INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

PULMONARY EDEMA develops in patients either by increased hydrostatic pressure in the pulmonary circulation, as occurs in patients with congestive heart failure, or by increased vascular permeability, as occurs in patients with acute respiratory distress syndrome (44). Clearance of lung edema from alveoli depends on the active transport of sodium (Na⁺) out of the air spaces by alveolar epithelial cells, and water follows osmotically (14, 24-27, 33, 35). Moreover, Matthay and Wiener-Kronish (27) found that patients with respiratory failure caused by pulmonary edema had better outcomes when the epithelial barrier was restored and clearance of lung edema occurred.

Na⁺ enters the alveolar epithelial cells predominantly via apical Na⁺ channels and is actively transported out of the cells by the basolaterally located Na⁺-K⁺-ATPase (10, 25, 26, 42). The Na⁺-K⁺-ATPase actively transports Na⁺ out of the cell and K⁺ into the cell at the expense of ATP hydrolysis (9). The functional Na⁺-K⁺-ATPase consists of a catalytic subunit () and a glycosylated -subunit (38). The -subunit contains binding sites for Na⁺, K⁺, and ATP as well as the cardiac glycoside ouabain, a specific inhibitor of Na⁺-K⁺-ATPase. There are three known isoforms of the -subunit (6, 29, 41). The -subunit is thought to anchor the isoform into the plasma membrane (28), and there are also three known -isoforms.

Previous studies (12) have shown that lung edema can be induced in rats by mechanical ventilation with high tidal volumes for as little as 20 min. A later study (11) showed that the size of the tidal volume affected the degree of edema. Importantly, when the chests of the rats were restricted, to achieve low tidal volumes with high airway pressures, there was no lung edema (13). These studies suggest that high levels of alveolar distension may contribute to the formation of alveolar edema. In a preliminary report (43), when rats were mechanically ventilated for 20 min with a high tidal volume, lung edema developed, and Na⁺-K⁺-ATPase activity in alveolar type II cells isolated after ventilation was higher compared with that in control rats and rats ventilated at a low tidal volume. However, it is not apparent whether overdistension of the alveolar epithelial cells stimulated the Na⁺-K⁺-ATPase or whether the presence of edema or other mechanisms within the lung was responsible.

Numerous studies in the last decade have demonstrated that cells derived from a variety of tissues respond to cyclic stretch. Cyclic stretch has been shown to stimulate short-term responses, such as increased intracellular Ca²⁺ release (7), elevated inositol 1,4,5-trisphosphate levels (15), and decreased prostanoid release (36) in airway epithelial cells, and long-term responses, such as increases in [³H]thymidine incorporation, cAMP levels, and the synthesis of surfactant-related phospholipids in fetal type II alveolar cells (37). Adult rat alveolar type II epithelial cells exposed to a single distension in culture exhibited elevated calcium mobilization and surfactant secretion (46), and sustained distension was shown to influence phenotypic expression (17). We have recently shown that cyclic stretch of rat pleural mesothelial cells stimulated an increase in endothelin-1 production (45). In this study, we tested the hypothesis that cyclic mechanical stretch regulates the activity of the Na⁺-K⁺-ATPase in alveolar epithelial cells.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell culture methods. A cell line of SV40-transformed murine lung epithelial cells (MLE-12) was generously provided by Dr. J. Whitsett, Cincinnati, OH (21). MLE-12 cells were derived from tumors of transgenic mice that bear a viral oncogene under transcriptional regulation of the promoter-enhancer region of the human surfactant protein C gene (21). Cells were grown in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum and were used in passages 8 through 14. We have not detected differences in Na⁺-K⁺-ATPase activity in this passage range. We have used the MLE-12 cell line as a model for alveolar type II epithelial cells. Both MLE-12 cells and primary cultures of alveolar type II epithelial cells express the Na⁺-K⁺-ATPase ₁-subunit, as determined by Western blot analysis, and ouabain inhibition curves generated similar IC₅₀ values (31). Cells were seeded onto collagen I-coated Silastic membranes in six-well plates (Flexcell, McKeesport, PA) at a density of 2 × 10⁵ cells/well. When the cells were ~80-90% confluent, the plates were used in experiments.

Cell stretching. Cells were subjected to cyclic stretch by using the Flexercell Strain Unit (Flexcell). The flexible cell-covered elastomer membranes were stretched by applying an oscillating vacuum to the underside of the membranes. A computer controlled the duration, amplitude, and frequency of the applied stretch. Cells were stretched at 30 cycles/min, with a maximum strain of 20% (10% mean strain). We estimate that this level of strain corresponds to elongation that may occur during normal tidal volume breathing. Comparisons were made between stretched cells and control cells cultured on the same plates but not subjected to cyclic strain. The vacuum manifold that held the plates was maintained in a 37°C, humidified incubator with 5% CO₂.

Na⁺-K⁺-ATPase activity. The effect of cyclic stretch on Na⁺-K⁺-ATPase activity was evaluated by measuring ouabain-sensitive K⁺ influx, as assessed by ⁸⁶Rb⁺ uptake in cells attached to collagen-coated elastic membranes. Cells were cultured as described above and either stretched or kept static. Before transport measurements, cells were washed and incubated in serum-free, HEPES-buffered DMEM for 10 min at 37°C in a bath that gyrated at 100 rpm. Cells were preincubated with or without 5 mM ouabain for 5 min. ⁸⁶Rb⁺ was added as RbCl at 1 mCi/ml. After a 5-min incubation, uptake was terminated by aspiration of the assay medium, and the plates were washed in ice-cold MgCl₂. Cells were then extracted with 0.5 N NaOH or 0.1% SDS, and aliquots were counted in a liquid scintillation counter to quantitate ⁸⁶Rb⁺ influx. Protein concentrations of the NaOH or SDS extracts were measured by the Lowry assay. Na⁺-K⁺-ATPase activity was calculated as ouabain-sensitive K⁺ influx (in nmol K⁺ · mg protein¹ · min¹). On average, Na⁺-K⁺-ATPase activity represented ~42% of the total K⁺ influx. The K⁺ concentration of DMEM was 10.2 mM. On each six-well plate, three wells were pretreated with ouabain for comparison with the remaining three untreated wells. The difference between each pair of wells (total flux minus ouabain-inhibited flux) was used as one measurement of ouabain-sensitive influx. Thus each plate yielded three measurements.

Measurement of ATP hydrolysis. Na⁺-K⁺ pump activity was assessed in MLE-12 cells by hydrolysis of ATP. Na⁺-K⁺-ATPase activity was determined after 15-min incubations with agonists, as previously described (4), as the rate of [³²P]ATP hydrolysis in suspended cells in buffer containing (in mM) 50 NaCl, 5 KCl, 10 MgCl₂, 1 EGTA, 50 Tris · HCl, 7 Na₂-ATP (Sigma Chemical, St. Louis, MO) and [³²P]ATP (Amersham, Arlington Heights, IL) in trace amounts (3.3 nCi/µl). Cells were transiently permeabilized by thermal shock to make them permeable to [³²P]ATP, and the reaction was carried out for 15 min at 37°C. The reaction was terminated by placement on ice and addition of TCA-charcoal to absorb nonhydrolyzed [³²P]ATP. The ouabain-sensitive ATPase activity was determined in buffer that lacked Na⁺ and K⁺ but included 2 mM ouabain. On average, the ouabain-sensitive ATPase activity was ~56% of the total. Nonspecific ATP hydrolysis was determined in the absence of cells. The ³²P_i liberated was measured by scintillation counting and was expressed as nanomoles P_i per milligram protein per minute.

Treatment protocols. Preliminary experiments demonstrated similar ouabain-inhibitable ⁸⁶Rb⁺ uptake in MLE-12 monolayers with different levels of confluence from 50 to 100% (data not shown). To determine the role of Na⁺ entry into the cells, monolayers (90% confluent) were incubated with or without agonist for 30 min before cyclic stretch at 30 cycles/min. Na⁺-K⁺-ATPase activity and/or expression was then evaluated after 30 min of stretch. Cells were treated with amiloride (1 µM) to block Na⁺ entry through Na⁺ channels, or Nystatin (5 µg/ml) to stimulate nonspecific Na⁺ entry into the cells, or gadolinium (50 nM) to block Na⁺ entry through stretch-activated ion channels.

Preparation of basolateral plasma membranes. MLE-12 cells were homogenized, and basolateral membranes were prepared as described previously (19).

Western blot analysis. Na⁺-K⁺-ATPase subunit abundance was determined by Western blot analysis. Cells were treated as described above, basolateral membranes were prepared, and then proteins were resolved by SDS-PAGE on a 10% polyacrylamide gradient gel. For each condition, 5 µg of protein were loaded for basolateral membranes, and 50 µg of protein were loaded for whole cell membranes. After electrophoresis was completed, proteins were transferred to nitrocellulose membranes (Hybond C, Amersham). After transfer was completed (3 h at 1 A), the membranes were quenched at room temperature for 1 h with 7.5% casein in PBS containing 0.1% Tween 20. Incubation with specific Na⁺-K⁺-ATPase ₁-antibody (30) was performed overnight at 4°C. The membranes were rinsed five times with PBS-1% Tween 20, followed by incubation with a horseradish peroxide-conjugated goat anti-mouse secondary antibody (Bio-Rad) for 1 h. Blots were developed, as previously described, with an enhanced chemoluminescence detection kit (Amersham) used as recommended by the manufacturer.

Statistics. Significant differences between two conditions were evaluated by using a t-test. Comparisons among three or more conditions were made by using repeated-measures analysis of variance and Tukey's modified t-test (the Bonferroni criterion). A value of P < 0.05 was judged significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characterization of Na⁺-K⁺-ATPase in MLE-12 cells. To demonstrate Na⁺-K⁺-ATPase activity in MLE-12 cells, K⁺ influx, measured as ⁸⁶Rb⁺ uptake, was measured in the presence of increasing doses of ouabain. As shown in Fig. 1, ouabain decreased uptake in a dose-dependent manner, and the shape of the curve is suggestive of a rodent ₁-subunit (31). The K⁺ influx was one-third to one-half of the total influx at saturating concentrations (1 mM). Expression of the ₁-subunit was verified by Western blot analysis (see Fig. 6 below). The ₂-subunit was not detected by Western blot analysis (data not shown).

View larger version (8K): [in this window] [in a new window]   Fig. 1.   Murine lung epithelial (MLE-12) cells express primarily the 1-subunit of Na+-K+-ATPase. Total 86Rb+ uptake was measured over 5 min in the presence of increasing doses of ouabain (n = 4 independent measurements). The curve is a nonlinear least squares fit of the data; values are means ± SE. The estimated IC50 value (2.4 × 105 M) and the shape of the curve suggest that only the 1-subunit is expressed.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Activity of MLE-12 Na+-K+-ATPase as a function of Na+, K+, and ATP concentrations. Initial rates of ATP hydrolysis were measured at the concentration of ATP and cations indicated. Inset: ATP hydrolysis as a function of cation concentration with varying doses of ATP. Isotonicity was maintained by exchanging NaCl and KCl.

Cyclic strain stimulates Na⁺-K⁺-ATPase activity. To determine the effect of short-term cyclic stretch on Na⁺-K⁺-ATPase activity, MLE-12 cells were stretched (10% mean strain) at 30 cycles/min, removed from the strain unit, rinsed, and exposed to ⁸⁶Rb⁺ for 5 min. As shown in Fig. 3, ouabain-sensitive K⁺ influx was not increased after 15 min of stretch but was significantly increased after 30 or 60 min of stretch. As an indication of cell injury, the concentration of lactate dehydrogenase in the media conditioned by the cells was not different in the stretched cells compared with controls, and, on the basis of similar cell counts, the cells did not detach from the Silastic membranes.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Cyclic stretch increased Na+-K+-ATPase activity after 30 min of stretch (30 cycles/min; 20% maximum strain, 10% mean strain). Cells were stretched, then incubated with or without 5 mM ouabain; 86Rb+ uptake was measured after 5 min. Bars represent means ± SE from 4 independent measurements. * Significant difference from control, P < 0.05.

Na⁺ influx stimulates Na⁺-K⁺-ATPase activity. To determine whether the stretch-induced increase in Na⁺-K⁺-ATPase activity was caused by a higher influx of Na⁺, cells were treated with either Nystatin (50 µg/ml) or amiloride (1 µM) for 30 min before and during 30 min of stretch. Nystatin permits nonspecific influx of intracellular Na⁺. Figure 4 shows that Nystatin significantly increased Na⁺-K⁺-ATPase activity in unstretched cells, but there was a further significant increase in activity after 30 min of stretch. Amiloride blocked entry of Na⁺ into the cells through amiloride-sensitive ion channels and thus decreased Na⁺-K⁺-ATPase activity in unstretched cells. Amiloride blocked the stretch-induced increase in activity. This suggests that the increase in activity was in large part caused by increased Na⁺ entry. To assess whether cyclic stretch increased Na⁺-K⁺-ATPase activity by Na⁺ entry into cells through stretch-activated ion channels, cells were treated with gadolinium (50 nM) for 30 min before and during 30 min of stretch to block stretch-activated, nonselective cation channels. As shown in Fig. 5, gadolinium significantly decreased Na⁺-K⁺-ATPase activity in unstretched cells and prevented the stretch-induced stimulation.

View larger version (13K): [in this window] [in a new window]   Fig. 4.   Na+-K+-ATPase activity was stimulated by Nystatin (5 µg/ml) and inhibited by amiloride (1 µM). Cells were treated with either Nystatin or amiloride 30 min before and during stretch. Amiloride blocked the stretch-induced increase in activity. Bars, means ± SE Na+-K+-ATPase activity from 4 measurements. * Significant difference from control; + cells treated with Nystatin and exposed to cyclic stretch were significantly different from unstretched (static) Nystatin-treated cells, P < 0.05.

View larger version (11K): [in this window] [in a new window]   Fig. 5.   Gadolinium (50 nM) inhibited Na+-K+-ATPase activity in control (static) and cyclically stretched cells (stretch). Cells were treated with gadolinium before and during 30 min of stretch. Bars, means ± SE from 4 measurements. * Significant difference from static control, P < 0.05.

Translocation of Na⁺-K⁺-ATPase to basolateral membrane. Because Nystatin treatment allows free influx of Na⁺, the Na⁺-K⁺-ATPase activity was measured under maximal reaction velocity conditions. Cyclic strain of Nystatin-treated cells stimulated a further increase in activity. To determine whether this increase was caused by increased expression of the Na⁺-K⁺-ATPase in the basolateral membrane, we isolated proteins from whole cells and from basolateral membranes and compared the expression of the ₁-subunit by Western blot analysis. As shown in Fig. 6, when cells were stretched for 30 min, there was no change in the expression of the ₁-subunit in lysates from whole cells, but there was a significant increase in ₁-subunit expression in the basolateral membranes. Furthermore, when cells were treated with Nystatin before and during cyclic stretch, there was an additional increase in ₁-subunit expression in basolateral membranes.

View larger version (71K): [in this window] [in a new window]   Fig. 6.   Cyclic stretch stimulated translocation of the 1-subunit of the Na+-K+-ATPase to the basolateral membrane. Control cells (top) or cells treated with Nystatin (50 µg/ml; bottom) were either stretched for 30 min or kept under static conditions before cell lysis. Basolateral membranes were prepared as described previously (15). Equal amounts of protein were loaded for comparisons between static and stretch under each condition, but the amount of protein loaded for whole membranes (50 µg) and basolateral membranes (5 µg) was not equivalent.

Cyclic stretch did not affect cAMP levels. Because elevated cAMP has been shown to stimulate Na⁺-K⁺-ATPase activity in alveolar epithelial cells (5), we hypothesized that cyclic stretch might stimulate increases in intracellular cAMP. We found that cAMP levels were unchanged by cyclic stretch or by treatment with Nystatin or amiloride (data not shown).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The accumulation of alveolar edema occurs in patients with congestive heart failure because of elevated hydrostatic pressure in the pulmonary circulation and in patients with acute respiratory distress syndrome because of increased microvascular permeability. In either case, restoration of the alveolar epithelium to improve edema clearance is associated with better clinical outcome in these patients, with reduced time of mechanical ventilation, and with decreased problems associated with extended stays in intensive care units (20, 27). Because clearance of alveolar edema is dependent on vectorial Na⁺ flux, the alveolar Na⁺-K⁺-ATPase has been recognized as an important contributor to clearance of lung edema fluid. Thus it is important to understand the factors that regulate the function of the Na⁺-K⁺-ATPase. The activity of the Na⁺-K⁺-ATPase has been shown to be stimulated by corticosteroids (3), catecholamines (2, 34, 40), and oxidants (16), and the mechanisms of regulation vary in different types of cells (4).

In this study, we demonstrated that short-term (30 min) cyclic stretch of MLE-12 cells stimulated an increase in Na⁺-K⁺-ATPase activity. Recently, Songu-Mize et al. (39) demonstrated that expression of both the ₁- and ₂-subunits of the Na⁺-K⁺-ATPase were significantly increased in aortic smooth muscle cells after 4 days of cyclic stretch. Ruiz-Opazo et al. (32) measured decreased in vivo gene expression of the ₂-subunit in the cardiac left ventricle and in the aorta in a rat pressure-overload model. Also, a preliminary report showed that alveolar type II cells isolated from rats that were mechanically ventilated with a high tidal volume for 20 min demonstrated higher Na⁺-K⁺-ATPase activity compared with similar cells from controls (43). On the basis of Western blot analysis (Fig. 6) and the ouabain inhibition curve shown in Fig. 1, we found that MLE-12 cells express only the ₁-subunit. Thus there was no possibility for a differential response to cyclic stretch by the two -subunits, as was demonstrated in aortic smooth muscle cells (39).

In the present study, alveolar epithelial cells that were cyclically stretched for 30-60 min exhibited elevated Na⁺-K⁺-ATPase activity, whereas cells stretched for 15 min were not significantly different from controls. The mean level of strain in these experiments was 10% (20% maximum). Assuming homogeneous expansion of the lung, we estimate that the circumferential strain in an alveolus would be ~10% during normal tidal volume breathing. Lung expansion from functional residual capacity to total lung capacity would lead to a strain level of 40% if functional residual capacity is defined as the baseline state. Thus the current experiments model the change in epithelial function from an unstretched condition to levels of stretch that would occur during normal breathing.

Treatment of the cells with Nystatin stimulated a two- to threefold increase in pump activity in static cells; this confirms that increased cellular Na⁺ influx alone might stimulate Na⁺-K⁺-ATPase function (Fig. 4). When cells were stretched in the presence of Nystatin, there was a further increase in Na⁺-K⁺-ATPase activity compared with that in unstretched cells. To determine whether the increased Na⁺-K⁺-ATPase activity was due to enhanced Na⁺ influx through Na⁺ channels, amiloride was used to block Na⁺ entry through amiloride-sensitive channels. There was no significant increase in ⁸⁶Rb⁺ uptake when cells were treated with amiloride and cyclically stretched (Fig. 4). These results suggest that in stretched cells the elevated Na⁺-K⁺-ATPase activity was partly caused by increased Na⁺ influx. Because increased Na⁺ influx may be caused by stretch-sensitive ion channels, we measured Na⁺-K⁺-ATPase activity in cells treated with gadolinium, which inhibits stretch-sensitive ion channels. Gadolinium significantly decreased ⁸⁶Rb⁺ uptake in unstretched cells and prevented any increase in response to cyclic stretch (Fig. 5). Taken together, these results suggest that Na⁺ entry through both amiloride- sensitive and stretch-activated, nonselective cation channels is increased by cyclic stretch in MLE-12 cells, which leads to elevated Na⁺-K⁺-ATPase activity. The elevated activity was not related to any changes in cAMP levels.

Because both amiloride and gadolinium inhibited the stretch-induced increase in activity to the same extent, the question arises as to whether amiloride and gadolinium inhibit the same channel. Recent studies by Awayda et al. (1) and Ismailov et al. (22) have demonstrated that a native epithelial Na⁺ channel, that is sensitive to amiloride, can be stretch-activated in planar lipid bilayers. As reviewed by Hamill and McBride (18), amiloride has been shown to block stretch-activated ion channels in several systems, but gadolinium is presently the inhibitor of choice for demonstrating the role of mechanogated ion channels in an experimental system. Gadolinium is a trivalent ion in aqueous solution and is a member of the lanthanide family. Its mechanism of action is not completely understood, but it is the most potent known blocker of mechanogated ion channels. Thus it is conceivable that amiloride and gadolinium block Na⁺ entry through the same channel, and future studies will be directed toward identifying the native epithelial Na⁺ channel in these cells. It should also be pointed out that gadolinium is not a specific antagonist, so inhibition of Na⁺ influx does not definitively identify stretch-activated ion channels (8). However, our results strongly suggest a stretch-induced increase in Na⁺ influx.

We also observed an additional increase in Na⁺-K⁺-ATPase activity when cells were treated with Nystatin and then subjected to cyclic strain. Because Nystatin would permit maximal pump activity, the additional increase suggested the increased expression of Na⁺-K⁺-ATPase. Figure 6 demonstrates that cyclic strain increased Na⁺-K⁺-ATPase ₁-subunit expression in the basolateral membrane but not in whole cells. Although Nystatin may have also stimulated translocation, cyclic strain further increased pump expression in the basolateral membrane (Fig. 6). These data suggest that there was no synthesis of new Na⁺-K⁺-ATPase ₁-subunit protein but rather recruitment and translocation of preformed Na⁺-K⁺-ATPase pumps from intracellular compartments into the basolateral membrane of MLE-12 cells. It has been demonstrated previously that increased Na⁺-K⁺-ATPase activity in muscle cells stimulated by insulin (23) or in alveolar type II cells stimulated by isoproterenol (5) was caused by translocation of Na⁺-K⁺-ATPase protein into basolateral membranes.

In summary, we have demonstrated that short-term cyclic stretch stimulates Na⁺-K⁺-ATPase activity, most likely by increasing intracellular Na⁺. Future work is warranted to determine whether long-term cyclic stretch leads to changes in Na⁺-K⁺-ATPase expression and activity in lung epithelial cells. Such a regulatory mechanism may be important to understand and manage patients with pulmonary edema who are undergoing mechanical ventilation.