Existing evidence supports the presence of active transport of Na+ across the mammalian alveolar epithelium and its upregulation by agents that increase cytoplasmic cAMP levels. However, there is controversy regarding the mechanisms responsible for this upregulation. Herein we present the results of various patch-clamp studies indicating the presence of 25- to 27-pS, amiloride-sensitive, moderately selective Na+ channels (Na+-to-K+ permeability ratio = 7:1) located on the apical membranes of rat alveolar type II (ATII) cells maintained in primary culture. The addition of terbutaline to the bath solution increased the open probability of single channels present in cell-attached patches of ATII cells without affecting their conductance. A similar increase in open probability was seen after the addition of protein kinase A, ATP, and Mg2+ to the cytoplasmic side of inside-out patches. Measurement of short-circuit currents across confluent monolayers of rat or rabbit ATII cells indicates that terbutaline and 8-(4-chlorophenylthio)-cAMP increase vectorial Na+ transport and activate Cl− channels. Currently, there is a controversy as to whether the cAMP-induced increase in Na+ transport is due solely to hyperpolarization of the cytoplasmic side of the ATII cell membrane due to Cl− influx or whether it results from simultaneous stimulation of both Cl− and Na+ conductive pathways. Additional studies are needed to resolve this issue.
- adenosine 3′,5′-cyclic monophosphate
- alveolar type II cells
- sodium channels
- patch clamp
the existence of active Na+ reabsorption across the adult alveolar epithelium in vivo in a number of species, including human, has been well documented (for a review, see Ref. 17). In brief, plasma or normal saline containing 5% bovine serum albumin instilled in the alveolar spaces of anesthetized animals or isolated perfused lungs is reabsorbed within a few hours. Coinstillation of amiloride or 5-(N-ethyl-N-isopropyl)-amiloride into the alveolar space or injection of ouabain into the circulation significantly decreases the rate of fluid reabsorption (16, 20, 29).
Additional insight into the nature of the transporters was derived from electrophysiological measurements performed on alveolar type II (ATII) cells. Patch-clamp studies (7, 30) demonstrated the presence of nonselective and moderately selective Na+ channels in the apical membranes of ATII cells maintained in primary culture. Furthermore, when grown to confluence and mounted in Ussing chambers, ATII cells generate a spontaneous potential difference and short-circuit current (I sc) that are inhibited to a large extent by amiloride and ouabain (4). Based on the results of these studies, we now believe that Na+ diffuses passively across the ATII apical membrane, mainly through cation channels (23), down a favorable electrochemical gradient maintained by Na+-K+-ATPase and then are actively transported across the basolateral membrane by the ouabain-sensitive Na+-K+-ATPase. K+ leaves ATII cells, driven by their favorable electrochemical gradient, through K+ channels located in the basolateral membrane, whereas Cl− crosses through the paracellular junctions in response to the transepithelial potential difference.
The importance of active Na+ transport in fluid clearance across the injured alveolar epithelium was demonstrated by two different studies. First, instillation of phenamil, an irreversible blocker of epithelial Na+ channels, into the lungs of rats exposed to hyperoxia resulted in higher levels of lung water compared with rats receiving vehicle alone (29). Second, a positive correlation has been established between active Na+ transport across the alveolar space of patients with acute lung injury and the rate of resolution of noncardiogenic pulmonary edema (18). The demonstration that Na+ transport across the alveolar epithelium in vivo and ex vivo, as well as across ATII cells, can be upregulated by β-agonists (2, 4, 5) has led to speculation that these agents may be useful in limiting alveolar edema and decreasing morbidity and mortality in patients with acute lung injury.
Presently, there is controversy concerning the mechanisms by which β-agonists such as terbutaline or lipid-soluble analogs of cAMP such as 8-(4-chlorophenylthio)-cAMP (8-CPT-cAMP) increase Na+transport across alveolar epithelial cells. Activation of β2-receptors, shown to be present on ATII cell surfaces (3), generally stimulates adenylate cyclase that, in turn, increases intracellular cAMP levels and activates Na+ channels in a number of epithelial cells and tissues (8, 12). However,I sc measurements across rabbit and rat ATII cells indicate that cAMP-induced responses are considerably more complex and involve both Na+ and Cl− conductive pathways (11, 19). Based on the results of their experiments in Ussing chambers, Jiang et al. (11) proposed that agents that increase cAMP activate apical cystic fibrosis transmembrane conductance regulator (CFTR)-type Cl− channels. Because the resting membrane potential of ATII cells is around −40 mV, stimulation of Cl− channels will result in influx of Cl−, hyperpolarization of the apical membrane, and creation of a favorable driving force for increased Na+transport. These investigators did not find any evidence of activation of Na+ channels by cAMP in their experimental model.
Herein, we present evidence indicating that increased levels of cAMP in the ATII cell cytoplasm activate amiloride-sensitive Na+channels. Results of patch-clamp studies convincingly show that β-agonists increase the open probability (P o) and mean open time (τ1)of amiloride-sensitive single Na+ channels located in the apical membranes of rat ATII cells maintained in primary culture without altering their single-channel unitary conductance. Furthermore, protein kinase A (PKA) in the presence of ATP phosphorylated a putative Na+-channel protein isolated from rabbit ATII cells and activated an amiloride-sensitive single channel when this protein was reconstituted in planar lipid bilayers. Finally, we believe that the results of I sc measurements across cultured rat and rabbit ATII cells obtained from a number of laboratories (4, 19) indicate that cAMP-stimulated vectorial Na+ transport cannot be explained solely by activation of Cl−channels.
SINGLE-CHANNEL RECORDINGS IN ATII CELLS: EFFECTS OFcAMP/PKA
Single channels with diverse biophysical properties have been detected in ATII cells isolated from the lungs of adult rabbits, rats, and guinea pigs (for a review, see Ref. 15). Yue et al. (30) isolated ATII cells from the lungs of rats by elastase digestion and plated them on fibronectin-treated coverslips for 12–24 h. In cells patched in the cell-attached mode, single-channel currents were observed for holding potentials between −80 and +30 mV, with a single-channel conductance of 27 ± 3 pS in 10–15% of successful patches (Figs. 1 and 2). The addition of 10 μM terbutaline (a potent β2-agonist known to increase intracellular cAMP levels) to the bath increased the τ1 and P o of these channels without affecting their unitary conductance (Table1). Pretreatment of ATII cells with propranolol (10 μM), a β2-antagonist, obviated the terbutaline-induced increases in P o and τ1. Because ATII cells in primary culture orient themselves so that their basal membranes are attached to the substratum and the apical membranes are pointing upward (14), channels in excised or cell-attached patches are likely to be located in their apical membranes.
Single-channel currents with a conductance of 25 ± 2 pS and Na+-to-K+ permeability ratio (PNa/PK) of 7:1 (Figs.3-5) were also recorded across 50% of ATII cells patched in the inside-out mode (30). The addition of 1 μM amiloride or 5-(N-ethyl-N-isopropyl)-amiloride to the pipette solution (150 mM sodium glutamate) blocked single-channel activity almost completely. The addition of 250 U/ml of PKA, 1 mM ATP, and 5 mM MgCl2 to the bath solution (150 mM sodium glutamate) increased the single-channel τ1 andP o without altering the unitary conductance (Figs.3-5, Table 1). The observed increase in P owith PKA may have been brought about by direct phosphorylation of Na+-channel proteins or phosphorylation of cytoskeletal proteins such as actin, ankyrin, spectrin, or fondrin interacting with Na+-channel proteins (22, 26).
Nonselective (P Na /P K = 1), voltage-independent, Ca+2-activated [intracellular Ca2+ concentration > 0.1 mM] cation channels with a conductance of 20.4 pS in symmetric NaCl (150 mM) solutions have also been identified in rat ATII cells patched in the inside-out mode after culture on collagen-coated coverslips for 24 and 72 h (7). The effects of cAMP/PKA on these channels have not been reported. However, the biophysical properties of these Ca+2-activated channels are almost identical to those recorded in fetal distal lung epithelial cells (21). Marunaka et al. (13) and Tohda et al. (28) reported that the addition of 10 μM terbutaline to the bath solution of fetal distal lung epithelial cells, patched in the cell-attached mode, increased the P o of the Ca+2-activated, nonselective cation channels. Furthermore, in the presence of brefeldin A, terbutaline did not alter the density of the Ca+2-activated cation channel, indicating that terbutaline may promote the trafficking of this cation channel to the apical cell surface (9).
RECONSTITUTION OF IMMUNOPURIFIED ATII CELL Na+-CHANNEL PROTEIN INTO PLANAR LIPID BILAYERS: EFFECTS OF PKA
A putative Na+-channel protein was isolated from freshly isolated rabbit ATII cells with ion-exchange chromatography followed by immunoaffinity purification in a column coated with a polyclonal antibody raised against purified bovine renal Na+-channel protein. The ATII Na+-channel protein consists of two peptides with molecular masses of ∼130 and 70 kDa (25). When this protein was reconstituted in lipid bilayers, single-channel currents with linear current-voltage relationships and a unitary conductance of 25 pS were seen (25), in agreement with what was observed in rat ATII cells patched in the cell-attached or inside-out modes (10, 30). The addition of PKA and ATP to the presumed cytoplasmic side of the bilayer increased the P o from 0.40 ± 0.14 to 0.8 ± 0.12 without altering the channel unitary conductance. In additional studies, Berdiev et al. (1) showed that PKA phosphorylated both the 135- and 70-kDa polypeptides of the immunopurified ATII Na+-channel protein. Studies in A6 cells have shown that PKA phosphorylates a subunit of the Na+-channel protein and that the level of phosphorylation correlates with vectorial Na+ transport (24).
CHANGES IN ISC ACROSS ATII CELL MONOLAYERS BYcAMP
Patch-clamp measurements provide definitive evidence for the existence of ion channels on cell membranes and considerable insight as to the factors responsible for their regulation. There is considerable interest in correlating these findings with macroscopic measurements of Na+ transport across confluent ATII monolayers. Jiang et al. (11) isolated and purified ATII cells from the lungs of adult rats and cultured them on large Transwell membrane filters using serum-free medium until they formed confluent monolayers. Six days later, ATII cells developed large I sc values, 60% of which were inhibited by 10 μM amiloride. The basolateral addition of terbutaline (2 μM) resulted in a rapid decrease inI sc followed by a gradual recovery to its baseline value (Fig. 6). When amiloride was added before terbutaline, I sc decreased but failed to return to its baseline value. Finally, when ATII monolayers were bathed in a Cl−-free solution, the response to terbutaline was almost completely abolished. Based on these findings, they proposed that terbutaline stimulated Cl− absorption through an apical Cl− conductance. The Cl−influx hyperpolarized the cell membrane, increasing the driving force for Na+ entry. Thus their measurements indicate that terbutaline stimulates Na+ transport without directly activating apical Na+ channels.
However, results from other laboratories cannot be explained by this model. For example, Cheek et al. (4) showed that the addition of terbutaline to the basolateral bath of an Ussing chamber containing confluent monolayers of rat ATII cells resulted in a transient decrease in I sc followed by a gradual increase to steady-state levels, surpassing the baseline (Fig.7). About 80% ofI sc after terbutaline stimulation was abolished by the addition of 10 μM amiloride to the apical compartment. The initial decrease in I sc is consistent with the findings of Jiang et al. (11) and is most likely due to activation of a Cl− conductance. However, Jiang et al.'s model cannot account for the increase in I scabove its initial value after terbutaline stimulation. Instead, Cheek et al.'s (4) data suggest that terbutaline stimulates Na+-conductive pathways secondary to an increase in cAMP. Indeed, this group reported that terbutaline stimulated the net apical-to-basolateral flux of 22NaCl, and the magnitude of stimulation was similar to the steady-state increase inI sc. An important experiment that needs to be done is to repeat these measurements in monolayers pretreated with amiloride and show that, under those conditions, the response ofI sc to terbutaline is blunted.
In another study, Nielsen et al. (19) assessed the effects of forskolin and 8-CPT-cAMP on the I sc of rabbit ATII cells grown to form tight confluent monolayers. Either forskolin (10 μM) or 8-CPT-cAMP (100 μM) produced an early biphasic change in theI sc followed by a slow steady-state increase to a value that was 3.4 ± 0.2 μA/cm2 higher than baseline, in agreement with the data of Cheek et al. (4) (Fig.8). In addition, Nielsen et al. (19) showed that the addition of forskolin to monolayers pretreated with amiloride resulted in a sustained increase in I sc. There are several explanations for these findings: first, forskolin may have reversed the effect of amiloride; second, it may have stimulated Na+ absorption across nonamiloride-sensitive pathways; and third, it may have induced Cl− secretion across cAMP-activated Cl− channels. In any event, their data cannot be explained by the model of Jiang et al. (11). Current experiments in a number of laboratories are attempting to identify the mechanisms involved and the factors that may reconcile these apparently contradictory results.
CONCLUSIONS AND UNANSWERED QUESTIONS
Measurement of single-channel activity in cell-attached and inside-out patches of dispersed ATII cells maintained in primary culture provides strong support for the hypothesis that agents that increase cAMP increase the activity of moderately selective, amiloride-sensitive Na+ channels located in their apical membranes. Whether this increase is brought up by insertion of new channels in the membrane from a cytoplasmic pool or by activation of previously inactive channels remains to be demonstrated.
However, there is a clear discrepancy among the aforementioned patch-clamp studies of Yue et al. (30) and results obtained across confluent monolayers of cultured ATII cells. There is clear evidence that these monolayers contain both amiloride-sensitive and CFTR-type Cl− channels in their apical membranes. Furthermore, all published studies agree that terbutaline, forskolin, or 8-CPT-cAMP activate Cl− currents across cultured ATII cells. However, the proposed hypothesis that increased Na+transport is due solely to an increased driving force for Na+ secondary to activation of Cl−channels cannot explain the sustained increase inI sc reported in at least two studies and contradicts the direct measurement of single-channel activation by terbutaline and PKA.
In trying to explain these apparently contradictory findings, one has to consider that cultured ATII cells undergo a number of important phenotypic changes and may be transformed to ATI cells as shown by their increased immunoreactivity to monoclonal antibodies raised against ATI cells (6). For example, it is possible that cultured but not freshly isolated rat ATII cells express functional CFTR. As shown by Stutts et al. (27), agents that increase cAMP activate amiloride-sensitive currents across cells transfected with rat epithelial Na+ channel but not rat epithelial Na+ channel and CFTR. One way to resolve this controversy will be to seed ATII cells on clear semipermeable filters and attempt to perform both patch-clamp and I sc measurements in the same filter.
We acknowledge the editorial assistance of Rebecca Todd.
Address for reprint requests and other correspondence: S. Matalon, Dept. of Anesthesiology, The University of Alabama at Birmingham, THT 940, 619 South 19th St., Birmingham, AL 35249-6810 (E-mail:).
This work was supported by National Heart, Lung, and Blood Institute Grants HL-31197 and HL-51173 and Office of Naval Research Grant N00014-97-1-0309.
V. G. Nielsen is the recipient of Grant-In-Aid 9810091SE from the American Heart Association (Southeast Affiliate).
- Copyright © 2000 the American Physiological Society