Amiloride-inhibitable Na+ conductive in alveolar type II pneumocytes

pathways

SADIS MATALON, ROBERT J. BRIDGES, AND DALE J. BENOS Departments of Anesthesiology, Physiology, and Biophysics, University of Alabama at Birmingham, Birmingham, Alabama 35223

MATALON, SADIS, ROBERT J. BRIDGES, AND DALE J. BENOS. Amiloride-inhibitable Na’ conductive pathways in alveolar type II pneumocytes. Am. J. Physiol. 260 (Lung Cell. Mol. Physiol. 4): L90-L96, 1991.-The purpose of these studies was to document the existence of electrogenic Na’ uptake by membrane vesicles of rabbit alveolar type II (ATII) cells and the extent to which this process was inhibited by amiloride. AT11 cells (>85% pure) were obtained by elastase digestion of lung tissue followed by Percoll centrifugation, and an enriched plasma membrane vesicle fraction was obtained by differential centrifugation. ““Na’ uptake into these vesicles was measured in the presence of a negative inside membrane potential, produced by the addition of the K+ ionophore valinomycin (10 )(LM) after all external K+ was removed. Electrogenic (valinomycin-sensitive) Na’ uptake (ELNa) was defined as the difference in uptake in the presence and absence of valinomycin. ELNa, normalized per milligram protein, was twice as high across AT11 cells than alveolar macrophage membrane vesicles, was inhibited by amiloride (50% inhibitory concentration = 10 PM), and was decreased in the presence of an outwardly directed proton gradient W-L, 6.8; pHur,t 7.8), suggesting that it was not mediated by Na’-H’ antiport. Furthermore, ELNa was equally inhibited by increasing concentrations of amiloride and benzamil but was more sensitive to 5-(N-ethyl-N-isopropyl)-2’-4’-amiloride in concentrations of lo-1,000 PM. These findings indicate that a fraction of Na’ transport across AT11 membrane vesicles occurs t.hrough a conductive pathway, probably a channel, that has different sensitivity to amiloride and its analogues than the previously described epithelial high amiloride-affinity Na’ channel.

benzamil;

lung; macrophages;

antiport;

valinomycin;

channel

FOR EFFICIENT GAS EXCHANGE to occur, the mammalian air spaces must be free of fluid. It has been thought that the clearance of fluid was passive and determined solely by the existing hydrostatic and oncotic pressure gradients acting across the mammalian blood-gas barrier. However, recent evidence in conscious animals, isolated perfused lungs, and cultured alveolar type II pneumocytes, indicates that active solute transport across the alveolar epithelium may play an important role in the maintenance of normal lung fluid balance. For example, Mason et al. (21) and Crandall et al. (9) have shown that alveolar type II (ATII) pneumocytes cultured on plastic supports form domes by a process linked to active Na’ transport from medium to substratum. Cheek et al. (8) demonstrated that rat epithelial cells cultured on tissue culture-treated Nucleopore filters formed tight monolayers and actively transported Na’ from the apical to I ,90

1040-O(iO5/9

1 $1.50 Copyright

basolateral sides under short-circuit conditions. The basal value of active Na’ transport was 80% inhibited by 10 PM amiloride and stimulated by terbutaline. The experiments of Basset et al. (2) and Matthay et al. (25) confirmed the presence of active Na+ transport across isolated and in situ lung lobes and indicate that active rather than passive solute transport may be the predominant pathway for fluid clearance across the adult lung. Active Na’ transport across epit.helial cells is a two step process: Na+ diffuses passively across the apical membrane down a favorable electrochemical gradient and then is actively transported across the basolateral membrane by the ouabain-sensitive Na’-K’-ATPase. Heretofore, a number of apical pathways for Na’ entry in mammalian lung epithelial cells have been described including a Na’-glucose cotransporter (18) and the ubiquitous Na+-H+ antiporter (27). The latter was only active when the internal cell pH was ~7.0. In addition to these pathways, the existence of amiloride-blockable Na’ channels have been demonstrated across a variety of epithelial cells (3, 14). Because channels are much more efficient pathways for Na’ transport and thus may permit cells to move large amounts of fluid, we investigated their possible existence in freshly isolated rabbit type II pneumocytes (ATII). This was done by first documenting the existence of electrogenic Na+ uptake into membrane vesicles of AT11 cells. We then quantified the extent to which electrogenic Na’ uptake was inhibited by amiloride and two of its structural analogues, benzamil and 5-(N-ethyl-N-isopropyl)Z’-4’-amiloride (EIPA), which have high inhibitory affinities for Na’ channels and Na’H+ antiports, respectively (3). We chose to perform our experiments in type II cells because they are easily isolated, they contain significantly more Na’-H’-ATPase activity than alveolar type I or endothelial cells (29), and, when cultured on porous supports, they generate a potential difference and actively transport sodium (8). These measurements were also repeated in membrane vesicles from alveolar macrophages (AM), an abundant nonepithelial lung cell. Our results indicate that a fraction of the electrogenic Na’ flux across ATII, but not macrophage derived vesicles, was inhibited by low concentrations of amiloride (0.1-l PM) and could not be enhanced by an outwardly directed pH gradient. These findings are consistent with the existence of a Na’ channel with low affinity to amiloride. This channel does not possessthe previously defined structural specificity for

I(:# 199 1 the American

Physiological

Society

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.

NA+

IJP’I‘AKE

IN

amiloride and its analogues (3,4) and seems to be present in a variety of epithelial cells (26). MATERIALS

AND

METHODS

General experimental conditions. All cells were isolated from Pasteurella-free New Zealand White adult rabbits, weighing between 1.9 and 2.1 kg. All rabbits were quarantined for _>l wk on arrival. Only healthy animals that were eating and drinking normally with no signs of respiratory illness were included in this study. Cell isolation. AMs were isolated by bronchoalveolar lavage of rabbit lungs as previously described (20). In brief, rabbits were killed with an intravenous injection of pentobarbital sodium, and the lungs were perfused free of blood with Ca’)+-free cold (4°C) phosphate-buffered saline (PBS; pH 7.4) and removed intact from the chest cavity. They were then lavaged gently with 300 ml PBS and 150 ml of Finkelstein balanced salt solution (FBSS; pH 7.4). The lavagate was pooled, filtered, and centrifuged at 300 g for 10 min to pellet the cells. The cell pellet, consisting of >99% macrophages as shown by the lack of staining by alkaline phosphatase (lo), was washed and resuspended in cold 4°C solution (referred to as the isolation buffer) (containing (in mM) 100 KSOd, 50 tris( hydroxymethyl)aminomethane [ (Tris)H,SO,], and 5 EGTA, pH 7.81. This procedure was repeated twice. Approximately 10 X lo7 macrophages were isolated per rabbit. Viability, assessed by trypan blue exclusion, was >95%. The cells were then processed for membrane vesicle formation as described below. AT11 cell pneumocytes were isolated from lung tissue by a modification of the Finkelstein procedure as previously described (20). In brief, after the initial lavage to remove macrophages, the lungs were inflated with 100 ml of Joklik modified minimum essential medium (JMEM) containing elastase (130 U; Worthington), and DNase (1 mg; Sigma; Bovine Pancreas DNase). Trypsin, which is ordinarily used in conjunction with elastase in this isolation procedure, was omitted from the instillate because it has been known to decrease short-circuit current across epithelial tissues (14). Proteolytic digestion was stopped after 30 min by the addition of 100 ml cold (4°C) JMEM containing 10% fetal bovine serum and 5 mg DNase. After removal of connective tissue, the lungs were minced with fine dissecting scissors; the minced tissue was stirred in a trypsinizing flask at (4°C) for 10 min. The tissue suspension was then filtered through nylon gauze of decreasing pore size (150,411, and 15 pm). AT11 cells were separated out of the crude cell suspension by discontinuous density gradient centrifugation with Percoll (Sigma); cell counts were determined using a hemocytometer. More than 90% of the cells were AT11 pneumocytes, as determined by phosphine-3R staining of lamellar bodies (20); cell viability was >95%. Approximately 20-30 x lo7 AT11 cells were obtained from each rabbit. They were washed extensively in isolation buffer (see below) and processed for vesicle formation. Isolation of membrane uesicles. The procedures described have been modified from those described by Garty et al. (12, 17) and Bridges et al. (4, 5). In brief,

AT11

VESICLES

L91

AMs, or AT11 cells, suspended in 5 ml of isolation buffer were homogenized in ice with a PT300 Brinkman Polytron (20,000 rpm; 6 15-s bursts). The cell homogenates were centrifuged at 1,000 g for 10 min, and the supernatant was recovered and centrifuged at 40,000 g for 60 min. The resulting pellet was resuspended in 300-500 ~1 of isolation buffer and used within 20-40 min for the Na’ uptake assay described below. Measurement of alkaline phosphatase. The alkaline phosphatase activity of the cell homogenate and membrane vesicles was according to the method of Torriani (31). Briefly, aliquots of homogenate or vesicular protein (20-200 ~1) were added in an 800~~1solution containing 700 ,uI of 100 mM Tris-HCl buffer (pH 9.0), 60 mM MgClz, 1% Triton X-100, and 100 ~1 of p-nitrophenyl phosphate (Sigma). The samples were incubated at 37°C for 20 min at which point the reaction was stopped by the addition of 1 ml of 0.5 N NaOH and the absorbance was read at 405 nm. Alkaline phosphatase activity was expressed as micromoles inorganic phosphate produced per hour per milligram protein. Protein concentration was assayed by the method of Bradford, using bovine serum albumin as the standard (5). “Na+ uptake into vesicles. “Na+ uptake was measured by the Garty-Rudy-Karlish method (12, 13, 17) as described by Bridges et al. (4, 5). In this method, the tracer uptake is measured in the presence of a negative inside membrane potential. This potential greatly increases the amount of “Na+ taken up through conductive pathways and prolongs its time course from milliseconds to minutes. A large negative inside membrane potential was established by a valinomycin-induced K+ diffusion potential. A chemical gradient for K+ was imposed by applying 100-150 ~1 (150-200 pg protein) of the vesicle solution to a cation-exchange column (Dowex-50x8-100, Tris form) and eluting the vesicles with 1.5-2.0 ml 8.5% sucrose into a test tube containing 5-7 ~1 1 M Tris base to ensure a final pH of 7.8. Aliquots were then added to vials containing carrier-free “‘Na” (New England Nuclear, final concentration 300 nM; -2 X 10" cpm) in a small volume (100-200 ~1) of 8.5% sucrose. Depending on the experimental requirements, the radioactive solution also contained valinomycin (10 PM), inhibitors (amiloride, benzamil, EIPA), or vehicle. Samples were removed from the radioactive suspension at designated times and placed on a second cation exchange column previously washed with 1 mg/ml bovine serum albumin (Sigma Chemical, St. Louis, MO) in 8.5% sucrose. The vesicles were eluted with 1.5 ml of 8.5% sucrose into scintillation vials for counting. After a 24-h dark-adaptation period, the vials were counted with a HRB-1214 Rack Beta Liquid Scintillation Counter. All uptake studies were conducted at room temperature (22OC). Protein concentration was assayed in the vesicles eluted from the first column by the method of Bradford, using bovine serum albumin as the standard (5). Time dependence of “Na+ uptake. To establish the time dependence of vesicular “Na+ uptake, vesicles obtained from two gradient-forming columns were pooled. One milliliter of aliquot was then added to 200 ~1 of ‘2Na+ sucrose solution containing either 10 PM valinomycin or an equal volume of vehicle (ethanol). Two

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.

L92

NA+

IJPTAKE

IN

hundred microliter aliquots were taken at 1, 5, 10, 30,60, and 1,200 min from the onset of incubation and eluted through the cation-exchange columns as described above. To assess the extent to which the resulting radioactivity was due to nonspecific binding of 22Na+ to the vesicular membranes, in one experiment, we added Triton X-100 (final concentration: 0.5%) to the valinomycin-containing vesicles 30 s before elution through the columns. Results were expressed as intravesicular 22Na+ radioactivity (cpm) per milligram protein per total added cpm. Inhibition of “Na+ uptake by amiloride and some of its analogues. As mentioned above, “Na+ uptake into AT11 and macrophage membrane vesicles was measured in the presence of different concentrations of amiloride (0. l500 PM). For ATII-derived vesicles, these measurements were repeated in the presence of benzamil hydrochloride (benzamil; 0.01-100 PM) and EIPA (0.1-500 PM). Stock solutions were prepared weekly by dissolving the aforementioned substances in dimethyl sulfoxide (DMSO) at a final concentration of 10 mM. They were stored at room temperature and protected from light. The concentration of the amiloride solution was verified by measuring its absorbance at 361 nm (E, = 0.02 M-l. cm-‘). Eighty microliters of the vesicle suspension (5-10 pg protein) were added to 150 ~1 sucrose solution containing “Na’ (2 X 10’ cpm), 10 PM valinomycin, and 2.3-12.3 ~1 (10 mM stock) of the appropriate inhibitor. Controls contained an equal amount of the vehicle (DMSO) and no inhibitor. The samples were vortexed and remained at room temperature for 10 min at which time they were eluted through cation-exchange columns and counted for ‘“Na+ radioactivity as described above. Dependence of 22Na+ uptake on vesicular pH. The aforementioned studies were conducted at similar intraand extravesicular pH values (pHi* = pHout 7.8). In a separate set of experiments, we measured 22Na+ uptake into AT11 vesicles and the extent that the uptake was inhibited by amiloride in the presence of an outwarddirected proton gradient (pHin 6.8; pHout 7.8). If 22Na+ was entering these vesicles through a Na’-H+ antiporter, the presence of a favorable H+ gradient should result in its increased intravesicular accumulation. AT11 cells were isolated as described above. After removal from the Percoll gradient, they were washed and homogenized in isolation buffer (pH 6.8). The extravesicular pH was maintained at 7.8 by adding a predetermined amount of Tris base to vesicles eluted through the cation exchange columns (see MATERIALS AND METHODS). 22Na’ uptake was measured in the presence and absence of 10 PM valinomycin and amiloride concentrations of 1 and 100 PM as described above. Statistics. All data are presented as the mean t SE of the mean. For each experimental condition, vesicles were derived from cells obtained from two rabbits to increase protein yield. The statistical significance of group means were assessedby Student’s t test or by analysis of variance followed by the Bonferroni modification of the t test (33). Results were considered significant if P < 0.05. RESULTS

In the first series of experiments, we quantified the time course of 22Na’ uptake into AT11 vesicles in the

AT11

VESICLES

presence and absence of 10 PM valinomycin (Fig. 1). AT11 vesicle preparations exhibited almost twice the alkaline phosphatase activity of the homogenate fraction [806 t 150 vs. 450 t 128 (means t SE) pmol inorganic phosphate h-’ .mg protein-‘; n = 4) indicating that they were enriched with apical membrane fractions. In the presence of a K+ diffusion potential and valinomycin, “Na+ uptake increased over a 20-min period, reached a plateau between 20 and 60 min, and then declined slowly to an equilibrium value. The magnitude of “Na’ uptake was comparable to what has been observed in toad bladder vesicles (17). In the absence of valinomycin, 2’2Na+ uptake was considerably less and increased- monotonically with time reaching the same equilibrium value. Only background levels of radioactivity were detected when Triton X-100 was added to the vesicular solution before its elution through the second cation-exchange column (data not shown). The transient nature of the uptake, its increase in the presence of valinomycin and a K+ diffusion potential, and its virtual disappearance when the vesicles were ruptured with Triton X-100 indicate that the measured radioactivi ty was due to intravesicular 22 N‘a+ and could not be accoun ted for by nonspecific binding of “Na+ to the vesicular membranes. 22Na+uptake into macrophage vesicles followed the same time course (Fig. 1). The kinetics of amiloride inhibition of the electrogenic component of 22Na+ uptake values into membrane vesicles from macrophages and AT11 cells are shown in Figs. 2-4. Figure 2 compares the magnitudes of “Na+ uptake in the presence and absence of valinomycin. All measurements were obtained 10 min after the addition of carrier-free 22Na+ to the appropriate vesicular solution. In the absence of valinomycin, mean values of “Na+ uptake across AT11 and macrophage derived vesicles were 12.8 and 6.2 (% of total radioactivity/mg protein), respectively. Instillation of 10 PM valinomycin increased ““Na’ uptake across AT11 and macrophage-derived vesicles to 27.4 and 12 (% of total radioactivity/mg protein), respectively. l

o------O

10

20

30

Time

40

50

60



AT11 -V

1200

(min)

FIG. 1. “Na’ uptake across alveolar type II (ATII) and macrophage (AM) vesicles in the presence (+V) and absence (-V) of valinomycin, as a function of time. Vesicles were prepared as described in text (pHiI, 7.8). “Na+ was added at t = 0. Equal amounts of vesicular 7.8, pHo,t protein were used for the measurement of 22Na+ uptake in bot,h AT11 and AM. Typical values from a single experiment. Similar results were obtained with three additional vesicular preparations.

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.

UPTAKE

NA+

IN

* f

6\”

-

II

0

II

- Val

+Val

uptake across alveolar type II (ATII) and macrophage (AM) vesicles in presence and absence of valinomycin measured at t = 10 min after addition of “Na+ into vesicular solution. Values are means * SE (n > 4). * Significantly different from corresponding macrophage value ( I3 < 0.05; t test). FIG.

t2. “Na’

157 ll

3 -1

0

a-

it

A-

I ’

0.1

. A

AT11 AM

I , - ..-*

Amiloride

1

, , * ““.’

10

- 7 ’ ““”

100

l

- ’ - *---A

1000

(PM)

3. Electrogenic “Na’ uptake into alveolar type II (ATII) and macrophage (AM) vesicles in presence of indicated amiloride concentrations. All measurements were made 10 min after addition of 22Na+ into vesicular solutions. * Significantly different than corresponding control (no amiloride present; one-way analysis of variance). Values are means t SE (n L 4). FIG.

3oT

AT11

L93

VESICLES

Values for this variable and its dependence on the external amiloride concentration, are shown in Fig. 3. As can across AT11 vesicles be seen, electrogenic 22Na+ transport in the absence of amiloride, normalized per milligram protein, was considerably higher (-130%) than the corresponding macrophage value. An important finding of this study is that 1 PM amiloride significantly decreased the value of electrogenic “Na+ across AT11 but not macrophage-derived vesicles. This reduction was the result of a decrease in ‘“Na+ uptake in the presence of valinomycin. In the absence of valinomycin amiloride in the range of l-100 PM did not reduce “Na+ uptake in a significant fashion. Assuming that the maximum value of electrogenic “Na+ uptake was the difference between the corresponding values at 500 and 0 PM amiloride, we calculated that 50% of the electrogenic “Na+ uptake was inhibited by -10 PM amiloride. To understand the nature of the amiloride-sensitive transporter, we performed two additional series of experiments using membrane vesicles from AT11 cells. In the first series, we compared the inhibition of the electrogenie ““Na’ uptake by amiloride and two of its analogues, benzamil and EIPA. As previously shown (3), compared with amiloride, benzamil has greatly enhanced inhibitory activity toward the Na+ channel and the Na+-Ca2+ exchanger, whereas EIPA has lower inhibitory activity toward the Na+ channel but a loo-fold higher inhibitory activity toward the Na’-H’ exchanger. The results are shown in Fig. 5. At concentrations of 0.1 and 1 PM, all compounds caused a similar degree of inhibition of Na+ uptake, whereas at concentrations X0 PM, EIPA blocked a higher fraction of the uptake than the other two inhibitors. It should be noted, that if the predominant pathway of “Na+ uptake across AT11 cells was a Na+-H+ antiporter, the difference in inhibition of 22Na+ uptake between amiloride and EIPA at the noted concentrations should have been considerably higher than measured. To investigate whether a fraction of the electrogenic (valinomycin-sensitive) component of 22Na+ uptake ocl -

m---H

Q)

5w/

0

0 amil. benz.

’ - I - --‘--’ ’ - --v---1 - - - -‘-‘.I - - --v--t 0.1 1 10 100 1000

Amiloride

(PM)

4. “Na’ uptake across alveolar type II (ATII) vesicles in presence (+ val) and absence of valinomycin (- val) for indicated amiloride concentrations. Measurements were taken 10 min after addition of different than corresponding ‘“Na’ in vesicular solution. * Significantly control (no amiloride present; one-way analysis of variance). Values are means & SE (n L 4).

*

FIG.

From these data, we calculated the electrogenic (valinomycin-sensitive) component of the “Na+ uptake by subtracting the 22Na+ uptake in the absence of valinomycin from that obtained when valinomycin was present.

0+-----d/



0

- ..‘.‘..’

0.1

Inhibitor

. . .‘....’

1

- . .“.“I

10

. .--I

100

1000

(PM)

‘2Na’ uptake by alveolar type II vesicles in 5. Electrogenic presence of different concentrations of amiloride, benzamil, and 5-(Nethyl-N-isopropyl)Z’-4’-amiloride (EIPA). Values are expressed as % of corresponding values in absence of inhibitors. Measurements were made 10 min after addition of 22Na+ into vesicular solutions. Values are means ~fr SE (n = 3). * EIPA value is significantly different than corresponding amiloride value (P < 0.05; two-way analysis of variance). FIG.

Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (137.154.019.149) on January 13, 2019.

L94

NA+

UPTAKE

IN

curred through a Na’-H’ exchanger in our system, we investigated the dependence of this variable on internal pH. The results are shown in Fig. 6. Lowering the internal pH from 7.8 to 6.8 (external pH 7.8) resulted in a large decrease of the magnitudes of “Na’ uptake both in the presence of valinomycin (from 27.2 to 10.6% of total radioactivity/mg protein) and in its absence (12.8-5.3). Mean electrogenic ‘“Na+ uptake decreased from 14.4 to 5.3 (% of total radioactivity/mg protein). Furthermore, when the internal pH was maintained at 6.8, amiloride (l-100 PM) did not decrease “Na+ uptake in either the presence or absence of valinomycin. These findings indicate that the amiloride-sensitive Na’ transport seen in AT11 vesicles could not have occurred through a Na’-H’ exchanger. DISCUSSION

Measurement of electrogenic Na+ uptake. In this series of experiments, accumulation of “Na+ into membrane vesicles was measured in the presence of a large negative inside potential that prolongs the time course of channelmediated “Na+ uptake from a few seconds to minutes. The limitations and assumptions of this technique have been previously reviewed by Garty et al. (12, 17) and Bridges et al. (4, 5). The membrane preparation was enriched in apical fractions as indicated by the doubling of the alkaline phosphatase value compared with the homogenate fraction. However, in addition to apical vesicles, it contained those from basolateral and intracellular membranes and particles. These structures may account for a fraction of the measured uptake. Although the purity of the apical membrane vesicles could be increased severalfold by Mn” precipitation (26), the use of this procedure was not possible in the present experiments due to the small amount of protein available. These limitations were circumvented by measuring the fraction of the electrogenic “Na+ transport that was inhibited by amiloride. At concentrations

Amiloride-inhibitable Na+ conductive pathways in alveolar type II pneumocytes.

The purpose of these studies was to document the existence of electrogenic Na+ uptake by membrane vesicles of rabbit alveolar type II (ATII) cells and...
2MB Sizes 0 Downloads 0 Views