INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

SEVERAL HORMONES MODULATE the activity of the renal epithelial Na⁺ channel (ENaC). This regulation is critical to maintain normal salt and fluid homeostasis, and slight alterations in these hormone-stimulated pathways may contribute to such common diseases as essential hypertension. Despite the importance of these processes, very little is known about the intracellular signaling pathways involved in hormonal regulation of epithelial Na⁺ reabsorption.

Aldosterone, a steroid hormone, mediates relatively long-term natriferic responses. Hormone binding to the mineralocorticoid receptor in responsive tissues is well documented. Likewise, it is known that the final outcome of this interaction is stimulation of Na⁺ transport via the amiloride-sensitive ENaCs (1-4, 8). However, the pathway(s) linking the aldosterone-receptor binding to activation of transport through functional channels within the apical membrane of the cells is unknown.

Phosphoinositide 3-kinases (PI 3-kinases) are enzymes that phosphorylate position 3 of the head group of the membrane lipid phosphatidylinositol (20). PI 3-kinases are a family of enzymes that can be distinguished by molecular characteristics, substrate specificity, and inhibitor sensitivities (13, 20, 22-24). Classic inhibitors of these enzymes are wortmannin, a fungal metabolite, which is an irreversible inhibitor effective in the nanomolar concentration range, and LY-294002, a structurally unrelated compound with a very high specificity for PI 3-kinase, which is a reversible inhibitor effective in the micromolar concentration range (23).

These enzymes have been implicated in a variety of diverse biochemical processes including membrane transport phenomena. Notable in this regard is the importance of the activation of PI 3-kinase for the insertion of the glucose transporters (GLUT-4) into the plasma membrane of adipocytes and skeletal muscle in response to insulin (6, 24). However, other peptide hormone-mediated transport events including insulin-stimulated Na⁺ transport via ENaC (14) and K⁺ uptake into fibroblasts via the Na⁺-K⁺-2Cl cotransporter (19) also require PI 3-kinase activation. In addition, these enzymes also appear to have a role in platelet-derived growth factor activation of the Na⁺/H⁺ exchanger (11), epidermal growth factor (EGF)-stimulated intestinal Na⁺ absorption (9), and EGF-mediated inhibition of Ca²⁺-dependent Cl secretion (21).

We have previously shown that the increase in aldosterone-stimulated Na⁺ transport is paralleled by an increase in the number of functional channels in the apical membrane (8). In this regard, the steroid hormone action is similar to the natriferic action of insulin, although the time course of the two hormones is very different. Insulin causes an increase in the number of functional channels within minutes, whereas aldosterone's action requires a longer time, consistent with the requirement for new protein synthesis (4). We have also previously demonstrated that insulin's action is dependent on the presence of PI 3-kinase (14). Given the importance of PI 3-kinase for peptide-mediated transport, we have investigated the role of this enzyme in steroid-mediated Na⁺ absorption.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell culture. For measurement of macroscopic rates of transport as well as lipid profile analysis, the A6 cells were grown at 27°C in DMEM (#91-5055EC; GIBCO BRL, Grand Island, NY) supplemented with 25 U/ml penicillin, 25 µg/ml streptomycin, and 10% calf serum in a humidified incubator gassed with 5% CO₂. The cells were subcultured onto 24-mm Transwell tissue-culture-treated inserts (Costar, Cambridge, MA) for at least 14 days to achieve confluence and used 14-21 days after seeding. For noise analysis, with the exception of three control tissues, A6 epithelia were grown on Transwell clear inserts at 28°C in a humidified incubator containing 1% CO₂. The growth medium was either the modified DMEM (as above) or a mixture of Ham's F-12 medium (N-6760; Sigma Chemical, St. Louis, MO) and L-15 Leibovitz medium (L-4386; Sigma Chemical) supplemented with 10% defined fetal bovine serum (FBS), 2.57 mM sodium bicarbonate, 3.84 mM L-glutamine, 96 U/ml penicillin, and 96 µg/ml streptomycin. Tissues were studied while continuously perfused with growth medium without FBS and glutamine.

Electrical measurements/noise analysis. Macroscopic rates of transcellular Na⁺ transport were measured as short-circuit currents (I_sc) as previously described (2, 8). After mounting in the chambers, the cells were incubated in serum-free media for 1-4 h to assure that a stable baseline was achieved before the addition of inhibitors and/or hormone.

In the noise experiments, blocker-sensitive Na⁺ currents (I_Na) were determined from the difference between the I_sc and the respective 100 µM amiloride-insensitive Na⁺ currents that average near 0.1 µA/cm² before and after treatment of the tissues with LY-294002. A pulse method of blocker-induced noise analysis was carried out as previously described (8), using the weak ENaC blocker 6-chloro-3,5-diaminopyrazine-2-carboxamide (CDPC; 27,788-6; Aldrich Chemical, Milwaukee, WI). Corner frequencies (f_c) and low-frequency plateaus (S_o) of the CDPC Lorentzians characterizing the blocker-induced current noise in power density spectra were determined by nonlinear curve fitting. Blocker on (k_ob) and off (k_bo) rate coefficients were calculated from the slopes and intercepts of rate-concentration plots using the time-filtered f_c at the respective 10 and 30 µM concentrations of CDPC. Single-channel currents (i_Na) were calculated in the usual way from the quotient [S_o(2f_c)²]/[4I_Na k_obB], where S_o were corrected for the shunting of current noise (~12%) through the apical membrane capacitance. Functional open-channel densities (N_o) were calculated as I_Na/i_Na. The tissues were studied in both their unstimulated and aldosterone-prestimulated states (0.27 µM, overnight). All experiments were carried out at ambient room temperature. Data are expressed as means ± SE.

Lipid analysis. To determine the phosphorylation of the inositol lipids, A6 cells were incubated for 16 h in serum-free media followed by a 2-h incubation in a nominally phosphate-free media containing H₃³²PO₄ before any experimental manipulations were performed. After hormone and/or inhibitor treatments, the lipids were extracted, separated, and analyzed by anion exchange chromatography as reported previously (14, 23).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

When A6 epithelia are challenged with LY-294002, a specific and reversible inhibitor of PI 3-kinase (23), inhibition of PI 3-kinase results in a marked decrease in basal Na⁺ transport and a complete abolition of aldosterone-stimulated Na⁺ transport as measured by amiloride-sensitive I_sc. To illustrate the time course of the basal inhibition, the data are plotted at 5-min intervals in the 30 min following the addition of LY-294002 (Fig. 1A). Such observations suggest that the products of the PI 3-kinase reaction, phosphoinositides, may be crucial intermediates in the natriferic pathway.

View larger version (27K): [in this window] [in a new window]   Fig. 1.   Effect of inhibitors of phosphoinositide 3-kinase (PI 3-kinase) on basal and aldosterone-stimulated Na+ transport. Macroscopic rates of ion transport were measured as short-circuit currents (Isc). A6 tissues were preincubated with 50 µM LY-294002 (A) or 100 nM wortmannin (B) for 30 min. Aldosterone (2.7 µM) was added to basolateral solution at time zero. After 5 h of aldosterone incubation, 10 µM amiloride was added to apical media to inhibit portion of current due to Na+ transport by epithelial Na+ channels (ENaCs). Results are means ± SE. For clarity, error bars are shown in one direction only.

Interestingly, wortmannin, a structurally unrelated inhibitor of PI 3-kinase (13), also inhibited basal current but did not affect the aldosterone-stimulated portion of Na⁺ transport (Fig. 1B). Thus the data suggest that the baseline rate of Na⁺ transport requires the constitutive activation of a wortmannin- and LY-294002-sensitive PI 3-kinase, whereas aldosterone's action involves an LY-294002-sensitive but wortmannin-insensitive form of the kinase.

Because intracellular signaling processes must occur before the final functional effect manifested by the cell, the phosphatidylinositol lipid profile was examined just before the onset of aldosterone-stimulated transport (20 min after hormone addition) and just after an aldosterone-induced increase in transport was clearly measurable (60 min). The lipid profiles of stimulated tissues were compared with control (non-hormone-treated) tissues that had been grown in parallel cultures (Fig. 2). Twenty minutes after the addition of aldosterone, there was an increase in the phosphoinositides phosphatidylinositol 3-phosphate [PtdIns(3)P] and phosphatidylinositol 3,4,5-trisphosphate [PtdIns(3,4,5)P₃]. Accurate quantitation of PtdIns(3)P as well as phosphatidylinositol 3,4-bisphosphate [PtdIns(3,4)P₂] was not possible in all experiments due to the proximity of these compounds to unidentified peaks that have a similar elution time. For this reason, the PtdIns(3,4,5)P₃ peak, which can be accurately quantitated, was used as the measure of PI 3-kinase activity in a more extensive series of experiments (Fig. 3).

View larger version (35K): [in this window] [in a new window]   Fig. 2.   Inhibition of aldosterone-stimulated PI 3-kinase in A6 cells. A: a single, representative experiment showing increase in 3-phosphorylated phosphatidylinositols in response to aldosterone (20 and 60 min of hormonal stimulation) and inhibition of this increased activity by LY-294002 after 60 min of hormone stimulation. B: an experiment demonstrating effect of LY-294002 on aldosterone-stimulated tissues 20 min after addition of the hormone. For inhibitor studies, 32PO4-labeled A6 cells were preincubated with 50 µM LY-294002 for 30 min before addition of aldosterone. Each panel represents lipid components from two 5-cm2 Transwell inserts of confluent A6 cells. Standards of [32P]phosphatidylinositol 3-phosphate (PI-3P), phosphatidylinositol 3,4-bisphosphate (PI-3,4P2), and phosphatidylinositol 3,4,5-trisphosphate (PIP3) were synthesized using purified bovine brain PI 3-kinase. For each experiment, standards that were run at time of sample analysis are shown. * Position of phosphatidylinositol 4,5-bisphosphate. CPM, counts/min.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Effect of aldosterone and LY-294002 on phosphoinositide production in A6 cells. Ratio of phosphatidylinositol 3,4,5-trisphosphate (PIP3) to phosphatidylinositol 4,5-bisphosphate (PIP2) was calculated for each sample and compared with same ratio in control (untreated) cells that were grown in parallel and assayed in tandem. Time indicates length of time of treatment with aldosterone. In inhibitor-treated samples, LY-294002 was added 30 min before addition of aldosterone. Each bar represents an individual sample. When LY-294002-treated samples are shown in same vertical space as aldosterone alone, these represent samples within same experiment where cells were grown, labeled, and analyzed in tandem.

To normalize for experimental variability due to sample recovery, the PtdIns(3,4,5)P₃ peak was compared with the PtdIns(4,5)P₂ peak in each sample. The amount of PtdIns(4,5)P₂ is not likely to change under these experimental conditions. The PtdIns(3,4,5)P₃/PtdIns(4,5)P₂ ratio from each sample was compared with the same ratio in a control sample that was grown, labeled, and processed in tandem (Fig. 3).

An increased level of PtdIns(3,4,5)P₃ is detectable before a measurable increase in aldosterone-stimulated Na⁺ transport (20 min) in each sample analyzed. After 60 min of hormone stimulation, PtdIns(3,4,5)P₃ remains elevated over control in all samples. With one exception, LY-294002 inhibits the production of PtdIns(3,4,5)P₃ below control levels even in the presence of aldosterone. Thus LY-294002 inhibited not only the aldosterone-mediated increase in the production of the phosphoinositides (Fig. 2 and 3) but also the Na⁺ transport response (Figs. 1 and 4).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Inhibition of Na+ transport by LY-294002. A pulse method of blocker-induced noise analysis was used to determine time-dependent changes of single-channel current (B) and functional open-channel density (No) of apical membrane ENaCs (C) in unstimulated and aldosterone-pretreated tissues in response to apical 10 µM LY-294002. Completely reversible inhibition of macroscopic rates of Na+ transport measured as amiloride-sensitive Isc (A) is due to relatively slow time-dependent changes of No. Values are means ± SE; n = 5 for unstimulated tissues; n = 7 for aldosterone-pretreated tissues.

Finally, we determined the underlying time-dependent changes of amiloride-sensitive Na⁺ transport (I_Na) in response to LY-294002 at the single-channel level. Noninvasive methods of blocker-induced noise analysis were used to monitor the changes of the single-channel currents (i_Na) and open-channel densities (N_o) in confluent short-circuited monolayers of A6 epithelia. For the purpose of these studies, the effects of the inhibitor on both unstimulated and aldosterone-pretreated tissues were examined (Fig. 4). Zero time values shown in Fig. 4 are typical of those reported for unstimulated and aldosterone-pretreated tissues where aldosterone causes a twofold or larger stimulation of transport by increase of N_o with no significant change of i_Na (8). From zero time values of I_Na that averaged 4.0 ± 0.1 µA/cm² in unstimulated tissues and 9.1 ± 0.9 µA/cm² in aldosterone-pretreated tissues, the amiloride-sensitive I_sc were inhibited by LY-294002, on average, to 33% and 38% of the zero time values, respectively, within 90 min (Fig. 4A), following, however, an initial small transient increase in transport (<5 min, data not shown). The inhibitory effect was relatively slow, with the mean I_Na response to 10 µM LY-294002 exhibiting quasiexponential decays with time constants of 22.4 and 24.4 min in control and aldosterone-pretreated tissues, respectively.

The changes of single-channel current caused by the inhibitor in the two groups of tissues are summarized in Fig. 4B. From zero time values that averaged ~0.39 pA, LY-294002 caused similar small time-dependent increases of i_Na. Consequently, the inhibitory effect of LY-294002 on Na⁺ transport is due to a decrease in the number of functionally open channels within the apical membrane of the cells. Underlying the trend in I_Na, N_o fell on average to 30% and 34% of the respective zero time values in control and aldosterone-pretreated tissues, respectively, after 90 min of exposure to LY-294002. Inhibition of the macroscopic rates of transport and open-channel densities is fully reversible on removal of the LY-294002 (Fig. 4).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Tight epithelia of the A6 cell line, derived from the kidney of Xenopus laevis, are commonly used as a model for the transport properties of the principal cells of the distal nephron. A6 cells contain ENaCs and respond to hormones known to regulate Na⁺ reabsorption in the mammalian distal nephron (12). In these cells, aldosterone stimulates an increase in transepithelial Na⁺ transport with a slow time course that is consistent with its action through new protein synthesis (3). The nature of the induced proteins as well as the identity of other intracellular regulators of the hormone-stimulated transport process are unknown.

Taken together, the data presented here indicate that PI 3-kinases are required for transcellular Na⁺ transport with specific involvement of apical membrane ENaCs. Analysis of the lipid profile in A6 cells that have been treated with aldosterone indicates that this steroid hormone mediates an increase in the amount of phosphoinositides, specifically PtdIns(3,4,5)P₃. This increase over the level found in control cells can be detected before the increase in the functional effect of enhanced Na⁺ transport, consistent with a regulatory role for the phosphoinositides in aldosterone-mediated natriferic activity. These results are similar to our previous studies that showed that insulin-mediated Na⁺ reabsorption via ENaC also involves the stimulation of an LY-294002-sensitive PI 3-kinase and that the PtdIns(3,4,5)P₃ formed by this enzyme is increased before the functional action on the channel (14). Insulin and aldosterone exhibit additive natriferic responses, suggesting that they are mediated by at least partially independent pathways (15). However, both hormone responses are a result of an increase in the number of active ENaCs in the apical membrane (4, 8). The similarities in PI 3-kinase inhibitor effects on the two hormone responses suggest that the enzyme may be mediating an effect at or near the channel.

PI 3-kinases play pivotal roles in a wide variety of metabolic processes. Specificity in the regulation of these diverse pathways may be explained, in part, by the growing number of PI 3-kinase variants that have been described recently (7, 14, 20, 22). Multiple variants can coexist in a single cell. In fact, four different isoforms of PI 3-kinase have been identified in adipocytes. Our results indicate that both basal Na⁺ transport as well as the aldosterone-stimulated portion of the natriferic activity are sensitive to inhibition by LY-294002, whereas only the basal transport is inhibited by wortmannin. Again, this is consistent with our previous studies showing that the insulin-stimulated increase in Na⁺ transport is a wortmannin-insensitive, LY-294002-sensitive process, whereas basal transport is sensitive to both inhibitors (14). These data suggest that at least two variants of PI 3-kinase are involved in transcellular Na⁺ transport.

LY-294002 inhibition of transport assessed macroscopically and at the single-channel level is fully reversible after removal of LY-294002, indicating the dynamic role of PI 3-kinase in regulation of transport. Notably, inhibition of transport is essentially the same for both control and aldosterone-pretreated tissues, indicating the importance of PI 3-kinase for functional expression of ENaCs.

The time course of the inhibitory effect in both basal and aldosterone-pretreated tissues is relatively slow. These results can be compared with the data of Kurashima et al. (10) who found that inhibitors of PI 3-kinase alter the constitutive endosomal recycling of the Na⁺/H⁺ exchanger NHE3. The time course of the inhibitory effect on the functional activity of the exchanger in intact cells was remarkably similar to the inhibition of Na⁺ transport in our study. It is also worth noting that ENaC has been shown to be a relatively short-lived protein with an estimated half time of ~1 h (18). The data obtained in the Kurashima studies suggested that the functional effects were likely due to a decrease in the cycling of intracellular Na⁺/H⁺ exchangers to the plasma membrane (10). Although the very low number of endogenous Na⁺ channels in our native epithelial cells makes similar studies examining recycling very difficult, we have examined the effect of LY-294002 on the number of active channels expressed on the apical membrane of A6 cells. When the inhibitory effect was assessed at a single-channel level, we found that the effect on macroscopic transport is clearly due to a decrease in the number of active channels in the apical membrane (Fig. 4). Although this result, combined with the relatively slow time course, is certainly consistent with a role for PI 3-kinase in dynamic recycling of Na⁺ channels, it should be noted that these biophysical measurements of active channels cannot distinguish between insertion of new channels from an intracellular pool and activation of quiescent channels that were preexisting in the membrane.

ENaC has been shown to be involved in rare forms of blood pressure aberrations (5, 17) and has been implicated as a potential factor in many forms of essential hypertension (16). Therefore, a complete elucidation of the hormonal pathways that regulate the activity of the channel is important to understand the etiology of pathological processes that may arise from aberrant regulation of this signaling. These studies provide evidence of a crucial role for phosphoinositide formation in aldosterone-stimulated Na⁺ transport and represent novel data regarding the intracellular pathway linking mineralocorticoid receptor binding to the final natriferic effect in transporting epithelial cells. Although PI 3-kinases have been linked to a variety of peptide factor transport responses, this is the first demonstration of a role for these enzymes in a steroid hormone response.