Acute Regulation of the Epithelial Na⁺ Channel by Phosphatidylinositide 3-OH Kinase Signaling in Native Collecting Duct Principal Cells

Abstract

Activity of the epithelial Na⁺ channel (ENaC) is limiting for Na⁺ reabsorption in the aldosterone-sensitive distal nephron. Hormones, including aldosterone and insulin, increase ENaC activity, in part by stimulating phosphatidylinositide 3-OH kinase (PI3-K) signaling. Recent studies in heterologous expression systems reveal a close spatiotemporal coupling between PI3-K signaling and ENaC activity with the phospholipid product of this kinase, PI(3,4,5)P₃, in some cases, directly binding the channel and increasing open probability (P_o). This study tested whether this tight coupling plays a physiologic role in modulating ENaC activity in native tissue and polarized epithelial cells. IGF-I was found to increase Na⁺ reabsorption across mpkCCD_c14 principal cell monolayers in a PI3-K–sensitive manner. Inhibition of PI3-K signaling, moreover, rapidly decreased Na⁺ reabsorption and ENaC activity in mpkCCD_c14 cells that were treated with corticosteroids and IGF-I. These decreases paralleled changes in apical membrane PI(3,4,5)P₃ levels, demonstrating tight spatiotemporal coupling between ENaC activity and PI3-K/PI(3,4,5)P₃ signaling within this membrane. For further probing of the mechanism underpinning this coupling, cortical collecting ducts (CCD) were isolated from rat and split open to expose the apical membrane for patch-clamp analysis. Inhibition of PI3-K signaling with wortmannin and LY294002 but not its inactive analogue rapidly and markedly decreased the P_o of ENaC. Moreover, IGF-I acutely increased P_o of ENaC in CCD principal cells in a PI3-K–sensitive manner. Together, these observations stress the importance of tight spatiotemporal coupling between PI3-K signaling and ENaC within the apical membrane of principal cells to the physiologic control of this ion channel.

Activity of the epithelial Na⁺ channel (ENaC) is limiting for Na⁺ transport across many epithelial tissues, including reabsorption across the renal collecting duct (reviewed in references [^1,2]). ENaC is one final effector of the renin-angiotensin-aldosterone system. Thus, ENaC functions as a critical component of the negative feedback pathway that couples renal Na⁺ handling to control of systemic fluid volume and BP. The importance of this channel and its proper regulation to human health and disease are apparent when considering that gain-of-function mutations in the channel itself and its upstream regulatory pathways cause improper renal salt conservation associated with hypertension (reviewed in references [^3–6]). In contrast, loss-of-function mutations in ENaC and in its regulatory pathways lead to inappropriate renal salt wasting. In addition to influencing BP, ENaC and its regulation by aldosterone allows Na⁺ reabsorption to be coupled to K⁺ secretion at the distal nephron, making this channel a therapeutic target for potassium-sparing diuretics. Inappropriate activation of ENaC in the collecting duct in response to stimulation of peroxisome proliferator–activated receptor-γ signaling, moreover, recently was implicated in the pathologic fluid retention that is associated with insulin-sensitizing thiazolidinediones (⁷). These examples also emphasize the need for a full understanding of the cellular signaling pathways and mechanisms that control ENaC activity.

Phosphatidylinositide 3-OH kinase (PI3-K) is central to many cellular signaling pathways that control ENaC activity (^8–10). In particular, PI3-K is an effector that is necessary for aldosterone and insulin control of the channel (^11–15). Both aldosterone and insulin stimulate PI3-K signaling and increase the levels of the phosphoinositide products of this kinase in renal epithelial cells (^13,16). Emerging evidence also suggests that PI3-K plays a role in IGF-I regulation of ENaC (¹⁷). One mechanism through which PI3-K enhances ENaC activity is by initiating a cellular signaling cascade that culminates in increased residency time of the channel in the apical membrane (reviewed in references [^5,18–20]). In this signaling pathway, serum and glucocorticoid-inducible kinase (Sgk) is positioned between PI3-K and the channel and serves to impede retrieval of ENaC. Sgk is phosphorylated in response to PI3-K signaling, and the levels of this kinase are regulated by aldosterone and other corticosteroids at the level of transcription. Extensive investigation has established the significance of this mechanism to control of the channel. However, it is clear that this mechanism does not account for all regulation of ENaC by corticosteroids and PI3-K signaling, because aldosterone also increases ENaC open probability (P_o) (^21,22). Moreover, active PI3-K is necessary for sustained Na⁺ reabsorption in response to prolonged exposure to aldosterone and insulin, although only aldosterone and not insulin induces expression of Sgk. Aldosterone does so only over a relatively short period of time (2 to 4 h), although, Na⁺ transport remains elevated and sensitive to PI3-K inhibition over the complete time period of exposure to steroid and insulin (^{10,12–14,23}). Inhibition of PI3-K, in addition, immediately decreases Na⁺ transport and ENaC activity under all conditions and in every system tested to date. It is unlikely that a mechanism that is predicated solely on activated Sgk's impeding ENaC retrieval would be capable of such a rapid and absolute response. Therefore, we hypothesized that PI3-K signaling influences ENaC activity through at least two mechanisms.

Recent studies have demonstrated that the phosphoinositide products of PI3-K bind to ENaC to influence the P_o of this channel (^24–26). These and other studies place ENaC into a category with several other ion channels, including KCNQ and inward-rectifier K⁺ channels, and TRP and Ca_v2 channels, as being directly sensitive to cellular phosphoinositide levels (reviewed in reference [²⁷]). The possible importance of this mechanism to regulation of ENaC in the collecting duct remains obscure. Here, we begin probing the physiologic importance of direct regulation of ENaC P_o by the phosphoinositide products of PI3-K in principal cells. Our major findings are that (1) there is close spatiotemporal coupling between PI3-K signaling/PI(3,4,5)P₃ levels and ENaC activity in the apical membrane of principal cells and (2) the P_o of ENaC within the apical membrane is dynamically coupled to PI3-K signaling with stimulation and inhibition of the phospholipid kinase, resulting in rapid increases and decreases, respectively, in channel P_o. These results are consistent with direct regulation of ENaC gating by PI(3,4,5)P₃ playing a physiologic role in modulation of Na⁺ transport across the collecting duct.

Materials and Methods

Chemicals and cDNA Constructs

All chemicals and materials were from Fisher Scientific (Waltham, MA), Sigma (St. Louis, MO), BioMol or CalBiochem (San Diego, CA) unless noted otherwise. The PI(3,4,5)P₃ reporter GFP-AktPH, used in this study, is a chimera that consists of the PI(3,4,5)P₃-binding plekstrin homology (PH) domain from Akt fused to green fluorescence protein (GFP). A PI(4,5)P₂ reporter that consisted of the PI(4,5)P₂-binding PH domain of PLC-δ1 fused to GFP served as a negative control. The cDNA that encoded these reporters were gifts from the T. Meyer laboratory (²⁸).

Cell and Tissue Culture

Immortalized mouse cortical collecting duct (mpkCCD_c14) principal cells were grown in defined medium on permeable supports (Costar Transwells, Canton, NY; 0.4 μm pore, 24 mm diameter) as described previously (²⁹). Cells were maintained with FBS and corticosteroids until they polarized and formed monolayers with high resistance and avid Na⁺ transport. In some instances, Na⁺ reabsorption was set to a basal level by culturing cells in medium that lacked FBS and corticosteroids for 48 h. In other experiments, before the addition of 100 ng/ml IGF-I (to the basolateral membrane; I-3769; Sigma), cells were pretreated for 30 min with inhibitors of PI3-K (LY294002; 50 μM), MEK1/2 (U0126; 500 nM), and Rho kinase (Y27632 and H-1152; 0.5 and 2.5 μM). For experiments that used both the PI3-K inhibitor LY294002 and the phosphatase and tensin homolog (PTEN) inhibitor bpV(pic) (³⁰), the former was used at 20 μM and the latter at 100 nM. For patch-clamp and fluorescence imaging studies, LY294002 and its inactive analogue LY30351 and wortmannin were used at 50, 50, and 0.2 μM, respectively. All inhibitors were from CalBiochem.

The isolated, split-open CCD preparation that was used in this study is similar to that described previously by the Wang laboratory (^31–33). Pathogen-free Sprague-Dawley rats of either gender (3 to 4 wk) were purchased from Charles River Laboratories (Wilmington, MA). Rats were allowed to settle upon arrival for up to 1 wk and then were maintained on a Na⁺-deficient diet (Harlan TEKLAD, Madison, WI; TD.90228; 0.01 to 0.02% Na⁺) for 5 to 7 d to increase the surface expression and activity of ENaC. In some cases, rats were maintained on normal chow (Harlan TEKLAD; TD.7912; 0.32% Na⁺) for the entire 2-wk period. Rats were killed by cervical dislocation, and the kidneys were immediately removed. Kidneys were cut into thin slices (<1 mm) and placed into ice-cold physiologic saline solution (pH 7.4). Collecting ducts were mechanically isolated from these slices by microdissection using watchmaker forceps under a stereomicroscope. Isolated CCD were allowed to settle onto 5 × 5-mm cover glass coated with poly-l-lysine. Cover glass that contained CCD were placed within a perfusion chamber mounted on an inverted Nikon TE300 (Melville, NY) and superfused with a physiologic saline solution buffered with HEPES (pH 7.4). CCD were split open with a sharpened micropipette controlled with a micromanipulator to gain access to the apical membrane. CCD were used within 1 to 2 h of isolation. Animal use and welfare adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals following a protocol that was reviewed and approved by the Institutional Laboratory Animal Care and Use Committee at the University of Texas Health Science Center at San Antonio.

Exogenous Expression of Protein

Plasmid cDNA that encode PI(3,4,5)P₃ and PI(4,5)P₂ reporters were introduced into mpkCCD_c14 principal cells within a confluent monolayer with a biolistic particle delivery system (Biolistic PDS-1000/He Particle Delivery System; Bio-Rad, Hercules, CA). Use of this system was described previously (^34,35). We closely followed these established protocols in our studies. In brief, mpkCCD_c14 cells were grown to confluence on permeable supports. After forming high-resistance monolayers that avidly transported Na⁺, cells were washed twice with physiologic saline, aspirated, and quickly bombarded (at the apical membrane) under vacuum with microcarriers that were coated with reporter cDNA. Medium was immediately returned to the cells, which where then placed within a tissue culture incubator for 2 to 3 d to allow expression of the PI(3,4,5)P₃ reporter. Bombardment had little disruptive effect on cellular and monolayer integrity as established by maintenance of Na⁺ transport and a high transepithelial resistance.

Electrophysiology

Transepithelial Na⁺ current across mpkCCD_c14 cell monolayers was calculated as described previously (³⁶). In brief, current was calculated using Ohm's law as the quotient of transepithelial voltage to transepithelial resistance under open circuit conditions using a Millicel Electrical Resistance System with dual Ag/AgCl pellet electrodes (Millipore Corp., Billerica, MA) to measure voltage and resistance.

Whole-cell macroscopic current recordings of ENaC in mpkCCD_c14 cells were made under voltage-clamp conditions using standard methods (^24–26). In brief, current recordings were made with a bath solution of (in mM) 160 NaCl, 1 CaCl₂, 2 MgCl₂, and 10 HEPES (pH 7.4) and a pipette solution of (in mM) 120 CsCl, 5 NaCl, 2 MgCl₂, 5 EGTA, 10 HEPES, 2.0 ATP, and 0.1 GTP (pH 7.4). Current recordings were acquired with an Axopatch 200B (Axon Instruments) patch-clamp amplifier interfaced via a Digidata 1322A (Molecular Devices, Sunnyvale, CA) to a PC that was running the pClamp 9.2 suite of software (Axon Instruments). Cells were clamped to a 20-mV holding potential with voltage ramps (500 ms) from 60 down to −100 mV used to elicit current. ENaC activity is reported as the amiloride-sensitive inward Na⁺ current at no applied potential (at 0 mV). For relative activity, current was normalized to maximum levels. Whole-cell capacitance and series resistances were routinely compensated.

For cell-attached patches that were made on the apical membranes of mpkCCD_c14 cells and principal cells in isolated, split-open collecting ducts, bath and pipette solutions were (in mM) 160 NaCl, 1 CaCl₂, 2 MgCl₂, and 10 HEPES (pH 7.4) and 140 LiCl, 2 MgCl₂, and 10 HEPES (pH 7.4), respectively. Current recordings were made using an Axopatch 200B. Currents were low-pass filtered at 100 Hz by an eight-pole Bessel filter (Warner Instruments, Hamden, CT) and digitized and stored on a PC using the Digidata 1322A interface. Current data were analyzed using pClamp 9.2. Channel activity defined as NP_o was calculated using the equation NP_o = Σ(t₁ + 2t₂ + … it_i), where t_i is the fractional open time spent at each of the observed current levels. P_o was estimated by normalizing NP_o for the observed number of channels within a patch. The error that is associated with this estimation of P_o increases as patches contain more channels and as P_o approaches either zero or unity (²¹). Only patches that contained five channels or fewer were used to estimate P_o. Single-channel conductance was calculated by recording at least three holding potentials.

Total Internal Reflection Fluorescence Microscopy

Fluorescence emissions from the PI(3,4,5)P₃ reporter GFP-AktPH at the apical membrane of mpkCCD_c14 cells within a confluent monolayer were collected using total internal reflection fluorescence (TIRF; also called evanescent-field) microscopy. TIRF generates an evanescent field that declines exponentially with increasing distance from the interface between the cover glass and plasma membrane, illuminating only a thin section (approximately 100 nm) of the cell that is in contact with the cover glass (^37,38). For these experiments, GFP-AktPH was introduced into polarized monolayers of mpkCCD_c14 cells that were grown on permeable supports with the particle delivery system described previously. Upon expression of the reporter, 5 × 5-mm sections of the support were excised, inverted, and placed onto cover glass coated with poly-l-lysine. This arrangement made it possible to visualize dynamic changes in the level of the PI(3,4,5)P₃ reporter at the apical membrane in real time in living cells.

All TIRF experiments were performed in the TIRF microscopy core facility housed within the Department of Physiology at the University of Texas Health Science Center, San Antonio (http://physiology.uthscsa.edu/tirf). We previously described imaging the GFP-AktPH reporter and other fluorophore-tagged proteins using this core facility (^24,39,40). The methods that were used in this study closely followed these published protocols. In brief, fluorescence emissions from GFP-AktPH reporter were collected using an inverted TE2000 microscope with through-the-lens (prismless) TIRF imaging (Nikon). Samples were viewed through a plain Apo TIRF ×60 oil-immersion, high-resolution (1.45 NA) objective. Fluorescence emissions were collected through a 535 ± 25-nm bandpass filter (Chroma Technology Corp., Rockingham, VT) by exciting GFP with an Argon-ion laser with an acoustic optic tunable filter (Pairie Technology, Middleton, WI) that was used to restrict excitation wavelength to 488 nm. Fluorescence images were collected and processed with a 16-bit, cooled charge-coupled device camera (Cascade 512F; Roper Scientific, Duluth, GA) interfaced to a PC that was running Metamorph software. This camera uses a front-illuminated electron-multiplying charge-coupled device with on-chip multiplication gain. Images were collected once per minute with a 100-ms exposure time. Images were not binned or filtered with pixel size corresponding to a square of 122 × 122 nm.

Statistical Analyses

All summarized data are reported as means ± SEM. Summarized data were compared with either the (two-tailed) t test or a one-way ANOVA in conjunction with the Dunnett or Student-Newman-Keuls posttest where appropriate. P ≤ 0.05 was considered significant. Macroscopic current are reported as relative to either control levels or current maximum within an experiment. Open circuit current and emissions from GFP-AktPH were normalized to starting levels. Vehicle treatment was used to quantify spontaneous decreases in fluorescence emissions (bleaching) over time. This value was subtracted from all fluorescence data. For presentation, current data from some cell-attached patches were subsequently software filtered at 20 or 50 Hz.

Results

PI3-K Signaling Is Necessary for IGF-I to Increase Na⁺ Transport

PI3-K is a critical component of the signaling cascades that are activated by aldosterone and insulin that are responsible for increasing ENaC activity and Na⁺ transport (^13,16). It is likely that PI3-K is also integral to increases in Na⁺ transport in response to IGF-I. We tested this idea here. As shown in Figure 1A, addition of 100 ng/ml IGF-I to polarized mpkCCD_c14 principal cells with steady-state basal transport rates significantly increased Na⁺ reabsorption in a time-dependent manner above basal levels and compared with vehicle. Significant increases were detectable after 30 min, the earliest time point measured, with a maximum reached by 2 to 3 h. Addition of amiloride to the apical membrane completely abolished the transepithelial current that was stimulated by IGF-I, indicating that this hormone increases Na⁺ reabsorption via ENaC. As summarized in Figure 1B, pretreating monolayers for 30 min with the PI3-K inhibitor LY294002 (50 μM) but not inhibitors of mitogen-activated protein kinase (U0126; 500 nM) and Rho (H-1152 [2.5 μM] and Y27632 [0.5 μM]) signaling disrupts IGF-I actions on Na⁺ reabsorption. Therefore, PI3-K is necessary to IGF-I–stimulated Na⁺ transport. Inhibition of PI3-K, in addition to abolishing the effects of IGF-I on Na⁺ transport, decreased basal transport at steady state. This suggests that some amount of PI3-K activity is required for Na⁺ reabsorption and that the activity of this phospholipid kinase is temporally coupled in a tight manner to the activity of the rate-limiting step in transepithelial Na⁺ reabsorption: ENaC activity.

Changes in Na⁺ Transport and ENaC Activity Parallel Changes in Apical Membrane PI(3,4,5)P₃ Levels

To test more directly the concept that Na⁺ reabsorption and ENaC activity are temporally coupled to the activity of PI3-K in a tight manner, we investigated the effects of inhibiting this kinase in polarized monolayers with high levels of Na⁺ transport at steady state. Transport was set high by culturing mpkCCD_c14 cells in the continued presence of corticosteroids and FBS. Figure 2A shows that within 15 min of addition of 50 μM LY294002, which is the earliest measurable time point, relative Na⁺ reabsorption significantly decreased to approximately 0.4, with the maximum decrease to approximately 0.2 reached by 45 min.

Decreases in the level of apical membrane PI(3,4,5)P₃ follow a similarly fast time course in confluent mpkCCD_c14 cells in response to inhibition of PI3-K. The representative fluorescence micrographs in Figure 2B show GFP emissions from the apical membrane of a mpkCCD_c14 principal cell that expressed the PI(3,4,5)P₃ reporter GFP-AktPH just before (1) and 5 (2) and 15 min (3) after addition of 50 μM LY294002. This cell was within a confluent monolayer cultured in the continued presence of corticosteroid and FBS that had high transepithelial resistance and avid Na⁺ reabsorption just before the experiment. Emissions from the apical membrane were isolated using TIRF microscopy. The graph in Figure 2C, summarizing such experiments (n = 9), shows the complete time course for decreases in apical membrane PI(3,4,5)P₃ in response to inhibition of PI3-K in mpkCCD_c14 cells. The relative decrease in apical membrane PI(3,4,5)P₃ in response to inhibition was rapid. In contrast, the inactive analogue LY303511 had no effect on apical membrane PI(3,4,5)P₃ levels (n = 9). Moreover, LY294002 during the course of these experiments had no effect on emissions from the negative control PI(4,5)P₂ reporter (control, n = 4). These results are consistent with ENaC activity being coupled not only in a tight temporal manner to PI(3,4,5)P₃ levels but also in a close spatial manner.

To define better the cause-and-effect relation between changes in the levels of PI(3,4,5)P₃ and Na⁺ transport, we next quantified transport while simultaneously manipulating the activity of proteins that are involved in both the synthesis and the degradation of this phosphoinositide. We used submaximal doses of LY294002 (20 μM) to slow but not completely abolish PI(3,4,5)P₃ synthesis, in combination with 100 nM of the PTEN inhibitor bpV(pic) to retard degradation of this phosphatidylinositide. The summary graph in Figure 3 compares decreases in Na⁺ reabsorption in mpkCCD_c14 monolayers that were treated with submaximal doses of LY294002 in the absence and presence of inhibited PTEN. Before these experiments, transport was set to a high level by constant exposure to corticosteroids and FBS. Inhibition of PTEN slowed and suppressed the effects of partial inhibition of PI3-K. These results are consistent with changes in the levels of PI(3,4,5)P₃ being causative for corresponding changes in transport.

Our ability to appreciate completely the temporal relation between changes in apical membrane PI(3,4,5)P₃ levels and ENaC activity with the experiments described were partially limited because quantifying transepithelial Na⁺ transport is an indirect measurement of ENaC activity and open-circuit current measurements provide suboptimal time resolution. To obviate these limitations, we directly monitored changes in ENaC activity in response to inhibition of PI3-K in mpkCCD_c14 cells in voltage-clamp experiments. Figure 4A shows a typical macroscopic current–voltage relation for the amiloride-sensitive Na⁺ current in a voltage-clamped mpkCCD_c14 cell before and after inhibition of PI3-K with 50 μM LY294002. As summarized in Figure 4B, inhibiting PI3-K significantly decreased ENaC activity. Figure 4C reports the time course of this effect. Shown in this figure is the increase in relative Na⁺ current carried by ENaC in a mpkCCD_c14 cells with no applied voltage as amiloride is washed. This is followed quickly by a decrease in ENaC activity upon addition of LY294002. The time course of decreasing ENaC activity upon inhibition of PI3-K closely parallels the decrease in apical membrane PI(3,4,5)P₃ levels for these cells as reported in Figure 2B. Real-time measurements of apical membrane PI(3,4,5)P₃ levels and ENaC activity demonstrate parallel changes upon inhibition of PI3-K signaling.

Tight Spatiotemporal Coupling between PI3-K Signaling and ENaC P_o in Native and mpkCCD_c14 Principal Cells

One mechanism by which PI3-K signaling influences ENaC activity is by PI(3,4,5)P₃ directly controlling channel gating (reviewed in reference [²⁷]). This mechanism is predicated on close spatiotemporal coupling between the channel and the second messenger. Because the previous results supported tight spatiotemporal coupling between ENaC activity and PI3-K signaling, we were interested in testing whether effects on ENaC P_o underlie this coupling and whether such a mechanism plays a role in physiologic regulation of the channel in native and cultured principal cells. To do this, we isolated collecting ducts and formed cell-attached seals on the apical membrane of principal cells to monitor directly changes in ENaC P_o in real time. Collecting ducts were isolated from salt-restricted (Na⁺-deficient diet for 1 wk) rats to set initial ENaC activity and P_o to a high level. Figure 5A shows current traces from a cell-attached patch on the apical membrane of a native principal cell in a freshly isolated rat collecting duct. The patched membrane was presented with test potentials that ranged from 0 to −60 mV. This seal contained at least three ENaC. Figure 5B shows the single-channel current–voltage relation for ENaC in rat principal cells in freshly isolated collecting ducts. With Li⁺ as the charge carrier, ENaC had a conductance of 9.8 ± 0.8 pS in this preparation (n = 6). The mean activity and P_o for ENaC in this preparation are 1.27 ± 0.18 and 0.44 ± 0.03 (n = 27), respectively. We also isolated and patched CCD from rats that were maintained on standard chow (0.32% Na⁺). In this group, ENaC P_o, as shown in Figure 5C, was significantly lower (0.17 ± 0.02; n = 13). Moreover, we observed fewer active ENaC in patches that were made on the apical membrane of principal cells within the CCD in rats that were maintained on normal chow (1.58; n = 13) compared with a Na⁺-deficient diet (3.43; n = 18). It is unclear whether this observation stems form only a decrease in P_o or also reflects a decrease in the number of channels in the membrane. Nevertheless, these values and observations agree with those published previously for similar preparations (^32,41,42).

Figure 6 shows representative current traces and summary graphs of P_o in paired experiments for ENaC in native principal cells before and after (5 to 10 min) treating with 0.2 μM wortmannin (Figure 6A), 50 μM LY294992 (Figure 6B), and 50 μM LY303511 (Figure 6C). Current traces are of ENaC in the apical membrane of principal cells in freshly isolated collecting ducts from salt-restricted rats. For these experiments, test potentials of −40 mV were applied to cell-attached patches of the apical membrane with ENaC activity and P_o continuously monitored. As is clear in these representative current traces and summary graphs, inhibition of PI3-K with two chemically and mechanistically distinct inhibitors, wortmannin and LY294002, decreased the P_o of native ENaC. In contrast, the inactive variant of LY294002, LY303511, had little effect on ENaC in this preparation. In response to wortmannin and LY294002, ENaC P_o significantly decreased from 0.42 ± 0.04 to 0.14 ± 0.02 (n = 8) and from 0.39 ± 0.04 to 0.10 ± 0.02 (n = 11), respectively. With LY303511, there was no difference in the P_o of 0.54 ± 0.03 and 0.49 ± 0.05 before and after addition, respectively (n = 8). Similar observations were made for cell-attached patches on mpkCCD_c14 cells, where 50 μM LY294002 significantly decreased ENaC P_o within 15 min from 0.41 ± 0.06 to 0.15 ± 0.06 (n = 6; data not shown). These results demonstrate that one effect of inhibiting PI3-K in native and cultured principal cells is to decrease the P_o of active ENaC within the apical membrane.

The representative current recordings in Figure 7 document the time course of inhibiting PI3-K activity on ENaC P_o in freshly isolated collecting duct (Figure 7A) and mpkCCD_c14 (Figure 7B) principal cells. The representative patch in Figure 7A (one of eight), formed on the apical membrane of a native principal cell, was clamped with a −40-mV test potential and contained at least two ENaC. A continuous trace before and after addition of 50 μM LY294002 is shown at the top. Segments before and after LY294002 are shown below at expanded time scales. As is apparent in this representative trace, inhibition of PI3-K resulted in a rapid decrease in ENaC P_o in this native preparation. Similarly, as shown by the representative current trace in Figure 7B (one of six), which is from a patch that was made on the apical membrane of a mpkCCD_c14 cell clamped to 0 mV, inhibition of PI3-K rapidly decreases ENaC P_o. Again, segments before and after LY294002 are shown below at expanded time scales. The results in Figures 6 and 7 demonstrating that the P_o of active ENaC within the apical membrane of native and cultured principal cells is tightly coupled to PI3-K activity are consistent with our previous findings in expression studies (^24,26).

The results in Figure 8 testing acute regulation of ENaC in CCD and mpkCCD_c14 principal cells by IGF-I are also consistent with tight coupling between PI3-K and ENaC activity. Figure 8A shows a representative current trace from a cell-attached patch that was made on a mpkCCD_c14 cell before and after addition of 100 ng/ml IGF-I. This patch contains at least three ENaC and was made on a cell that was cultured in minimal medium to set ENaC activity initially to a basal level. As shown in the continuous trace at the top and in segments before and after IGF-I shown below at expanded time scales, IGF-I rapidly increases ENaC P_o. In mpkCCD_c14 cells with basal levels of ENaC activity, IGF-I, as documented in Figure 8B, more than doubled P_o from 0.12 ± 0.03 to 0.26 ± 0.03 (n = 4) within 10 min. A similar observation was made in principal cells from isolated collecting ducts from rats that were maintained on standard chow. Figure 8C shows a representative current trace from a cell-attached patch that was made on the apical membrane of a principal cell in a CCD that was freshly isolated from a rat that was maintained on standard chow. This patch contains at least three ENaC. Shown at the top is a continuous trace with segments before and after addition of IGF-I shown below at expanded time scales. In this native preparation, IGF-I significantly increased ENaC P_o from 0.16 ± 0.02 to 0.35 ± 0.04 (n = 10; Figure 8D) within 10 min. As documented by the representative trace in Figure 8E (one of three), IGF-I stimulation of ENaC P_o in native CCD principal cells was sensitive to inhibition of PI3-K with 50 μM LY294002. For these experiments, CCD that were isolated from rats that were maintained on normal chow were treated with 100 ng/ml IGF-I for 20 min before formation of cell-attached patches on the apical membrane of principal cells. Subsequent to seal formation, LY294002 was applied to the bath. The continuous current trace is shown at the top. A region after treatment with LY294002 is shown below at an expanded time scale.

Although not a focus of this investigation, we did observe a renal outer medullary K⁺ channel–like small-conductance K⁺ (SK) channel in some patches that were made on the apical membrane of principal cells in isolated split-open rat collecting ducts. As reported previously for this preparation (³³), inhibition of PI3-K increased the activity of this channel (data not shown).

Discussion

As discussed at the beginning of this article, PI3-K is recognized to play a central role in regulation of ENaC. It serves as a downstream effector and possible point of convergence for several natriferic hormones, including aldosterone and insulin, both of which increase activity of this phospholipid kinase in renal epithelial cells (^13,14,16). One mechanism by which PI3-K controls ENaC activity is through a signaling cascade that ultimately increases the residency time of the channel in the apical membrane by slowing channel retrieval (reviewed in references [^5,18–20]). However, by its nature, being dependent on trafficking, such a mechanism must be relatively slow to develop and abate and cannot account for all of the effects, particularly the acute actions, of PI3-K signaling on the channel. Therefore, we hypothesized and tested the idea that PI3-K impinges on ENaC activity through at least one other mechanism. We found that there is tight spatiotemporal coupling between PI3-K signaling and Na⁺ reabsorption, with the P_o of ENaC within the apical membranes of principal cells changing in parallel with the levels of PI(3,4,5)P₃ within these membranes.

ENaC, similar to some other types of ion channels, senses membrane phosphoinositide levels and responds with changes in channel P_o through direct interactions with these molecules (see reference [²⁷]). Our laboratory showed that ENaC physically interacts with the phosphoinositide products of PI3-K and that this interaction stabilizes ENaC gating to increase P_o (^24,26). These findings are consistent with such a mechanism contributing to regulation of ENaC activity in both the immortalized mouse principal cell line and principal cells from freshly isolated rat collecting ducts studied here.

We found the observations that decreases in apical membrane PI(3,4,5)P₃ levels in response to inhibiting PI3-K parallel decreases in Na⁺ reabsorption, ENaC activity, and ENaC P_o significant. Such tight spatiotemporal coupling between the levels of the phosphoinositide and P_o of ENaC within the apical membrane readily explains why we and other investigators observed an almost instantaneous but persistent decrease in channel activity and Na⁺ transport upon PI3-K inhibition. This is not to say that membrane levels of ENaC might not also ultimately decrease in the presence of inhibited PI3-K but rather that the initial event is a decrease in P_o. Interpreting our results in the context of others allows us to propose that two discrete mechanisms serve to decrease ENaC activity upon inhibition of PI3-K: An initial event, described here, whereby channel P_o decreases rapidly as a result of the loss of direct regulation by PI(3,4,5)P₃ and a slower developing phase whereby apical membrane levels of ENaC decrease as a result of cessation of impeded channel retrieval.

Our observation that IGF-I stimulates Na⁺ reabsorption across principal cells is consistent with that recently published by Gonzalez-Rodriguez et al. (¹⁷). Also consistent with the findings of this group is the observation that PI3-K is necessary for IGF-I actions on Na⁺ transport. We advance this understanding by demonstrating here that IGF-I also acutely increases the P_o of ENaC within the apical membrane of principal cells in a PI3-K–dependent manner. That inhibiting PI3-K signaling blocks hormone-sensitive as well as basal and sustained Na⁺ transport and ENaC P_o, though, does somewhat limit interpretation, because it is not clear whether an increase in PI3-K activity is necessary for hormonal control of the channel or rather that active PI3-K is permissive for hormone action on the channel. This is related to the question of whether direct regulation of ENaC P_o by PI(3,4,5)P₃ underpins mediated channel modulation in response to hormonal input or rather serves a permissive role that allows channels that are targeted to the membrane to gate. More simply put, do the direct effects of PI(3,4,5)P₃ on ENaC P_o play a role in regulated or only constitutive channel modulation?

The finding that apical membrane PI(3,4,5)P₃ levels decrease in a rapid and marked manner upon inhibiting PI3-K supports the idea that there is significant phospholipid phosphatase activity countering PI3-K activity in renal principal cells. Therefore, a reasonable scenario that is consistent with our results is that hormones, which stimulate PI3-K activity, such as insulin, IGF-I, and aldosterone, increase both the membrane residency time and P_o of ENaC to increase activity of this channel. With such a scenario, then, in the absence of hormone, there would be fewer channels in the apical membrane, and the channels that are there would have a lower P_o. This prediction is consistent with our results describing the frequency of observing ENaC in apical membranes of CCD principal cells in salt-restricted rats compared with rats that were maintained on normal chow, as well as with channel P_o in these rats. Moreover, this prediction may possibly settle some of the earliest controversies regarding regulation of ENaC by aldosterone in principal cells, in which aldosterone action was sometimes attributed to an increase in the number of channels in the membrane (⁴¹) and at other times to an increase in P_o (²¹). Put another way, activation of PI3-K signaling possibly increases the residency time of ENaC within the apical membrane and enables these channels to open. Although further proof is required before fully accepting such bimodal regulation, this study, by demonstrating that PI3-K signaling acutely regulates the P_o of ENaC within the apical membrane of principal cells, is a strong first step.

Disclosures

None.

Figure 1: Phosphatidylinositide 3-OH kinase (PI3-K) is necessary for IGF-I to increase Na⁺ reabsorption across principal cells. (A) Time course for IGF-I–dependent increases in relative Na⁺ transport across monolayers of mpkCCD_c14 principal cells. Current relative to starting levels. Na⁺ reabsorption was initially set to basal levels by culturing monolayers in the absence of corticosteroids and insulin for 48 to 72 h. IGF-I was added at time 0. (B) Effects of IGF-I on relative Na⁺ reabsorption in the absence (control) and presence of inhibitors of Rho kinase (Y27632 and H-1152), Mek1/2 (U0126), and PI3-K (LY294002) signaling. Inhibitors were added 30 min before IGF-I treatment. All other conditions are the same as in A.

Figure 2: Tight spatiotemporal coupling between Na⁺ reabsorption and PI(3,4,5)P₃ levels in the apical membrane of principal cells. (A) Time course of decreases in relative Na⁺ transport in response to PI3-K inhibition with LY294002. Na⁺ reabsorption was initially set to a high level by culturing monolayers in the continued presence of corticosteroids, insulin, and FBS. LY294002 and vehicle (control) were added at time 0. All other conditions are the same as in Figure 1A. (B) Fluorescence micrographs of emissions from the GFP-AktPH PI(3,4,5)P₃ reporter in the apical membrane of a mpkCCD_c14 principal cell within a confluent monolayer before (1) and 5 (2) and 15 (3) minutes after treatment with LY294002. Monolayers were maintained in the continued presence of corticosteroids, insulin, and FBS. Emissions were collected in a paired manner using total internal reflection fluorescence (TIRF) microscopy. (C) Decay in relative apical membrane PI(3,4,5)P₃ levels in mpkCCD_c14 cells within confluent monolayers upon inhibiting PI3-K with LY294002 (•). PI(3,4,5)P₃ levels within the apical membrane monitored with the GFP-AktPH reporter and TIRF microscopy. Values are corrected for a small but constant decay in signal over time as a result of photobleaching, which was established by following changes in emission during this time course in the presence of vehicle. Also shown are the effects or rather lack of effects of the inactive analogue LY303511 (▪) on PI(3,4,5)P₃ levels and the negative control that LY294002 has no effect on PI(4,5)P₂ levels (control; ▴) in this preparation.

Figure 3: PI(3,4,5)P₃ is the second messenger coupling Na⁺ reabsorption to PI3-K signaling in principal cells. The time course of decreases in relative Na⁺ transport in response to partial inhibition of PI3-K with submaximal doses of LY294002 in the absence and presence of simultaneous inhibition of phosphatase and tensin homolog with bpV(pic) is shown. Na⁺ reabsorption was initially set to a high level by culturing monolayers in the continued presence of corticosteroids, insulin, and FBS. LY294002 and bpV(pic) were added at time 0. All other conditions are the same as in Figure 1A.

Figure 4: Epithelial Na⁺ channel (ENaC) activity is dynamically coupled to PI3-K signaling in principal cells. (A) Representative current–voltage relation of the amiloride-sensitive macroscopic current in a voltage-clamped mpkCCD_c14 principal cell before (control) and after treatment with LY294002. (B) Effect of inhibiting PI3-K on relative ENaC activity in voltage-clamped mpkCCD_c14 cells. (C) Representative diary plot showing the time course of LY294002 actions on relative ENaC activity in a typical voltage-clamped mpkCCD_c14 cell. At the beginning of this experiment, amiloride was washed from the bathing solution to allow ENaC to activate. LY294002 and amiloride were then added to the bath sequentially. In this typical voltage-clamp experiment, current at time t (I) is normalized to the maximum current (I_max) observed.

Figure 5: ENaC in the apical membrane of a principal cell in an isolated split-open collecting duct. (A) Current traces from a representative cell-attached patch that contained at least three ENaC. This patch was made on the apical membrane of a principal cell in an isolated split-open collecting duct that was isolated from a salt-restricted rat. Current was recorded at test potentials (−V_p) that ranged from 0 to −60 mV. Inward Li⁺ current is downward, and the dashed lines show the 0 current level (closed state) at each voltage. (B) Single-channel current–voltage relation for ENaC in cell-attached patches that were made on the apical membrane of principal cells in isolated split-open rat collecting ducts. (C) Comparison of mean open probability (P_o) for ENaC in cell-attached patches that were made on the apical membranes of principal cells in cortical collecting ducts (CCD) that were isolated from rats that were maintained on Na⁺-deficient (0.01%) and standard (0.32%) chow.

Figure 6: The P_o of ENaC within the apical membrane of principal cells in isolated split-open collecting ducts is dynamically coupled to PI3-K signaling. Current traces (left) are from representative cell-attached patches that contained ENaC and were made on the apical membrane of principal cells in isolated split-open rat collecting ducts before and after treatment with wortmannin (A), LY294002 (B), and LY303511 (C). Summary graphs of P_o before and after wortmannin, LY294002, and LY303511 are shown to the right. Patches were held at test potentials of −40 mV during the course of these experiments. All other conditions are the same as in Figure 5A.

Figure 7: Inhibition of PI3-K results in rapid and marked decreases in the P_o of ENaC within the apical membrane of cultured and native principal cells. (A) Continuous current trace from a representative cell-attached patch that contained at least two ENaC and were made on the apical membrane of principal cells in isolated split-open rat collecting ducts before and after treatment with LY294002. Areas before (1) and after (2) treatment are shown below with an expanded time scale. This patch was held at a −40-mV test potential during the course of the experiment. Top dashed line denotes closed current level. Middle and bottom dashed lines denote the first and second open current levels, respectively. All other conditions are the same as in Figure 5A. (B) Continuous current trace from a representative cell-attached patch that contained at least four ENaC and was made on the apical membrane of a mpkCCD_c14 cell before and after treatment with LY294002. This patch was held at a 0-mV test potential during the course of the experiment. All other conditions are the same as in A.

Figure 8: IGF-I acutely increases the P_o of ENaC in cultured and native principal cells in a PI3-K–dependent manner. (A) Continuous current trace from a representative cell-attached patch that contained at least three ENaC and was made on the apical membrane of a mpkCCD_c14 cell before and after treatment with IGF-I. Patches were made on cells that were cultured in minimal medium to set initial ENaC activity to basal levels. This patch was clamped to −60 mV. All other conditions are the same as in Figure 7A. (B) Paired experiments showing the acute (10 min) effects of IGF-I on ENaC P_o in mpkCCD_c14 cells with basal channel activity. (C) Continuous current trace before and after IGF-I from a representative cell-attached patch that contained at least two ENaC and was made on the apical membrane of a principal cell in a CCD that was isolated from a rat that was maintained on standard chow. All other conditions are the same as in Figure 7A. (D) Paired experiments showing the acute (10 min) effects of IGF-I on ENaC P_o in CCD principal cells with basal channel activity. (E) Continuous current trace before and after LY294002 from a representative cell-attached patch that contained several ENaC and was made on the apical membrane of a principal cell in a CCD that was isolated from a rat that was maintained on standard chow. Twenty minutes before seal formation, this CCD was treated with IGF-I. Shown below at an expanded time scale is a segment of the trace after treatment with LY294002. All other conditions are the same as in Figure 7A.

This research was supported by National Institutes of Health grants RO1DK59594 and R01DK070571, American Heart Association grant EIA 0640054N (to J.D.S.), and a National Kidney Foundation Research Fellowship and American Heart Association grant SDG 0730111N (to A.S.).

Jorge Medina is recognized for excellent technical support. We thank Dr. W.-H. Wang and his laboratory for helping to establish the isolated split-open collecting duct preparation in our laboratory.