INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

CHLORIDE SECRETION across the airway epithelium provides the driving force for fluid secretion into the airways of the fetal lung and is essential for normal lung development. Lung fluid secretion rate decreases significantly a few days before birth (4). At the same time, increases in Na⁺ and fluid absorption occur across the pulmonary epithelium (37). Before and shortly after birth, airway and alveolar fluid is removed, allowing for efficient gas exchange. Active Na⁺ transport across the alveolar epithelium is thought to play a major role in this transition (8, 37, 39, 40), and there is evidence to suggest that fluid clearance is enhanced by -adrenergic receptor agonists in both fetal (10, 22, 37, 41, 54) and adult lung (6, 23, 48). These changes in transepithelial transport correlate with an increase in plasma epinephrine concentration and an increase in -adrenergic receptor expression in pulmonary epithelial tissue late in gestation (10, 15, 55, 56). In addition, levels of epithelial Na⁺ channel (ENaC) mRNA and of Na⁺-K⁺-ATPase mRNA, protein, and activity increased late in gestation and remained high in adult lung (28, 38, 53). These results support the idea that stimulation of fluid absorption by -adrenergic receptor agonists results from an increase in net Na⁺ absorption (47). This hypothesis is supported by experiments using isolated perfused rat lungs (17, 26), fetal lung epithelium (34, 36, 51), and isolated rat alveolar type II cells (14, 49, 58) and by in vivo experiments (7, 24, 29). In isolated perfused rat lung (21, 29, 30) and in vivo studies (24, 29), fluid absorption was considerably reduced by Na⁺ channel blockers. Channels with high and low affinity for amiloride coexist in the apical membrane of fetal rat lung epithelial cells (32), with different cation selectivity [permeability ratio (P_Na/P_K) = 0.9 (42), P_Na/P_K > 10 (52, 53)]. In adult rat, high-affinity amiloride-sensitive Na⁺ channels have been identified in cultured alveolar type II cells (14, 16, 46). In addition, low-affinity amiloride-sensitive Na⁺ channels are present in freshly isolated (33) and cultured (57, 58) adult rat alveolar type II cells.

In contrast to the results of experiments using fetal airway and distal lung epithelial cells (3, 5, 30, 50), adult alveolar type II cells absorb Cl in response to -adrenergic stimulation. This conclusion was based on isotopic flux measurements using monolayers of adult rat alveolar type II cells (31). The specific mechanisms involved in -adrenergic stimulation of Cl absorption are presently unknown. The objective of this study was to determine the effect of -adrenergic receptor stimulation and 8-(4-chlorophenylthio)adenosine 3',5'-cyclic monophosphate (CPT-cAMP) on Cl and Na⁺ transport properties of cultured primary adult alveolar epithelial cells.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Materials. Male Sprague-Dawley rats weighing 150-174 g were purchased from Harlan (Indianapolis, IN). Elastase was purchased from Worthington Biochemical (Freehold, NJ). Rat IgG, DNase I, nonessential amino acids, BSA, L-glutamine, HEPES, trypsin inhibitor, terbutaline, and glibenclamide were obtained from Sigma Chemical (St. Louis, MO). DMEM-Ham's F-12 nutrient mixture in a 1:1 ratio (DMEM/F-12) and penicillin-streptomycin were purchased from GIBCO BRL (Grand Island, NY). Nitex mesh (120 and 40 µm) was purchased from Tetko (Elmsford, NY). Tissue culture-treated Transwell polycarbonate filters were obtained from Corning Costar (Cambridge, MA). PBS was obtained from Celox Laboratories (Oakdale, MN). Amiloride was obtained from Merck Sharp & Dohme Research Laboratories (West Point, PA). Benzamil, 5-(N-ethyl-N-isopropyl)-amiloride (EIPA), N-phenylanthranilic acid [also called diphenylamine-2-carboxylic acid (DPC)], 5-nitro-2-(3-phenylpropylamino) benzoic acid (NPPB), and CPT-cAMP were purchased from RBI (Natick, MA). All chemical reagents used for electron microscopy were obtained from Electron Microscopy Sciences (Fort Washington, PA). All other chemicals were purchased from Sigma Chemical.

Cell preparation and culture. Alveolar epithelial cells were isolated from adult rat lungs using a modification of the protocol described by Borok et al. (9). Rats were anesthetized with an intraperitoneal injection of pentobarbital. Lungs were perfused with solution 2 (in mM: 140 NaCl, 5 KCl, 2.5 NaH₂PO₄, 1.3 MgSO₄, 2.0 CaCl₂, 6 glucose, and 10 HEPES). After removal, the lungs were repeatedly lavaged with solution 1 (in mM: 140 NaCl, 5 KCl, 2.5 NaH₂PO₄, 6 glucose, and 10 HEPES) and solution 2 to eliminate macrophages. The lungs then were filled with elastase-containing solution (2.7 U/ml in solution 2) and were incubated at 37°C for 30 min in a shaker bath. Elastase was neutralized by stop solution [2 mM EDTA, 1% BSA, 0.1% soybean trypsin inhibitor, and 0.15/ml DNase I in a buffered saline solution containing (in mM) 136 NaCl, 2.2 Na₂HPO₄, 5.3 KCl, 10 HEPES, and 5.6 glucose]. Finely minced tissues were filtered through 120- and 40-µm Nitex mesh. Cells were further purified by panning on IgG-coated petri dishes to remove remnant macrophages and suspended directly in serum-free DMEM/F-12 medium supplemented with 1.25 mg/ml BSA, 0.1% nonessential amino acids, 2.0 mM glutamine, 100 U/ml sodium penicillin G, and 100 µg/ml streptomycin. The cells were then seeded onto Transwell membrane filters (4.52 cm², 0.4-µm pore size) at a density of 1.5 × 10⁶ cells/cm² to prepare confluent monolayers. The medium was changed every other day. The resistance of the monolayers was measured using an epithelial voltohmmeter (WPI, New Haven, CT). High-resistance monolayers formed on day 4 or 5. All the measurements were performed on day 5, 6, or 7 following isolation.

Ussing chamber measurements. High-resistance monolayers on Transwell inserts were mounted in Ussing chambers and bathed with identical solutions [either serum-free DMEM/F-12 medium or Cl-free Ringer solution containing (in mM) 120 sodium methanesulfonate, 10 potassium methanesulfonate, 30 mannitol, 3 calcium gluconate, 0.7 MgSO₄, 20 NaHCO₃, and 0.3 NaH₂PO₄, at 37°C, bubbled with 95% O₂-5% CO₂] on each side to eliminate any chemical driving force across the monolayer. Monolayer potential difference (luminal side as reference), short-circuit current (I_sc), and resistance were measured with voltage-clamp circuitry from JWT Engineering (Overland Park, KS). Workbench data acquisition software (Kent Scientific) was used to record the data. To measure apical membrane conductance, amphotericin B (10 µM) was added to the serosal solution to eliminate the basolateral membrane as a barrier to ion movement. Potassium methanesulfonate Ringer solution (in mM: 120 potassium methanesulfonate, 30 mannitol, 3 calcium gluconate, 0.7 MgSO₄, 20 KHCO₃, 0.3 KH₂PO₄, and 10 NaCl) was used as the serosal solution and either serum-free DMEM/F-12 medium or sodium methanesulfonate Ringer solution (in mM: 120 sodium methanesulfonate, 30 mannitol, 3 calcium gluconate, 0.7 MgSO₄, 20 NaHCO₃, and 0.3 NaH₂PO₄) was used to bathe the apical surface of the monolayer. Cl concentration ([Cl]) in the basolateral solution was manipulated by replacing potassium methanesulfonate with equimolar KCl to observe changes in reversal potentials (E_rev). For the halide permeability experiments, the basolateral solution contained (in mM) 70 potassium methanesulfonate, 30 mannitol, 3 calcium gluconate, 0.7 MgSO₄, 20 KHCO₃, 0.3 KH₂PO₄, and 50 KCl. The apical solution was similar to the basolateral solution, except that 50 mM KCl was replaced with 50 mM KSCN, KCl, KBr, or KI. A World Precision Instrument (Sarasota, FL) epithelial voltage clamp was used in combination with a Dagan LM-12 analog-to-digital interface (Dagan, Minneapolis, MN) and pCLAMP software (Axon Instruments) to run the voltage-clamp protocol and record the resulting currents. Current-voltage (I-V) relationships of apical membranes were obtained by imposing a voltage step command protocol (see Figs. 4, 11, and 13) with a holding potential of 0 mV. The currents before and after addition of compound were subtracted to obtain the compound-sensitive components of the current. The halide permeability ratios P_X/P_Cl (where X is SCN, Br, or I) were calculated from reverse potential (E_rev) measurements using the Goldman-Hodgkin-Katz equation E_rev = (RT/zF) · ln[(P_X · [X])/(P_Cl · [Cl])], where R is the gas constant, T is absolute temperature, z is valence, and F is Faraday's constant.

Electron microscopy. Alveolar epithelial cells were grown to confluence on the Transwell membrane filters and mounted in Ussing chambers to measure ion transport activities on day 7. After the experiments, the monolayers were fixed with 2% glutaraldehyde in 0.1 M sodium cacodylate for 30 min at room temperature. The monolayers then were washed four times with 0.1 M sodium cacodylate (pH 7.4) and postfixed with 1% OsO₄ in cacodylate for 1 h at room temperature. After three quick rinses with 0.1 M sodium cacodylate, the membranes were removed from the inserts with a scalpel and cut into strips ~2 × 0.5 mm. Strips were dehydrated with ethanol and embedded in epon. Thin sections were stained with uranyl acetate and lead citrate and examined using a Jeol CX100 electron microscope.

Statistics. All data are presented as means ± SE, and n is the number of monolayers studied. The IC₅₀ values for amiloride, benzamil, EIPA, NPPB, glibenclamide, and DPC were determined by using a four-parameter logistic function to fit the data. The IC₅₀ of each compound was derived from the equation used to fit the concentration-response relationship. Differences between means were analyzed by using either a paired or unpaired t-test as appropriate.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Cell characterization and basal bioelectric properties. Isolated alveolar type II cells undergo a number of phenotypic changes with time in culture, including decreased ability to secrete surfactant lipid (20), gradual loss of lamellar bodies (19), and expression of type I cell surface epitopes (18). These changes are due to their gradual transformation to alveolar type I cells (13, 18). Previous studies have shown that cultured alveolar epithelial cells can be successfully maintained under completely defined serum-free conditions (9). Cultured alveolar epithelial cells maintained some alveolar type II cell morphological characteristics for at least 7 days following isolation under the culture conditions used in these studies. The cells exhibited an uneven cuboidal shape and had microvilli associated with the apical membrane (Fig. 1), but lamellar bodies could not be clearly distinguished. Alveolar epithelial cells reached confluence on Transwell membrane filters within 5 days after isolation and exhibited mean maximum transepithelial resistance of 3,898 ± 194  · cm² on days 6 and 7 (Fig. 2). The mean I_sc was 1.96 ± 0.15 µA/cm² (n = 38), and the mean potential difference was 1.33 ± 0.18 mV (n = 38, luminal side as reference).

View larger version (68K): [in this window] [in a new window]   Fig. 1.   Morphology of alveolar epithelial cells cultured on Transwell membrane filters 7 days following isolation. Cells maintained some type II cell characteristics, exhibiting an uneven cuboidal shape and containing microvilli. Scale bar, 1 µm.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Time course of transepithelial resistance development of alveolar epithelial monolayers (n = 12). Mean maximum transepithelial resistance was 3,898 ± 194  · cm2.

Effects of Na⁺ channel blockers on I_sc. Figure 3 shows the effects of Na⁺ channel blockers on the basal I_sc of intact alveolar epithelial monolayers mounted in Ussing chambers and bathed with the same serum-free DMEM/F-12 medium on both apical and basolateral sides. In Fig. 3A, addition of amiloride (20 µM) to the apical solution immediately inhibited 59.95 ± 2.60% of the basal I_sc (n = 10), after which the current remained relatively constant. The concentration-response relationships for amiloride and related analogs are shown in Fig. 3B. The rank order of potency for inhibition by these compounds was benzamil  amiloride > EIPA, with mean IC₅₀ values of 0.50, 0.96, and 260 µM, respectively. To further characterize the properties of the amiloride-sensitive conductance, the basolateral membrane barrier was eliminated by treatment with amphotericin B. After permeabilization, an initial determination of the apical membrane I-V relationship was made, followed by addition of 20 µM amiloride to the apical solution. The I-V relationship for the amiloride-sensitive current is reported in Fig. 4. The E_rev for the current was 46.46 ± 3.44 mV, indicating a high selectivity for Na⁺ over K⁺ (12.5:1). Treatment of monolayers with terbutaline (2 µM) produced no significant change in either the slope or the E_rev (49.58 ± 4.29 mV) of the amiloride-sensitive I-V relationship, indicating that terbutaline does not increase the amiloride-sensitive Na⁺ conductance of the apical membrane (Fig. 4).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Effects of Na+ channel blockers on short-circuit current (Isc). Monolayers were mounted in Ussing chambers and bathed on both apical and basolateral sides with same serum-free DMEM/F-12 medium. A: representative Isc tracing showing apical addition of amiloride (20 µM; n = 10). Initial mean potential difference and conductance values were 1.53 ± 0.40 mV and 10.13 ± 2.22 mS (n = 10; luminal side as reference), respectively. B: apical addition of Na+ channel blockers decreases Isc in a dose-dependent fashion, with rank order of potency of benzamil  amiloride > 5-(N-ethyl-N-isopropyl)-amiloride (EIPA). IC50 values were 0.50, 0.96, and 260 µM for benzamil (n = 4), amiloride (n = 5), and EIPA (n = 5), respectively. Initial mean potential difference and conductance values for benzamil, amiloride, and EIPA were 1.03 ± 0.41 mV and 10.75 ± 3.08 mS (n = 4), 1.05 ± 0.26 mV and 11.15 ± 2.45 mS (n = 4), and 0.90 ± 0.14 mV and 10.43 ± 2.39 mS (n = 4), respectively.

View larger version (26K): [in this window] [in a new window]   Fig. 4.   Current-voltage (I-V) relationship for amiloride-sensitive conductance in apical membrane. Experiments were performed with amphotericin-permeabilized monolayers mounted in Ussing chambers and bathed with potassium methanesulfonate Ringer solution on basolateral side and serum-free DMEM/F-12 medium on apical side. A: representative tracing of amiloride-sensitive current in absence of terbutaline recorded in 5-mV step increments from 60 to +80 mV. B: mean reversal potentials (Erev) for amiloride-sensitive current in absence () or presence () of terbutaline were 46.47 ± 3.44 (n = 9) and 49.58 ± 4.29 mV (n = 6), respectively. Amiloride (20 µM) was applied to apical bathing solution, and terbutaline (2 µM) was applied to basolateral bathing solution. Initial mean potential difference and conductance values were 2.80 ± 0.87 mV and 4.25 ± 0.544 mS (n = 9) for control and 1.50 ± 0.27 mV and 5.20 ± 0.72 mS (n = 6) for terbutaline, respectively.

Effects of terbutaline and CPT-cAMP on I_sc. Addition of the -adrenergic agonist terbutaline (2 µM) to the basolateral bath caused a rapid decrease in I_sc, followed by a slow rebound toward the initial basal current (Fig. 5A). The mean peak current drop stimulated by terbutaline was 3.2 ± 0.42 µA (n = 7). Subsequent addition of amiloride (20 µM) to the apical bathing solution blocked 62.5 ± 1.8% (n = 5) of the remaining current. Pretreatment with 20 µM amiloride significantly decreased the amplitude of the terbutaline-sensitive current response (from 3.2 ± 0.42 µA to 1.06 ± 0.10 µA, n = 5, P < 0.01) and completely eliminated the recovery of the I_sc to its initial level (Fig. 5B). Addition of 100 µM CPT-cAMP to basolateral bath produced an I_sc response that was identical to that of terbutaline (Fig. 6). The peak inward current produced by CPT-cAMP was 1.88 ± 0.26 µA (n = 5). Addition of 20 µM amiloride to the apical solution following CPT-cAMP produced a response similar to that shown in Fig. 5A. When serum-free DMEM/F-12 medium was replaced with Cl-free Ringer solution, the response to either terbutaline or CPT-cAMP was totally inhibited, indicating that the I_sc decrease was Cl dependent (Fig. 7). Subsequent addition of 20 µM amiloride to the apical solution under Cl-free conditions produced a response similar to that observed with serum-free DMEM/F-12 medium. Note, however, that the amiloride-insensitive I_sc under Cl-free conditions was significantly reduced compared with the amiloride-insensitive I_sc observed in serum-free DMEM/F-12 medium (Fig. 5A). The amiloride-sensitive I_sc was not significantly different between cultures maintained in serum-free DMEM/F-12, normal Ringer solution, and Cl-free Ringer solution (6.18 ± 0.50, 5.22 ± 0.81, and 5.85 ± 0.98 µA, respectively; Fig. 8A). The amiloride-insensitive I_sc was not significantly different between normal Ringer solution (1.33 ± 0.46 µA) and Cl-free Ringer solution (0.57 ± 0.22 µA). However, the magnitude of amiloride-insensitive I_sc was significantly different between serum-free DMEM/F-12 medium (4.24 ± 0.56 µA) and normal Ringer solution (1.33 ± 0.46 µA) or Cl-free Ringer solution (0.57 ± 0.22 µA). When nonessential amino acids were added to cultures bathed in Cl-free Ringer solution, the amiloride-insensitive residual I_sc increased to a level similar to that observed in serum-free DMEM/F-12 medium (3.26 ± 0.84 µA, Fig. 8B), suggesting that amino acid-coupled Na⁺ transport was responsible for a large portion of the amiloride-insensitive I_sc. A similar response was observed when nonessential amino acids were added to cultures bathed in normal Ringer solution. To determine whether a portion of the basal I_sc was due to Cl secretion, the loop diuretics bumetanide and furosemide were added to the basolateral solution in an attempt to block Cl entry into the cells. No response was observed following treatment with up to 200 µM bumetanide and 200 µM furosemide, indicating the absence of loop diuretic-sensitive Cl secretion as a component of the basal I_sc.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Representative Isc tracings showing effects of basolateral administration of terbutaline (2 µM) on Isc with or without pretreatment with apical amiloride (20 µM). Monolayer filters were mounted in Ussing chambers and bathed on both apical and basolateral sides with same serum-free DMEM/F-12 medium. A: addition of terbutaline produced a rapid decrease in Isc, followed by a slow increase back to basal Isc. Further addition of amiloride blocked most of remaining Isc (n = 5). Initial mean potential difference and conductance values were 1.69 ± 0.38 mV and 7.76 ± 1.42 mS (n = 5), respectively. B: in presence of amiloride, addition of terbutaline produces a rapid sustained decrease in Isc without secondary recovery phase (n = 5). Initial mean potential difference and conductance values were 1.92 ± 0.82 mV and 7.74 ± 2.42 mS (n = 5), respectively.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Representative Isc tracing showing effects of basolateral administration of 8-(4-chlorophenylthio) (CPT)-cAMP (8 cpt-cAMP; 100 µM) on Isc. Monolayer filters were mounted in Ussing chambers and bathed on both apical and basolateral sides with same serum-free DMEM/F-12 medium. Addition of CPT-cAMP produces a rapid decrease in Isc, followed by a slow increase back to basal Isc. Further addition of amiloride blocked most of remaining Isc (n = 5). Initial mean potential difference and conductance values were 1.28 ± 0.52 mV and 13.45 ± 6.17 mS (n = 5), respectively.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Representative Isc tracings showing effects of basolateral administration of 2 µM terbutaline (A) or 100 µM CPT-cAMP (B) on Isc under Cl-free conditions. Monolayers were mounted in Ussing chambers and bathed with same Cl-free Ringer solution on both apical and basolateral sides. No response was produced by either compound, whereas addition of amiloride (20 µM) to apical solution produced a sustained decrease in Isc (n = 4 for each compound). Initial mean potential difference and conductance values were 1.9 ± 0.89 mV and 5.83 ± 2.56 mS (n = 4) for terbutaline and 1.05 ± 0.56 mV and 8.34 ± 3.05 mS (n = 4) for CPT-cAMP, respectively.

View larger version (18K): [in this window] [in a new window]   Fig. 8.   A: amiloride-sensitive and amiloride-insensitive Isc when monolayers were mounted in Ussing chambers and bathed on both apical and basolateral sides with same serum-free DMEM/F-12 medium (n = 10), normal Ringer (n = 9), or Cl-free solution (n = 7). Amiloride-sensitive Isc was not significantly different between groups (6.18 ± 0.53, 5.22 ± 0.81, and 5.85 ± 0.98 µA, respectively). Magnitude of amiloride-insensitive Isc was significantly different between serum-free DMEM/F-12 medium (4.24 ± 0.56 µA) and normal Ringer (1.33 ± 0.46 µA) or Cl-free solution (0.57 ± 0.22 µA) but not significantly different between normal Ringer and Cl-free Ringer solution. Initial mean potential difference and conductance values for serum-free medium, normal Ringer, and Cl-free solution were 1.53 ± 0.40 mV and 10.13 ± 2.22 mS (n = 10), 1.04 ± 0.28 mV and 8.33 ± 1.30 mS (n = 9), and 1.44 ± 0.46 mV and 8.06 ± 2.73 mS (n = 7), respectively. B: representative Isc tracing showing effects of nonessential amino acids on Isc. Nonessential amino acids (1.4 mM at each arrow) were added to both apical and basolateral solutions (n = 5). Monolayers were mounted in Ussing chambers and bathed with same Cl-free Ringer solution on both apical and basolateral sides. Initial mean potential difference and conductance values were 1.10 ± 0.41 mV and 8.70 ± 2.86 mS (n = 5), respectively.

Properties of terbutaline- or CPT-cAMP-activated Cl current. To further define the properties of the terbutaline- or CPT-cAMP-activated current, amphotericin B was added to the basolateral solution to eliminate the basolateral membrane as a resistive barrier. The effects of amiloride and terbutaline on current flow across the apical membrane are shown in Figs. 9 and 10, respectively. The amplitude of the current activated by apical addition of 2 µM terbutaline (14.06 ± 4.26 µA) was not significantly different from the terbutaline response of monolayers pretreated with 20 µM amiloride (15.29 ± 3.38 µA). The magnitude of the current blocked by amiloride (6.63 ± 1.23 µA) also was not significantly different after pretreatment of the monolayers with terbutaline (7.63 ± 0.92 µA). Replacement of Cl with methanesulfonate in both apical and basolateral solutions completely inhibited the terbutaline-sensitive current response but had no effect on the amiloride-sensitive current (Fig. 7). The I-V relationships for the terbutaline- and CPT-cAMP-sensitive difference currents are shown in Fig. 11. Figure 11A shows a representative tracing of the terbutaline-sensitive current obtained by subtracting the basal current recorded at each voltage step from the current recorded after addition of terbutaline (2 µM) to the basolateral solution. The terbutaline- and CPT-cAMP-sensitive currents exhibited nearly linear I-V relationships, with mean E_rev of 28.42 ± 2.68 (n = 12) and 26.32 ± 2.50 mV (n = 5), respectively (Fig. 11B). Increasing [Cl] in the basolateral solution from 10 to 35 mM shifted the E_rev of the terbutaline-sensitive current to more depolarized voltages (from 28.42 ± 2.68 to 12.51 ± 1.90 mV; Fig. 12), indicating that the terbutaline-sensitive current was carried by Cl. To characterize the anion permeability properties of the terbutaline-sensitive conductance, a series of biionic experiments was performed to determine the halide permeability sequence (Fig. 13). The mean E_rev of the terbutaline-sensitive I-V relations for SCN, Br, Cl, and I were 13.99 ± 2.67, 9.20 ± 1.15, 0.20 ± 0.22, and 3.04 ± 0.76 mV, respectively. The halide permeability sequence was SCN > Br > Cl > I (relative permeabilities for SCN, Br, Cl, and I were 1.76, 1.45, 1, and 0.89, respectively).

View larger version (14K): [in this window] [in a new window]   Fig. 9.   Representative Isc tracings showing effects of basolateral addition of 2 µM terbutaline followed by apical addition of 20 µM amiloride (A) or apical addition of 20 µM amiloride followed by basolateral addition of 2 µM terbutaline (B). Experiments were performed with amphotericin-permeabilized monolayers mounted in Ussing chambers and bathed with potassium methanesulfonate Ringer solution on basolateral side and serum-free DMEM/F-12 medium on apical side. Voltage across monolayers was clamped at 0 mV. Initial mean potential difference and conductance values were 1.08 ± 0.42 mV and 6.67 ± 1.62 mS (A; n = 8) and 1.11 ± 0.42 mV and 5.46 ± 1.23 mS (B; n = 8).

View larger version (27K): [in this window] [in a new window]   Fig. 10.   Terbutaline (Terb)- and amiloride (Amil)-sensitive current responses obtained from amphotericin-permeabilized monolayers voltage clamped at 0 mV. Pretreatment with amiloride (n = 8) did not significantly affect magnitude of terbutaline-sensitive current (15.29 ± 3.38 µA with pretreatment vs. 14.06 ± 4.26 µA without pretreatment). In addition, pretreatment with terbutaline (n = 8) did not significantly affect amiloride-sensitive current (7.63 ± 0.92 µA with pretreatment vs. 6.63 ± 1.23 µA without pretreatment). Initial mean potential difference and conductance values were same as in Fig. 9.

View larger version (25K): [in this window] [in a new window]   Fig. 11.   I-V relationship for terbutaline- and CPT-cAMP-sensitive conductance in apical membrane. Experiments were performed with amphotericin-permeabilized filters mounted in Ussing chambers and bathed with potassium methanesulfonate Ringer solution on basolateral side and serum-free DMEM/F-12 medium on apical side. A: representative tracing of terbutaline-sensitive current recorded in response to 10-mV voltage step increments from 70 to +80 mV. B: I-V plot for terbutaline- and CPT-cAMP-sensitive currents with mean Erev of 28.42 ± 2.68 (n = 12) and 26.32 ± 2.50 mV (n = 5), respectively. Terbutaline (2 µM) and CPT-cAMP (100 µM) were applied to basolateral bathing solution after pretreatment of monolayers with apical amiloride (20 µM). Initial mean potential difference and conductance values were 1.71 ± 0.39 mV and 3.67 ± 1.28 mS (n = 12) for terbutaline and 1.76 ± 0.60 mV and 6.63 ± 1.79 mS (n = 5) for CPT-cAMP, respectively.

View larger version (14K): [in this window] [in a new window]   Fig. 12.   Increasing basolateral Cl concentration ([Cl]) produced a shift in terbutaline-sensitive Erev. Experiments were performed with amphotericin-permeabilized filters mounted in Ussing chambers and bathed with potassium methanesulfonate Ringer solution on basolateral side and serum-free DMEM/F-12 medium on apical side. Changes in [Cl] were achieved by replacing potassium methanesulfonate with equimolar KCl. Terbutaline (2 µM) was applied to basolateral bathing solution after pretreatment of monolayers with apical amiloride (20 µM). Erev were plotted against logarithm of basolateral [Cl]. Linear regression analysis was used to fit data (R = 0.983, n = 5-12). Initial mean potential difference and conductance values for 10, 20, and 35 mM Cl solutions were 1.71 ± 0.39 mV and 3.67 ± 1.28 mS (n = 12), 2.28 ± 0.81 mV and 4.56 ± 0.973 mS (n = 5), and 2.50 ± 0.65 mV and 3.25 ± 0.22 mS (n = 5), respectively.

View larger version (30K): [in this window] [in a new window]   Fig. 13.   Halide permeability of terbutaline-sensitive current. Experiments were performed with amphotericin-permeabilized filters mounted in Ussing chambers and bathed with modified potassium methanesulfonate Ringer solution on basolateral side and modified potassium methanesulfonate Ringer solution with 50 mM KCl replaced with 50 mM KSCN, KCl, KBr, or KI on apical side. Terbutaline (2 µM) was added to basolateral solution. Mean Erev for SCN, Br, Cl, and I were 13.99 ± 2.67, 9.20 ± 1.15, 0.20 ± 0.22, and 3.04 ± 0.76 mV, respectively. Initial mean potential difference and conductance values for 50 mM KSCN, KCl, KBr, and KI were 0.80 ± 0.34 mV and 11.41 ± 5.78 mS, 0.24 ± 0.31 mV and 4.59 ± 1.10 mS, 0.05 ± 0.44 mV and 10.87 ± 4.22 mS, and 0.14 ± 0.26 mV and 10.55 ± 3.42 mS, respectively (n = 5 for each).

Blocker pharmacology. NPPB, glibenclamide, and DPC were used to block the terbutaline-sensitive current in amphotericin-permeabilized monolayers. Apical addition of NPPB, glibenclamide, or DPC following pretreatment with amiloride and terbutaline revealed that these compounds could inhibit the terbutaline-activated current in a concentration-dependent manner but that they had no effect on the amiloride-sensitive current. Figure 14A shows a representative trace of the effect of glibenclamide on terbutaline-activated current in amphotericin-permeabilized monolayers. The IC₅₀ values for NPPB, glibenclamide, and DPC were 12, 110, and 640 µM, respectively (Fig. 14B).

View larger version (19K): [in this window] [in a new window]   Fig. 14.   Effects of Cl channel blockers on terbutaline-sensitive current. Experiments were performed with amphotericin-permeabilized filters mounted in Ussing chambers and bathed with potassium methanesulfonate Ringer solution on basolateral side and serum-free DMEM/F-12 medium on apical side. Monolayer voltage was clamped at 0 mV. 5-Nitro-2-(3-phenylpropylamino) benzoic acid (NPPB), glibenclamide, and N-phenylanthranilic acid (DPC) were added to apical solution. A: representative tracing of effect of apical glibenclamide (200 µM at each arrow) on terbutaline-activated current. B: concentration-response relationship for NPPB, glibenclamide, and DPC inhibition of terbutaline-sensitive current. IC50 values for NPPB, glibenclamide, and DPC were 12 (n = 6), 110 (n = 6), and 640 µM (n = 9), respectively. Initial mean potential difference and conductance values for NPPB, glibenclamide, and DPC were 1.53 ± 0.40 mV and 9.29 ± 2.10 mS (n = 6), 1.31 ± 0.40 mV and 10.70 ± 3.11 mS (n = 6), and 1.49 ± 0.57 mV and 12.07 ± 1.91 mS (n = 9), respectively.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Cultured alveolar epithelial cells used in this study were grown on large Transwell membrane filters (4.5 cm²), where they form confluent monolayers with a high transepithelial resistance. The basal electrical properties of these monolayers agreed with those reported in earlier studies using similar cell isolation procedures (2, 5, 14, 35). The cells exhibited apical microvilli and an uneven cuboidal shape similar to that of native alveolar epithelial cells. In addition, a significant portion of the basal I_sc was blocked by addition of amiloride to the apical solution, consistent with previous results using cultured adult rat alveolar epithelial cells (9, 14, 31).

In our initial experiments, we examined the effects of amiloride analogs on basal I_sc and observed that the rank order of potency was benzamil > amiloride > EIPA. The IC₅₀ values and rank order of potency for these inhibitors were similar to results reported in previous studies (9) and indicated the presence of high-affinity, amiloride-sensitive Na⁺ channels active under basal conditions. To determine the Na⁺/K⁺ permeability ratio of the apical Na⁺ conductance, we examined the amiloride-sensitive I-V relationship using amphotericin-permeabilized monolayers. The E_rev of the amiloride-sensitive current was +46 mV. The estimated P_Na/P_K ratio calculated from the constant field equation was 12.5:1, a value similar to that of the cloned rat -ENaC (P_Na/P_K = 10:1) expressed in Xenopus oocytes (12). Thus alveolar epithelial cells in culture for a period of 6 days or more express a Na⁺ conductance that exhibits a relatively high selectivity for Na⁺. However, in freshly isolated adult rat alveolar epithelial cells, low-affinity amiloride-sensitive Na⁺ channels were previously observed (33). These channels had relatively low selectivity for Na⁺ over K⁺ and were effectively blocked by EIPA over a concentration range similar to that of amiloride (33). These differences indicate that culture conditions may significantly affect the expression of Na⁺ channels with different selectivity and pharmacological properties.

Previous studies with adult rat alveolar epithelial cells in culture demonstrated that stimulation with terbutaline increased net Na⁺ and Cl absorption (31). These flux experiments were performed under short-circuit conditions, indicating that a transcellular pathway for Cl absorption was activated by -adrenergic stimulation. In the present study, basolateral addition of terbutaline produced an immediate decrease in current, followed by a slow recovery back to the initial I_sc. This effect was also observed when monolayers were treated with CPT-cAMP, a membrane-permeant, nonmetabolizable analog of cAMP. The results obtained from experiments with terbutaline were similar to those previously reported by Cheek et al. (14). In the present study, replacement of Cl in both apical and basolateral solutions with methanesulfonate eliminated the effects of terbutaline and CPT-cAMP, suggesting that the decrease in I_sc was dependent on extracellular [Cl]. The amiloride-insensitive I_sc in serum-free DMEM/F-12 medium was presumably due to Na⁺-amino acid cotransport. Na⁺-amino acid cotransport in alveolar epithelial cells was previously reported by Brown et al. (11). When nonessential amino acids were added into normal Ringer solution or Cl-free Ringer solution, the nearly zero I_sc was increased to a level similar to that observed in serum-free DMEM/F-12 medium. Active Cl secretion dependent on bumetanide-sensitive Na⁺-K⁺-2Cl cotransport was not detected in our studies. The I-V relationships for the terbutaline- and CPT-cAMP-activated conductances had similar E_rev, suggesting Cl as the current-carrying ion. Changing the basolateral [Cl] from 10 to 20 and then to 35 mM shifted the E_rev toward 0 mV, thus confirming the presence of a terbutaline- and cAMP-dependent Cl conductance in the apical membrane.

To determine the selectivity properties of the terbutaline- and cAMP-dependent Cl conductance, we performed a series of biionic anion substitution experiments using SCN, Br, and I as replacement anions and measured the shift in E_rev for each replacement anion and determined permeability ratios relative to Cl. The order of selectivity was SCN > Br > Cl > I for these experiments. In addition, a linear I-V relationship was observed under asymmetrical Cl conditions. The anion selectivity properties of the terbutaline-sensitive anion conductance observed in this study, along with the linear character of the I-V relationship in asymmetrical Cl-containing solutions, were similar to those of CFTR (1). Moreover, apical addition of NPPB, glibenclamide, or DPC produced inhibition of the terbutaline-activated Cl current in a concentration-dependent manner. The IC₅₀ values for these compounds were consistent with inhibition of CFTR (27). These experiments indicate that terbutaline, presumably acting through cAMP, activates Cl channels in the apical membrane with functional and pharmacological properties similar to those of CFTR.

To investigate the possibility that terbutaline directly increases Na⁺ absorption by increasing apical membrane Na⁺ permeability, we measured the effects of terbutaline and CPT-cAMP on I_sc when Cl was replaced with methanesulfonate. Under these conditions, neither terbutaline nor CPT-cAMP produced any change in I_sc. In addition, the amiloride-sensitive I_sc after terbutaline in Cl-free solution was not significantly different from the amiloride-sensitive I_sc measured in Cl-containing solution without terbutaline. Moreover, no change in amiloride-sensitive current was observed in amphotericin-permeabilized monolayers voltage clamped at 0 mV following treatment with terbutaline. Finally, no significant change in the slope or E_rev of the amiloride-sensitive I-V relation was detected after treatment with terbutaline. These results indicate that the amiloride-sensitive Na⁺ conductance in these cells is not directly activated by terbutaline.

The data presented in this study suggest the following model to explain the mechanism of -adrenergic stimulation on transepithelial Na⁺ and Cl transport in alveolar epithelial cells (Fig. 15). Terbutaline binds to -adrenergic receptors located in the basolateral membrane that are coupled to adenylyl cyclase, resulting in an increase in intracellular cAMP. We propose that cAMP, presumably acting through protein kinase A, activates a CFTR-like Cl channel in the apical membrane and that Cl enters the cell. Cl influx is consistent with the initial decrease in I_sc and occurs as a result of a substantially depolarized apical membrane due to the constitutive influx of Na⁺ through amiloride-sensitive Na⁺ channels. The time-dependent recovery of the I_sc after terbutaline is due to an increase in Na⁺ absorption. This increase, however, results from an increase in driving force for Na⁺ uptake as a consequence of membrane hyperpolarization produced by Cl influx. This would explain the observation that I_sc recovery is completely blocked by pretreatment with amiloride. It would also explain how terbutaline could increase net Na⁺ absorption under conditions in which no change in apical Na⁺ conductance could be observed. The Na⁺ that enters the cell across the apical membrane is transported across the basolateral membrane by the Na⁺-K⁺-ATPase. Cl, entering across the apical membrane, is transported across the basolateral membrane by unknown transport pathways. One possibility is KCl cotransport, since it could couple Cl efflux to an outwardly directed K⁺ concentration gradient and has been shown to mediate electroneutral Cl efflux across the basolateral membrane of Cl-absorbing epithelia (45). Another possibility is a basolateral Cl channel, as proposed for sweat duct epithelial cells (44). For this to be feasible, however, the basolateral Cl conductance must have an E_rev that is more depolarized than the basolateral membrane potential.

View larger version (20K): [in this window] [in a new window]   Fig. 15.   Cell model showing proposed mechanisms for Cl and Na+ transport across adult rat alveolar epithelial cells. PKA, protein kinase A.

In conclusion, we propose that the net effect of -adrenergic stimulation is to produce a tight coupling between Na⁺ and Cl influx across the apical membrane that results in an increase in net NaCl absorption. This increase in salt absorption sets up the osmotic driving forces required for movement of fluid across the epithelium. We believe that the mechanism for transcellular Cl transport outlined above provides an explanation for the previously reported increase in net Cl absorption stimulated by -adrenergic agonists (31). Whether this mechanism reflects the actions of -adrenergic agonists in vivo is presently unknown. The transport properties of cultured epithelial cells may not reflect those of alveolar type II cells in vivo. However, the results of this study do suggest some interesting experiments that should provide some insight into the physiological role of CFTR in alveolar fluid absorption.