manipulation but has the advantage of controlling levels of the hormone. We first examined the dose-response relationship between infused aldosterone and the amount of cleaved γENaC in WT mice (fig 3A). The highest dose (48 μg/d) was chosen based on previous studies in mice and rats (10, 23). We found that a lower dose of 2.4 μg/d still produced a maximal effect, so this was used in subsequent blots to normalize the data. This relationship was approximately hyperbolic with a half-maximal effect at an infusion rate of 0.5 μg/day (fig 3B). Measurements of plasma aldosterone as a function of infusion rate showed considerably more variability, but results from the lowest infusion rates indicated that the half-maximal response occurred at plasma concentrations of ~ 1 ng/ml or 2.7 nM (fig. 3C), less than twice the baseline level, and consistent with occupancy of the high-affinity mineralocorticoid receptor (2).

Figure 3.

Effect of aldosterone infusion on γENaC processing. Mice on control diet were implanted with osmotic minipumps infusing aldosterone at different rates. A. Western blot of kidney microsomes from individual animals infused with 0, 0.24, 0.8, 2.4 or 7 μg/day. Each lane was loaded with 60 μg total protein. B. Band density for cleaved γENaC as a function of the aldosterone infusion rate. Data were normalized to values at infusion rates of 2.4 μg/day and fit with a hyperbolic function using the equation $\mathrm{A}\left(\mathrm{r}\right)={\mathrm{A}}_0+(\left({\mathrm{A}}_{\mathrm{m}\mathrm{a}\mathrm{x}}–{\mathrm{A}}_0\right)/\left(1+{\mathrm{r}}_{0.5}/\mathrm{r}\right)$ where ${\mathrm{A}}_0=0.39,{\mathrm{A}}_{\mathrm{m}\mathrm{a}\mathrm{x}}=1.07$ and ${\mathrm{r}}_{0.5}=0.8\mathrm{\mu }\mathrm{g}/\mathrm{d}\mathrm{a}\mathrm{y})$ C. Plasma aldosterone as a function of infusion rate. At low rates the relationship was approximately linear with ${\mathrm{P}}_{\text{aldo}}=0.66+0.34•\mathrm{r}$. All mice were males.

We then compared the effects of a low rate (0.3 μg/day) and a high, supramaximal rate (48 μg/day) of aldosterone infusion on ENaC processing (fig 4). At low rates we observed no significant change in the abundance of cleaved γENaC in Liddle mice compared with WT but there was a decrease in the amount of the full-length form which led to an increased in the cleaved/full-length ratio. This pattern was similar to that observed in untreated mice on control diet (fig 1), suggesting that there is no marked increase in the sensitivity of the Liddle mice to aldosterone. Females had a higher abundance of both full-length and cleaved γENaC. We did not run parallel untreated controls and cannot say whether this reflects a higher baseline expression or an increased response to aldosterone. With high rates of infusion, we observed both more full-length (50%; p = 0.0007) and cleaved (35%; p = 0.003) γENaC in the Liddle mice; the cleaved/full length ratio was unchanged. This could contribute the increase in channel activity found under these conditions (10) although this contribution is modest compared with the change in ENaC currents (6-fold).

Figure 4.

Effects of aldosterone infusion on WT and Liddle mice. Animals were implanted with osmotic minipumps to infuse aldosterone at rates of either 0.3 (A) or 48 μg/day (B). Each lane was loaded with 50 μg (A) or 80 μg (B) protein from an individual animal. The sex of individual animals is marked on each lane (m = male, f = female). Band densities were normalized to those of the WT cleaved form. Males and females were normalized separately. The ratios of cleaved/full-length are plotted as absolute values. Bar graphs plot mean ± SEM for n = 4 pairs. (A) and n = 10 pairs (B). p-values from Student t-tests are shown above the bars.

Liddle mutations can disrupt the binding sites for the E3 ubiquitin ligase Nedd4–2 on the C-termini of the βENaC and γENaC subunits (13, 24). With the same Liddle mouse model used here, we verified that truncation of the βENaC C-terminus diminished ubiquitination of γENaC (25). These findings are consistent with the idea that Liddle mutations could increase the lifetime of ENaC at the cell surface. To test this, we followed the amount of cleaved γENaC during acute Na⁺ repletion of mice on a low-Na diet. Figure 5 shows the time course in WT mice after introduction of a high-Na (5%) diet together with saline drinking water. The abundance of the cleaved form of γENaC decreased following an exponential curve with a time constant of ~4 hours. The abundance of full-length γENaC increased more slowly, with a time constant of ~20 hrs. These kinetics are similar to those observed in rats after removal of aldosterone and treatment with the mineralocorticoid-receptor (MR) antagonist spironolactone (20). In that paper we argued that the relatively rapid decrease in cleaved γENaC reflected a short lifetime of the fully processed channels.

Figure 5.

Kinetics of ENaC processing during salt repletion. Mice were fed a low-Na⁺ diet for 1 week. At time 0 they were offered a high- Na⁺ diet as well as saline drinking water. A. Western blot of kidney microsomes at different times after the initiation of the Na⁺ repletion trial. Each lane was loaded with 60 μg of protein from a different animal. Blots were stained for γENaC. B. Abundance of cleaved and full-length forms of γENaC at different times after the initiation of Na⁺ repletion, normalized to values at t = 0. Data for the cleaved form were fit with the equation:

$\mathrm{A}\left(\mathrm{t}\right)=1-\mathrm{B}•(1–\mathrm{e}\mathrm{x}\mathrm{p}(-\mathrm{t}/\mathrm{\tau }\left)\right)$ with B = 0.8 and τ = 5.2 hours.

Data for the full-length form were fit with the equation

$\mathrm{A}\left(\mathrm{t}\right)=1+(\mathrm{A}\mathrm{m}\mathrm{a}\mathrm{x}–1)•(1-\mathrm{e}\mathrm{x}\mathrm{p}(\mathrm{t}/\mathrm{\tau }\left)\right)$ with A_max = 5.0 and τ = 23 hours.

All mice were male.

To see if the Liddle mutation decreased the rate of degradation, we maintained WT and Liddle mice on a low-Na diet for 1 week and then followed the Na⁺ repletion phase for either 3 or 6 hours. The average excretion of Na⁺ in the urine in both protocols, taken as an indicator of total Na⁺ consumption, was similar in WT (243 ± 44 μmoles) and Liddle (233 ± 28 μmoles). The Na-repleted mice also gained similar amounts of body weight (WT 1.0 ± 0.2 g; Liddle 1.2 ± 0.2 g). However, in the 3-hour protocol the Liddle mice excreted, more Na⁺ in the first 3 hours (233 ± 42 vs 134 ± 9 μmoles). We do not know how that would impact the responses to Na⁺ repletion.

There was no measurable difference in the rates at which the cleaved form decreased, and the cleaved/full-length ratios were also similar (fig 6). ANOVA indicates significant variation with time (p = 0.02) but not with genotype (p = 0.86). The decrease cleaved γENaC at 6 hrs is significant for both WT (p = 0.02) and Liddle (p = 0.01) mice.

Figure 6.

Decreases in cleaved γENaC during Na⁺ repletion of WT and Liddle mice. A. Western blot of kidney microsomes before and 3 or 6 hours after initiation of Na⁺ repletion. Each lane was loaded with 60 μg of protein from a different animal. All mice were male. B. Band density of cleaved γENaC and full-length forms are normalized to values for Na⁺ depleted mice. The ratios of cleaved/full-length are plotted as absolute values. Data represent means ± SEM for n = 4–8 animals. p-values from Student t-tests comparing t = 0 and t = 6 hrs are shown above the bars.

DISCUSSION

Liddle syndrome is characterized by hypertension, hypokalemia and metabolic alkalosis; all these symptoms can be explained by hyperactivity of ENaC (9). In the mouse model that we have studied, these effects were observed only when the animals were fed a high-salt diet (16). Paradoxically, direct measurements of channel activity in isolated CCDs revealed increases only under conditions of elevated aldosterone including exogenous hormone administration, a low-Na diet, or a high-K diet (10). With a reduced level of dietary Na intake, Na⁺ reabsorption in upstream segments likely increases (26, 27), limiting Na⁺ delivery to the ENaC-expressing parts of the renal tubule. This effect may be exaggerated in the Liddle mice, mitigating the impact of the gain-of-function mutation on Na⁺ balance. Exactly where along the nephron these compensatory actions take place is unknown.

ENaC processing

Our studies reveal several aspects of ENaC processing as assessed by the cleavage of the γENaC subunit. Previous work showed that cleavage involves furin, mainly in the Golgi apparatus, together with other proteases that appear to work in the extracellular (luminal) space such as prostasin, kallikrein, and, under pathological conditions, plasmin (28, 29). The appearance of the cleaved form likely reflects movement of the channels toward the apical surface stimulated by aldosterone (5) and correlates strongly with ENaC surface expression (20).

We previously showed that the time course of increased ENaC processing in response to an increase in aldosterone levels occurred within a few hours (20). This resembles the time course of the physiological actions of the hormone and consistent with the idea that increased processing mediates, at least in part, its well established natriferic and kaliuretic effects. Here we show that processing can be activated by low nanomolar increases in plasma aldosterone, consistent with actions mediated through the high-affinity mineralocorticoid receptor (2).

Not all nephron segments regulate ENaC in the same way. Nesterov and colleagues showed that a population of cells near the transition between the DCT and CNT have constitutive ENaC activity that can be observed even when circulating aldosterone levels are low (21). These cells are likely identified with the “late DCT” in micropuncture studies, where amiloride blocks Na⁺ reabsorption (30), and the “DCT2” segment where both NCC and ENaC are expressed (31, 32). Further studies with adrenalectomized animals suggested that aldosterone-independent transport could be supported by glucocorticoids binding to mineralocorticoid receptors based on the premise that these cells lack the enzyme 11β-HSD2 that is thought to keep local corticosterone concentrations low (33). Data in figure 2 show that Na⁺ depletion can trigger enhanced ENaC processing even in nephron segments close to or within the late DCT as judged by tubule morphology and the expression of the NaCl cotransporter NCC. This implies that such cells can respond to aldosterone. Consistent with this idea, a recent study using single-nucleus RNAseq in the DCT reported that a minority population of cells expressed all 3 ENaC subunits as well as the mineralocorticoid receptor and 11β-HSD2 (34). The simplest interpretation is that the nephron-segment differences in activity reflect different degrees of activation of processed channels.

We recently showed that removal of aldosterone stimulation of rat kidneys in vivo led to a rapid decrease in the abundance of processed ENaC, with a time constant of around 2–5 hours (20). We suggested that this reflects rates of degradation of the cleaved forms of the subunits. This was accompanied by a slower increase in unprocessed channels, presumably through replenishment of a pool of newly synthesized channels. Here we show similar time courses in mouse kidneys in response to a sudden change in salt intake. The amounts of cleaved γENaC declined rapidly with a time constant of 5 hours, while the levels of full-length subunits increased more slowly. In this protocol, changes could arise both from decreased aldosterone levels and from a phenomenon attributed to feedback inhibition of channels during a Na overload (35). In any case, the data are consistent with the proposed kinetic model in which aldosterone promotes the relatively slow step of channel processing.

Effects of Liddle mutation

Liddle syndrome involves gain-of-function mutations in ENaC that account for the symptoms of hypertension, hypokalemia, and metabolic alkalosis in humans (9). A mouse model of Liddle syndrome featuring a truncating mutation in the C-terminal of βENaC produces a phenotype that mimics that of the human disease although this became apparent only when the animals were fed with a high-salt diet (16). In this model, increased ENaC activity was confirmed using electrophysiological measurements in renal tubules and distal colon (10, 15). Effects of the mutation on ENaC processing and trafficking have not been studied in detail. Comparison of ENaC protein in WT and Liddle mice shown here provides insights into the potential mechanisms underlying channel hyperactivity.

A prevalent view of the effects of Liddle mutations entails the abrogation of binding sites for the ubiquitin ligase Nedd4–2 on the C-termini of βENaC or γENaC (12, 13). Inhibition of channel ubiquitination could then increase the lifetime and therefore the abundance of the channels. Measurements of the surface expression of WT and mutant channels in heterologous expression systems in vitro supported this idea (36, 37). Furthermore, we confirmed ubiquitination of the α and γENaC subunits in vivo, and decreased ubiquitination in Liddle mice (25, 38). However, measurements in Xenopus oocytes showed that the mutation also increased activity of channels already at the surface (14).

Here we show that in Liddle mice infused with high levels of aldosterone hadincreased abundance of both cleaved and full-length ENaC compared to WT (fig 4), consistent with the idea of longer protein lifetimes. However, this effect – about a 35% increase - was modest compared with the 6-fold stimulation of channel activity observed in animals under similar conditions (10). This suggests an additional mechanism involved in the regulation of ENaC, perhaps at the level of increased activity of fully processed channel protein.

In mice with normal to high levels of Na⁺ intake – in the latter condition the mouse phenotype resembles that of the human disease (16) - the amounts of cleaved γENaC were actually lower in Liddle mice. These decreases may result from decreased levels of aldosterone in the Liddle mice (14–16). Since these decreases presumably arise from increased channel activity and increased urinary Na⁺ retention, these observations also indicate that the main effect of the mutation is to enhance the activity of processed channels, rather than increasing the abundance of the processed forms.

Our kinetic scheme (20) predicts that an increase in channel lifetime should slow the decrease in the abundance of cleaved γENaC in Na⁺-repleted animals. In Na⁺ depleted animals offered Na⁺ in their food and drinking water, cleaved γENaC decreased rapidly but this decline was similar in WT and Liddle mice. Again, this suggests that the dominant effect of the mutation is on channel activity rather than channel processing.

Limitations

An important limitation of these studies is that we did not directly assess the surface expression of ENaC. Although we were previously able to make such measurements in mouse kidneys using in situ biotinylation (39), with this strain we could not achieve adequate delivery of biotin to the apical surface through perfusion. We also attempted biotinylation of tubular suspension, which worked for basolateral proteins but not ENaC, and of isolated cells, in which no ENaC expression could be documented. We therefore cannot distinguish increases in the activity of processed ENaC from a mechanism in which the Liddle mutation increases the fraction of fully processed channels in the membrane itself relative to the number in the underlying apical cytoplasm.

Conclusions

We show that ENaC processing is stimulated by Na⁺ depletion/aldosterone in the late DCT/early CNT as well as in downstream segments. It is driven by modest changes in aldosterone secretion and can be reversed rapidly with Na⁺ repletion. A Liddle syndrome mutation engineered in mice has only modest effects on channel processing and maturation. The hyperactivity of these mutant channels is likely due either to changes in the distribution of fully processed channels or to activation of channels already in the membrane.