Transepithelial transport by pulmonary alveolar ... - Semantic Scholar

2 downloads 0 Views 2MB Size Report
Jun 25, 1982 - Transepithelial transport by pulmonary alveolar type II cells in primary culture. (sodium transport/pulmonary edema). ROBERT J. MASON*t, ...
Proc. Nati Acad. Sci. USA

Vol. 79, pp. 6033-6037, October 1982 Medical Sciences

Transepithelial transport by pulmonary alveolar type II cells in primary culture (sodium transport/pulmonary edema)

ROBERT J. MASON*t, MARY C. WILLIAMS*, JONATHAN H. WIDDICOMBE*, MARTIN J. SANDERSt, DAYTON S. MISFELDTt, AND LEONARD C. BERRY, JR.* *Cardiovascular Research Institute, Department of Medicine, Physiology, and Anatomy, University of California, San Francisco, California 94143; and *Department of Medicine, Palo Alto Veterans Administration Hospital, Stanford University, Palo Alto, California 94305

Communicated by John A. Clements, June 25, 1982

monolayer from the growth surface, tight junctions between epithelial cells, and attachment points to the culture surface to prevent the fluid from flowing underneath the monolayer. Dome formation is a sensitive indicator of active transport but is difficult to study quantitatively (12). For more quantitative studies, monolayers of epithelial cells can be mounted in Ussing type chambers and the electrical properties of the monolayers can be determined by the methods used for excised epithelia in vitro. In this paper, we describe dome formation in primary cultures of alveolar type II cells from adult rat lungs and the electrical properties of type II cells maintained on collagen-coated Millipore filters. Our results indicate that alveolar type II cells have the ability to actively transport sodium and suggest that active as well as passive forces are important in keeping the alveoli relatively free of fluid. These observations have been reported in abstract form (13, 14).

ABSTRACT Fluid and electrolyte transport by epithelial cells in vitro can be recognized by the ability of cultured cells to form domes and by the electrical properties of monolayer cultures. Pulmonary alveolar epithelial cells are thought to be partially responsible for fluid movement in the fetal lung, but their role in electrolyte transport in the adult lung is not known. We isolated alveolar type II cells from adult rat lung and maintained them on plastic culture dishes alone, on plastic culture dishes coated with an extracellular matrix, and on collagen-coated Millipore filters. Numerous large domes were formed on culture dishes coated with the extracellular matrix; smaller domes were formed on uncoated plastic culture dishes. Sodium butyrate (3 mM) stimulated dome formation. Transmission electron microscopy showed that the epithelial cells had flattened but still retained lamellar inclusions and that the cells were polarized with microvilli on the apical surface facing the culture medium. The electrical properties of the monolayers maintained on collagen-coated Millipore filters were tested in two laboratories. The transepithelial potential differences were 0.7 ± 0.1 mV (24 filters, seven experiments) and 1.3 ± 0.1 mV (13 filters, two experiments) apical side negative, and the corresponding resistances were 217 + 11 ohm.cm2 and 233 ± 12 ohm.cm2. Terbutaline (10 FM) produced a biphasic response with a transient decrease and then a sustained increase in potential difference. Amiloride (0.1 mM) completely abolished the potential difference when it was added to the apical side but not when it was added to the basal side, whereas 1 mM ouabain inhibited the potential difference more effectively from the basal side. Thus, type II cells form a polarized epithelium in culture, and these cells actively transport electrolytes in vitro.

MATERIALS AND METHODS Cell Isolation and Culture. Alveolar type II cells were isolated from adult Sprague-Dawley male rats by tissue dissociation with elastase and partial purification on a metrizamide density gradient (15). The cells were suspended at 106/ml in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 2 mM glutamine, gentamicin at 10 pg/ml, penicillin at 100 units/ml, and streptomycin at 50 ug/ml with and without amphotericin at 2.5 kug/ml. All observations were made in the presence and absence of amphotericin; there was no apparent effect of this concentration of amphotericin on dome formation or on the electrical properties ofthe monolayers. Extracellular matrices were prepared in 35-mm plastic culture dishes from confluent cultures of bovine corneal endothelial cells by the method of Gospodarowicz (16). Extracellular matrices prepared from L-2 cells, A549 cells, and lung fibroblasts (American Type Culture Collection) were made by growing the cells in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, glutamine, and antibiotics and by extracting the monolayers with 20 mM ammonium hydroxide (17). Dome Formation and Analysis. The partially purified type II cells were plated at a density of 2 x 105/cm' in 35-mm culture dishes that did or did not contain an extracellular matrix. After the type II cells had adhered overnight, the monolayers were rinsed with medium and the nonadherent cells were discarded. The resultant adherent cells cultured directly on plastic dishes were >90% type II cells and those cultured on extracellular matrices made from bovine corneal endothelial cells

The alveolar space of adult mammalian lungs is lined by only a thin film of fluid. Although the amount of fluid is not known precisely, one estimate is that 20 ml of alveolar fluid is distributed over a surface area of about 70 m2 (1). When lungs are fixed for electron microscopy by vascular perfusion, the only areas that appear to contain fluid are the corners ofthe alveoli, where the radius of curvature is short (2). At the corners, the net force of surface tension, which is a function of the radius ofcurvature and the actual surface tension, is directed to draw fluid into the alveolus (3-7). How this force is counterbalanced is not known. Although most investigators have tended to discount the possibility of active transport across alveolar epithelium (7, 8), recent observations in vivo suggest that there may be active transport of fluid from the airspace into the interstitium (9, 10). Cell culture techniques have been used to study transepithelial fluid movement. Epithelial cells that transport fluid form domes or hemicysts in culture (11, 12). Dome formation is thought to require active transport to create the force to lift the The publication costs ofthis article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement' in accordance with 18 U. S. C. §1734 solely to indicate this fact.

t Present address: ver,

6033

CO 80206.

Dept. of Medicine, National Jewish Hospital, Den-

6034

Proc. Natl. Acad. Sci. USA 79 (1982)

Medical Sciences: Mason et aL

>95% type II cells as judged by the modified Papanicolaou stain or phosphine 3R (18). After domes had formed, the monolayers were fixed with 1.25% glutaraldehyde in phosphate-buffered saline, stained with a polychrome stain of azure II and methylene blue (19), and counted under a dissecting microscope. Monolayers were fixed and embedded for electron microscopy as described (20). In some experiments, the cells were stained with tannic acid prior to dehydration (21). Electrical Characterization of the Monolayers. Type II cells were maintained on Millipore filters that were coated with rat tail collagen and then fixed with glutaraldehyde (22). Because adherence to floating collagen gels is enhanced with porcine serum and dexamethasone (23), the type II cells maintained on the collagen-coated filters were cultured with 10% porcine serum and 0.1 AM dexamethasone. The spontaneous potential difference across the monolayers was low and therefore the electrophysiologic studies were verified independently in two laboratories (at the University of California and at Stanford University). The filters were mounted in Ussing type chambers for measurements of resistance and transepithelial potential difference (22, 24, 25). In the University of California laboratory, potential differences were measured with 3 M KCI/agar bridges positioned 2 mm from either side of the tissue. Tissue resistance was calculated from the change in potential difference produced by passing 90-_uA.cm-2 current pulses of 1-sec duration. Current was passed via 150 mM NaCI/agar bridges positioned 1.5 cm from either side of the tissue. Collagen coated filters alone had an insignificant resistance and no potential difference. With no tissue present, the resistance of the solution between the 3 M KCI/agar bridges was determined at the start and end of each experiment and was 12-18 ohm-cm2. With tissues in place, the change in potential difference accounted for by the solution resistance was only about 7% of the total potential difference change and was routinely corrected for in determining tissue resistance. Similar techniques were used in the Stanford laboratory (22, 25). Materials. Reagents were obtained as reported (15, 23). Denis Gospodarowicz kindly supplied fibroblast growth factor for culturing corneal endothelial cells. Terbutaline was provided by Astra Pharmaceutical Products (Framingham, MA). were

RESULTS After 2 to 3 days in culture, type II cells maintained directly on plastic culture dishes formed a few small domes. However, type II cells maintained on an extracellular matrix formed by corneal endothelial cells adhered better, flattened more rapidly, and formed numerous large domes, typically 1 mm in diameter. A low-power light micrograph ofdomes formed on an extracellular matrix is shown in Fig. 1. When the medium was removed, the domes were visible to the unaided eye and appeared like grains ofsand sprinkled on the culture dish. The domes formed by type II cells on the corneal extracellular matrix are larger than those formed by MDCK cells grown on a plastic culture surface (our observations and refs. 26 and 27). Type II cells formed fewer and smaller domes when they were cultured on extracellular matrices prepared from lung epithelial cells (L-2 cells or A549 cells), lung fibroblasts, or primary cultures of type II cells. When L-2 cells, A549 cells, and lung fibroblasts were cultured on extracellular matrices prepared from corneal endothelial cells, these cell types did not form domes. Although dome formation is generally thought to be a result of active fluid transport (11, 12) we wanted to show that dome formation

was

reversibly blocked by inhibitors of transport,

that

dome formation could be stimulated by pharmacologic agents, and that the fluid under a dome was different from the culture

:

V.|. I:

.F

i'..V ,

4

FIG. 1. Light micrograph of domes in primary culture. Type I cells were grown for 3 days on culture dishes coated with corneal extracellular matrix. The domes were fixed with glutaraldehyde and stained with a polychrome stain (19). (x40.)

medium. Dome formation was inhibited by 1 mM ouabain and by 0.1 mM amiloride. However, overnight incubation was necessary to cause these effects. The cells appeared damaged, and the effects were irreversible. In a series of experiments to evaluate drugs for their ability to stimulate dome formation, we found that 3 mM sodium butyrate greatly enhanced dome formation (Table 1). Butyrate has been used to stimulate differentiated functions in a variety ofcell types, but the mechanism of its action is not known. David Warnock (University of California, San Francisco) sampled dome fluid by micropuncture and showed that the bicarbonate concentration under the domes was the same as that of the medium. He verified that he had sampled the contents of a dome by adding radioactive inulin to the culture medium; no inulin was found in his samples. Concentrations ofother electrolytes were not determined. Because these observations were made with primary cultures, we needed to establish that the cells in the domes had the morphologic characteristics of type II alveolar epithelial cells. Phosphine 3R is a fluorescent compound that causes the lamellar bodies of type II cells to fluoresce intensely (18). The cells in the domes appeared like type II cells in culture and were not distinguishable from the other type II cells in the monolayer (Fig. 2). Electron micrographs showed that the cells formed a polarized epithelium with microvilli on the apical surface and tight junctions between the epithelial cells (data not shown) and that cells contained membrane-bound lamellar inclusions that were structurally similar to those of type II cells of intact lung (Figs. 3 and 4). Table 1. Effects of drugs on dome formation Domes per dish, no. Addition Exp. 1 Exp. 2 Exp. 3 3± 1 74 ± 10 13 ± 2 None Phorbol 12-myristate 2± 1 14 ± 5 66 ± 15 13-acetate(1OnM) 8± 2 79 ± 8 9± 3 Cholera toxin (0.1 Ag/ml) 122 ± 29 1±1 10 ± 2 Terbutaline (10 uM) 47 ± 10* 194 ± 23* 120 ± 8* Sodium butyrate (3 mM) Cultures were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum for overnight adherence and then rinsed and incubated with fresh medium containing various drugs. Fresh medium and drugs were again added 24 hr later and the monolayers were fixed with glutaraldehyde after incubation for 48 hr with drugs. Results represent mean ± SEM of four separate dishes for each of three independent experiments. *P < 0.01 by Friedman's two-way analysis of variance of ranks and the Newman-Keuls multiple comparison procedure (28).

Proc. Natl. Acad. Sci. USA 79 (1982)

Medical Sciences: Mason et al.

6035

A

at

4.

.

V

k

.

A.

-I h o

z

4

in

B oll ~ ~

~

~

r >A..

FIG. 3. Light micrographs of type II cells attached to substratum and in a dome. Type II cells were grown for 3 days on a coverslip covered with corneal extracellular matrix. (A) Type II cells on the flat portion of the monolayer. (B) Type II cells in a dome. Type II cells appear as flattened epithelial cells with apical microvilli and lamellar inclusions near the nucleus. The extracellular matrix can be seen beneath the cells. (x850.)

P < 0.05 by paired t test). Amiloride (0.1 mM), which blocks membrane sodium channels (29), rapidly brought the potential difference to zero when it was added to the apical side but not

FIG. 2. Dome stained with phosphine 3R. Type II cells were cultured on a glass coverslip coated with corneal extracellular matrix. The lamellar bodies of type II cells fluoresce intensely after incubation with phosphine 3R (18). These two micrographs show a small dome and the monolayer focused at different levels. (x380.)

To show that dome formation was due to electrolyte transport, type II cells were maintained on Millipore filters coated with rat tail collagen and studied in Ussing type chambers. The spontaneous potential difference showed that the apical surface was consistently negative with respect to the basolateral surface. The transepithelial potential differences, measured in two laboratories, were 0.7 ± 0.1 mV (24 filters, seven experiments) and 1.3 ± 0.1 mV (13 filters, two experiments), and the corresponding resistances were 217 ± 11 and 233 ± 12 ohm.cm2. As shown in Fig. 5, terbutaline, a 3-adrenergic agonist, produced a transient decrease and then a prolonged increase in the potential difference. In the experiment shown, the transepithelial resistance was not changing at the time when terbutaline was added. In a number of other experiments, however, terbutaline was added at a time when the resistance was decreasing. Because of this complicating factor, we determined the effect of terbutaline, not on the potential difference, but on the apparent short-circuit current, where short-circuit current is given by the ratio of open-circuit potential difference to transepithelial resistance. Terbutaline was found to increase short-circuit current by 28%, from 3.58 ± 0.30 to 4.58 ± 0.30 uAAcm-2 (n = 21;

40 /'

+Afwi ~£ vb t# J,.I -

=.%e

i

'-1'

-e

-w

O

w;..==.=s.£a==~

'lh- msig

..

FIG. 4. Transmission electron micrographs of cultured type II cells. A large lamellar inclusion is present in the cytoplasm and the extracellular matrix is visible beneath the cell. The apical surface is covered by microvilli that contain an array of microfilaments (Inset). (Main, x11,400; Inset, x39,900.)

Medical Sciences: Mason et al.

6036

Proc. Natl. Acad. Sci. USA 79 (1982)

1.0

or that the domes are relatively permeable to bicarbonate over the time span required for dome formation. We tested several drugs for their ability to stimulate or to inhibit dome formation. From other studies, butyrate (27), terbutaline (32), and cholera

Os

AL

I

E

-

I

I

I

I

_

AL

w

U

z w

1.01..

w

IL

z

I

I

I

I

0

I

w

1.0F 0 a-

As

AL

I

-20

I

I

I

0

20

40

TIME

I

t

60

(min)

FIG. 5. Potential differences across monolayers of type II cells. The type II cells were cultured for 7 days on collagen-coated Millipore filters and then placed in Ussing type chambers. At time zero, 10 I-M terbutaline was added. At the end of the experiment, 1 mM ouabain (0) or 0.1 mM amiloride (A) was added to the apical/luminal side (L) or to the basal/serosal side (S).

when it was added to the basal side. On the other hand, 1 mM ouabain, which inhibits Na+, K+-ATPase, inhibited the potential difference more effectively when it was added to the basal side. The potential difference was not affected by adding phlorizin or furosemide to either side or by adding glucose to the apical side. Thus, both morphologic and electrophysiologic data show that type II cells formed a polarized epithelium in culture. The negative spontaneous apical surface potential difference and the effect ofamiloride are consistent with sodium resorption as the dominant transport process.

DISCUSSION Both dome formation and the electrical properties of the monolayers indicate that type II alveolar epithelial cells actively transport fluid and electrolytes. Although most cells in monolayer cultures do not form domes, epithelial cells that transport fluid-e.g., epithelial cells from the kidney, mammary gland, colon, trachea, and choroid plexus-form domes in monolayer culture (11, 12). Definitive proof ofactive transport is, however, more difficult to establish. Ouabain and amiloride inhibited dome formation, but in our experiments these drugs appeared to be cytotoxic. We measured the bicarbonate ion concentration under the domes to see whether the pH was different or whether there was a gross change in electrolyte concentration. These experiments were designed to see whether type II cells reabsorb bicarbonate, which might account for the acid pH of alveolar fluid (30, 31). The bicarbonate concentration was the same as that of the culture medium, which indicates that the transport process is not selective or restrictive for bicarbonate

toxin might be expected to increase dome formation and phorbol 12-myristate 13-acetate (33) to decrease dome formation. Because dome formation is a complex process of active transport, interaction with the growth surface, maintenance of tight junctions, and other aspects of cellular metabolism, stimulation of dome formation cannot be attributed simply to increased active transport. The butyrate effect, however, indicates that dome formation can be stimulated by drugs. An unlikely alternative explanation for dome formation is hydrolysis of the extracellular matrix and the creation of an osmotic pressure gradient. The electrical properties of the monolayer and the observation of dome formation by cells maintained on plastic alone strongly weigh against this alternative explanation. The electrical properties of monolayer cultures of alveolar type II cells are similar to some other transporting epithelial cells in culture. The apical side is electrically negative. The spontaneous potential difference and transepithelial resistance are similar to those of MDCK cells (25, 34, 35) and LLC-PK1 renal cells (22) but less than those of mammary cells on floating collagen gels (36) and toad bladder cells (37). As expected from studies of epithelial cells in vitro, amiloride was most active when it was added to the apical side and ouabain was most active when it was added to the basal side of the monolayer (38, 39). In almost all epithelial cells, the Na',K+-ATPase is found predominately on the basolateral membrane. Dome formation had not been described in cultures of type II cells before these experiments were completed. This is presumably because dome formation requires a complete monolayer of nearly pure type II cells, the domes take several days to form, and the domes formed on tissue culture plastic are small. The use of the extracellular matrix increases the size of the domes dramatically. The matrix improves the plating efficiency and cellular spreading and thereby the formation of the monolayer. Corneal extracellular matrices have been used primarily to improve the proliferative response of endothelial cells and smooth muscle cells (16, 17). Independently, Goodman et aL (40) recently reported dome formation and inhibition in primary cultures of alveolar type II cells. Because of the heterogeneity of lung cells, the complete array of physiological functions of individual pulmonary cell types is difficult to define from studies with intact lung. We propose that transepithelial transport and dome formation in vitro are distinctive functions of alveolar type II cells in addition to their major function of synthesis, storage, and secretion of surfaceactive material. In this regard, two epithelial cell lines, L-2 cells and A549 cells, that have some similarities to type II cells, do not form domes in culture (41). The type II cells in our cultures are, however, large flat cells and, in this regard, are similar to type I epithelial cells. Nevertheless, the cells in our cultures have microvilli on their surface and do not have the array of small pinocytotic vesicles under their plasma membrane that are typically seen in type I cells in intact adult rat lungs. Hence, although type II cells have been shown to differentiate into type I cells in vivo and in the process develop into "transitional" cells, which still contain lamellar inclusions but are beginning to flatten (42), we will refer to the cells in our cultures simply as type II cells, until definitive biochemical or antigenic markers for type I cells are discovered and can be tested with these cells. Recent studies in vivo and theoretical considerations support the concept that the alveolar epithelium actively transports fluid and electrolytes from the alveolar side (apical side) to the interstitium (basal side). In fetal lung at the beginning of labor,

Medical Sciences: Mason et al. there is not only a cessation of secretion but net resorption of the alveolar fluid, and this resorption is stimulated by a-adrenergic agonists (32). In the fetal lung, the alveolar side is electrically negative and the potential difference across the epithelium is increased with 3-adrenergic agonists (9). Matthay et aL (10) found that protein-rich fluid instilled into the distal lung ofanesthetized sheep was absorbed against a large oncotic pressure gradient. They concluded that either the interstitial pressure was very negative or there was active fluid resorption. In the initial physiologic considerations of pulmonary surfaceactive material, Pattle (4) and Clements (5) noted that surface tension would tend to draw fluid into the alveolus and that this effect was minimized by a low surface tension at the air/liquid interphase. The initial calculations of the force due to surface tension assumed the radius of an entire alveolus (5). Guyton and Moffatt (6) showed that its force becomes much greater as the radius decreases and, if one assumes that the radius is as small as 0.5 pum (2) and the surface tension is 10 mN/m, the force is enormous [greater than 300 mm Hg (1 mm Hg = 133 Pa)]. Thus, although the magnitude of this force is not known and will vary with alveolar geometry and surface tension during the respiratory cycle, the force at the corners ofthe alveoli due to surface tension is probably significantly larger than 4 to 5 mm Hg (5, 43). The increased surface tension in disease states has been proposed as a mechanism for the formation ofpulmonary edema (5, 44). We propose that active sodium transport by the alveolar epithelium is one of the forces that counterbalance the effect of surface tension. Thus, theoretical considerations, recent physiologic studies in vivo, and our studies with type II cells in primary culture, all suggest that active as well as passive forces are involved in the regulation of alveolar fluid in mammalian lungs. We thank Dr. Denis Gospodarowicz for providing fibroblast growth factor and showing us how to prepare extracellular matrices, Dr. David Warnock for carrying out the micropuncture and for determining the bicarbonate concentration in the dome fluid, Lynne D. Calonico for photographing the domes, and Yuki Kubo-Hendricks and Marcia Hansen for technical assistance. This work was supported by National Institutes ofHealth Grants HL-24075 and HL-24136 and National Cancer Institute Grant 5 T32 CA 90287.

1. Macklin, C. C. (1954) Lancet i, 1099-1104. 2. Gil, J., Bachofen, H., Gehr, P. & Weibel, E. R. (1979) J. Appl

Physiol 47, 990-1001. Wilson, T. A. (1981) J. Appl Physiol, 50, 222-224. Pattle, R. E. (1958) Proc. R. Soc. London Ser. B 148, 217-240. Clements, J. A. (1961) Arch. Environ. Health 2, 280-283. Guyton, A. C. & Moffatt, D. S. (1981) Prog. Resp. Res. 18, 62-75. Staub, N. C. (1980) Prog. Cardiovasc. Dis. 32, 53-80. Snashall, P. D. & Hughes, J. M. B. (1981) Rev. Physiol Biochem. Pharmacol. 89, 5-62. 9. Olver, R. E., Ramsden, C. A. & Strang, L. B. (1981) 1. Physiol (London) 319, 38P-39P. 10. Matthay, M. A., Landolt, C. & Staub; N. C. (1982)J. Appi. Physiol 53, 96-104. 11. Lever, J. E. (1979)J. Supramol Struct. 12, 259-272. 3. 4. 5. 6. 7. 8.

Proc. Natl. Acad. Sci. USA 79 (1982)

6037

12. Handler, J. S., Perkins, F. M. & Johnson, J. P. (1980) Am. J. Physiol 238, F1-F8. 13. Mason, R. J. & Williams, M. C. (1981) Am. Rev. Resp. Dis. 123, 216 (abstr.). 14. Mason, R. J., Williams, M. C., Widdicombe, J. H., Nathanson, I. T., Sanders, M. & Misfeldt, D. (1981) J. Cell Biol 91, 415a (abstr.). 15. Dobbs, L. G., Geppert, E. F., Williams, M. C., Greenleaf, R. D. & Mason, R. J. (1980) Biochim. Biophys. Acta 618, 510-523. 16. Gospodarowicz, D. & Ill, C. (1980)J. Clin. Invest. 65, 1351-1364. 17. Gospodarowicz, D. & Ill, C. (1980) Proc. Nati Acad. Sci. USA 77, 2726-2730. 18. Mason, R. J. & Williams, M. C. (1976) Am. Rev. Resp. Dis. 113, 47 (abstr.). 19. Humphrey, C. D. & Pittman, F. E. (1974) Stain Tech. 49, 9-14. 20. Williams, M. C. (1977) J. Cell Biol 72, 260-277. 21. Simionescu, N. & Simionescu, M. (1976) J. Cell Biol 72, 608-621. 22. Misfeldt, D. S. & Sanders, M. J. (1981) J. Membr. Biol 59, 13-18. 23. Geppert, E. F., Williams, M. C. & Mason, R. J. (1980) Exp. Cell Res. 128, 363-374. 24. Widdicombe, J. H. & Welsh, M. J. (1980) Am. J. Physiol 239, C112-C117. 25. Misfeldt, D. S., Hamamoto, S. T. & Pitelka, D. R. (1976) Proc. Natl Acad. Sci. USA 73, 1212-1216. 26. Leighton, J., Brada, Z., Estes, L. W. & Justh, G. (1969) Science 163, 472-473. 27. Lever, J. E. (1979) Proc. Natl Acad. Sci. USA 76, 1323-1327. 28. Zar, J. H. (1974) Biostatistical Analysis (Prentice-Hall, Engle-

wood Cliffs, NY).

29. Biber, U. L. T. (1979) in Amiloride and Epithelial Sodium Transport, eds. Cuthbert, A. W., Fanelli, G. M., Jr., & Scriabine, A. (Urban and Schwarzenberg, Baltimore), p. 61-77. 30. Nielson, D. W., Goerke, J. & Clements, J. A. (1981) Proc. Natl Acad. Sci. USA 78, 7119-7123. 31. Adamson, T. M., Boyd, R. D. H., Platt, H. S. & Strang, L. B. (1969) J. Physiol (London) 204, 159-168. 32. Walters, D. V. & Olver, R. E. (1978) Pediatr. Res. 12, 239-242. 33. Ojakian, G. K. (1981) Cell 23, 95-103. 34. Cereijido, M., Meza, I. & Martinez-Palomo, A. (1981) Am. J. Physiol 240, C96-C102. 35. Cereijido, M., Robbins, E. S., Dolon, W. J., Rotunno, C. A. & Sabatini, D. D. (1978) J. Cell Biol 77, 853-880. 36. Bisbee, C. A., Machen, T. E. & Bern, H. A. (1979) Proc. Natl Acad. Sci. USA 76, 536-540. 37. Handler, J. S., Steele, R. E., Sahib, M. K., Wade, J. B., Preston, A. S., Lawson, N. L. & Johnson, J. P. (1979) Proc. Natl Acad. Sci. USA 76, 4151-4155. 38. Mills, J. W., MacKnight, A. D. C., Dayer, J. & Ausiello, D. A. (1979) Am. J. Physiol. 236, C157-C162. 39. Cereijido, M., Ehrenfeld, J., Meza, I. & Martinez-Palomo, A. (1980) J. Membr. Biol. 52, 147-159. 40. Goodman, B. E., Fleischer, R. S. & Crandall, E. D. (1982) Fed. Proc. Fed. Am. Soc. Exp. Biol 41, 1245 (abstr.). 41. Mason, R. J. & Williams, M. C. (1980) Biochim. Biophys. Acta 617, 36-50. 42. Evans, M. J., Cabral, L. J., Stephens, R. J. & Freeman, G. (1975) Exp. Mol Pathol 22, 142-154. 43. Staub, N. C. (1974) Physiol Rev. 54, 678-811. 44. Albert, R. K., Lakshminarayan, S., Hildebrandt, J., Kirk, W. & Butler, J. (1979) J. Clin. Invest. 63, 1015-1018. 45. Erhlich, B. (1981) Nature (London) 243, 1004-1005.