INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE EPITHELIAL SODIUM CHANNEL (ENaC) resides in the apical membrane of Na⁺-transporting epithelia in the distal nephron, distal colon, lung, and other epithelia (14). In its fully functional state, it is composed of three homologous subunits (4, 5, 25). Its major function is to provide the rate-limiting step to transepithelial Na⁺ transport. Thus it plays a central role in regulating fluid homeostasis and blood pressure. The recent demonstration that activating and inactivating mutations in this channel produce hypertension and hypotension, respectively, prove its central importance in these functions (32).

Na⁺ absorption through ENaC is regulated through a number of mechanisms including steroid-induced channel synthesis, phosphorylation by intracellular kinases, methylation, and ionic effects (14). The role of kinases in the rapid (minutes) regulation of ENaC function has been well recognized for many years, but the molecular mechanisms remain elusive. One possibility is that direct phosphorylation of one or more of the ENaC subunits is responsible for regulation of its activity. A recent report demonstrated that two of the three ENaC subunits ( and , but not ) can be phosphorylated in vivo by treatment with protein kinase A (PKA), protein kinase C (PKC), insulin, or aldosterone (35). The demonstration of direct phosphorylation of ENaC subunits does not exclude the possibility that phosphorylation of other proteins participates in ENaC regulation. Intracellular kinases might phosphorylate proteins that associate with ENaC, or phosphorylation cascades might produce effects that are only distantly related to channel phosphorylation.

One of the difficulties encountered by investigators attempting to dissect the mechanisms responsible for kinase-mediated regulation of ENaC is that specific kinase activators and inhibitors have different effects in different tissues. For example, whereas PKC is generally considered to be an inhibitor of ENaC (13, 16, 24), this deduction overlooks the significant differences in response to phorbol esters in different tissues (6, 11).

The purpose of the present experiments was to begin to develop a system where kinase activity could be systematically studied to dissect the molecular mechanisms responsible for ENaC regulation. We chose the Xenopus oocyte because of its versatility and simplicity in evaluating effects of heterologously expressed proteins. The results provide a basis for suspecting a complex network of kinases interacting to regulate ENaC activity.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Channel expression in Xenopus oocytes. The human ENaC (hENaC) -, -, and -subunits used in these experiments have been described previously (28, 40). Specifically, we used the -1 subunit as defined by Thomas et al. (40). The coding regions for the hENaC subunits were subcloned into the PGEM-HE plasmid. This plasmid has been engineered for use in the Xenopus oocyte cRNA expression system (23). It contains a T7 promoter site upstream from the 5'-untranslated region (UTR) and 3' UTR of Xenopus -globin mRNA. The hENaC subunit coding sequence was inserted into the multiple cloning site between the -globin UTRs. We have found that generating hENaC cRNA using the PGEM-HE construct results in enhanced hENaC expression in oocytes compared with cRNA containing no UTRs (unpublished observations). Plasmids were amplified using the JM109 strain of Escherichia coli (Promega), purified by CsCl banding, and linearized with a specific restriction enzyme (Nhe I or Sph I) that cuts just downstream to the -globin 3' UTR. The mMessage mMachine (Ambion) T7 in vitro transcription kit was used to produce capped cRNA from each construct. The integrity of the cRNAs was evaluated by agarose gel electrophoresis and quantitated by densitometry. The cRNAs were diluted in water so that 50-nl injections with a Drummond Nanoject oocyte injector carried 1 ng of each subunit.

Xenopus oocyte handling and current measurement. Mature female Xenopus laevis were housed in the University of Iowa animal facility in dechlorinated tap water at 18-20°C. Stage V and VI oocytes were removed from toads that were anesthetized in an ice-cold 2 mg/ml tricaine solution. The oocytes were defolliculated with collagenase and stored overnight at 18°C in frog Ringer solution consisting of 115 mM NaCl, 2.5 mM KCl, 1.8 mM CaCl₂, 10 mM HEPES, 5 mM sodium pyruvate, and 100 U/ml penicillin-streptomycin (pH 7.35). After 12-24 h of recovery from the collagenase treatment, healthy oocytes were injected with cRNA from -, -, and -hENaC. Ringer solution was changed daily. Whole cell hENaC currents were measured 48-72 h after cRNA injections. Measurements were made in frog Ringer solution using an OC-725C oocyte voltage-clamp amplifier (Warner Instruments). The pCLAMP software suite (Axon Instruments) was used for amplifier control and data collection/analyses. All recordings were performed at room temperature. Amiloride-sensitive currents were derived by subtracting currents recorded in 10-33 µM amiloride from preamiloride currents. Whole cell capacitance was determined electronically using the automated voltage-step protocol and current transient analysis of the pCLAMP program.

Short-circuit current measurements. M-1 cells were obtained from Dr. Géza Fejes-Tóth (Dartmouth Medical School) and cultured as previously described (31). Briefly, cells were seeded at confluent density on Millicell PCF filters pretreated with human placental collagen. The filters were grown 3 days in DMEM/Ham's F-12 supplemented with insulin (5 µg/ml), transferrin (5 µg/ml), triiodothyronine (5 nM), hydrocortisone (50 nM), sodium selenite (10 nM), gentamicin (50 µg/ml), BSA (10 g/l), and dexamethasone (5 nM). The monolayers were grown 1 day in the same medium without albumin and steroids and then 1 day in albumin- and serum-free media with the addition of 100 nM aldosterone plus 100 nM dexamethasone. Measurements of transepithelial voltage, resistance (R_t), and short-circuit current (I_sc) were conducted at 37°C as described (31). The bathing solution used for examining the staurosporine effect contained (in mM) 140 NaCl, 5 KCl, 10 HEPES, 1 mM MgCl₂, 1.5 mM CaCl₂, 1 mM Na₂HPO₄, and 5 mM glucose. Benzamil-sensitive currents were derived by subtracting currents recorded in 10 µM benzamil from prebenzamil currents.

Materials. Mature female X. laevis were purchased from Xenopus I (Dexter, MI) or Nasco (Fort Atkinson, WI). Benzamil, phorbol 12-myristate 13-acetate (PMA), and 4-PMA were purchased from RBI (Natick, MA). Staurosporine, chelerythrine chloride, calphostin C, H-8, KN-93, SB-203580, and 1-oleoyl-2-acetyl-sn-glycerol (OAG) were obtained from Biomol (Plymouth Meeting, PA). All other chemicals were purchased from Sigma Chemical (St. Louis, MO). The Millicell filters were purchased from Millipore (Bedford, MA).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PKC stimulation enhances ENaC currents. Our initial intent was to study the mechanism of PKC inhibition of ENaC currents expressed in Xenopus oocytes as was reported in Ref. 1. To accomplish this, we injected the cRNAs for hENaC , , and  into Xenopus oocytes and recorded currents while applying 80-100 nM PMA in the bathing solution. Surprisingly, when the magnitude of the whole cell currents was <2 µA at 60 mV, PMA caused a reproducible increase in current. A representative example of this effect is shown in Fig. 1A. Peak enhancement typically occurred ~7-9 min after exposure to PMA. Figure 1B demonstrates that the percent increase in current induced by PMA was inversely related to the magnitude of the current measured before PMA addition. The smallest baseline currents (<500 nA) were typically at least doubled in magnitude while larger currents were only modestly increased if there was any increase at all. The results were similar whether human or rat ENaC cRNA was injected. Water-injected oocytes had no amiloride-sensitive currents and did not respond to PMA (data not shown). It is important to note that 10 µM amiloride consistently reduced the baseline and PMA-enhanced inward currents from hENaC-expressing oocytes to magnitudes similar to the amiloride-insensitive currents recorded from water-injected oocytes (from 0 to 200 nA at 60 mV). These observations strongly suggest that both the baseline currents and the PMA-enhanced currents were carried through hENaC.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Phorbol 12-myristate 13 acetate (PMA) enhances whole cell currents recorded from Xenopus oocytes injected with cRNA for human epithelial Na+ channel (hENaC) -, -, and -subunits. A: each data point represents average current during a 500-ms hyperpolarization to 60 mV from a holding voltage of 0 mV. By convention, the flow of positive ions into the oocyte is shown as negative current. Current values were recorded 5 s apart; 2 min were allowed for establishment of baseline current values and run-down rate; 80 nM PMA and 10 µM amiloride were added to the bathing solution at the indicated times. B: relationship between the PMA-induced current increase and pre-PMA current magnitude.

View larger version (10K): [in this window] [in a new window]   Fig. 2.   4-PMA does not stimulate hENaC currents. Whole cell inward currents were recorded at 60 mV every 5 s. Each data point represents average current during a 500-ms hyperpolarization. Additions were made to the bathing solution as indicated by arrows; 100 nM 4-PMA did not enhance current in the same oocyte where subsequent exposure to 80 nM PMA caused a typical increase in current. This was a consistent finding in 7/7 experiments.

Effect of kinase inhibitors. The PMA and OAG results suggested that PKC might be involved in the minute-to-minute regulation of hENaC activity in Xenopus oocytes. However, since PMA treatment enhanced hENaC current only when the baseline current was <2 µA, we suspected that PKC might be maximally active in oocytes expressing large hENaC currents. We also considered the possibility that other kinases might play a role in regulating (increasing) hENaC currents. This notion derives in part from experiments showing variability in the response to PKC agonists (6, 11). We therefore performed experiments using kinase inhibitors with varying specificities for PKC and other known kinases. A kinase inhibitor profile for hENaC activity would serve two purposes. The PKC inhibitors would test the hypothesis that PKC was involved in acute regulation of hENaC activity, and other kinase inhibitors could identify molecules that might also participate in the regulation of hENaC.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Selected kinase inhibitors reduce amiloride-sensitive currents. Currents were recorded at 60 mV from oocytes incubated in various kinase inhibitors for 30-90 min. Concentrations were 100 nM staurosporine, 10 µM chelerythrine, 100 µM SB-203580, 1 µM calphostin C, 1 µM H-8, and 1 µM KN-93. Each bar represents the average of the amiloride-sensitive inward currents that were normalized to the inhibitor-free control currents of that day. Numbers in parentheses represent numbers of oocytes used to calculate average and SE. * Significantly different from control (P < 0.05).

Staurosporine-sensitive current. The striking inhibitory effect of staurosporine on hENaC currents in Xenopus oocytes prompted further investigation. First, we asked whether the reduction in inward current at 60 mV was due to the inhibition of the Na⁺-selective hENaC. In the vehicle control experiments, nearly all of the current was amiloride sensitive; therefore, it was reasonable to conclude that the staurosporine-sensitive channel was hENaC. A further test of this conclusion was the determination of the current-voltage relationships and calculated reversal potentials (E_rev) for the control and staurosporine-treated whole cell currents. As shown in Fig. 4, the reversal potential of the control currents is approximately +5 mV. Oocytes treated with staurosporine had smaller inward currents at all voltages negative to +5 mV and smaller outward currents at all voltages positive to +10 mV. The extrapolated E_rev for the staurosporine group was approximately 2 mV. This current reduction profile and E_rev shift toward negative values caused by staurosporine are consistent with the inhibition of a Na⁺-selective channel. These results eliminate the possibility that endogenous K⁺ or Cl channel activities or alterations in intracellular ion concentrations could explain the staurosporine-induced inward current reductions at 60 mV.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Staurosporine reduces currents at all voltages and shifts the reversal potential in the negative direction. Whole cell currents were elicited with various voltage steps from a holding potential of 0 mV. Representative current traces for 2 groups, control (B) and 100 nM staurosporine (C), are shown at right. Dashed lines, zero current level. A: comparison of average current-voltage (I-V) relationships (control, n = 47; staurosporine, n = 43). Extrapolated reversal potential, defined as voltage where the I-V relationship intercepts the abscissa, is shifted toward negative values in the staurosporine-treated group.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Staurosporine does not inhibit ROMK1 Ba2+-sensitive K+ currents. Oocytes were injected with 1 ng of rat ROMK1 cRNA. After 72 h incubation in frog Ringer solution, 2.5 mM Ba2+-sensitive currents were measured in a modified frog Ringer solution containing 60 mM KCl and 0.3 mM CaCl2. Average I-V relationship for 5 control oocytes is compared with that for 5 oocytes incubated for 30-60 min in 100 nM staurosporine.

View larger version (13K): [in this window] [in a new window]   Fig. 6.   Staurosporine inhibition of hENaC currents is immediate. Currents were elicited with voltage steps to 60 mV from a holding potential of 0 mV every 5 s. Each data point represents average current during a 500-ms hyperpolarization. After a 3-min control period, during which a constant rate of rundown was established, 100 nM staurosporine was added to the bathing solution. Dashed line, best-fit regression line to the pretreatment current values; , extrapolated staurosporine-free current value at the time of the 33 µM amiloride addition.

Staurosporine-sensitive region of hENaC. We hypothesized that the staurosporine effect was mediated via an intracellular portion of the hENaC complex. We examined the COOH terminus because this region is involved in regulating ENaC activity (33, 34, 37). First, we mutated each of the three subunits to remove the COOH-terminal portions beginning just downstream of the M2 membrane-spanning domain. These three truncations,  S594X,  R566X, and  K576X represent similar structural modifications and allow expression of channel activity. Figure 7 shows that truncation of the - or -subunit resulted in enhanced baseline currents. These results are consistent with reports demonstrating that elimination of the PPPXY motif increases Na⁺ transport and causes Liddle's syndrome (33, 34, 37). However, in contrast to  and , truncation of the COOH terminus of  resulted in control currents that were not larger in magnitude than wild-type currents. This result is consistent with our previous report (37). Staurosporine inhibited currents when  or  were truncated. However, staurosporine did not inhibit hENaC currents when a truncated  construct was expressed. These results suggest that the staurosporine-sensitive region is located in the COOH terminus of the -subunit and not the - or -subunit.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Truncation of intracellular COOH terminus of -hENaC, but not - or -hENaC, eliminates the staurosporine effect. Amiloride-sensitive currents were measured from 4 groups of cDNA-injected oocytes 48 h after nuclear injection. Open bars, average staurosporine-free currents at 60 mV that have been adjusted for rundown as shown in Fig. 6; solid bars, average currents 10 min after bath application of 100 nM staurosporine. Numbers in parentheses represent number of oocytes used to calculate average and SE. * Statistical significance between control and staurosporine currents at P < 0.05 using a paired t-test.

Staurosporine inhibits M-1 cell I_sc. The M-1 mouse collecting duct epithelial cell line has been used extensively to study amiloride-sensitive Na⁺ transport through ENaC (7, 21, 22, 38). Based on the strong and rapid inhibitory effect of staurosporine on heterologously expressed hENaC, we hypothesized that there were staurosporine-sensitive kinases involved in short-term ENaC regulation in cells that naturally express ENaC. We therefore asked whether staurosporine inhibited Na⁺ transport in the M-1 cell line by measuring I_sc while applying staurosporine. Figure 8 demonstrates the dose-dependent inhibition of Na⁺ transport in M-1 cell monolayers by staurosporine.

View larger version (11K): [in this window] [in a new window]   Fig. 8.   Staurosporine inhibits M-1 cell monolayer short circuit currents (Isc). Each bar represents average fraction of benzamil-sensitive current in each group. Values were determined 30 min after solution change containing no staurosporine (control), 100 nM staurosporine, or 1 µM staurosporine. Baseline currents were 10-20 µA/cm2, and postbenzamil currents were 1-2 µA/cm2. Numbers in parentheses represents numbers of monolayers in each group. * Significant difference from control (P < 0.05) using ANOVA and Newman-Keuls multiple comparisons.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The present results demonstrate that kinases endogenous to the Xenopus oocyte can participate in the regulation of hENaC activity. Activation of the PKC system stimulates the activity of hENaC (Fig. 2), while several kinase inhibitors, including chelerythrine, a relatively specific inhibitor of PKC, inhibit hENaC activity (Fig. 3). Staurosporine, a nonspecific kinase inhibitor, had the most dramatic effect of the inhibitors tested and produced a rapid reduction in Na⁺ current (Fig. 6). The effect of staurosporine seems to be relatively specific for hENaC activity, as there was no effect on the ROMK1 K⁺ channel (Fig. 5). This kinase regulation appears to be effected via a specific region of the hENaC complex. Whereas truncation of the COOH-terminal regions of the - and -subunits did not alter the inhibitory effect of staurosporine (Fig. 7), truncation of the analogous region of the -subunit eliminated this staurosporine effect. We infer from these results that a domain within this region of the -subunit plays an important role in the kinase-mediated regulation of hENaC.

These results might appear to conflict with those of Awayda et al. (1) who reported that phorbol esters inhibit ENaC currents in oocytes. However, the time course of these studies were different; these authors examined the effects after 30 min of exposure, whereas our data report the effects after only 10 min of exposure. It is likely that the timing and specific experimental conditions influence the nature of the reaction to phorbol esters. The response to phorbol esters is also somewhat dependent on the specific oocyte. As demonstrated in Fig. 1, those with larger hENaC currents show a smaller response to phorbol esters. We interpret these results to indicate that there may be a variety of factors in the oocyte that influence the magnitude of hENaC currents. The effect of an inhibitor of p38 MAP kinase, SB-302580 (Fig. 3), supports this interpretation. In addition, it has been recently reported that overexpression of a steroid-induced kinase, sgk, increases ENaC currents in oocytes (9, 29). The activity of such kinases might vary from oocyte to oocyte.

What is the relevance of effects of protein kinases on hENaC function in Xenopus oocytes to mammalian epithelial cells? This question touches on the relevance of all heterologous expression systems and must be addressed in any effort to integrate isolated molecular actions into models of cell and organ function. In this regard, staurosporine inhibits Na⁺ current in the cortical collecting duct (CCD) cell line, M-1 (Fig. 8). Thus its actions in the oocyte might have relevance to regulation of ENaC in intact cells. However, the action of kinases on ENaC function is much more varied than is commonly appreciated. For example, activation of PKA in rat CCD produces a sustained increase in Na⁺ transport, whereas similar maneuvers applied to the rabbit CCD produces inhibition of Na⁺ transport (3, 8). In addition, and perhaps more relevant for the present experiments, phorbol esters can either increase or decrease Na⁺ transport depending on the tissue examined (11). These divergent effects appear to be tissue, rather than species, specific. Amphibian skins from two different species demonstrate enhancement of Na⁺ transport in response to phorbol esters, whereas urinary bladders from these species respond by decreasing Na⁺ transport (6). One possible explanation for this difference in response is different PKC isoforms in the various tissues (6). It is also possible that activation of the resident PKC might participate in a cascade with other endogenous protein kinases to produce the final effect on ENaC function.

What inferences can we make about the mechanism(s) of kinase regulation from these experiments? First, it appears that there is endogenous kinase activity (in the oocyte and in M-1 cells) that participates in maintaining an active ENaC; the response to the protein kinase inhibitors demonstrates this point. Second, PKC may play a role in this activity, but it cannot explain the entire effect. The response to phorbol esters is modest and dependent on the baseline current. Furthermore, relatively specific inhibitors of PKC did not reduce the Na⁺ current to values similar to those of the more nonspecific kinase inhibitor, staurosporine. Third, this effect of staurosporine is not mediated via generalized endocytosis (41). If endocytosis were a prominent feature, we would have expected that all membrane proteins, including ROMK1, would be reduced in activity. Furthermore, the oocyte capacitance was not altered by staurosporine. Fourth, this kinase effect is not mediated solely through the COOH termini of - or -hENaC; truncation of these regions did not prevent staurosporine from inhibiting the Na⁺ current. Thus the recent demonstration that these regions can be phosphorylated (35) apparently does not explain the actions of staurosporine reported here. Finally, in contrast to the - and -subunits, the COOH terminus of the -subunit does seem to play a role in producing the staurosporine effect. This result is even more intriguing in the context of the failure of kinases to phosphorylate this region under the same conditions where - and -subunits were phosphorylated (35). Thus these kinase-dependent effects may not be mediated directly via phosphorylation of ENaC.

A review of the literature on the effects of mutations of the -ENaC subunit identifies clues as to the location of the region that might be responsible for the staurosporine effect. First, two groups have reported that mutation of the key amino acids of the COOH-terminal PPPXY motif of any one of the three subunits will produce an increase in Na⁺ current (33, 37). We note that this result may not be as simple as it first appears, as not all investigators agree about the importance of this motif in the -subunit (36). Nevertheless, this motif within the -subunit seems to participate in restraining ENaC activity. Second, a truncation of the rENaC -subunit at the P646 position (a COOH-terminal deletion beginning 25 amino acids upstream of the PPPXY motif) produces an increase in current (34). It seems likely that this effect is due to the elimination of the PPPXY motif. Third, a truncation of the hENaC -subunit at the S594 position (just 3' to the second membrane-spanning domain) produces a decrease in Na⁺ current (Fig. 7 and Ref. 37). Integrating these results leads to the conclusion that the effect of staurosporine may be mediated through the region of the -subunit somewhere between the second membrane-spanning region and the PPPXY motif. Inspection of this region does not provide obvious candidate domains.

In conclusion, we have demonstrated that kinase inhibitors can reduce Na⁺ transport in M-1 cells and ENaC activity in Xenopus oocytes. These results suggest that endogenous kinases stimulate ENaC activity. The specific kinase(s) responsible for this effect is not clear at present, but the aggregate data suggest that several are involved. The action of these kinases appears to be mediated through the COOH-terminus of the -subunit.